3.0. System Software 3.1. BASIC 10.0 C64DX BASIC 10.0 3.1.1. INTRODUCTION This section lists BASIC 10.0 commands, statements, and functions in alphabetical order. It gives a complete list of the rules (syntax) of BASIC 10.0, along with a concise description of each. COMMAND AND STATEMENT FORMAT The commands and statements presented in this section are governed by consistent format conventions designed to make them as clear as possible. In most cases, there are several actual examples to illustrate what the actual command looks like. The following example shows some of the format conventions that are used in the BASIC commands: EXAMPLE: DLOAD <"program name"|(file_name_var)> [,U#] [,D#] | | | | | | | | keyword argument (if any) optional arguments The parts of the command or statement that the user must type in exactly as they appear are in capital letters. Words that don't have to be typed exactly, such as the name of the program, are not capitalized. When quote marks (" ") appear (usually around a program or file name), the user should include them in the appropriate place according, to the format example. KEYWORDS, also called RESERVED WORDS, appear in uppercase letters. THESE KEYWORDS MUST BE ENTERED EXACTLY AS THEY APPEAR. However, many keywords have abbreviations that can also be used. Keywords are words that are part of the BASIC language that the computer understands. Keywords are the central part of a command or statement. They tell the computer what kind of action to take. These words cannot be used as variable names. ARGUMENTS (also called parameters) appear in lower case. Arguments are the parts of a command or statement; they complement keywords by providing specific information about the command or statement. For example, a keyword tells the computer to load a program, while the argument tells the computer which specific program to load and a second argument specifies which drive the disk containing the program is in. Arguments include filenames, variables, line numbers, etc. SQUARE BRACKETS [] show OPTIONAL arguments. The user selects any or none of the arguments listed, depending on the requirements. ANGLE BRACKETS <> indicates that the user MUST choose one of the arguments listed. VERTICAL BAR | separates items in a list of arguments when the choices are limited to those arguments listed, and no other arguments can be used. Then the vertical bar appears in a list enclosed in SQUARE BRACKETS, the choices are limited to the items in the list, but still have the option not to use any arguments. ELLIPSIS ..., a sequence of three dots, means that an option or argument can be repeated more than once. QUOTATION MARKS " " enclose character strings, filenames, and other expressions. When arguments are enclosed in quotation marks in a format, the quotation marks must be included in a command file or statement. Quotation marks are not conventions used to describe formats; they are required parts of a command or statement. PARENTHESES () When arguments are enclosed in parentheses in a format, they must be included in a command or statement. Parentheses are not conventions used to describe formats; they are required parts of a command or statement. VARIABLE refers to any valid BASIC variable name such as X, A$, or T%. EXPRESSION means any valid BASIC expression, such as A+B+2 or .5*(X+3). 3.1.2. ALPHABETICAL LIST OF COMMANDS, FUNCTIONS, and OPERATORS * Token = AC multiplication + Token = AA addition - Token = AB subtraction / Token = AD division < Token = B3 less-than = Token = B2 equal > Token = B1 greater-than ^ Token = AE exponentiation (PI) Token = FF return value of PI ABS Token = B6 absolute function AND Token = AF logical AND operator APPEND Token = FE,0E append file ASC Token = C6 string to PETSCII function ATN Token = C1 trigonometric arctangent function AUTO Token = DC auto line numbering BACKGROUND Token = FE,3B background color BACKUP Token = F6 backup diskette BANK Token = FE,02 memory bank selection BEGIN Token = FE,18 start logical program block BEND Token = FE,19 end logical program block BLOAD Token = FE,11 binary load file from diskette BOOT Token = FE,1B load & run ML, or BASIC autoboot BORDER Token = FE,3C border color BOX Token = E1 draw graphic box BSAVE Token = FE,10 binary save to disk file BUMP Token = CE,03 sprite collision function BVERIFY Token = FE,28 verify memory to binary file CATALOG Token = FE,0C disk directory CHANGE Token = FE,2C edit program CHAR Token = E0 display characters on screen CHR$ Token = C7 PETSCII to string function CIRCLE Token = E2 draw graphic circle CLOSE Token = A0 close channel or file CLR Token = 9C clear BASIC variables, etc. CMD Token = 9D set output channel COLLECT Token = F3 validate diskette (chkdsk) COLLISION Token = FE,17 enable BASIC event COLOR Token = E7 set screen colors CONCAT Token = FE,13 concatenate two disk files CONT Token = 9A continue BASIC program execution COPY Token = F4 copy a disk file COS Token = BE trigonometric cosine function CUT Token = E4 cut graphic area DATA Token = 83 pre-define BASIC program data DCLEAR Token = FE,15 mild reset of disk drive DCLOSE Token = FE,0F close disk channel or file DEC Token = D1 decimal function DEF Token = 96 define user function DELETE Token = F7 delete BASIC lines or disk file DIM Token = 86 dimension BASIC array DIR Token = EE disk directory DISK Token = FE,40 send disk special command DLOAD Token = F0 load BASIC program from disk DMA Token = FE,1F define & execute DMA command DMA Token = FE,21 " DMA Token = FE,23 " DMODE Token = FE,35 set graphic draw mode DO Token = EB start BASIC loop DOPEN Token = FE,0D open channel to disk file DPAT Token = FE,36 set graphic draw pattern DSAVE Token = EF save BASIC program to disk DVERIFY Token = FE,14 verify BASIC memory to file ELLIPSE Token = FE,30 draw graphic ellipse ELSE Token = D5 if/then/else clause END Token = 80 end of BASIC program ENVELOPE Token = FE,0A define musical instrument ERASE Token = FE,2A delete disk file ERRS Token = D3 BASIC error function EXIT Token = FD exit BASIC loop EXP Token = BD exponentiation function FAST Token = FE,25 set system speed to maximum FILTER Token = FE,03 set audio filter parameters FIND Token = FE,2B hunt for string in BASIC program FN Token = A5 define user function FOR Token = 81 start BASIC for/next loop FOREGROUND Token = FE,39 set foreground color FRE Token = B8 available memory function GCOPY Token = FE,32 graphic copy GENLOCK Token = FE,38 set video sync mode GET Token = Al receive a byte of input GO Token = CB program branch GOSUB Token = 8D program subroutine call GOTO Token = 89 program branch GRAPHIC Token = DE set graphic mode HEADER Token = F1 format a diskette HELP Token = EA display BASIC line causing error HEX$ Token = D2 return hexadecimal string function HIGHLIGHT Token = FE,3D set highlight color IF Token = 8B if/then/else conditional INPUT Token = 85 receive input data from keyboard INPUT# Token = 84 receive input data from channel (file) INSTR Token = D4 locate a string within a string INT Token = B5 integer function JOY Token = CF joystick position function KEY Token = F9 define or display function key LEFT$ Token = C8 leftmost substring function LEN Token = C3 length of string function LET Token = 88 variable assignment LINE Token = E5 draw graphic line, input line LIST Token = 9B list BASIC program LOAD Token = 93 load program from disk LOCATE Token = E6 (currently unimplemented) LOG Token = BC natural log function LOOP Token = EC end of do/loop LPEN Token = CE,04 lightpen position function MID$ Token = CA substring function MONITOR Token = FA enter ML Monitor mode MOUSE Token = FE,3E set mouse parameters MOVSPR Token = FE,06 set sprite position and speed NEW Token = A2 clear BASIC program area NEXT Token = 82 end of for-next loop NOT Token = A8 logical complement function OFF Token = FE,24 (subcommand) ON Token = 91 multiple branch or subcommand OPEN Token = 9F open I/O channel OR Token = B0 logical or function PAINT Token = DF graphic flood-fill PALETTE Token = FE,34 set palette color PASTE Token = E3 draw graphic area from cut buffer PEEK Token = C2 return memory byte function PEN Token = FE,33 set graphic pen color PIC Token = FE,37 graphic subcommand PLAY Token = FE,04 play musical notes from string POINTER Token = CE,0A address of string var function POKE Token = 97 change memory byte POLYGON Token = FE,2F draw graphic polygon POS Token = B9 text cursor position function POT Token = CE,02 return paddle position PRINT Token = 99 display data on text screen PRINT# Token = 98 send data to channel (file) PUDEF Token = DD define print-using symbols QUIT Token = FE,1E (currently unimplemented) RCLR Token = CD (currently unimplemented) RDOT Token = D0 (currently unimplemented) READ Token = 87 read program pre-defined program data RECORD Token = FE,12 set relative disk file record pointer REM Token = 8F BASIC program comment RENAME Token = F5 rename disk file RENUMBER Token = F8 renumber BASIC program lines RESTORE Token = 8C set DATA pointer, subcommand RESUME Token = D6 resume BASIC program after trap RETURN Token = 8E end of subroutine call RGR Token = CC (currently unimplemented) RIGHT$ Token = C9 rightmost substring function RMOUSE Token = FE,3F read mouse position RND Token = BB pseudo random number function RREG Token = FE,09 return processor registers after SYS RSPCOLOR Token = CE,07 return sprite color function RSPPOS Token = CE,05 return sprite position function RSPRITE Token = CE,06 return sprite parameter function RUN Token = 8A run BASIC program from memory or disk RWINDOW Token = CE,09 return text window parameter function SAVE Token = 94 save BASIC program to disk SCALE Token = E9 (currently unimplemented) SCNCLR Token = E8 erase text or graphic display SCRATCH Token = F2 delete disk file SCREEN Token = FE,2E set parameters or open graphic screen SET Token = FE,2D set system parameter, subcommand SGN Token = B4 return sign of number function SIN Token = BF trigonometric sine function SLEEP Token = FE,0B pause BASIC program for time period SLOW Token = FE,26 set system speed to minimum SOUND Token = DA perform sound effects SPC Token = A6 skip spaces in printed output SPRCOLOR Token = FE,08 set multicolor sprite colors SPRDEF Token = FE,1D (currently unimplemented) SPRITE Token = FE,07 set sprite parameters SPRSAV Token = FE,16 set or copy sprite definition SQR Token = BA Square root function STEP Token = A9 for-next step increment STOP Token = 90 halt BASIC program STRS Token = C4 string representation of number function SYS Token = 9E call ML routine TAB Token = A3 tab position in printed output TAN Token = C0 trigonometric tangent function TEMPO Token = FE,05 set tempo (speed) of music play THEN Token = A7 if/then/else clause TO Token = A4 (subcommand) TRAP Token = D7 define BASIC error handler TROFF Token = D9 BASIC trace mode disable TRON Token = D8 BASIC trace mode enable TYPE Token = FE,27 display sequential disk file UNTIL Token = FC do/loop conditional USING Token = FB define print output format USR Token = B7 call user ML function VAL Token = C5 numeric value of a string function VERIFY Token = 95 compare memory to disk file VIEWPORT Token = FE,31 (currently unimplemented) VOL Token = DB set audio volume WAIT Token = 92 pause program pending memory condition WHILE Token = ED do/loop conditional WIDTH Token = FE,1C (currently unimplemented) WINDOW Token = FE,1A set text screen display window XOR Token = CE,08 logical xor function 3.1.3. BASIC 10.0 COMMAND AND FUNCTION DESCRIPTION ABS -- Absolute value function ABS (expression) The ABSolute value function returns the unsigned value of the numeric expression. X = ABS(1) Result is X = 1 X = ABS(-1) Result is X = 0 AND -- Boolean operator expression AND expression The AND operator returns a numeric value equal to the logical AND of two numeric expressions, operating on the binary value of signed 16-bit integers in the range (-32768 to 32767). Numbers outside this range result in an 'ILLEGAL QUANTITY' error. X = 4 AND 12 Result is X=4 X = 8 AND 12 Result is X=8 X = 2 AND 12 Result is X=0 In the case of logical comparisons, the numeric value of a true situation is -1 (equivalent to 65535 or $FFFF hex) and the numeric value of a false situation is zero. X = ("ABC"="ABC") AND ("DEF"="DEF") Result is X=-l (true) X = ("ABC"="ABC") AND ("DEF"="XYZ") Result is X= 0 (false) APPEND -- Open a disk file and prepare to append data to it APPEND# logical_file_number, "filename" [,Ddrive] [Udevice] Opens filename for writing, and positions the file pointer at the end of the file. Subsequent PRINT# statements to the logical_file_number will cause data to be appended to the end of this file. If the file does not exist, it will be created. APPEND#1, "filename" APPEND#1, (file$), ON U(unit) ASC -- PETSCII value function ASC (string) This function returns the PETSCII numeric value of the first character of a string. The PETSCII value of an empty (null) string is zero. This function is the opposite of the CHR$ function. Refer to the Table of PETSCII Character Codes. X = ASC("ABC") Result is X=65 X = ASC("") Result is X=0 ATN -- Arc tangent function ATN (expression) This function returns the angle whose tangent is the value of the numeric expression, measured in radians. The result is in the range of -PI/2 to PI/2 radians. X = ATN(45) Result is X=1.54050257 To get the arc tangent of an angle measured in degrees, multiply the numeric expression by PI/180. AUTO -- Enable or disable automatic line numbering AUTO [increment] Turns on the automatic line numbering feature which eases the job of entering programs by typing the line numbers for the user. As each program line is entered by pressing the next line number is printed on the screen, with the cursor in position to begin typing that line. The increment parameter refers to the increment between line numbers. AUTO with no increment given turns off auto line numbering. AUTO mode is also turned off automatically when a program is RUN. This statement is executable only in direct mode. AUTO 10 automatically numbers line in increments of ten. AUTO 50 automatically numbers line in increments of fifty. AUTO turns off automatic line numbering. BACKGROUND -- Set the background color of the display BACKGROUND color Sets the screen background color to the given color. The color given must be in the range (0-15). See the Color Table. BACKUP -- Backup an entire disk from one drive to another BACKUP Dsource_drive TO Ddestination_drive [Udevice] This command copies all the files on a diskette to another on a dual drive system only. It cannot backup diskettes using CBM serial bus type drives, for example. If the destination diskette is unformatted, BACKUP will automatically format it. BACKUP copies every sector, so any data already on the destination diskette will be overwritten. To copy specific files from one drive to another, use the COPY command. NOTE: This command can only be used with a dual disk drive, such as the built-in C64DX drive and optional F016-type expansion drive. To backup diskettes using different drives, such as the built-in drive and a 1581-type serial bus drive, use a utility program. BACKUP D0 to D1 Copies all files from the disk in drive 0 to the disk in drive 1. BACKUP D0 TO D1, ON U9 Copies all files from drive 0 to drive 1 in disk drive unit 9. BANK -- Set the memory bank number for PEEK, POKE, SYS, WAIT, LOAD, SAVE BANK memory_bank [*** THIS COMMAND MIGHT CHANGE ***] This command should be used before and BASIC command that has an address parameter. The address parameters are limited to the range (0-65535, $0000-$FFFF hex). The BANK command tells the computer which 64K byte memory bank the location you want is in. The memory_bank parameter is number from 0-255. Refer to the System memory map to see what is in each bank. A BANK number greater than 127 (i.e., has its most significant bit set) means "use the current system configuration", and must be used to access an I/O location. BASIC defaults to BANK 128. For examples, see PEEK, POKE, etc. BEGIN/BEND -- Extend an IF clause over more than one line BEGIN/BEND are used to define a block of code which is considered by the IF statement to be one statement. The normal usage of IF/THEN/ELSE would be along the following lines: IF boolean THEN statement(s) : ELSE statement(s) The main restriction is that the entire body of the IF/THEN/ELSE construct can only occupy one line. BEGIN/BEND allows either the 'THEN' or the 'ELSE' clause to run on for more than one line. IF boolean THEN BEGIN: statements... statements... statements... BEND : ELSE BEGIN statements... statements... BEND Remember, however, that this is only a way to extend the body for more than one line: all other 'IF/THEN' rules apply. For example: 100 IF x=1 THEN BEGIN : a=5 110 : b=6 120 : c=7 130 BEND : print "ah-ha!" In the above example, "ah-ha!" would be printed ONLY if the expression 'x=1' is TRUE, because the print statement is on the same logical line as the THEN clause. It is bad practice to GOTO a line in the middle of a BEGIN-BEND block. If BEGIN or BEND is encountered outside of an active IF statement, it is ignored. BLOAD -- loads a binary disk file into memory BLOAD "filename" [,Bbank] [,Paddress] [Udevice] Used to load a machine language program or other binary data (such as display pictures or sprite data) into memory. If a load address is not given, the load address given in the disk file will be used. If a bank number is not given, the bank given in the last BANK statement will be used. If a load overflows a bank (that is, the load address exceeds 65535 ($FFFF)), an 'OUT OF MEMORY' error is reported. Also see the LOAD command. BLOAD "sprites", P(dec("600")), B0 BOOT -- Load and execute a program BOOT BOOT SYS BOOT filename [,Bbank] [,Paddress] [,Ddrive] [Udevice] BOOT without a filename given causes the computer to look for a BASIC program called AUTOBOOT.C65* on the indicated diskette, LOAD it and RUN it (just like RUN "AUTOBOOT.C65*"). BOOT with a filename given will cause the executable binary file to be BLOADed and executed beginning at the load address. If a load address is not given, the file will be loaded and execution begun at the address stored on disk. BOOT SYS is a special command that copies the "home" sector (the very track and sector) of the C64DX built-in drive into memory at address $400 to $5FF (one physical sector, 512 bytes) and perform a machine language JSR (Jump SubRoutine) to it. It has the same function as turning on your C64DX while holding down the ALT key. It is used to boot an alternate operating system from either a CBM 3.5" diskette or an MSDOS (720K) diskette. If used in a BASIC program, and it fails, the system can be corrupted. BOOT SYS does *not* use the normal DOS to access the disk. BOOT Loads & runs BASIC program called AUTOBOOT.C65* on system disk. BOOT U9 Loads & runs BASIC program called AUTOBOOT.C65* on disk unit 9. BOOT "ml" Load & executes machine language program called ML, starting at address stored on disk. BORDER -- Set the exterior border color of the display BORDER color Sets the screen border color to given color. The color must be in the range (0-15). See the Color Table. BOX -- Draw a 4-sided graphical shape BOX x0,y0, x1,y0, x0,y1, x1,y1 [,solid] Requires two line segments to be specified, the order of which determines the shape drawn. The shape is drawn in the currently specified PEN color, on the currently specified SCREEN. The above command will draw the following shape: |0,<=0 +--------------------+ |1,<=0 | | | | |0,<=1 +--------------------+ |1,<=1 But if the order of the coordinates were given as: BOX x0,y0, x1,y0, x1,y1, x0,y1 a "bowtie" shape would be drawn. See the sample program at SCREEN. BSAVE -- Save an area of memory in binary disk file BSAVE "[@]filename", Pstart_adr TO Pend_adr [,Bbank] [,Ddrive] [Udevice] BSAVE copies an area of memory into a binary disk file called "filename", starting at start_adr and ending at end_adr-1 (i.e. end_adr must be one more than actual last address saved). If a bank number is not given, the bank given in the last BANK statement will be used. end_adr must be greater than start_adr, and area to be saved must be limited to the indicated memory bank. You cannot save data from more than one bank at a time. start_adr is saved on disk as the load address. If filename already exists on the designated diskette, memory is NOT saved and a 'FILE EXISTS' error is reported. Preceding the filename with an '@'-sign will allow you to overwrite an existing file, but see the cautions at DSAVE. BSAVE "sprites", P(dec("600")) TO P(dec("800")), B0 BUMP -- Sprite collision function BUMP (type) This function return a numeric summary of sprite collisions accumulated since the last time the BUMP function was used. You can use the COLLISION command to set up a special routine in your program to receive control whenever a sprite BUMPs into something, but a particular COLLISION does not have to be enabled to use BUMP. See the COLLISION command. To evaluate sprite collisions, where a BIT position (0-7) in the numeric result corresponds to a sprite number (0-7): BIT position: 7 6 5 4 3 2 1 0 | | | | | | | | BUMP value in binary: 0 0 0 0 0 1 0 1 = 5 decimal BUMP(1) returns a value representing sprite-to-sprite collisions. BUMP(2) returns a value representing sprite-to-data collisions. X = BUMP(1) Result is X=3 if sprites 0 & 1 collided, as shown above. (binary 101 = 5 decimal). Note that more than one collision can be recorded, in which case you should evaluate a sprite's position using the RSPPOS function to figure out which sprite collided with what. BUMP is reset to zero after each use. BVERIFY -- Compare a binary disk file to an area of memory BVERIFY "filename" [,Paddress] [,Bbank] [,Ddrive] [Udevice] BVERIFY compares a binary disk file called "filename" to an area of memory. In direct mode, if the areas contain the same data the message "OK" is displayed, and if the data differs the message 'VERIFY ERROR' is displayed. In program mode, an error is generated if a mismatch is found otherwise the program continues normally. The comparison starts with the address given, else it starts at the address stored on disk. The comparison ends when the last byte is read from the disk file. If a bank number is not given, the bank given in the last BANK statement will be used. The ending address is determined by the length of the disk file. The comparison halts on the first mismatch or at the end of the file. The area to be compared must be confined to the indicated memory bank. BVERIFY "sprites", P(dec("600")), B0 CATALOG -- see DIR (DIRECTORY) command CHANGE -- Find text in a BASIC program and change it. CHANGE :string1: TO :string2: [,line_range] CHANGE "string1" TO "string2" [,line_range] This is a direct (edit) mode command. CHANGE looks for all occurrences of string1 in the program, displays each line containing string1 with the target string highlighted, and prompts the user for one of the following: Y Yes, change it and look for more N No, don't change it, but look for more * Yes, change all occurrences from here on Exit command now, don't change anything Any character can be used for the string delimiter, but there are side effects: see comments at FIND command. If the line number range is not given (see LIST for description of range parameter), the entire program is searched. CHAR -- Draw a character string on a graphic screen CHAR column, row, height, width, direction, "string" [,charsetadr] [*** THIS IS SUBJECT TO CHANGE ***] CHAR displays text on a graphic screen at a given location. The character height, width, and direction are programmable. The parameters are defined as: column: Character position: For 320 wide screens, 0-39 For 640 wide screens, 0-79 row: Pixel line: For 200 line screens, 0-199 For 400 line screens, 0-399 height: Multiple of 8-bit character height: 1= 8 pixels high, 2= 16 pixels, etc. width: Multiple of 8-bit character width: 1= 8 pixels high, 2- 16 pixels, etc. direction: Bit mask: B0= up B1= right B2= down B3= left The string can consist of any printable character, as defined by the VIC character set. Non-text characters are ignored. If the address of the character set is not given, the upper/lower ROM character set is used ($29800). CHAR 18,96, 1,1,2, "C65D", DEC("9000") The above example will draw the characters "C65D" in the center of a 320x200 pixel screen using the system's uppercase/graphic character set. CHR$ -- Character string function CHR$ (value) This function returns a string of one character having the PETSCII value specified. This function is the opposite of the ASC function. It's often used in PRINT strings to output data that is not visible, such as control codes and escape sequences. Refer to the Table of PETSCII Character Codes. PRINT CHR$(27)"Q"; CHR$(27) is the escape character. This statement performs the clear-to-end-of-line escape function. CIRCLE -- Draw a circle on a graphic screen CIRCLE x_center, y_center, radius [,solid] The CIRCLE command will draw a circle with the given radius centered at (x_center,y_center) on the current graphic screen. The circle will be filled (i.e., a disc) if SOLID is non-zero. CIRCLE 160,100,50 The above example will draw a circle in the center of a 320x200 pixel screen (160,100) having a radius of 50 pixels. The aspect ratio of the screen may cause it to appear as an ellipse, however. See also the ELLIPSE command. CLOSE -- Close a logical I/O channel CLOSE logical_channel_number This command closes the input/output channel associated with the given logical_channel_number, established by an OPEN statement. In the case of buffered output (such as the serial bus or RS232) any data in the device's buffer will be transmitted before the channel is closed. Refer to specific I/O operations for details. The logical_channel_number is required; to close all channels on a given device, use the DCLOSE command. Note that RUN, NEW, and CLR commands will initialize the logical channel tables but will not actually close any channels. CLR -- Clear program variables CLR This statement initializes BASIC's variable list, setting all numeric variables to zero and string variables to null. It also initializes the DATA pointer, BASIC runtime stack pointer (i.e., clears all GOSUBs, DO/LOOPs, FOR/NEXT loops, etc.), and clears any user functions (DEF FNx). Any OPEN channels are forgotten (but a CLOSE is not performed; don't use if there are any open disk output files). A CLeaR is automatically performed by a RUN or a NEW command. CMD -- Set default output channel CMD logical_channel_number [,string] CMD changes the default output device, normally the screen, to that specified. The logical_channel_number can be any previously OPENed write channel, such as one to a disk file, printer, or RS232. When redirected via CMD, all output which normally would go to the screen (such as PRINT commands, LIST output, DIRECTORY lists, etc.) is sent to another device or file. The redirection is terminated by CLOSE-ing the CMD channel or executing a PRINT# to the CMD channel. Some output devices require a PRINT# to be performed before the CMD channel is closed, such as printers, to cause the device's buffer to be flushed (i.e., displayed). Any system error will redirect output back to the system default, normally the screen, but will not flush nor close the output channel. If the optional string is given, it is output immediately after the CMD device is established. This feature is normally used to set up printers (eg., set printer modes via escape codes) or to identify the output (eg., title printouts). OPEN 4,4 OPENS device #4, which is the printer. CMD 4 All normal output now goes to the printer. LIST The LISTing goes to the printer. PRINT#4 Set output back to the screen. CLOSE 4 Close the printer channel. COLLECT -- Check (validate) disk, delete bad files and free lost sectors COLLECT [Ddrive] [Udevice] Refer to the DOS 'V'alidate command. This command will cause the DOS to recalculate the Block Availability Map (BAM) of the diskette in the indicated drive, allocating only those sectors being used by valid, properly closed files. All other sectors are marked as "free" and improper files are automatically deleted. Note: COLLECT should be used with extreme care, and MUST NOT be used on diskettes with special boot sectors or direct access (eg., random) files. In any case, be sure the diskette has been BACKUP-ed first. COLLISION -- Setup subroutine to handle special events COLLISION type [,linenumber] [*** THIS MIGHT CHANGE ***] COLLISION is used to handle "interrupt" situations in BASIC, such as sprites bumping into things or lightpen triggers. When the specified situation occurs, BASIC will finish processing the currently executing instruction and perform an automatic GOSUB to the linenumber given. When the subroutine terminates (it must end with a RETURN) BASIC will resume processing where it left off. Interrupt handling continues until a COLLISION of the same type but without any linenumber is specified. More than one type interrupt may be enabled at the same time, but only one interrupt can be handled at a time (i.e., no recursion and no nesting of interrupts). The type interrupt can be: 1 = Sprite to sprite collision 2 = Sprite to display data collision 3 = Light pen Note that what caused an interrupt may continue causing interrupts for some time unless the situation is altered or the interrupt is disabled. This is especially true for BASIC, which is slow to respond to interrupts. Use the BUMP and RSPPOS functions to evaluate the results of sprite collisions, and the LPEN function to evaluate the position of a light pen. 10 COLLISION 1,90 20 SPRITE 1,1 : MOVSPR 1,100,100 : MOVSPR 1,0#5 30 SPRITE 2,1 : MOVSPR 2,100,150 : MOVSPR 2,180#5 40 DO : PRINT : LOOP 50 END 90 PRINT"BUMP! ";:RETURN In this example, sprite-to-sprite collisions are enabled (line 10), and two sprites are turned on, positioned, and made to move (lines 20 & 30). One sprite moves up and the other moves down while the program does nothing other than print blank lines to the screen (line 40). When the sprite collide, the subroutine at line 90 is called, it prints "BUMP!", and the computer goes back to printing blank lines. COLOR -- Enable or disable screen color (character attribute) control COLOR COLOR turns on or turns off the screen editor's attribute handler. When colors are turned off, whatever character attributes are being currently displayed (text color, underline, flash, etc.) are "stuck". The main purpose for doing this is to speed up screen handling (writing to the screen or scrolling the screen) about two times, since the screen editor no longer has to manipulate the attributes. Note that only FOREGROUND colors (and special VIC attributes) are affected. To change screen colors, use the following commands: FOREGROUND color# Set Foreground color (text) HIGHLIGHT color# Set Highlight color (text) BACKGROUND color# Set VIC background color BORDER color# Set VIC border color CONCAT -- Concatenate (merge) two sequential disk files CONCAT "file1"[,Ddrive1] TO "file2"[,Ddrive2] [Udevice] CONCAT merges two SEQuential files, appending the contents of "file1" to "file2". Upon completion, "file2" contains the data of both files, and "file1" is unchanged. Both files must exist on drives of the the same unit, and pattern matching is not allowed. Some disk drives handle CONCAT differently; refer to the DOS manual for specific details. CONT -- Continue program execution CONT CONTinue is used to re-start a BASIC program that was halted by a STOP or END statement, or interrupted by the key. The program will resume at the statement following the STOP or END instruction, or at the statement after the one that was interrupted by the key. CONT is typically used during program debugging. You can look at and alter variables while the program is halted. Programs halted as a result of an untrapped error condition cannot be CONTinued. Programs that have been edited in any way cannot be restarted. Any error condition that occurs since the program was halted will prevent it from being restarted. Programs that cannot be restarted via CONT can be restarted with a GOTO, as long as you don't need to resume execution in the middle of a line of commands and you recall where the halt occurred. Note that the key can interrupt some commands in mid-execution, such as file I/O, drawing commands, etc. In such cases, programs may not run correctly after a CONTinue. COPY -- Copy disk files COPY ["file1"][,Dd1] TO ["file2"][,Dd2] [Udevice] COPYs a disk file to another disk file. On single drive units, the filenames must be different. On dual drive units, copying can be done between two drives on the same unit, and the filenames can be the same or different. Pattern matching can be used. Copying files from one unit to a different unit cannot be done: use a copy utility program in such cases. Only legal type files can be copied; direct access data, boot sectors, and partitions cannot be copied. Refer to the DOS manual for your disk drive specific details. COPY "file1" TO (F2$) Copies "file1" to another file whose name is in F2$ on the same drive. Names must differ. COPY "file1",D0 TO D1,U9 Copies "file1" from unit 9 drive-0 to unit 9 drive-1. COPY D0 TO D1 Copies all files from drive-0 to drive-1 on the same unit. COPY "???.src",D0 TO "*",D1 Copies all files on drive-0 matching the pattern to a file of the same name on drive-1. COS -- Cosine function COS (expression) This function returns the cosine of X, where X is an angle measured in radians. The result is in the range -1 to 1. X = COS (pi) Result is X=-1 To get the cosine of an angle measured in degrees, multiply the numeric expression by pi/180. CUT -- Cut a graphic area into a temporary structure CUT x,y,dx,dy [*** NOT YET IMPLEMENTED ***] DATA -- Define program constant data to be accessed by READ command DATA [list of constants] DATA statements store lists of data that will be accessed during program execution by a READ statement. The DATA statement can appear anywhere in the program, and it is never executed. BASIC keeps a pointer to the earliest un-READ DATA statement, and data is read sequentially from first item in a DATA statement to the last item, from the earliest DATA statement in the program to the last DATA statement in the program. The list of constants can contain both numeric data (integer or floating point) and string data, but cannot contain expressions which must be evaluated (such as 1+2, DEC("1234"), or CHR$(13)). Items are separated by commas. String data need not be enclosed in quotes unless it contains certain characters, such as spaces, commas, colons, graphic characters, or control codes. If two commands have nothing between them, the data will be READ as 0 if numeric or a null string. The RESTORE command allows you to position BASIC's data pointer to a specific line number. If the program tries to read more DATA than exists in the program, an 'OUT OF DATA' error results. If a READ statement's variable type does not agree with the DATA being read, a 'TYPE MISMATCH' error results. DATA 100, 200, FRED, "HELLO, MOM", , 3.14, ABC123, -1.7E-9 DCLEAR -- Clear all open channels on disk drive DCLEAR [Ddrive] [Udevice] DCLEAR sends the indicated disk drive an 'I'nitialize command. This clears all open channels, closes all open files, and causes the DOS to re-read the diskette's Block Allocation Map (BAM). Note that DCLEAR DOES NOT close open channels on the computer's side (see the DCLOSE command). There are some other side affects caused by this command with different types of drives -- refer the DOS manual for your disk drive for specific details. DCLOSE -- Close a disk file, or close all channels on a device DCLOSE [#logical_file_number] [Udevice] DCLOSE is intended to close a file opened with the DOPEN command. Specific files can be closed by specifying a logical_file_number, or all files on a particular drive can be closed by not specifying a particular logical_file_number. It is possible to close channels on non-disk devices with this command by specifying only the device number. DCLOSE#1 Closes the file associated with logical logical file number 1. DCLOSE Closes all files currently open on the default system drive. DCLOSE U(U2) Closes all channels open to device U2. DEC -- Decimal value function DEC (hex_string) This function return the decimal value of a string representing a hexadecimal number in the range "0000" to "FFFF". The result is in the range 0-65535. If the string contains a non-hexadecimal digit or is more than four (4) characters in length an 'ILLEGAL QUANTITY' error is reported. VIC = DEC("D000") Result is VIC=53248, the address of the VIC chip DEF FN -- Define function DEF FN name(numeric_variable) = numeric_expression Define a user-written numeric function. The DEF FNx statement must be executed before the function can be used. Once a function has been defined, it can be used like any other numeric variable. The function name is the letters FN followed by any legal floating point (non-integer) variable name. A function can be defined only in a program. The numeric_variable is a "dummy" variable. It names the variable the numeric_expression which will be replaced when the function is used. It's not required to be used in the numeric_expression, and its value won't be changed by the function call. The numeric_expression performs the calculations of the function. It is any legal numeric expression that fits on one line. Variables used in the expression have their value at the time the function is used. Functions can be used only by the program which defines them. If one program chains to another program, the first program's functions cannot be used (usually a 'SYNTAX ERROR' results). Similarly, if the program is moved in any way after the function is defined, the function cannot be used. 10 DEF FN R(MAX) = INT(RND(0)*MAX)+1 20 INPUT "MAXIMUM"; MAX 30 PRINT FN R(MAX) In this example, we've defined a function which will return a pseudo random number between 1 and whatever MAX is. Instead of using the expression INT(RND(0)*MAX)+1 every time a random number is needed, we can now use FN R(MAX). When we use FN R(x), the value of 'x' will be substituted everywhere MAX is used in the function definition. 10 DEF FN I(X) = X+1 20 DEF FN L(Z) = LEN(A$) 30 DEF FN AVG(N) = (TOT*CNT+N)/(CNT+1) DELETE -- Delete lines of BASIC program, or Delete disk files DELETE [startline] [-[endline]] DELETE "filespec" [,Ddrive] [Udevice] [,R] There are two forms of DELETE. The first form is used in direct mode to remove lines from a BASIC program: DELETE 75 Deletes line 75 DELETE 10-50 Deletes line 10 through 50 inclusive. DELETE -50 Deletes all lines from the beginning of the program up to and including line 50. DELETE 75- Deletes all lines from 75 to the end of the program. The second form is used in program or direct mode to delete a disk file. See the SCRATCH command. DELETE "myfile" Deletes the file MYFILE on the system drive. DIM -- Declare array dimensions DIM variable(subscripts) [,variable(subscripts)]... Before arrays of variables can be used, the program must first execute a DIM statement to establish DIMensions of that array (unless there are 11 or fewer elements in the array). The statement DIM is followed by the name of the array, which may be any legal variable name. Then, enclosed in parentheses, put the number (or numeric variable) of elements in each dimension. An array with more than one dimension is called a matrix. Any number of dimensions may be used, but keep in mind that the whole list of variables being created takes up space in memory, and it is easy to run out of memory if too many are used. To figure the number of variables created with each DIM, multiply the total number of elements in each dimension of the array. Note: each array starts with element 0, and integer arrays take up 2/5ths of the space of floating point arrays. More than one array can be dimensioned in a DIM statement by separating the arrays by commas. If the program executes a DIM statement for any array more than once, the message 'REDIM'D ARRAY' is reported. It is good programming practice to place DIM statements near the beginning of the program. 10 DIM A$(40),B7(15),CC%(4,4,4) | | | 41 elements 16 elements 125 elements DIRECTORY -- List the files of a diskette DIR DIRECTORY ["filespec"] [,R] [,Ddrive] [Udevice] A directory is a list of the names of the files that are on a diskette. The directory listing consists of the name of the diskette, the names, sizes, and filetypes of all the files on a diskette, and the remaining free space on the diskette. The filespec is used to specify a pattern match string to view selected files. Not all disk drives support the same options or filespecs; refer to your DOS manual for details. The C64DX allows you to print DIR listings without having to 'load' the directory; see example below. The commands DIR, DIRECTORY, and CATALOG have the exact same function. They can be used in direct or program mode. DIRECTORY List all files on the diskette in the default system drive DIR "*.src", U9 Lists the all the files ending with ".src" on unit 9. DIR "*,=p",R List all the deleted but recoverable PRG-type files on the system drive. OPEN 4,4:CMD 4:DIR:CLOSE 4 Print DIR listing to printer unit 4. The following program can be used to load the directory into variables for use within a program. In this case, the filename is simply printed to the screen: 10 OPEN 1,8,0,"$0:*,P,R" open dir as a file 20 : IF DS THEN PRINT DS$: GOTO 100 abort if error 30 GET#1,X$,X$ trash load address 40 DO read each line 50 : GET#1,X$,X$: IF ST THEN EXIT trash links, check EOF 60 : GET#1,BL$,BH$ get file size 70 : LINE INPUT#1, F$ get filename & type 80 : PRINT LEFT$(F$,18) print filename 90 LOOP loop until EOF 100 CLOSE 1 close dir DISK -- Send a disk command DISK "command_string" [Udevice] The DISK command is used to send special commands to the DOS via the disk drive's command channel. The DISK command is analogous to the following BASIC code: OPEN 1,n,15: PRINT#1,"command_string": CLOSE 1 Not all disk drives understand the same commands. Refer to your DOS manual for commands and command syntax for your drive. Note that the drive number, if any, must be included in the command_string. DISK "U0>10" Renumber system drive to 10. DISR "U0>V"+chr$[0) Turn off write verify DISK "S0:file",U(n) Scratch "file" on unit n DLOAD -- Load a BASIC program file from disk DLOAD "filename" [,Ddrive] [Udevice] This command copies a BASIC program from disk into the BASIC program area of the computer. It can then be edited, DSAVEd, or RUN. Used in program mode, it overlays the current program in memory and begin execution automatically at the first line of the new program. Variable definitions will be left intact, but any open data files and the disk command channel will be automatically closed. This is called CHAINING. See also RUN. Use BLOAD to load binary or machine language data. DLOAD "myprogram" Searches the default system disk drive for the BASIC program "myprogram", loads it, and relinks it. DLOAD (F$),U9 LOADs a program whose name is in F$ from disk unit 9. DMA -- Perform a DMA operation DMA command [,length,source(l/h/b),dest(l/h/b),subcmd,mod(l/h) [,...]] [*** THIS COMMAND IS SUBJECT TO CHANGE ***] The DMA command defines and executes a Direct Memory Access operation. The parameters are used to construct a DMA list, which is then passed to the DMA processor for execution. Refer to the DMA chip specification for details. Chained DMA commands are not allowed, but multiple DMA commands can be given and the DMA handler will set up and execute each one, one at a time. Refer to the system memory map to find out where things are. Because this command directly accesses system memory, extreme care should be taken in its use. Changing the wrong memory locations can crash the computer (press the reset button to reboot). DMA 3, 2000, ASC("+"), 0, DEC("800"), 0 Fill screen with '+' DMA 0, 2000, DEC("800"), 0, DEC("8000"), 1 Copy screen to $18000 DMODE -- Set graphic display mode DMODE jam, comp, inverse, stencil, style, thickness [*** THIS COMMAND IS SUBJECT TO CHANGE ***] jam 0-1 complement 0-1 inverse 0-1 stencil 0-1 style 0-3 thickness 1-8 DO/LOOP/WHILE/UNTIL/EXIT -- Program loop definition and control DO [UNTIL boolean_expression | WHILE boolean_expression] . . statements [EXIT] . LOOP [UNTIL boolean_expression | WHILE boolean_expression] Performs the statements between the DO statement and the LOOP statement. If no UNTIL or WHILE modifies either the DO or the LOOP statement, execution of the intervening statements continues indefinitely. If an EXIT statement is encountered in the body of a DO loop, execution is transferred to the first statement following the nearest LOOP statement. Do loops may be nested, following the rules defined for FOR-NEXT loops. If the UNTIL parameter is used the program continues looping until the boolean argument is satisfied (becomes true). The WHILE parameter is basically the opposite of the UNTIL parameter: the program continues looping as long as the boolean argument is TRUE. An example of a boolean argument is A=1, or G>65. DO UNTIL X=0 or X=1 This loop will continue : statements until X=0 or X=1. If LOOP X=0 or 1 at beginning the loop won't execute. 10 A$="": DO GETKEY A$: LOOP UNTIL A$="Q" This will loop until the user types 'Q' 10 DOPEN#1,"FILE" This program will 20 C=0 count the number of 30 DO: LINEINPUT#1,A$: C=C+1: LOOP UNTIL ST lines in FILE 40 DCLOSE#1 50 PRINT"FILE CONTAINS";C;" LINES." DOPEN -- Open a disk file DOPEN#lf,"filename[,]"[,L[reclen]] [,W] [,Ddrive] [Udevice] This command OPENs a file on disk for reading or writing. If is the logical file number, which you will use in PRINT#, INPUT#, GET#, RECORD#, and DCLOSE# commands to reference the channel to your file. The filename is required. The defaults are to OPEN a SEQuential file for Reading, in which case the file must exist or a 'FILE NOT FOUND' error results. To create an file and write to it, use the 'W'rite option. 'FILE EXISTS' error is report if an output file already exists. To read or write a RELative file, use the 'L'ength option. The 'reclen' record length is required only when creating a relative file. For more information regarding Relative files, see the RECORD command and refer to your DOS manual. See also APPEND. See the OPEN command for a discussion about channel and device numbers. DOPEN#1,"readfile" Opens sequential READFILE for reading. DOPEN#1,"writefile",W Creates & opens seq WRITEFILE for writing. DOPEN#1,"file,P",U(u) Opens a PRoGram type file for reading on unit U DOPEN#1,(rf$),L Open existing relative file whose name's in RF$ DOPEN#a,"rel",L80 Create a relative file with record length of 80 DPAT -- Set graphic draw pattern DPAT type [, # bytes, byte1, byte2, byte3, byte4] [*** THIS COMMAND IS SUBJECT TO CHANGE ***] type 0-63 # bytes 1-4 byte1 0-255 byte2 0-255 byte3 0-255 byte4 0-255 DSAVE -- Save a BASIC program into a disk file DSAVE "[@]filename" [,Ddrive] [Udevice] This command copies a BASIC program in the computer's BASIC memory area into a PRoGram-type disk file. If the file already exists, the program is NOT stored and the error message 'FILE EXISTS' is reported. If the filename is preceded with an '@', then if the file exists it will be replaced by the program in memory. Because of some problems with the 'save-with-replace' option on older disk drives, using this option is not recommended if you do not know what disk drive is being used. Use the DVERIFY to compare the program in memory with a program on disk. To save a binary program, use the BSAVE command. DSAVE "myprogram" Creates the PRG-type file MYPROGRAM on the default system disk and copies the BASIC program in memory into it. DSAVE "@myprogram" Replaces the PRG-type file MYPROGRAM with a new version of MYPROGRAM. If MYPROGRAM doesn't exist, it's created. DSAVE (F$),U9 Saves a program whose name is in F$ on disk unit 9. DVERIFY -- Compare a program in memory with one on disk DVERIFY "filename" [,Ddrive] [Udevice] This command is just like a DLOAD, but instead of LOADing the BASIC program file into computer memory the data is read from disk and compared to computer memory. If there's any difference at all a 'VERIFY ERROR' is reported. Note: If the BASIC program in memory is not located at the same address as the version on disk was SAVEd from, the files will not match even if the program is otherwise identical. The comparison ends when the last byte is read from the disk file. Use the BVERIFY command to compare memory with binary files. DVERIFY "myprogram" Good: SEARCHING FOR 0:myprogram Bad: SEARCHING FOR 0:myprogram VERIFYING VERIFYING OK ?VERIFY ERROR ELLIPSE -- Draw an ellipse on a graphic screen ELLIPSE x_center, y_center, x_radius, y_radius [,solid] The ELLIPSE command will draw an ellipse with the given radii centered at (x_center,y_center) on the current graphic screen. The ellipse will be filled (i.e., a disc) if SOLID is non-zero. ELLIPSE 160,100,65,50 The above example will draw an ellipse in the center of a 320x200 pixel screen (160,100) having radii of (65,50) pixels. The aspect ratio of the screen may cause it to appear as an circle, however. See also the CIRCLE command. ELSE -- See IF/THEN/ELSE END -- Define the end of program execution END The END statement terminates program execution. It does not close channels or files, and it does not clear any variables or reset any pointers. An END statement does not need to be put at the last line of a program. The CONTinue command can be used to resume execution with the next statement following the END statement. See also the STOP command. ENVELOPE -- Define musical instrument envelopes ENVELOPE n, [,[atk] [,[dec] [,[sus] [,[rel] [,[wf] [,pw] ]]]]] n............... Envelope number (0-9) atk ............ Attack rate (0-15) dec ............ Decay rate (0-15) sus ............ Sustain rate (0-15) rel ............ Release rate (0-15) wf ............. Waveform: 0 = triangle 1 = sawtooth 2 = pulse (square) 3 = noise 4 = ring modulation pw ............. Pulse width (0-4095) [*** THIS COMMAND IS SUBJECT TO CHANGE ***] A parameter that is not specified will retain its current value. Pulse width applies to pulse waves (wf=2) only and is determined by the formula (pwout = pw/40.95 %), so that pw = 2048 produces a square wave and values of 0 or 4095 produce constant DC output. The C64DX initializes the ten (10) tune envelopes to: n A D S R wf pw instrument ------------------ ------------- ENVELOPE 0, 0, 9, 0, 0, 2, 1536 piano ENVELOPE 1,12, 0,12, 0, 1 accordion ENVELOPE 2, 0, 0,15, 0, 0 calliope ENVELOPE 3, 0, 5, 5, 0, 3 drum ENVELOPE 4, 9, 4, 4, 0, 0 flute ENVELOPE 5, 0, 9, 2, 1, 1 guitar ENVELOPE 6, 0, 9, 0, 0, 2, 512 harpsichord ENVELOPE 7, 0, 9, 9, 0, 2, 2048 organ ENVELOPE 8, 8, 9, 4, 1, 2, 512 trumpet ENVELOPE 9, 0, 9, 0, 0, 0 xylophone ERASE -- Delete disk files ERASE "filespec" [,Ddrive] [Udevice] [,R] This command is identical to DELETE and SCRATCH. See the SCRATCH command for details. ERASE "myfile" Deletes the file MYFILE on the system drive. ERR$ -- Error message function ERR$ (error_number) This function returns a string which is the BASIC error message corresponding to the given error_message. If the given number is too small (less than 1) or too large (greater than 41) an 'ILLEGAL QUANTITY' error is reported. This function is usually used to display a BASIC error condition in a TRAP routine, using the BASIC error word ER as the error_number. Note that when ER=-1, no BASIC error has occurred and ERR$(-1) results in an 'ILLEGAL QUANTITY' error. See the example at TRAP. EXIT -- See DO/LOOP/WHILE/UNTIL/EXIT EXP -- Function to return e^x EXP (number) This function returns the numeric value of e (2.71828183), the base of natural logarithms) raised to the power of given number. If the number is greater than 88.0296919 an 'OVERFLOW' error is reported. X = EXP(4) Result is X=54.5981501 FAST -- Set system speed to 3.58MHz FAST is the default state of the system. FAST is used to restore this state following direct access of "slow" I/O devices such as the SID sound chips. FETCH -- (see the DMA command) FILTER -- Define sound filter parameters FILTER [freq] [,[lp] [,[bp] [,[hp] [,res] ]]] freq ..... Filter cut-off frequency (0-2047) lp ....... Low pass filter on (1) off (0) bp ....... Band pass filter on (1), off(0) hp ....... High pass filter on (1), off(0) res ...... Resonance (0-15) [*** THIS COMMAND IS SUBJECT TO CHANGE ***] Unspecified parameters result in no change to the current value. The filter output modes are additive. For example, how low pass and high pass filters can be selected to produce a notch (or band reject) filter response. For the filter to have an audible effect at least one filter output mode must be selected and at least one voice must be routed through the filter. FIND -- Find text in a BASIC program. FIND :string: [,line_range] FIND "string" [,line_range] This is a direct (edit) mode command. FIND looks for all occurrences of string in the program and displays each line containing string, with string highlighted. Use the key to slow the display, or the key to pause the display. Press to cancel. Any character can be used for the string delimiter, but there are side effects. Using a non-quote delimiter will cause the string to be tokenized, and FIND will find only tokenized strings in the program that match. Using a quote character as the delimiter will cause the string to be interpreted as plain PETSCII, and any matches found will therefore be plain PETSCII. Searching for some tokens such as DATA statements may require the use of colons as delimiters due to the special affect these commands have upon the interpreter. If the line number range is not given (see LIST for description of range parameter), the entire program is searched. FN xx -- User defined function FN xx (expression) The result of this numeric function is determined by the BASIC program in a DEF FN statement. See the example at DEF FN. FOR/TO/STEP/NEXT -- Program loop definition and control FOR index = start TO end [STEP increment] | NEXT index [,index] This command group performs a series of instructions a given number of times. The loop index is a floating point (non-integer) variable which will initially be set to the start value and be incremented by the STEP increment when the NEXT statement is encountered. The loop continues until the index exceeds the end value at the NEXT statement. The start, end, and increment values can be numeric variables or expressions. If the STEP increment is not specified, it is assumed to be one (1). The STEP increment can be any value, positive, negative, or non-integer. If the STEP increment is negative, the loop continues until the index is less than the end value at the NEXT statement. Note that, regardless of the start, end, or increment values, the loop will always execute at least once. The index can be modified within the loop, but it is bad practice to do so. It is also bad practice to GOTO a line inside a loop structure, or to similarly jump out of a loop structure (which can cause an 'OUT OF MEMORY' error). Loops may be nested. If too many are nested, an 'OUT OF MEMORY' error is reported (depends upon stack size, room for about 28 nested loops). The index variable can be omitted from the NEXT statement, in which case the NEXT will apply to the most recent FOR statement. If a NEXT statement is encountered and there is no preceding FOR statement, the error 'NEXT WITHOUT FOR' is reported. 10 FOR L = 1 TO 10 20 PRINT L 30 NEXT L 40 PRINT "I'M DONE! L = "L This program prints the numbers from one to ten, followed by the message I'M DONE! L = 11. 10 FOR L = 1 TO 100 20 FOR A = 5 TO 11 STEP .5 30 NEXT A 40 NEXT L This program illustrates a nested loop. FOREGROUND -- Set the text color of the display FOREGROUND color Sets the text color to the given color index. Color must be in the range (0-15). See the Color Table. COLOR must be ON (see the COLOR command). FRE -- Free byte function FRE (x) This function returns the number of available ("free") bytes in a specified area. PRINT FRE(0) Shows the amount of memory left in the program area, C64DX bank 0 X = FRE(1) X= the amount of available memory in variable area C64DX bank 1. This causes a "garbage collect" to occur, a process which compacts the string area. X = FRE(2) X= the number of expansion RAM banks present. GCOPY -- Copy a graphic area [*** NOT YET IMPLEMENTED ***] GENLOCK -- Enable or disable video sync mode & colors GENLOCK ON [,color#]... GENLOCK OFF [,color#,R,G,B]... To enable video sync mode and specify which colors are affected, use the GENLOCK ON command, and list the palette color indices (0-255) which will display external video. To disable video sync mode and restore the associated palette colors use the GENLOCK OFF command, and list the color index and its RGB values to restore them (see the SET PALETTE command for details). Also see the PALETTE RESTORE command. GET -- Get input data from the keyboard GET variable_list The GET statement is a way to get data from the keyboard one character at a time. When the GET is executed, the character that was typed is received. If no character was typed, then a null (empty) character is returned, and the program continues without waiting for a key. There is no need to hit the key, and in fact the key can be received with a GET. The word GET is followed by a variable name, usually a string variable. If a numeric were used and any key other than a number was hit, the program would stop with an error message. The GET statement may also be put into a loop, checking for an empty result, that waits for a key to be struck to continue. The GETKEY statement could also be used in this case. This statement can only be executed within a program. 10 DO: GET A$: LOOP UNTIL A$ ="A" This line waits for the A key to be pressed to continue. GETKEY -- Get input character from keyboard (wait for key) GETKEY variable_list The GETKEY statement is very similar to the GET statement. Unlike the GET statement, GETKEY waits for the user to type a character on the keyboard. This lets it be used easily to wait for a single character to be typed. This statement can only be executed within a program. 10 GETKEY A$ This line waits for a key to be struck. Typing any key will continue the program. GET# -- Get input data from a channel (file) GET# logical_channel_number, variable_list Used with a previously OPENed device or file to input one character at a time. Otherwise, it works like the GET statement. This statement can only executed within a program. 10 GET#1,A$ GO64 -- Exit C64DX mode and switch to C64 mode GO64 This statement switches from C64DX mode to C64 mode. The question 'ARE YOU SURE?' (in direct mode only) is posted for the user to respond to. If Y and return is typed then the currently loaded program is lost and control is given to C64 mode. This statement can be used in direct mode or within a program. GOSUB -- Call a BASIC subroutine GOSUB line This statement is like the GOTO statement, except that the computer remembers from where it came. When a line with a RETURN statement is encountered, the program jumps back to the statement immediately following the GOSUB. The target of a GOSUB statement is called a subroutine. A subroutine is useful if there is a section of the program that can be used by several different parts of the program. Instead of duplicating the section over and over, it can be set up as a subroutine and called with a GOSUB statement from different parts of the program. This also make the main part of your program much more readable. See also the RETURN statement. Variables are shared with the main program and all subroutines. You can pass information to, and get information back from, subroutines by using variables as messengers. GOSUB statements can be nested. That is, one subroutine can call another subroutine, and the computer automatically keeps track of all the calls. It's important not to jump into or out of subroutines since this can confuse the computer. If too many GOSUBs are nested (usually cause by jumping out of them) an 'OUT OF MEMORY' error is reported because the computer ran out of room to keep track of all the calls. 10 DIR : GOSUB 100 show directory, check status 20 GOSUB 200 print gap 30 LIST "PROGRAM": GOSUB 100 show listing, check status 40 GOSUB 200 print gap 50 etc... 90 END 99 : 100 REM SUBROUTINE TO CHECK DISK STATUS 110 IF DS THEN GOSUB 200: PRINT "DISK ERROR: ";DS$ 120 RETURN 199 : 200 REM SUBROUTINE TO PRINT A SPACER ON THE SCREEN 210 PRINT 220 FOR I=1 TO 39: PRINT"-";: NEXT 230 PRINT 240 RETURN GOTO -- Transfer program execution to specified line number GOTO line_number GO TO line_number After a GOTO statement is executed, the next line to be executed will be the one with the line number following the word GOTO. When used in direct mode, GOTO line number allows starting of execution of the program at the given line number without clearing the variables. 10 PRINT"COMMODORE" 20 GOTO 10 The GOTO in line 20 makes line 10 repeat continuously until STOP is pressed. GRAPHIC -- select graphic mode GRAPHIC CLR GRAPHIC command# [,args] Basically this is a modified C64-type SYS command, minus the address. In the C64DX system, this will represent the ML interface, not the BASIC 10.0 interface which is implemented in the development system. [*** THIS COMMAND IS SUBJECT TO CHANGE ***] GRAPHIC CLR initializes (warm-starts) the BASIC graphic system. It clears any existing graphic modes, screens, etc. and allows a program to commence graphic operations from scratch. HEADER -- Format a diskette HEADER "diskname" [,Iid] [,Ddrive] [Udevice] The HEADER command prepares a new diskette for use, sometimes called FORMATting a diskette. There are two types of "newing" a diskette -- a long form and a quick (or short) form. You must use the long form when preparing a new diskette for its first use. Thereafter you can use the quick form. WARNING: Formatting a diskette (long or short) will destroy all existing data on the diskette! In direct mode, you are asked to confirm what you are doing with 'ARE YOU SURE?'. Type 'Y' and press return to proceed, or TYPE ANY OTHER CHARACTER AND PRESS RETURN TO CANCEL the command. In program mode there is no confirmation prompt. The long HEADER form requires a diskname and an ID. The diskette will be completely (re)sectored, zeros written to all blocks, and a new system track (directory, BAM, etc.) will be created. HEADER "newdisk",I01 prepares a new diskette The short HEADER form is performed when the ID option is omitted. The diskette is assumed to have been previously formatted, and only a new system track (directory, BAM, etc.) is installed. This is roughly equivalent to deleteing all the files, but much quicker. HEADER "makelikenew" re-news an working diskette The diskname is limited to 16 characters and the ID string to two characters. The same rules apply for the diskname as for a filename. Some Disk Systems use the ID string to tell if you have swapped a diskette in a drive, so it's recommended that the ID string be unique for each of your diskettes. Some more examples: HEADER "QUICK" HEADER "MYDISK", I23 HEADER "RECS", I"FB", U9 HEADER (FILES), I(ID$), U(UNIT) HELP -- Show the BASIC line that cause the last error HELP The HELP command is used after an error has been reported in a program. When HELP is typed, the line where the error occurred listed, with the portion containing the error highlighted. Print ERR$(ER) for the error message, and print EN or EL for the error number and error line, respectively. HELP can be used in direct mode or in program mode. Note that, in the case of many I/O errors, there is no associated BASIC error. Check ST or DS$ errors in these cases. HEX$ -- Hexadecimal value function HEX$ (decimal_expression) This function returns a 4-character string that represents the hexadecimal value of the numeric decimal expression. The expression must be in the range (0-65535, $0000-$FFFF hex) or an 'ILLEGAL QUANTITY' error is reported. PRINT HEX$(10) The string "000A" is printed. PRINT RIGHT$(HEX$(10),2) The string "0A" is printed. HIGHLIGHT -- Set the text highlight color of the display HIGHLIGHT color Sets the highlight color to the given color index. The color value must be in the range (0-15). See the Color Table. COLOR must be ON (see the COLOR command). The highlight color is used in HELP messages and FIND/CHANGE strings. IF/THEN/GOTO/ELSE -- Conditional program execution IF expression [:ELSE else_clause] IF...THEN lets the computer analyze a BASIC expression preceded by IF and take one of two possible courses of action. If the expression is true, the statement following THEN is executed. This expression can be any BASIC statement. If the expression is false, the program goes directly to the next line, unless an ELSE clause is present. The ELSE clause, if present, must be in the same line as the IF-THEN part. When an ELSE clause is present, it is executed when the THEN clause isn't executed. In other words, the ELSE clause executes when the expression is FALSE. See BEGIN/BEND to spread the IF statement out over several lines. An ELSE statement is matched to the closest THEN statement in the case of nested IF/THEN statements. The expression being evaluated may be a variable or formula, in which case it is considered true if nonzero, and false if zero. Usually expressions involve relational operators =, <, >, <=, >=, <>. 50 IF X>0 THEN PRINT "X>0": ELSE PRINT "X<=0" If X is greater than 0, the THEN clause is executed, and the ELSE clause isn't. If X is less than or equal to 0, the ELSE clause is executed and the THEN clause isn't. INPUT -- Get input from the keyboard [LINE] INPUT ["prompt"<,l;>] variable_list The INPUT statement pauses the BASIC program, prints the prompt string if present, prints a question mark and a space, and waits for data to be typed by the user, terminated by a return character. If the prompt string ends with a comma instead of a semicolon, a question mark and space is not printed. Input is gathered and assigned to variables in the variable_list. The type of variable must match the type of input typed or a 'TYPE MISMATCH' error is reported. Separate data items typed by the user must be separated with commas. String data with imbedded spaces or commas must be surrounded with quotes. If insufficent data to satisfy the variable-list is typed, two question marks are displayed by the computer to prompt for additional data to be input. If the computer does not understand the input (such as the user typing cursor up or down keys) the computer responds with the message 'REDO FROM START?' and waits for acceptable data to be entered. Input is limited to 160 characters (two screen lines in 80-column mode), which is the size of the input buffer. The INPUT statement can only be executed from within a program. LINE INPUT allows the program to input a string which includes any PETSCII character (including colons, commas, imbedded spaces, etc.) up to but not including a null or return character. There should be only one string-type variable name in the variable_list in this case, but if there are more the computer prompts as usual with two question marks for more data to assign to the additional variables. 10 INPUT "WHAT'S YOUR FIRST NAME AND AGE"; NA$,A 20 PRINT "YOUR NAME IS ";NAS;" AND YOU ARE";A;" YEARS OLD" The above INPUT is the traditional BASIC form. 10 LINE INPUT "WHAT'S YOUR ADDRESS"; AD$ 20 PRINT "YOUR ADDRESS IS: ";AD$ The above INPUT allows an entire line of data to be assigned to a string variable, including commas and other common punctuation marks. 10 INPUT "ENTER YOUR NAME HERE: ", NA$ The above INPUT suppresses the traditional '? ' prompt by using a comma instead of a semicolon after the prompt string. To suppress the '?' without a prompt string, make the prompt string null. INPUT# -- Input data from an I/O channel (file) [LINE] INPUT#logical_channel_number, variable_list The INPUT# command works like the INPUT command, except no prompt string is allowed and input is gathered from a previously OPENed channel or file. This command can only be used in a program. The logical channel number is the number assigned to the device (file) in an OPEN (or DOPEN) statement. Items in the variable list must agree with the type of data input, or a 'FILE DATA ERROR' will resuit. On the C64DX, an End Of File (EOF) condition or bad I/O status will terminate input, as if a return character was received. It's good practice to examine the I/O status byte (and the DS disk status for file I/O) after every I/O instruction to check for problems or errors. 10 DOPEN#1, "FILE" This program will 20 C=0 count the number of 30 DO: LINE INPUT#1,AS: C=C+1: LOOP UNTIL ST lines in FILE 40 DCLOSE#1 50 PRINT"FILE CONTAINS";C;" LINES." INSTR -- Get the location of one string inside another string INSTR (string_1, string_2 [,starting_position]) This function searches for the first occurrence of string_2 in string_1 and returns its location. A value of zero (0) is returned if no match is found, if either string is null (empty), or if string_2 is longer than string_1. If the starting position is given, the search begins at that location, otherwise the search begins at the first character of string_1. The strings can be literals, variables, or string expressions. X = INSTR("123456","4") Result is X= 0 X = INSTR("123456","X") Result is X= -1 X = INSTR("123123","2") Result is X= 123 X = INSTR("123123","2",3) Result is X=-124 INT -- Greatest integer function INT (expression) This function returns the greatest integer less than or equal to the numeric expression. JOY -- Joystick function JOY (port) This function returns the state of a joystick controller in the specified port. When port=l returns position of joystick 1 When port=2 returns position of joystick 2 The value returned is encoded as follows: Fire = 128 + 1 8 2 7 0 3 6 4 5 A value of zero (0) means that the joystick is not being manipulated. A value of 128 or more means that the fire button is being pressed. The possible vales returned are: 0 No activity 128 fire 1 up 129 fire + up 2 up + right 130 fire + up + right 3 right 131 fire + right 4 right + down 132 fire + right + down 5 down 133 fire + down 6 down + left 134 fire + down + left 7 left 135 fire + left 8 left + up 136 fire + left + up KEY -- Enable, disable, display, or define function keys KEY ON KEY OFF KEY [key#, string] There are 14 function keys available on the C64DX (seven unshifted and seven shifted). The user can assign a string consisting of BASIC commands, control codes, escape functions, or a combination of each to function key. The data assigned to a key is typed out when that key is pressed, just as if the characters were typed one by one on the keyboard. The user can enable ("turn on") or disable ("turn off") the function keys. When they are disabled, pressing a function key return that key's normal character code instead of the string assigned to it. This includes the HELP and (shifted) RUN keys. It is also possible to redefine the HELP and (shifted) RUN keys, as function keys 15 and 16, respectively. The system has default assignments for all function keys. KEY with no parameters displays a listing of the current assignments for all the function keys. The maximum length for all the definitions together is 240 characters. If an assignment would be too big to fit, an 'OUT OF MEMORY' error is reported and the assignment is not made. KEY 2, "DIR U9"+CHR$(13) This causes the computer to display the directory from disk unit #9 when function key 2 is pressed. This is equivalent to typing 'DIR U9' and pressing the key directly. The CHR$(13) is the character for . Other often used control codes are CHR$(141) for 'shifted RETURN', CHR$(27) for 'ESCape', and CHR$(34) to incorporate a double quote into a KEY string. KEY 2, "DIR"+CHR$(34)+"*=P"+CHR$(34)+CHR$(13) This is equivalent to typing DIR"*=P" and pressing at the keyboard. Note the way quotes can be incorporated into an assignment. When function key 2 is pressed, a directory of all program files on the default system disk will be displayed. KEY OFF This turns off function key strings. Pressing a function key now would return the character codes associated with F-keys as on the VIC-20 and C64 computers. KEY ON would re-enable function key strings, unchanged from their previous assignments. To restore the system default assignments, reset the computer. LEFT$ -- Get the leftmost characters of a string LEFT$ (string,count) This function returns a string containing the leftmost 'count' number of characters of the string expression. Count is an numeric expression in the range (0-255). If count is greater than the length of the string, the entire string will be returned. If count is zero, a null (empty) string will be returned. A$ = LEFT$("123ABC",3) Result is A$="123" LEN -- Get the length of a string LEN (string) This function returns the number of characters in a string expression. Nonprinting characters and blanks are counted. A = LEN("ABC") Result is A=3 LET -- Assign a value to a variable [LET] variable = expression The LET command is optional, since the equal sign by itself is understood by the computer to mean assignment. Multiple assignments on LET statements are not allowed. 10 LET A=1: LET B=A+1: LET C$=" THREE" 20 : D=1: E=D+1: F$=" THREE" 30 PRINT A;B;C$ 40 PRINT D;E;F$ Output: 1 2 THREE 1 2 THREE LINE -- Draw a line on a graphic screen LINE x0, y0, x1, y1 LINE draws a line on the currently defined graphic screen with the currently defined draw modes. The line is draw from (x0,y0) to (x1,y1). LIST -- List a BASIC program from memory or disk LIST [startline] [- [endline] ] LIST "filename" [,Ddrive] [<,|ON>Udevice] LIST is used to view part or all of a BASIC program in memory or all of a BASIC program on disk (without affecting the program that is currently in memory). The display can be slowed down by holding down the key or it can be paused by pressing the key or . A listing that is paused can be restarted by pressing again or by pressing . The display can be stopped by pressing . If the word LIST is followed by a line number, the computer shows only that line number. If LIST is typed with two numbers separated by a dash, the computer shows all lines from the first to the second line number. If LIST is typed followed by a number and just a dash, it shows all lines from that number to the end of the program. And if LIST is typed, a dash, and then a number, all lines from the beginning of the program to that line number are LISTed. By using these variations, any portion of a program can be examined or easily brought to the screen for modification. LIST can be used in direct mode or in a BASIC program. LIST Shows entire program. LIST 100- Shows from line 100 until the end of the program. LIST 10 Shows only line 10. LIST -100 Shows lines from the beginning until line 100. LIST 10-200 Shows lines from 10 to 200, inclusive. LOAD -- Load a program or data into memory from disk LOAD "filename" [,device_number [,relocate_flag]] This command loads a file into the computer's memory. The filename must be given, and pattern matching may be used. In the case of dual drive systems, the drive number must be part of the filename. If a device number is given, the file is sought on that unit, which must be a disk drive. If a device number is not given, the default system drive is used. See also DLOAD and RUN commands. The relocate_flag is used to LOAD binary files. If the relocate_flag is present and non-zero, the file will be copied into memory starting at the address stored on disk when the file was SAVEd. See BLOAD. Do not use the relocate_flag to load BASIC programs: they will be automatically relocated to the start of the BASIC program area and relinked. To compare a program in memory to a disk file, use the VERIFY or DVERIFY command. To compare a binary file, use BVERIFY. See the discussion at DLOAD regarding CHAINING programs. LOAD "PROG" Loads BASIC program PROG from the system drive. LOAD FILE$,DRV Loads a program whose name is in the variable called F$ from the unit whose number is in DRV. LOAD "0:PROG" 8 Loads BASIC program PROG from unit 8, drive 0. LOAD "BIN",8,1 Loads a binary file into memory. LOCATE -- [*** NOT YET IMPLEMENTED ***] LOG -- Get the natural logarithm of a number LOG (number) This function returns the natural logarithm of a numeric expression. A natural log is a log to the base e (2.71828183). See the EXP function. To convert to log base 10, divide by LOG(10). A = LOG(123) Result is A=4.81218436 A = LOG(123) / LOG(10) Result is A=2.08990511 LOOP -- See DO/LOOP/WHILE/UNTIL/EXIT LPEN -- Get the position of a lightpen PEN (position) This function returns the current position of a lightpen on the screen. When position=0, the X position is returned, and when position=1 the Y position is returned. Note that lightpen coordinates, like sprite coordinates, are offset from the normal graphic coordinate map. This means you have to calculate where the lightpen is with respect to the screen display. The electronics of each lightpen also introduces a skew which must be factored into your calculations. The X resolution is limited to every 2 pixels, and will always be an even number in the approximate range (60-320). The Y position is in the approximate range (50-250). If either the X or the Y position is zero, the lightpen is off-screen. Note that a lightpen COLLISION need not be enabled to use LPEN. A bright background color, such as white, is usually required to stimulate the light pen. Lightpens only work in game port 1. 10 TRAP 40 We're done if STOP key 15 BACKGROUND 1 Make background color white 16 FOREGROUND 0 Make text color black 20 COLLISION 3,100 Enable lightpen interrupt 30 DO:LOOP Hang here until done 40 END Done 100 COLLISION 3 Got one, don't want more 110 PRINT LPEN(0),LPEN(1) Display lightpen position 120 COLLISION 3,100 Re-enable interrupt 130 RETURN MID$ -- Substring function MID$ (string, position [,length]) This function can appear on the left or the right side of an assignment statement: Case 1: string_var = MID$ (string_expression, position [,length]) This form returns a piece of another string. The function returns a string of the specified length taken from the string_expression beginning at the indicated position. The position must be in the range (1-255), one (1) being the first character. The length can be any number in the range (0-255), or it can be omitted. If the position specified is greater than the number of characters in the string_expression, a null (empty) string is returned. If the length is greater than the number of characters from the given position to the end of the string_expression, or the length is omitted, then all the rightmost characters beginning at the position are returned. A$ = MID$("TICTACTOE",4,3) Result is A$="TAC" A$ = MID$("TICTACTOE",4) Result is A$="TACTOE" A$ = MID$("TICTACTOE",10,1) Result is AS="" (empty) Case 2: MID$ (string_var, position [,length]) = string_expression This form replaces a portion of the string contained in string_var with data from another string_expression, beginning at the specified position in the string_var. If the length is given only, that many characters from the string_expression are taken, otherwise all the characters in the string_expression will replace characters in the string_var beginning at the position specified. The there are too many characters to fit in the string_var, an 'ILLEGAL QUANTITY' error is reported. If the length given is zero, no characters will be replaced. A$="TICTACTOE": MID$(A$,4,3)="123456" Result is A$="TIC123TOE" A$="TICTACTOE": MID$(A$,4) ="123456" Result is A$="TIC123456" A$="TICTACTOE": MID$(A$,5) ="123456" Result is 'ILLEGAL QUANTITY' MONITOR -- Enter the built-in machine language monitor SEE SECTION 3.2 ON THE C64DX MONITOR. MOUSE -- Enable or disable the mouse driver MOUSE ON [,port [,sprite [,position] ] ] MOUSE OFF port = joyport 1, 2, or either (both) (1-3) sprite = sprite pointer (0-7) position = initial pointer location (x,y) normal, relative, or angular coordinate defaults to sprite 0, port 2 ???? add min/max x/y positions [*** THIS COMMAND IS SUBJECT TO CHANGE ***] Mouse ON enables the built-in mouse driver. The user must load a pointer into the proper sprite area ($600-$7FF). The driver assumes the "hot point" is the top left corner of the sprite, and does not allow this point to leave the screen. Mouse OFF will turn off the driver and the currently associated sprite. Use the RMOUSE function to get the current pointer position and button status. See the sample program at RMOUSE. MOVSPR -- Position sprite or set sprite in motion MOVSPR sprite <,x,y> Use the SPRITE command to turn on a sprite and MOVSPR to position it. Sprites are numbered 0-7. The sprite's position can be specified using one of the following coordinate types: [+/-]x, [+/-]y = [relative] position x#y = angle and speed x;y = distance and angle Angles are specified as 0-360 degrees, with 0 being straight up. Speeds are specified as a number of pixels per frame, 0-255. Sprites are moved through each pixel so that collisions are accurately detected. NEW -- Delete program in memory and clear all variables NEW [RESTORE] This command erases the entire program in memory and clears all variables and open channels (but it does NOT properly close open disk write files -- used DCLOSE or DCLEAR beforehand). NEW also resets the runtime stack pointer (clears GOSUB & FOR/NEXT stacks), the DATA pointer, and the PRINTUSING characters. The BASIC program in memory is lost unless it was previously SAVEd to disk. If you have not entered or loaded any BASIC programs since typing NEW, the RESTORE option will recover the BASIC program in memory. But if the BASIC environment has been changed in any way, the program may not be restored correctly. If BASIC can tell something's wrong, it will report 'PROGRAM MANGLED'. NEW can be used in direct (edit) mode or in a program. When it's encountered in a program, the program terminates. NEXT -- See FOR/NEXT/STEP and RESUME NOT -- Get the complement of a number NOT (expression) The NOT function returns the complement of an integer in the range (-32768 to 32767). The function operates on the binary value of signed 16-bit integers. An expression outside of this range will cause an 'ILLEGAL QUAUTITY' error. X = NOT(5) Result is X=-6 X = NOT(-6) Result is X=5 NOT is often used in logical comparisons (such as an IF statement) to invert the result, since -1 (true) is the result of NOT(0) (false), and 0 (false) is the result of NOT(-1) (true). X = NOT("ABC"="ABC") AND ("DEF"="DEF") Result is X= 0 (false) X = NOT("ABC"="ABC") AND ("DEF"="XYZ") Result is X=-1 (true) OFF -- Subcommand used with various BASIC commands. ON -- Computed GOTO/GOSUB ON expression line_number_list This is a variation of the IF GOTO statement that branches to one of several line numbers based upon the value of an expression. The integer value of the evaluated expression determines which line number in the line_number_list gets control. If the expression evaluates to one, the first line number in the list gets control, if it's two the second line number gets control, and so on. Fractional parts of the value are truncated (for example, 2.9 becomes 2). If the value is zero or greater than the number of items in the list the computer takes none of the branches and continues on with the next statement. If the value is negative, an 'ILLEGAL QUANTITY ERROR' is reported. The ON/GOSUB statement must call the first line number of a subroutine and the subroutine must end with a RETURN statement. After executing the subroutine, control is returned to the statement following the ON/GOSUB statement. 10 INPUT"ENTER A NUMBER 1-3: ",X 20 ON X GOTO 100, 200, 300 30 PRINT"TOO LOW OR TOO HIGH": RUN 100 PRINT"ONE": RUN 200 PRINT"TWO": RUN 300 PRINT"THREE": RUN OPEN -- Open a channel to a device or disk file OPEN logical_chnl_num, device_number [,secondary_adr [,]] Before a program can access a device or a file, an I/O channel must be opened to it to communicate through. When something is opened, you associate a logical channel number with it, and it is with this number that all other I/O statements access the device or file. The OPEN command can be used in direct (edit) mode or in a program. The channel number, device number, and optional secondary address are integers from 0-255. Refer to the device's manual for more information about what (if any) secondary addresses it uses. channel: 0-127 return = output return character only 128-255 return = output return + linefeed device: 0 Keyboard 1 Default system drive whatever its number is (see SET DEF) 2 RS232 3 Screen 4-7 Serial bus (usually reserved for printers) 8-31 Serial bus (usually reserved for disk drives) The filespec is the file name in the case of disk files (refer to your DOS manual for details). Typically, the filename is a string having the the following form: [[@|S]drive:] filename [,type] [,mode] An example would be 0:MYFILE,SEQ,READ to open the sequential file MYFILE for reading on drive 0. Disk drives usually support some kind of filename pattern matching. Most disk drives support the following file types and modes (can be abbreviated to first character): types: 'S'equential 'P'rogram 'R'elative 'U'ser modes: 'R'ead 'W'rite 'L'ength (for relative type files) Some channels or devices accept a command string instead of a filename when they are opened. An example would be the disk command channel or the RS232 open/setup command. Refer to the device's documentation. OPEN 1,8,15,"I" Open CBM disk command channel & send it the 'I'nitialize command. OPEN 4,4,7 Open CBM printer channel in upper/lower case mode. OPEN 128,2,2,CHR$(14) Open a 9600 8N1 RS232 channel and translate CR into CRLF on output. See also DOPEN, DCLOSE, CLOSE, CMD, GET#, INPUT#, and PRINT# statements and I/O status variables ST, DS, and DS$. OR -- Boolean operator expression OR expression The OR operator returns a numeric value equal to the logical OR of two numeric expressions, operating on the binary value of signed 16-bit integers in the range (-32768 to 32767). Numbers outside this range result in an 'ILLEGAL QUANTITY' error. X = 4 OR 8 Result is X=12 In the case of logical comparisons, the numeric value of a true situation is -1 (equivalent to 65535 or $FFFF hex) and the numeric value of a false situation is zero. X = ("ABC"="ABC") OR ("DEF"="DEF") Result is X=-l (true) X = ("ABC"="ABC") OR ("DEF"="XYZ") Result is X=-1 (true) X = ("ABC"="XYZ") OR ("DEF"="XYZ") Result is X= 0 (false) PAINT -- Fill a graphics area with color PAINT x,y, mode [,color] x,y coordinate to begin fill at mode 0: fill area to edge = color 1: fill area to edge=same as color at x,y PAINT fills an enclosed graphic area starting at the given coordinate with the color of the currently defined PEN. The mode parameter identifies the region to be filled. [*** THIS COMMAND IS NOT YET IMPLEMENTED ***] PALETTE -- Define a color PALETTE [screen#|COLOR], color#, red, green, blue PALETTE RESTORE screen# 0-1 color# 0-255 red 0-15 green 0-15 blue 0-15 The PALETTE command can be used to define a color for a logical graphic screen, set an absolute color, or restore the C64DX VIC-III default colors. PALETTE can be used in direct mode or in a program. The VIC-III pre-defines the first 16 colors to the usual C64-type colors, but you can change them with the PALETTE COLOR command or restore them all with the PALETTE RESTORE command. See the sample program after the SCREEN command. PASTE -- Put a CUT graphic area on the screen PASTE x,y [*** NOT YET IMPLEMENTED ***] PEEK -- Function returning the contents of a memory location PEEK (address) This function returns the contents of a memory location. The address must be an integer in the range of 0-65535 ($0-$FFFF) and the value returned will be an integer in the range of 0-255 ($0-$FF). Use the BANK command to specify which 64K memory bank the address is in. Note that a BANK number greater than 127 (i.e., a bank number with the most significant bit set) must be used to address an I/O location, such as the VIC chip or color memory. Refer to the system memory map for details. PEEK uses the DMA device to access memory. Use the POKE command to change the contents of a memory location. BANK 0: X = PEEK (208) Reads the keyboard buffer index. If it's empty, X will be zero, otherwise X will be the number of characters in it. PEN -- Specify a pen color for drawing on graphic screen PEN pen, color pen 0-2 color 0-255 Before you can draw anything on a graphic screen, you have to tell BASIC what color your PENs are. You should first define what your colors are using the PALETTE command, then use PEN to associate those colors with a PEN. Whatever graphic commands you use after a PEN command will use the PEN you specified. PEN 0,1 Put color 1 "ink" into draw pen 0 See the sample program after the SCREEN command. PIC -- Graphic picture subcommand PLAY -- Play a musical string PLAY "[Vn,On,Tn,Un,Xn,elements]" [*** WILL CHANGE TO ADD 2ND SID SUPPORT ***] The PLAY command lets you select a voice, octave, instrument, volume, filter, and musical notes. All these parameters are packed into a string (spaces are allowed for readability). On = Octave (n=0-6) Tn = Tune envelope # (n=0-9) 0= piano (defaults) 1= accordion 2= calliope 3= drum 4= flute 5= guitar 6= harpsichord 7= organ 8= trumpet 9= xylophone Un = Volume (n=0-9) Vn = Voice (n=1-3) Xn = filter on (n=1), off (n=0) Elements: A,B,C,D,E,F,G ... Notes, may be preceded by: # ................. Sharp S ................. Flat . ................. Dotted W ................. Whole note H ................. Half note Q ................. Quarter note I ................. Eighth note S ................. Sixteenth note R ................. Rest M ................. Wait for all voices playing to end (a measure) Once the music string starts PLAYing, the computer will continue with the next statement. The music will continue to play automatically. Using the 'M'easure command will cause the computer to wait until the music has up to that point has been played out. Use the TEMPO command to alter the tempo (speed) of PLAY. Note that the VOLume command can change a PLAY string's volume setting. POINTER -- Get the address of a variable descriptor POINTER (variable_name) This function returns the address of an entry in the variable table. If the value returned is zero, the variable is currently undefined. The variable table is normally in the second RAM bank (BANK 1). See the section on variable storage for details. Note that, while the location of a string descriptor will not change, the location of the actual string in memory changes all the time. Also, when working with an array name you must specify a particular element, to which POINTER will return a pointer to that element's descriptor and not to the array descriptor. 10 A$="FRED" Define A$ 20 DESC=POINTER (A$) Lookup A$ in variable table 30 BANK1: PRINT PEEK(DESC) Displays the length of A$ PORE -- Write a byte to memory location POKE address, byte [,byte ...] POKE is used to write one or more bytes into one or more memory locations. The address must be an integer in the range of 0-65535 ($0-SFFFF) and the value to be written must be an integer in the range of 0-255 ($0-$FF). If more than one byte is given, it will be written into successive memory locations. Use the BANK command to specify which 64K memory bank the address is in. Note that a BANK number greater than 127 (i.e., a bank number with the most significant bit set) must be used to address an I/O location, such as the VIC chip or color memory. Refer to the system memory map for details. Also note that, unlike previous CBM computers, POKEs to a ROM location will not "bleed through" into a corresponding RAM location. POKE uses the DMA device to access memory. Use the PEEK function to read a byte from a memory location. Because this command directly accesses system memory, extreme care should be taken in its use. Altering the wrong memory location can crash the computer (press the reset button to reboot). BANK 0: POKE 208,0 Resets location 208 ($000D0), clearing the keyboard buffer. BANK 128: POKE DEC("D023"),1,2,3 Sets the VIC extended background colors to 1, 2, and 3 respectively POLYGON -- Draw a regular n-sided figure on a graphic screen POLYGON x,y, xradius,yradius, [solid], angle,drawsides,sides,subtend x,y = center of polygon x,yradius = radii of polygon solid = solid flag (0-1) angle = starting angle (0-360) drawsides = # of sides to draw (3-127) sides = # sides of polygon (drawsides<=sides) subtend = subtend flag (0-1) POS -- Get the column number of the cursor POS(0) This function returns the current text column the cursor is in, with respect to the currently defined window (see RWINDOW). It's usually used to format text printed to the screen. The argument (0) is not used for anything. POS will not work as expected if text output is redirected to a disk file or the printer. 10 MAXCOL = RWINDOW(l) 20 FOR ADR=DEC("600") TO DEC("7FF") 30 PRINT HEX$(PEEK(ADR));" "; 40 IF POS(0) > (MAXC0L-5) THEN PRINT 50 NEXT This example illustrates one way to format output to the screen, keeping the last item on a line from being split between two lines, regardless of the window size (as long as the window size is at least 4 characters wide). It dumps the data for the first sprite in hex. POT -- Paddle function POT (paddle) This function returns the state of a game paddle (POTentiometer) controller in one of the two game ports. paddle=1 ..... Position of paddle #1 (port 1, paddle "A") paddle=2 ..... Position of paddle #2 (port 1, paddle "B") paddle=3 ..... Position of paddle #3 (port 2, paddle "A") paddle=4 ..... Position of paddle #4 (port 2, paddle "B") The value returned by POT ranges from 0 to 255. Any value greater than 255 means that the fire button is also pressed. Paddles are read "backwards" from normal things like volume knobs or faucets. A value of 255 means the paddle has been turned counterclockwise as far as it will go ("off"), and a value of 0 means the paddle has been turned clockwise as far as it will go "on"). Note that some paddles are "noisy" and their output must be averaged or "damped" to prevent whatever they are controlling from jittering. 10 SPRITE 1,1 Turn on a sprite 20 DO Begin a loop 30 X=POT(3) Read paddle "A" in port 2 40 MOVSPR 1,300-(X AND 254),200 Move the sprite 50 LOOP UNTIL X>255 Loop until button pressed 60 SPRITE 1,0 Turn off sprite This sample program turns on a sprite and lets you move it horizontally with a paddle. If you press the paddle's fire button, it turns off the sprite and the program ends. The calculations in line 40 do several things all at once -- they mask the fire button and "damp" the output to reduce jitter by masking the least significant bit (the X AND 254 part) and invert the output so that turning the paddle to the right makes the sprite go right (subtracting result from 300). PRINT -- Display data on text screen PRINT [expression_list] [<,|;>] PRINT will evaluate each item in the expression_list and pass the results to the system screen editor to display on the screen. If a screen window is defined, the output will be confined to the window. PRINT can be used to send control codes and escape sequences to the screen editor to do such things as set windows, change TAB stops, change text colors or set reverse field, or choose cursor styles. See the section on Editor modes for details. PRINT can be followed by any of the following: Numeric or string expressions 12, "HELLO", 1+1, "S"+STR$(I) Variable names A, B, A$, X$ Functions ABS(33), HEX$(160) Punctuation marks ;, Nothing Numeric values are always followed by a space. Positive numbers are preceded by a space, and negative numbers are preceded by a minus sign ('-'). Scientific notation is used when a number is less than 0.01 or greater than or equal to 999999999.2 . A semicolon (';') or space between list items causes the next item to be printed immediately following the previous item. A comma (',') causes the next item to be printed at the next comma stop (similar to TAB stops, but every 10 spaces). These rules apply to the next print statement, if the expression_list ends with either a semicolon or a comma, otherwise a return is printed. Note that floating point variable names should not be separated from the next variable name with a space, and constants should not be preceeded or followed by a space. For formatted PRINT output, see the PRINT USING command. PRINT "HELLO" HELLO A$="THERE": PRINT "HELLO ";A$ HELLO THERE A=4:B=2: PRINT A+B 6 J=41: PRINT J;: PRINT J-1 41 40 C=A+B:D=B-A: PRINT A;B;C;D 4 2 6 -2 C=A+B:D=B-A: PRINT A,B,C,D 4 2 6 -2 A=1:B=2:AB=3: PRINT A B 3 PRINT 1 2 3, 1 2 3 +1 123 124 PRINT 0.009, 0.01 9E-03 .01 PRINT 999999999; 999999999.2 999999999 1E+09 The CMD command can be used to redirect PRINT output to a device or file. Also see the POS, SPC, TAB functions, CHAR and PRINT USING. PRINT# -- Send data to an I/O channel (file) PRINT#logical_channel_number [,expression_list] [<,|;>] This command is used to send (transmit) data to a device or file. The logical_channel number is the number assigned to the device (file) in an OPEN (or DOPEN) statement. The output is otherwise identical to that of a PRINT statement, including the comma and semicolon conventions. Note that certain screen-oriented functions, such as TAB and SPC do not have the same effect as they do with screen I/O. It's good practice to examine the I/O status byte (and the DS disk status for file I/O) after every I/O instruction to check for problems or errors. For formatted output, use the PRINT# USING command. 10 OPEN 1,8,15 Initialize disk drive 20 PRINT#1,"I" (same as DCLEAR) 30 CLOSE 1 10 DOPEN#1,"NEWFILE",W Create a SEQ file 20 FOR I=1 TO 10 30 PRINT#1, I, STR$(I) Write numbers 1-10 to it 40 NEXT 50 DCLOSE#1 10 OPEN 2,2,2,CHR$(12) Open 1200 baud RS232 channel 20 PRINT#2, "ATDT,5551212" Send modem a Hayes dial command PRINT USING -- Output formatted data to the screen, device, or file PRINT [#logical_channel_number,] USING format; expression_list [<,|:>] Read about the PRINT and PRINT# commands first for information regarding the syntax of the expression list and, for device output, establishing the logical_channel_number. The items in the expression list must be separated by commas (','). The format is defined in a string literal or string variable and is described below. See the PUDEF command for specifing special formatting characters. The various formatting characters are: CHARACTER SYMBOL NUMERIC STRING ---------------- ------ ------- ------ Pound sign # X X Plus sign + X Minus sign - X Decimal Point . X Comma , X Dollar Sign $ X Four Carets ^^^^ X Equal Sign = X Greater Than Sign > X The pound sign ('#') reserves room for a single character in the output field. If the data item contains more characters than the number of pound signs in the format field, the entire field will be filled with asterisks ('*'). 10 PRINT USING "####";X For these values of X, this format displays: A = 12.34 12 A = 567.89 568 A = 123456 **** For a STRING item, the string data is truncated at the bounds of the field. Only as many characters are printed as there are pound signs in the format item. Truncation occurs on the right. The plus ('+') and minus ('-') signs can be used in either the first or last position of a format field but not both. The plus sign is printed if the number is positive. The minus sign is printed if the number is negative. If a minus sign is used and the number is positive, a blank is printed in the character position indicated by the minus sign. If neither a plus sign nor a minus sign is used in the format field for a numeric data item, a minus sign is printed before the first digit or dollar symbol if the number is negative and no sign is printed if the number is positive. This means that one more character is printed if the number is positive. If there are too many digits to fit into the field specified by the pound sign and +/- signs, then an overflow occurs and the field is filled with asterisks ('*'). A decimal point ('.') symbol designates the position of the decimal point in the number. There can be only one decimal point in any format field. If a decimal point is not specified in the format field, the number is rounded to the nearest integer and printed without any decimal places. When a decimal point is specified, the number of digits preceding the decimal point (including the minus sign, if the number is negative) must not exceed the number of pound signs before the decimal point. If there are too many digits an overflow occurs and the field is filled with asterisks ('*'). A comma (',') allows placing of commas in numeric fields. The position of the comma in the format list indicates where the commas appears in a printed number. Only commas within a number are printed. Unused commas to the left of the first digit appear as the filler character. At least one pound sign must precede the first comma in a field. If commas are specified in a field and the number is negative, then a minus sign is printed as the first character even if the character position is specified as a comma. FIELD EXPRESSION RESULT COMMENT ------ ----------- ------ ----------------------------- ##.# -.1 -0.1 Leading zero added ##.# 1 1.0 Trailing zero added #### -100.5 -101 Rounded to no decimal places ###. 10 10. Decimal point added #$## 1 $1 Leading dollar sign #### -1000 **** Overflow because 4 digits and minus sign don't fit in field A dollar sign ('$') symbol shows that a dollar sign will be printed in the number. If the dollar sign is to float (always be placed before the number), specify at least one pound sign before the dollar sign. If a dollar sign is specified without a leading pound sign, the dollar sign is printed in the position shown in the format field. If commas and/or a plus or minus sign is specified in a format field with a dollar sign, the program prints a comma or sign before the dollar sign. The four up arrows or carets symbol is used to specify that the the number is to be printed in E format (scientific notation). A pound sign must be used in addition to the four up arrows to specify the field width. The arrows can appear either before or after the pound sign in the format field. Four carats must be specified when a number is to be printed in E format. If more than one but fewer than four carats are specified, a syntax error results. If more than four carats are specified only the first four are used. The fifth carat is interpreted as a no text symbol. An equal sign ('=') is used to center a string in a field. The field width is specified by the number of characters (pound sign and =) in the format field. If the string contains fewer characters than the field width, the string is centered in the field. If the string contains more characters that can be fit into the field, then the rightmost characters are truncated and the string fills the entire field. A greater than sign ('>') is used to right justify a string in a field. 5 X=32: Y=100.23: A$="TEST" 10 PRINT USING "$##.## ";13.25,X,Y 20 PRINT USING "###>#";"CBM",A$ When this is RUN, the following output appears on the screen: $13.25 $32.00 $***** CBM TEST $***** is printed instead of Y because Y has 5 digits, which exceeds the format specification. The second line asks for the strings to be right justified, which they are. PUDEF -- Redefine PRINT USING symbols PUDEF definition_string PUDEF allows redefinition of up to 4 symbols in the PRINT USING statement. Blanks, commas, decimal points, and dollar signs can be changed into some other character by placing the new character in the correct position in the PUDEF definition_string. Position 1 is the filler character. The default is a space character. Place another character here to be used instead of spaces. Similarly, Position 2 is the comma character. Default is a comma. Position 3 is the decimal point. Position 4 is the dollar sign. 10 PUDEF "*" PRINTs * in the place of blanks. 20 PUDEF " @" PRINTs @ in place of commas. QUIT -- [*** UNIMPLEMENTED ***] RCLR -- Get the current screen color RCLR(source) [*** CURRENTLY UNIMPLEMENTED ***] This function returns the color assigned to source as an number in the range of 0-15. The color sources are: 0 = background 1 = foreground 2 = multicolor 1 3 = multicolor 2 4 = border 5 = highlight color RDOT -- Get the current position or color of the pixel cursor RDOT(source) [*** CURRENTLY UNIMPLEMENTED ***] This function returns information about the current pixel location. 0 = current X position 1 = current Y position READ -- Read data from DATA statements READ variable_list READ statements are used along with DATA statements. READ statements read data from DATA statements into variables, just like an INPUT statement reads data typed by the user. READ statements can be used in direct or program mode, but DATA statements must be in a program. The variable types in the variable_list must match the type of DATA being read, or a 'TYPE MISMATCH' error is reported. If there are insufficient data in the program's DATA statements to satisfy all of the variables in the READ statement, an 'OUT OF DATA' error is reported. The computer maintains a pointer to the next DATA item to be read by a READ statement. Initially this pointer points to the beginning of the program. As each variable in a READ statement is filled, the computer moves the DATA pointer to the next DATA item. If all of a READ statement's variables are filled before all of the data has been read from a DATA statement, the next READ statement will begin reading data at the point where the previous READ stopped. The DATA pointer can be changed by the RESTORE command. It can be reset back to the beginning of the program, or pointed to a specific line number. See RESTORE. 10 DATA 100, 200, FRED, "HELLO, MOM", , 3.14, ABC123, -1.7E-9 20 READ X,Y 30 READ NAME$, MSG$, NULL$ 40 READ PI, JUNK$, S 50 RESTORE RECORD -- Specify a relative disk file record number RECORD #logical_channel_number, record [,byte] This command allows you to access any part of any record in a RELative type disk file. If the byte parameter is omitted, the access pointer is pointed at the first byte of the specified record number. Before you can use RECORD, you must OPEN a file. See OPEN and DOPEN for instructions. Also refer to your DOS manual for an explanation of RELative type files. 10 INPUT"ENTER RELATIVE FILENAME: ",F$ get name of existing file 20 DOPEN#1, (F$),L: PRINT DS$ open it & display disk status 30 R=1: INPUT"ENTER RECORD NUMBER: ",R get a record number 40 B=1: INPUT"ENTER BYTE (RETURN): ",B get byte number, if any 50 RECORD#1, R,B position file pointer 60 INPUT#1, REC$ read the record 70 PRINT REC$ display the record 80 PRINT "CONTINUE? (Y/N)" 90 GETKEY A$: IF A$="Y" THEN 30 100 DCLOSE#1 close the file REM -- Place an explanatory remark or comment in a program REM plain text message The REMark command is just a way to leave a note to whomever is reading a LISTing of the program. It might explain a section of the program, give information about the author, etc. REM statements in no way effect the operation of the program, except to add length to it (and therefore slow it down a little). No other executable statement can follow a REMark on the same line. 10 REM THIS PROGRAM WAS WRITTEN ON 2/14/91 BY F.BOWEN 20 REM SAMPLE PROGRAM 30 : 40 DIR :REM DISPLAY THE DISK DIRECTORY 50 LIST "SAMPLE PROGRAM" :REM DISPLAY THIS PROGRAM 60 END RENAME -- Rename a disk file RENAME "oldname" TO "newname" [,Ddrive] [Udevice] The RENAME command changes the name of a file in the disk directory. Pattern matching is not allowed, and "newname" must be a valid filename that does not already exist on the disk. The file being renamed does not need to be open. RENAME "TEST" TO "FINALTEST" RENAME (OLD$) TO (OLD$+".OLD") ON U(DEV) RENUMBER -- Renumber the lines of a BASIC program RENUMBER [new_starting_line [,[increment] [,old_starting_line]]] Renumber is used to resequence the line numbers of a BASIC program in memory. All or part of a program can be renumbered. The RENUMBER command first scans the program to make sure all the line numbers referenced in commands (such as GOTO, GOSUB, TRAP, etc.) exist, that new line numbers are in the legal range, and that changing the program would not overflow the available memory. An 'UNRESOLVED REFERENCE', 'LINE NUMBER TOO LARGE', or 'OUT OF MEMORY' error is reported if there's a problem, and RENUMBER is automatically canceled without having changed anything. If the program passes all the checks, RENUMBER changes the specified line numbers and updates all references to the old numbers throughout the program and relinks the program. The new_starting_line is the number of the first line in the program after renumbering. It defaults to 10. The increment is the spacing between line numbers (eg., 10, 20, 30 would mean an increment of 10). It also defaults to 10. The old_starting_line is the line number in the program where you want renumbering to begin. RENUMBER can be used in direct (edit) mode only. Note that line number zero (0) is a valid line number. RENUMBER Renumbers the entire program. After renumbering, the first line will be 10 the second 20, etc. through the end of the program. RENUMBER ,1 Renumbers the entire program as above, but in increments of one. The first line will be 10, the second 11, etc. RENUMBER 100, 5, 80 Starting at line 80, renumbers the program. Line 80 becomes line 100, and lines after that are numbered in increments of 5, through the end of the program. RENUMBER ,,65 Starting at line 65, renumbers lines in increments of 10, starting at line 10 through the rest of the program. RESTORE -- Position READ pointer at specific DATA statement RESTORE [line] The computer maintains a pointer to the next DATA item to be read by a READ statement. Initially this pointer points to the beginning of the program. The DATA pointer can be changed by the RESTORE command. Using RESTORE without specifying a line number will reset the DATA pointer back to the beginning of the program. If a line number is specified, the DATA pointer is pointed to that line. The line does not have to contain a DATA statement. When the computer executes the next READ statement, it will look for the next DATA item starting at the line the DATA pointer is at. See the READ command an example. RESUME - Resume program execution after error TRAP RESUME [line|NEXT] Used to return to execution after TRAPping an error. If a line number is given, the computer performs a 'GOTO line' and resumes execution at that line. RESUME NEXT resumes execution at the statement following the one that cause the error. RESUME without any parameters will resume execution at the statement that cause the error. If the computer encounters a RESUME statement outside of a TRAP routine or if a TRAP was not in effect a 'CAN'T RESUME' error is reported. RESUME can only be used in program mode. 10 TRAP 90 20 FOR I=-5 TO 5 30 PRINT 5/I 40 NEXT 50 END 60 : 90 PRINT ERR$(ER): RESUME NEXT RETURN -- Return from subroutine or event handler RETURN This statement is associated with the GOSUB (GO SUBroutine) statement. When a subroutine is called by a GOSUB statement, the computer remembers where it's at before it calls the subroutine. When the computer encounters a RETURN statement, it returns to the place it last encountered a GOSUB and continues with the next statement. If there wasn't a previous GOSUB, then a 'RETURN WITHOUT GOSUB' error is reported. RETURN is also used by event handlers, set up by the COLLISION command. See COLLISION. RGR -- Get the current graphic mode RGR(0) [*** CURRENTLY UNIMPLEMENTED ***] This function returns current graphic mode. A result of zero means the display is text, a non-zero result means it's graphic. RIGHT$ -- Get the rightmost characters of a string RIGHT$ (string,count) This function returns a string containing the rightmost 'count' number of characters of the string expression. Count is an numeric expression in the range (0-255). If count is greater than the length of the string, the entire string will be returned. If count is zero, a null (empty) string will be returned. A$ = RIGHT$("123ABC",3) Result is A$="ABC" RMOUSE -- Get the mouse position and button status RMOUSE [Xposition [,Yposition [,button]]] X,Yposition = current position of mouse pointer sprite Button = current status of mouse buttons 0 = no button 1 = right button 128 = left button 129 = both buttons RMOUSE is a command which retrieves a mouse's current position and the state of its buttons, and places this information into the specified numeric variables. If a mouse is not installed, "-1" is returned for all variables. If both ports are enabled, buttons from each port are merged. Use the MOUSE command to turn a mouse on or off. 10 MOUSE ON, 2, 1 Turn mouse on, port 2, sprite 1 20 DO Begin loop 30 RMOUSE X, Y, B Get mouse position & buttons 40 PRINTUSING"### ";X,Y,B Show " " " 50 LOOP UNTIL B=129 Loop until user presses both buttons 60 MOUSE OFF Turn mouse off RND -- Get a pseudo-random number RND (type) The RND function returns a pseudo RaNDom number between 0 and 1. The random sequence returned is determined by the type parameter: type = 0 Returns a random number based upon the system clock. type < 0 Negative numbers "seed" the random number generator, defining a new but reproducible random sequence. type > 0 Positive numbers draw the next random number from the sequence defined by the last "seed" value. This lets a programmer use a reproducible sequence while debugging (fixing) a program, so that random errors can be easily reproduced. Once the program has been fixed, it can be "seeded" such that a random sequence is used every time the program is run. 10 DO 20 INPUT "SEED"; S 30 IF S=0 THEN END 40 FOR I=1 TO S 50 PRINT INT(RND(1)*6)+1, INT(RND(1)*6)+1 60 NEXT 70 LOOP The above program will demonstrate the results of seeding the random number generator. It lets you specify a positive or negative seed value, and then prints the first S random pairs of that sequence. Enter a zero to end the program. The calculations in line 50 make the random numbers be integers from 1 to 6, like dice. Type in a negative dice from that sequence. Every time you enter "-1", for example, you will roll the same numbers: first roll 2 and 6 second 6 and 1 third 1 and 1 fourth 1 and 4 fifth 5 and 5 Games and statistical programs should use RND(0) for true randomness or seed the generator with a random number, such as RND(-TI). The general form for getting random integers using RND is: INT( RND(0) * MAX ) + 1 where MAX is the highest number you can get. This gives you numbers as low as 1 and as high as MAX. For dice, MAX is 6 (or 12 if you want to simulate rolling two dice at once). For cards, MAX is 52. INT( RND(0) * 16) This form will return integers from zero to 15, which is useful for generating random colour values, for example. RREG -- Get register data after a SYS call RREG [a_reg] [,[x_reg] [,[y_reg] [,[z_reg] [,status] ]]] Following a SYS call, the RREG command retrieves the contents of the microprocessor's registers and puts them into the specified numeric variables. See the sample program at SYS. RSPCOLOR -- Get multicolor sprite colors RSPCOLOR (multicolor#) Returns the current colors for multicolor sprites. Color values range from 0-15. Use RSPRITE function to get the foreground sprite color. multicolor# = 1 gets multicolor #1 multicolor# = 2 gets multicolor #2 See SPRITE and SPRCOLOR. RSPPOS -- Get the location and speed of a sprite RSPPOS (sprite,parameter) The RSPPOS function returns the current X or Y position of a sprite and its speed, set by the MOVSPR command. A sprite does not have to be on to use RSPPOS. The sprite number must be in the range of 0-7, and the parameter is: 0 to get current X position 1 to get current Y position 2 to get current speed (0-255) RSPRITE -- Get information about a sprite RSPRITE (sprite,parameter) The RSPRITE function returns the current state of a sprite, set by the SPRITE command. The sprite number must be in the range of 0-7, and the parameter is: 0 to see if it's turned on (1)=yes (0)=no 1 to get sprite foreground color (0-15) 2 to get priority over background (1)=yes (0)=no 3 to get X-expansion factor (1)=yes (0)=no 4 to get Y-expansion factor (1)=yes (0)=no 5 to get multicolor factor (1)=yes (0)=no RUN -- execute BASIC program RUN [line #] RUN "filename" [,Ddrive] [Udevice] RUN executes the BASIC program that is currently in memory. The program has to be LOADed (DLOAD) or manually typed in before it can be executed. If a line number is specified, execution begins at that line. If a filename is specified, the program is automatically loaded from disk into memory and executed. RUN can be used in a program. RUN clears all variables and open channels (but it does NOT properly close open disk write files -- used DCLOSE or DCLEAR beforehand). RUN also resets the runtime stack pointer (clears GOSUB & FOR/NEXT stacks) the DATA pointer, and the PRINT USING characters. To start a program without initializing everything, use GOTO. RUN Starts the program at the first line. RUN 100 Starts the program at line 100. RUN "TEST" Loads the program TEST from the, default system disk and starts the program at the first line. RWINDOW -- Get information about the current text window RWINDOW (parameter) This is a function that returns information about the current console text display. The parameter is specified as: 0 to get the maximum line # in the current window 1 to get the maximum column # in the current window 2 to get the screen size, either 40 or 80 columns SAVE -- Save a BASIC program in memory to disk SAVE "[[@]drive:]filename" [,device_number] This command copies a BASIC program in the computer's BASIC memory area into a PRoGram-type disk file. If the file already exists, the program is NOT stored and the error message 'FILE EXISTS' is reported. If the filename is preceded with an '@0:', then if the file exists it will be replaced by the program in memory. Because of some problems with the 'save-with-replace' option on older disk drives, using this option is not recommended if you do not know what disk drive is being used (DELETE the file before SAVEing). Pattern matching is not allowed. In the case of dual drive systems, the drive number must be part of the filename. Use the VERIFY or DVERIFY command to compare the program in memory with a program on disk. To save a binary program, use the BSAVE command. SAVE "myprogram" Creates the PRG-type file MYPROGRAM on the default system disk and copies the BASIC program in memory into it. SAVE "@0:myprogram" Replaces the PRG-type file MYPROGRAM with a new version of MYPROGRAM. If MYPROGRAM doesn't exist, it's created. SAVE F$,9 Saves a program whose name is in F$ on disk unit 9. SCALE -- Set the logical dimension of the graphic screen [*** NOT YET IMPLEMENTED ***] SCNCLR -- Clear a text or graphic screen SCNCLR [color] This command will clear the current text window if [color] omitted, otherwise it will clear the current graphic screen using the given color value. See also SCREEN CLR. SCNCLR Clears the text screen. If a window is defined it clears only the window area. SCNCLR 0 Clears the current graphic screen with color 0. SCRATCH -- Delete files from disk directory Recover accidentally deleted files SCRATCH "filespec" [,Ddrive] [Udevice] [,R] SCRATCH, ERASE, or DELETE are different names of the same command. They are used to delete a file from a disk directory, or optionally to recover if possible an accidentally deleted file. The diskette must not be 'write protected', or a 'WRITE PROTECT ON' error is reported. WARNING: Deleting a file will destroy all existing data in that file. Be extremely careful if you are using pattern matching, which can delete any or all files. In direct mode, you are asked to confirm what you are doing with 'ARE YOU SURE?'. Type 'Y' and press to proceed, or type any OTHER CHARACTER and press to cancel the command. In program mode there is no confirmation prompt. Upon completion, in direct mode only, the computer will display the number of files deleted. Refer to your disk manual for other details. Different disk drives implement slightly different pattern matching rules or support features such a specially protected files. If the 'R'ecover option is present and the DOS supports it, a deleted file can be recovered if nothing else has been written to the diskette since the file was accidentally deleted. You will still be asked to confirm the operation, and upon completion the computer will display the number of files restored. SCRATCH "oldfile" Deletes the file OLDFILE from the disk in the default system drive. SCRATCH "file.*" Deletes all files beginning with FILE. SCRATCH (F$), U(DD) Deletes the file whose name is in F$ from the disk in device DD. SCRATCH "SAVEME", R Attempt to recover the program SAVEME. SCREEN -- Graphic command The SCREEN command is used to initiate a graphic command. It always precedes another command word which identifies the graphic operation to be performed: SCREEN CLR - Set graphic screen color SCREEN CLR color# Clears (erases) the currently opened graphic screen using the given color value. Use SCNCLR to clear a text screen. See also SCNCLR. SCREEN DEF - Define a graphic screen SCREEN DEF screen#, width, height, depth screen# 0-1 width 0=320, 1=640, 2=1280 height 0=200, 1=400 depth 1-8 bitplanes (2-256 colors) Defines a logical screen (numbered 0 or 1), specifies its size and how many colors (bitplanes) it has. It does not allow access to the screen and it does not display the screen. The screen must be defined before it is opened for viewing and/or drawing to. SCREEN SET - Set draw and view screens SCREEN SET DrawScreen#, ViewScreen# draw screen # 0-1 view screen # 0-1 This command specifies which logical screen is to be viewed and which logical screen is to be accessed by the various draw commands. The screen must be defined and opened first. Both the draw and the view screen can be, and usually are, the same logical screen. For double buffering, they are different. SCREEN OPEN - Open a screen for access SCREEN OPEN screen# [,error_variable] screen# 0-1 error_variable [*** NOT YET IMPLEMENTED ***] This command actually sets up the screen and allocates the necessary memory for it. If it's the view screen it will be displayed. If it's the draw screen, it can now be drawn to. If there is not enough memory for the screen, 'NO GRAPHICS AREA' is reported and the screen is not opened. SCREEN CLOSE - Close a screen SCREEN CLOSE screen# screen# 0-1 This command closes a logical screen, ending access to it by the draw commands if it's the draw screen and restoring the text screen if it's the view screen. SCREEN CLOSE deallocates any memory reserved for the screen. SAMPLE GRAPHIC PROGRAM: 1 TRAP 170 in case of error want text screen 10 GRAPHIC CLR initialize graphics 20 SCREEN DEF 1,0,0,2 define a 320x200x2 graphic screen 30 SCREEN OPEN 1 open it 40 PALETTE 1,0, 0, 0, 0 define screen 1 color 0 = black 50 PALETTE 1,1, 15, 0, 0 define screen 1 color 1 = red 55 PALETTE 1,2, 0, 0,15 define screen 1 color 2 = blue 60 PALETTE 1,3, 0,15, 0 define screen 1 color 3 = green 70 SCREEN SET 1,1 make it the view screen 80 SCNCLR 0 clear screen with palette color 0 90 BORDER 0 set border color to color 0 100 PEN 0,1 make draw pen = color 1 (red) 110 LINE 100,100, 150,150 draw a diagonal red line 120 PEN 0,2 make draw pen = color 2 (blue) 130 BOX 50,50, 50,80, 80,50, 80,80 draw a blue box 140 PEN 0,3 make draw pen = color 3 (green) 150 CHAR 25,50, 1,1,2, "WORDS" draw green text 160 SLEEP 5 pause for 5 seconds 170 SCREEN CLOSE 1 close graphic, get text screen 180 PALETTE RESTORE restore normal system colors 190 BORDER 6 restore normal border color 200 END SET -- Set various system parameters The SET command is used to set a system parameter. It always precedes another command word which identifies the parameter to be changed: SET DEF - Set default system disk drive SET DEF device The BASIC DOS commands default to disk unit 8. Use SET DEF to change which device these commands default to. This command does not renumber a disk device, use SET DISK for that. Commands which specify a device will still access the device they specified. A program can be made more "user friendly" by either not specifying a drive (thus using the user's preferred drive) or by specifying device 1. Device number 1 means "use the system default drive, whatever its number is." 10 DIR gets directory of device 8 20 DIR U1 gets directory of device 8 30 DIR U10 gets directory of device 10 40 SET DEF 10 change the default drive to unit 10 50 DIR gets directory of device 10 60 DIR U1 gets directory of device 10 70 DIR U8 gets directory of device 8 SET DISK - Change a disk device number SET DISK oldnumber TO newnumber Use this command to renumber (change) a disk drive's unit number. Not all drives can be renumbered -- refer to your disk drive manual for details. This command sends to the disk's command channel the conventional CBM serial disk drive "M-W" command. See also the DISK command, which lets you send any command to a disk drive. SET DISK 8 TO 10 Change unit 8's number to 10 Because the built-in C64DX drives always take precedence over serial bus drives, this is one way to get the built-in drive "out of the way" so that you can access a serial bus drive #8. SGN -- Get the sign of a number SGN (expression) The SiGN function returns the sign of a numeric expression as follows: If the expression is < 0 (negative) .... returns -1 If the expression is = 0 (zero) ........ returns 0 If the expression is > 0 (positive) .... returns 1 SIN -- Sine function SIN (expression) This function returns the sine of X, where X is an angle measured in radians. The result is in the range -1 to 1. X = SIN (pi/4) Result is X=0.707106781 To get the sine of an angle measured in degrees, multiply the numeric expression by pi/180. SLEEP -- Pause program execution of a specified period of time SLEEP seconds Temporarily suspends execution of your program for 1 to 65535 seconds. SLOW -- Set system speed to 1.02MHz SLOW is used primarily to directly access "slow mode only" devices such as the SID sound chips. FAST is the default system speed. SOUND -- Produce sound effects SOUND v, f, d [,[dir] [,[m] [,[s] [,[w] [,p] ]]]] v = voice (1-6) f = frequency (0-65535) d = duration (0-32767) dir = step direction (0(up), 1(down), or 2(oscillate)) default=0 m = min frequency (0-65535) default=0 s = sweep (0-65535) default=0 w = waveform (0=triangle,1=saw,2=square,3=noise) default=2 p = pulse width (0-4095) 50% duty cycle=default=2048 The sound command is a fast and easy way to create sound effects and musical tones. The first three parameters are required to select the voice, frequency, and duration of the tone. The duration is specified in "jiffies" (60 jiffies = 1 second). Optionally, you can specify a waveform and, for square waves, the pulse width. The SOUND command can sweep a voice through a series of equally-spaced frequencies. The direction of the sweep, minimum and maximum frequencies can be programmed. If time expires before the sweep is done, the sound stops. If the minimum or maximum frequency is reached before time expires, the sound repeats. For programming details, refer to the SID hardware documentation. Use the VOLume command to change the volume of the sound. Note that the TEMPO command affects PLAY strings only, not SOUND effects. FREQout = ( f * 0.0596 ) Hz PWout = ( p / 40.95 ) % Each voice can be programmed separately and played simultaneously for a wide variety of sound effects. Once a sound effect is initiated, BASIC execution continues with the next statement while the sound plays out, allowing you to combine and control graphics, animation, and sound from a BASIC program. The examples below include information about how to generate precise tones for exact times, but for most casual users trial and error are perfectly acceptable! (Note that the values used are for 60Hz (NTSC) systems): Using voice 1, emit a square-wave, 440Hz tone for 1 second. Note that 440Hz = 7382 * 0.0596 using the above formula. SOUND 1, 7382, 60 Using voice 2, sweep from 100Hz (m=1638) to 440Hz (f=7382) in increments of 1Hz (s=17). The time required to do this can be calculated as t=(f-m)/s, so t=336 jiffies. SOUND 2, 7382, 336, 0, 1678, 17 Using voice 3, make a neat sound using an oscillating sweep (dir=2) and a sawtooth waveform (w=1) for 3 seconds (t=180). SOUND 3, 5000, 180, 2, 3000, 500, 1 SPC -- Space PRINT output SPC (number) The SPaCe function is used to format PRINTed data to the screen, a printer, or a file. It specifies the number of spaces to be skipped, from 0 to 255. A semicolon (';') is always assumed to follow SPC, even if it appears at the end of a print line. The SPC function works a little differently on screen, printer and disk output. On the screen, SPC skips over characters already on the screen, which is not the case with printer and disk output. On printers, if the last character on a line is skipped, the printer will automatically perform a carriage return and linefeed. PRINT "123";SPC(3);"456" Displays '123 456' PRINT "X";SPC(5) :PRINT"X" Displays 'X X' See also the TAB function. A better way to format PRINT output is with PRINT USING. SPRCOLOR -- Set multicolor sprite colors SPRCOLOR [sprite_mc1] [,sprite_mc2] Use the SPRITE command to set up a multicolor sprite, and use SPRCOLOR to set the additional colors. Note that these colors are common to all multicolor sprites. The color values must be in the range (0-15). Use the RSPCOL0R function to get the current multicolor sprite colors, and RSPRITE to get the current sprite foreground color. SPRDEF -- Define a sprite pattern [*** NOT EXPECTED TO BE IMPLEMENTED ***] SPRITE -- Turn a sprite on or off, and set its characteristics SPRITE number [,[on] [,[color] [,[priority] [,[x_exp] [,[y_exp] [,mode] ]]]]] The SPRITE command allows you set all of the characteristics of a sprite. Use the MOVSPR command to position it or set it in motion. Use the SPRCOLOR to set the multicolor sprite colors, if you are using multicolor sprites. All the parameters except the sprite number are optional. If you don't specify a parameter then it won't be changed. number = sprite number (0-7) on = enable (1) or disable(0) color = sprite foreground color (0-15) priority= sprite to display data priority: 0 means sprite goes over screen data 1 means sprite goes under screen data x,y_exp = sprite expansion on (1) or off (0) mode = sprite mode: 0 high resolution 1 multicolor The SPRITE command does not define a sprite. The sprite definitions must be loaded into the sprite area first ($600-$7FF). Use the BLOAD and BSAVE commands. [*** THIS MAY CHANGE ***] A sprite is 24 pixels wide and 21 pixels high. Each sprite definition requires 63 ($40 hex) bytes: $600 Sprite 0 definition $640 Sprite 1 definition $680 Sprite 2 definition $6C0 Sprite 3 definition $700 Sprite 4 definition $740 Sprite 5 definition $780 Sprite 6 definition $7C0 Sprite 7 definition Use the RSPRITE function to read a sprite's characteristics, or the RSPPOS function to read a sprite's position. The RSPCOLOR function is used to get the current multicolor sprite colors. 10 BLOAD"sprite 1 data", Load sprite 1's definition P(DEC("640")) 20 SPRITE 1, 1, 2 Turn it on, make it red 30 MOVSPR 1, 24, 50 Put it at top-leftmost corner 40 SPRSAV 1, 2 Copy sprite 1 definition to 2 50 SPRITE 2, 1, 7 Turn on sprite 2 make it yellow 60 MOVSPR 2, 320, 229 Put it at bottom-rightmost corner 70 BSAVE"sprite 2 data", Save sprite 2 P(DEC("680")) TO P(DEC("6C0")) 80 SPRITE 1, 0 Turn off sprite 1 90 SPRITE 2, 0 Turn off sprite 2 SPRSAV -- Copy a sprite definition SPRSAV source, destination Use this command to copy a sprite's data (shape) to another sprite or into a string variable, or copy a shape from a string variable into a sprite. You can have many different sprite shapes in memory at one time, all stored in strings. This makes it possible to animate sprites from BASIC by quickly "flipping through" shapes, using each shape like a frame from a movie film. SPRSAV 0, A$ copy the data (shape) of sprite 0 into A$ SPRSAV A$, 2 copy the data (shape) in A$ into sprite 2 SPRSAV 1, 2 copy the data (shape) in sprite 1 to sprite 2 STASH -- (see the DMA command) SQR -- Square root function SQR (number) This function returns the SQuare Root of the given numeric expression. The numeric expression must not be negative or an 'ILLEGAL QUANTITY' error is reported. A = SQR(10) Result is A = 3.16227766 STEP -- See FOR/NEXT/STEP STOP -- Halt program execution When STOP is executed, the computer immediately stops running the program and reports 'BREAK IN LINE xx'. No variables are cleared and files are not closed. This command is usually used while debugging (fixing) a BASIC program, since it lets you stop at a specific place, examine variables, change variables, and restart the program where it was halted (see CONTinue command) or some other line (see GOTO). In many cases, you can even change the program and use GOTO to resume execution with variables and open channels intact. SWAP -- (see the DMA command) STR$ -- Get the string representation of a number STR$ (number) The STRing function returns a string identical to PRINT's output of the given numeric expression. See PRINT for details regarding the format of numeric output. STR$ is the opposite of VAL. A$ = STR$(123) Result is A$ = " 123" A$ = STR$(-123) Result is A$ = "-123" A$ = STR$(.009) Result is A$ = " 9E-03" SYS -- Call a ROM routine or user machine language routine SYS address [,[a] [,[x] [,[y] [,[z] [,s] ]]]] This statement performs a call to a machine language routine at the specified address (range 0-65535, $3000-$FFFF) in a memory bank set up previously by the BANK command. The microprocessor's registers are loaded with the values specified in the parameters following the address (if given) and a JSR (Jump SubRoutine) instruction is performed. When the called routine ends with an RTS (ReTurn from Subroutine), the microprocessor's registers are saved and control is returned to the BASIC program. The microprocessor's registers can be examined with the RREG command. Because this command instructs the computer's microprocessor (CPU) to perform something, extreme care should be taken in its use. It can easily crash the computer if you do something wrong (press the reset button to reboot). Also see the BOOT SYS command. BANK 128: SYS DEC("FF5C") Call the Kernel's PHOENIX routine. BANK 128: SYS DEC("FF81") Reset the Screen Editor 10 BANK 128 20 BLOAD"user routine",P(DEC("1800")) Load a user routine 30 SYS DEC("1800"), areg, xreg Call it with args in A and X 40 RREG areg, xreg, , , sreg Get args back in A, X, and S 50 carry = (sreg AND 1) Get carry flag from S 60 PRINT "ACCUMULATOR = ";HEX$(areg) Display registers 70 PRINT "X REGISTER = ";HEX$(xreg) 80 PRINT "CARRY FLAG = ";carry See the USR function for another way to call machine language routines. TAB -- Space PRINT output TAB (number) The TAB function is used to format PRINTed data to the screen, a printer, or a file. It's primarily for screen text output, moving the cursor to the specified column (plus one) as long as the current print position is not already beyond that point (for example, if the current print position is the first column, TAB(1) would print subsequent text beginning in column 2). If the current print position is already beyond the column specified by the TAB function, nothing is done. For disk and printer output, TAB works exactly like the SPC function (see SPC). A semicolon (';') is always assumed to follow TAB, even if it appears at the end of a print line. PRINT "TEXT";TAB(10);"HERE" Result is 'TEXT HERE' PRINT "TEXT";SPC(10);"HERE" Result is 'TEXT HERE' The above examples illustrate the difference between TAB and SPC. See also the SPC function. A better way to format PRINT output is with PRIUT USING. Don't confuse the TAB function with the TAB character, CHR$(9), which is used to format data using the programmable TAB stops. TAN -- Tangent function TAN (expression) This function returns the tangent of the numeric expression, measured in radians. If the result overflows, TAN(pi/2) for example, an 'OVERFLOW' error is reported. X = TAN(1) Result is X=1.55740772 To get the tangent of an angle measured in degrees, multiply the numeric expression by pi/180. TEMPO -- Set the tempo (speed) of a PLAY string TEMPO rate Use this command to adjust the tempo (speed) of music playback by the PLAY command. The rate determines the duration of a whole note. The default is 12, making a whole in 4/4 time last 2 seconds. The formula is: duration = 24/rate The higher the rate, the faster the note. The range is (1-255). THEN -- See IF/THEN/ELSE TO -- See FOR/NEXT/STEP. Also used as a subcommand. TRAP -- Define an BASIC error handler TRAP [line_number] When turned on, TRAP intercepts all BASIC execution error conditions except 'UNDEF'D STATEMENT ERROR'. Even the STOP key can be TRAPped. When an error occurs, BASIC saves the error's location, line number, and error number. If TRAP is not set, BASIC returns to direct mode and displays the error message and line number. If TRAP is set, BASIC performs a GOTO to the line number specified in the TRAP statement and continues executing. Your BASIC error handling routine can examine the error number, message, and the line number where the error occurred and determine the proper course of action. The system error words are: ER Error Number EL Error Line (line where the error occurred) ERR$() Error Message If ER is -1, then a BASIC error did not occur. The error routine should check the disk status words, in case they were the cause of the error: DS Disk Error Number DS$ Disk Error Message Refer to the list of BASIC and Disk error messages in the appendix. Note that an error in your TRAP routine cannot be trapped. The RESUME statement can be used to resume execution -- see RESUME. TRAP with no line number specified turns off error TRAPping. 10 TRAP 90 enable trapping 20 FOR I=-5 TO 5 30 PRINT 5/I error when I=0 40 NEXT 50 TRAP turn trapping off 60 END 70 : 90 PRINT ERR$(ER): RESUME NEXT error routine TROFF -- Turn off trace mode TRON -- Turn on trace mode TROFF TRON Trace mode is used while debugging (fixing) a BASIC program. TRON enables tracing, and TROFF disables tracing. When the program is run and trace mode is on, the line number of the command that is being executed is displayed on the screen. If there are three commands on the line, the line number will be displayed three times, once each time one of the commands is executed. Trace mode lets you know what the computer is doing. Trace mode works even when a graphic screen is being displayed, but the line number is still displayed on the text screen so you won't be able to see it until the graphic screen is turned off. If your program is doing alot of PRINT statements, the display can seen a little confusing. Trace mode can be set in direct mode to trace the entire program, or it can be turned on and off from within your program to let you trace only selected portions of the program. Trace mode has no effect on commands entered in direct (edit) mode. The NEW command disables trace mode, but RUN and CLR do not. 10 FOR I=-5 TO 5 15 TRON 20 PRINT 5/I 25 TROFF 30 NEXT TYPE -- Display the contents of a sequential disk file TYPE "filename" [,Ddrive] [<,|ON>Udevice] Use this command to print the contents of a PETSCII data file on the screen. The file must contain lines no longer than 255 characters long and terminated by a return character (CHR$(13)). Lines too long result in a 'STRING TOO LONG' error. TYPE "readme" display the contents of the README file on the screen The command sequence below will print the contents of the README file on a CBM serial bus printer in upper/lower case mode. OPEN 4,4,7: CMD4: TYPE"readme": CLOSE4 UNTIL -- See DO/LOOP/WHILE/UNTIL/EXIT USR - Call a user defined machine language function USR (expression) When this function is used, the program jumps to a machine language subroutine whose starting address must be POKEd into system memory (BANK 128) at address 760 (low byte) and 761 (high byte), or $2F8 hex. The floating point value of the numeric expression is passed to the routine in the Floating point ACCumulator (FACC), and the value to be returned is taken from the FACC when the routine ends. If the USR vector is not set up prior to making the USR call, an 'UNDEF'D FUNCTION' error is reported. The routine must be located in the system bank. The BANK command does not affect USR. Using this method of calling a machine language routine requires a fair amount of set up and a good knowledge of the lower level math routines built into BASIC. See the SYS command, which is more commonly used to call a machine language routine. The following program illustrates the basic steps required for installing a USR routine and calling it: 10 BANK 128 System bank for poke & load 20 UV = DEC("1800") Where my routine is 30 BLOAD "my user routines",P(UV) Load my routine 40 POKE DEC("2F8"), UV AND 255, UV / 256 Set up USR address 50 x = USR(123): PRINT X Call my routine with the the value 123, get back and print whatever my routine leaves in FACC The following program actually works. It points the USR vector to the BASIC math jump table entry for the routine which inverts the sign of the number in the FACC. Type in positive & negative numbers: 10 BANK 128 System bank for poke 20 POKE DEC("2F8"), DEC("33"), DEC("7F") Set up USR address 30 DO: INPUT"SIGNED NUMBER"; N Get number input 40 : PRINT USR(N) Display USR output 50 : LOOP UNTIL N=0 End if user types zero USING -- See PRINT USING VAL -- Get the numerical value of a string VAL (string) The VALue function converts a string into a number. The conversion starts with the first character and ends at the end of the string or the first character that is not allowed in normal number input. Spaces are ignored. If the first character of the string is not a legal character, a zero is returned. The VAL function works the same way the INPUT and READ commands do. VAL is the opposite of STR$. X = VAL(" 123") Result is X = 123 X = VAL("-123") Result is X = -123 X = VAL(" 9E-02") Result is X = .09 VERIFY -- Compare a program or data in memory with a disk file VERIFY "filename" [,device_number [,relocate_flag]] This command is just like a LOAD command, except instead of putting the data read from a file into memory, the computer compares it to what is already in memory. If there's any difference at all a 'VERIFY ERROR' is reported. The filename must be given, and pattern matching may be used. In the case of dual drive systems, the drive number must be part of the filename. If a device number is given, the file is sought on that unit, which must be a disk drive. If a device number is not given, the default system drive is used. See also DVERIFY. Note: If the BASIC program in memory is not located at the same address as the version on disk was SAVEd from, the files will not match even if the program is otherwise identical. The relocate_flag is used to VERIFY binary files. If the relocate_flag is present and non-zero, the file will be compared to memory starting at the address stored on disk when the file was SAVEd. The memory bank used is the bank given in the last BANK statement. The ending address is determined by the length of the disk file. The comparison halts on the first mismatch or at the end of the file. The area to be compared must be confined to the indicated memory bank. Do not use the relocate_flag to verify BASIC programs. See also BVERIFY. VERIFY "myprogram" Good: SEARCHING FOR 0:myprogram Bad: SEARCHING FOR 0:myprogram VERIFYING VERIFYING OK ?VERIFY ERROR VERIFY "PROG" Compares BASIC program in memory to file PROG on the default system disk. VERIFY FILE$,DRV Compares program in memory to a program whose name is in the variable FILE$ on the unit whose number is in DRV. VERIFY "0:PROG",8 Compares memory to BASIC program PROG on unit 8, drive 0. BANK 128 Compares a binary file into memory. The VERIFY "BIN",8,1 address used comes from the disk file, but you must specify the memory bank. VIEWPORT -- [*** CURRENTLY UNIMPLEMENTED ***] VOL -- Set audio volume level VOL volume [*** THIS COMMAND WILL CHANGE ***] This statement sets the volume level for SOUND and PLAY statements. VOLUME can be set from 0 to 15, where 15 is the maximum volume. A volume of 0 turns sound output off. VOLume affects all 3 voices. Note that PLAY strings can change the volume, too. WAIT -- Pause BASIC program until a memory state satisfied WAIT address, and_mask [,xor_mask] The WAIT statement causes program execution to be suspended until data at a specified memory location matches a given bit pattern. It's used to pause your program until an event occurs. The event could be an I/O state (such as a fire button or peripheral port change), a hardware state (such as the raster position or RS232 status), or memory change caused by an interrupt event (such as a keyboard scan). The WAIT statement tells the computer to read (PEEK) a memory location (0-65535) and AND the value it got with the number in and_mask (0-255). If the result is zero, repeat the operation until the result is not zero. This is like the following BASIC instructions, but much faster: DO: result = PEEK(address): LOOP UNTIL (result AND and_mask) <> 0 This works if the state you are WAITing for is non-zero (a one or "high" state). If you want to wait for a zero state (a "low" state), you need to use the xor_mask option to "flip" the bits of the result. Note that it's possible to "hang" your program indefinitely if the state you are waiting for never happens or you specify the wrong data. Press the STOP and RESTORE keys at the same time to get control back. Be sure to use the BANK command before you tell the computer to WAIT to specify which 64K memory bank the address is in. Note that a BANK number greater than 127 (i.e., a bank number with the most significant bit set) must be used to address an I/O location, such as the VIC chip. Refer to the system memory map for details. 10 BANK 128 Wait for the VIC raster to be 20 WAIT DEC("D011"), 128 offscreen (want RC8 = 1) 10 BANK 128 Wait for the VIC raster to be 20 WAIT DEC("D011"), 128, 128 onscreen (want RC8 = 0) 10 BANK 128 20 WAIT DEC("D3"), 1 Wait for user to press key 30 WAIT DEC("D3"), 2 Wait for user to press key 40 WAIT DEC("D3"), 4 Wait for user to press key 50 WAIT DEC("D3"), 8 Wait for user to press key WHILE -- See DO/LOOP/WHILE/UNTIL/EXIT WIDTH -- [*** CURRENTLY UNIMPLEMENTED ***] WINDOW -- Set a text window WINDOW left_column, top_row, right_column, bottom_row [,clear] This command defines a logical text screen window. All text I/O will be confined to this window. The row parameters must be in the range (0-24), and the column parameters must be in the range (0-79) for 80-column screens or (0-39) for 40-column screens. The parameters are always referenced to the physical screen (i.e., you cannot define a window within a window). If the clear flag is given, the new window area will be cleared after it's set up. Use the RWINDOW function to get the current window size. You are responsible for saving and restoring screen data in all windows because the WINDOW command simply sets the window margins. The WINDOW command does not draw a border around a window. All color commands and screen modes (such as scroll disable, TAB stops, etc.) are global. Two consecutive "home" characters will reset the window definition back to the physical screen. WINDOW 0,0,39,24 Define a window in 80-column mode that is the left half of the screen. WINDOW 40,0,79,24 Define a window in 80-column mode that is the right half of the screen. WINDOW 0,0,79,12 Define a window in 80-column mode that is the top half of the screen. WINDOW 0,13,79,24 Define a window in 80-column mode that is the bottom half of the screen. WINDOW 20,6,59,12,1 Define a window in 80-column mode in the center of the screen and clear it. The window is 12 characters high and 40 characters wide. PRINT CHR$(19)CHR$(19); Reset the window back to full screen in either 40 or 80-column mode and put the cursor in top left corner. XOR -- Exclusive-Or function XOR (number,number) The XOR function returns a numeric value equal to the logical XOR of two numeric expressions, operating on the binary value of the unsigned 16-bit integers in the range (0 to 65535). Numbers outside this range result in an 'ILLEGAL QUANTITY' error. X = XOR(4,12) Result is X= 8 X = XOR(2,12) Result is X=14 3.1.4. VARIABLES The C64DX uses three types of variables in BASIC: floating point X integer X% string X$ Normal NUMERIC VARIABLES, also called floating point variables, can have any from up to nine digits of accuracy. When a number becomes larger than nine digits can show, as in +10 or -10, the computer displays it in scientific notation form, with the number normalized to 1 digit and eight decimal places, followed by the letter E and the power of ten by which the number is multiplied. For example, the number 12345678901 is displayed as 1.23456789E+10. INTEGER VARIABLES can be used when the number is a signed whole number from +32767 to -32768. Integer data is a number like 5, 10, or -100. Integers take up less space than floating point variables, particularly when used in an array. STRING VARIABLES are those used for character data, which may contain numbers, letters, and any other character that the computer can make. An example of string data is "Commodore C64DX". VARIABLE NAMES may consist of a single letter, a letter followed by a number, or two letters. Variable names may be longer than 2 characters, but only the first two are significant. An integer is specified by using the percent (%) sign after the variable name. String variables have a dollar sign ($) after their names. EXAMPLES: Numeric Variable Names: A, A5 , BZ Integer Variable Names: A%, A5%, BZ% String Variable Names : A$, A5$, BZ$ ARRAYS are lists of variables with the same name, using an extra number (or numbers) to specify an element of the array. Arrays are defined using the DIM statement, and may be floating point, integer, or string variable arrays. The array variable name is followed by a set of parentheses () enclosing the number of the variable in the list. EXAMPLE: A(7), BZ%(11), A$(87) Arrays can have more than one dimension. A two dimensional array may be viewed as having rows and columns, with the first number identifying the row and the second number identifying the column (as if specifying a certain grid on the map). EXAMPLE: A(7,2), BZ%(2,3,4), Z$(3,2) RESERVED VARIABLE NAMES are names that are reserved for use by the computer, and may not be used for another purpose. These are the variables DS, DS$, ER, ERR$, EL, ST, TI, and TI$. KEYWORDS such as TO and IF or any other names that contain KEYWORDS, such as RUN, NEW, or LOAD cannot be used. ST is a status variable for input and output (except normal screen/keyboard operations). The value of ST depends on the results of the last I/O operation. In general, if the value of ST is 0 then the operation was successful. TI and TI$ are variables that relate to the real-time clock built into the C64DX. The system clock is reset to zero when the system is powered up or reset, and can be changed by the user or a program. TI$="hh:mm:ss.t" Allows optional colons to delimit parameters and allows input to be abbrieviated (eg., TI$="h:mm" or even TI$=""), defaulting to "00" for unspecified parameters. 24-hour clock (00:00:00.0 to 23:59:59.9). TI 24-hour TOD converted into tenths of seconds. The value of the clock is lost when the computer is turned off. It starts at zero when the computer is turned on, and is reset to zero when the value of the clock exceeds 23:59:59.9. The variable DS reads the disk drive command channel, and returns the current status of the drive. To get this information in words, PRINT DS$. These status variables are used after a disk operation, like DLOAD or DSAVE, to find out why the error light on the disk drive is blinking. ER, EL, and ERR$ are variables used in error trapping routines. They are usually only useful within a program. ER returns the last error encountered since the program was RUN. EL is the line where the error occurred. ERR$ is a function that allows the program to print one of the BASIC error messages. PRINT ERR$(ER) prints out the proper error message. 3.1.5. OPERATORS The BASIC OPERATORS include ARITHMETIC, RELATIONAL, and LOGICAL OPERATORS. The ARITHMETIC operators include the following signs: + addition - subtraction * multiplication / division ^ raising to a power (exponentiation) On a line containing more than one operator, there is a set order in which operations always occur. If several operators are used together, the computer assigns priorities as follows. First exponentiation, then multiplication and division, and last, addition and subtraction. If two operators have the same priority, then calculations are performed in order from left to right. If these operations are to occur in a different order, BASIC 10.0 allows giving a calculation a higher priority by placing parentheses around it. Operations enclosed in parentheses will be calculated before any other operation. Make sure that the equations have the same number of left and right parentheses, or a SYNTAX ERROR message is posted when the program is run. There are also operators for equalities and inequalities, called RELATIONAL operators. Arithmetic operators always take priority over relational operators. = is equal to < is less than > is greater than <= or =< is less than or equal to >= or => is greater than or equal to <> or >< is not equal to Finally, there are three LOGICAL operators, with lower priority than both arithmetic and relational operators: AND OR NOT These are most often used to join multiple formulas in IF...THEN statements. When they are used with arithmetic operators, they are evaluated last (i.e., after + and -). If the relationship stated in the expression is the true the result is assigned an integer of -1 and if false an integer of 0 is assigned. There is also an XOR function. EXAMPLES: IF A=B AND C=D THEN 100 requires both A=B & C=D to be true IF A=B OR C=D THEN 100 allows either A=B or C=D to be true A=5:B=4:PRINT A=B displays 0 A=5:B=4:PRINT A>3 displays -1 PRINT 123 AND 15:PRINT 5 OR 7 displays 11 and 7 3.1.6. ERROR MESSAGES 3.1.6.1. BASIC ERROR MESSAGES The following error messages are displayed by BASIC. Error messages can also be displayed with the use of the ERR$() function. The error number refers only to the number assigned to the error for use with this function. In direct mode, DOS error messages (DS$) are automatically displayed. They are described in the section after this one. ERROR# ERROR NAME DESCRIPTION ---------------------------------------------------------------------- 1 TOO MANY FILES There is a limit of 10 files OPEN at one time. 2 FILE OPEN An attempt was made to open a file using the number of an already open file. 3 FILE NOT OPEN The file number specified in an I/O statement must be opened before use. 4 FILE NOT FOUND No file with that name exists on the specified drive. 5 DEVICE NOT PRESENT The required I/O device not available. 6 NOT INPUT FILE An attempt made to read data from a file that was opened for writing. 7 NOT OUTPUT FILE An attempt was made to write data to a file that was opened for reading. 8 MISSING FILE NAME Filename was missing in command. 9 ILLEGAL DEVICE NUMBER An attempt was made to use a device improperly (SAVE to the screen, etc) or an illegal device number was specified. 10 NEXT WITHOUT FOR Either loops are nested incorrectly, or there is a variable name in a NEXT statement that doesn't correspond with one in FOR. 11 SYNTAX ERROR A statement is unrecognizable by BASIC. This could be because of missing or extra parenthesis, parameters, delimiters, or a misspelled keyword. 12 RETURN WITHOUT GOSUB A RETURN statement was encountered when no GOSUB statement was active. 13 OUT OF DATA A READ statement was encountered with no DATA left unREAD. 14 ILLEGAL QUANTITY A number used as an argument is outside the allowable range (too big or too small) 15 OVERFLOW The result of a computation is larger than the largest number allowed (1.701411834E+38) 16 OUT OF MEMORY There is not enough memory for the program, or variables, or there are too many DO, FOR or GOSUB statements in effect. 17 UNDEF'D STATEMENT A line number referenced doesn't exist. 18 BAD SUBSCRIPT The program tried to reference an element of an array out of the range specified by a DIM statement, a missing DIM statement or a mistyped function name. 19 REDIM'D ARRAY An array can only be DIMensioned once. 20 DIVISION BY ZERO Division by zero is illegal. 21 ILLEGAL DIRECT Command is only allowed to be used in a program. 22 TYPE MISMATCH A numeric variable was used in place of a string variable or vice versa. 23 STRING TOO LONG An attempt was made to assign more than 255 characters to a string, or enter more than 160 characters from the keyboard, or to input more than 255 characters from a file. 24 FILE DATA The wrong type of data was read from a file. 25 FORMULA TOO COMPLEX An expression is too complicated for BASIC to process all at one time. Break it into smaller pieces or use fewer parentheses. 26 CAN'T CONTINUE The CONT command does not work if the program was not RUN, there was an error or a line has been edited. 27 UNDEFINED FUNCTION An attempt was made to use a user defined function that was never defined. 28 VERIFY The program on disk does not match the program in memory. 29 LOAD There was a problem loading. 30 BREAK The program was halted by the STOP key or a STOP statement. 31 CAN'T RESUME A RESUME statement was encountered without a TRAP in effect, or an error occurred in the trap handler itself. 32 LOOP NOT FOUND The program encountered a DO statement and cannot find the corresponding LOOP. 33 LOOP WITHOUT DO A LOOP was encountered without a DO statement active. 34 DIRECT MODE ONLY A command was used in a program that can only be used in direct mode. 35 NO GRAPHICS AREA A graphics command was used before a graphics screen was defined and opened. 36 BAD DISK A BOOT SYS command failed because the disk could not be read. 37 BEND NOT FOUND A BEND statement not found for BEGIN. 38 LINE NUMBER TOO LARGE A line number cannot exceed 64000. 39 UNRESOLVED REFERENCE Renumber failed because a referenced line number does not exist. 40 UNIMPLEMENTED COMMAND The given command is not currently implemented in this computer. 41 FILE READ There was a problem reading data from a disk file. Similar to LOAD ERROR. 3.1.6.2. DOS ERROR MESSAGES The following error messages are returned through the DS and DS$ variables. If a disk command is type in direct mode, these messages will be displayed automatically. NOTE: DOS message numbers less than 20 are advisory and are not necessarily errors. DOS messages may vary slightly depending upon the drive model. Refer to your DOS manual for details. ERROR # DESCRIPTION ------- -------------------------------------------------------------- 00: OK (no error) 01: FILES SCRATCHED (not an error) The following number (track) tells how many files were deleted by the scratch command. 02: PARTITION SELECTED (not an error) The requested disk partition (subdirectory) has been selected. 03: FILES LOCKED The requested file(s) have been locked. 04: FILES UNLOCKED The requested file(s) have been unlocked. 05: FILES RESTORED The requested file(s) have been recovered (undeleted). 20: READ ERROR (block header not found) The disk controller is unable to locate the header of the requested data block. Caused by an illegal sector number, or the header has been destroyed. 21: READ ERROR (no sync character) The disk controller is unable to detect a sync mark on the desired track. Caused by misalignment of the read/write head, no diskette is present, or unformatted or improperly seated diskette. Can also indicate a hardware failure. 22: READ ERROR (data block not present) The disk controller has been requested to read or verify a data block that was not properly written. This error occurs in conjunction with the BLOCK commands and indicates an illegal track and/or sector request. 23: READ ERROR (checksum error in data block) This error message indicates that there is an error in one or more of the data bytes. The data has been read into the DOS memory, but the checksum over the data is in error. This message may also indicate grounding problems. 24: READ ERROR (byte decoding error) The data or header has been read into the DOS memory, but a hardware error has been created due to an invalid bit pattern in the data byte. This message may also indicate grounding problems. 25: WRITE ERROR (write-verify error) This message is generated if the controller detects a mismatch between the written data and the data in the DOS memory. 26: WRITE PROTECT ON This message is generated when the controller has been requested to write a data block while the write protect switch is depressed. 27: READ ERROR This message is generated when a checksum error is in the header. 28: WRITE ERROR This error message is generated when a data block is too long. 29: DISK ID MISMATCH This message is generated when the controller has been requested to access a diskette which has not been initialized. The message can also occur if a diskette has a bad header. 30: SYNTAX ERROR (general syntax) The DOS cannot interpret the command sent to the command channel. Typically, this is caused by an illegal number of file names, or patterns are illegally used. For example, two file names appear on the left side of the COPY command. 31: SYNTAX ERROR (invalid command) The DOS does not recognize the command. The command must start in the first position. 32: SYNTAX ERROR (invalid command) The command sent is longer than 58 characters. 33: SYNTAX ERRROR (invalid file name) Pattern matching is invalidly used in the OPEN or SAVE command. 34: SYNTAX ERROR (no file given) The file name was left out of the command or the DOS does not recognize it as such. 39: SYNTAX ERROR (invalid command) This error may result if the command sent to the command channel (secondary address 15) is unrecognized by the DOS. 40: UNIMPLEMENTED COMMAND Command is not implemented at this time. 41: FILE READ The file cannot be read. 50: RECORD NOT PRESENT Result of disk reading past the last record through INPUT# or GET# commands. This message will also occur after positioning to a record beyond end_of_file in a relative file. If the intent is to expand the file by adding the new record (with a PRINT# command), the error message may be ignored. INPUT and GET should not be attempted after this error is detected without first repositioning. 51: OVERFLOW IN RECORD PRINT# statement exceeds record boundary. Information is truncated. Since the carriage return which is sent as a record terminator is counted in the record size, this message will occur if the total characters in the record (including the final carriage return) exceeds the defined size. 52: FILE TOO LARGE Record position within a relative file indicates that disk overflow will result. 53: BIG RELATIVE FILES DISABLED 60: WRITE FILE OPEN This message is generated when a write file that has not been closed is being opened for reading. 61: FILE NOT OPEN This message is generated when a file is being accessed that has not been opened in the DOS. Sometimes, in this case, a message is not generated: the request is simply ignored. 62: FILE NOT FOUND The requested file does not exist on the indicated drive. 63: FILE EXISTS The file name of the file being created already exists on the diskette. 64: FILE TYPE MISMATCH The requested access mode is not possible using the filetype given. 65: NO BLOCK The sector you tried to allocate with the B-A command was already allocated. The track and sector numbers hold the next higher, available track and sector. If the track number is zero, no higher sectors are free (try a lower track & sector). 66: ILLEGAL TRACK AND SECTOR The DOS has attempted to access a track or block which does not exist in the format being used. This may indicate a problem reading the pointer of the next block. 67: ILLEGAL SYSTEM T OR S This special error message indicates an illegal system track or sector. 70: NO CHANNEL The requested channel is not available, or all channels are in use. A maximum of five sequential files may be opened at one time to the DOS. Direct access channels may have six opened files. 71: DIRECTORY ERROR The BAM is corrupted. Try initializing the disk. 72: DISK FULL Either the blocks on the diskette are used or the directory is at its entry limit. DISK FULL is sent when two blocks are available to allow the current file to be closed before its data is lost. 73: DOS MISMATCH (also the powerup message) Initially given at powerup to identify the drive. On some drives this message is given as an error to indicate the media was formatted by an incompatible DOS. 74: DRIVE NOT READY An attempt has been made to access the Floppy Disk Drive without any diskette present. 75: FORMAT ERROR 76: CONTROLLER ERROR The DOS has determined that the hardware is malfunctioning. 77: SELECTED PARTITION ILLEGAL An attempt was made to access a partition as a subdirectory, but it has no directory track or does not meet the criteria of a directory partition. 78: DIRECTORY FULL There is no more room in the directory sector for another file entry. Delete a file to make room, or change disks. 79: FILE CORRUPTED The DOS has determined that a file is bad, probably having bad links. Prepare a new disk and copy the good files to it. Could be the result of an unsuccessful file recovery. 3.2. MACHINE LANGUAGE MONITOR 3.2.1. INTRODUCTION The MONITOR is a built in machine language program that lets the user easily write machine language programs. The C64DX MONITOR includes a machine language monitor, an assembler, and a disassembler. Machine language programs written using the MONITOR can run by themselves, or be used as very fast 'subroutines' for BASIC programs. Care must be taken to position the assembly language programs in memory so that the BASIC program does not overwrite them and the proper memory is in context at all times (including during interrupts). 3.2.2. MONITOR COMMANDS A ASSEMBLE - Assemble a line of 4502 code C COMPARE - Compare two sections of memory D DISASSEMBLE - Disassemble a line of 4502 code F FILL - Fill a section of memory with a value G GO - Start execution at specified address H HUNT - Find specified data in a section of memory L LOAD - Load a file from disk M MEMORY - Dump a section of memory R REGISTERS - Display the contents of the 4502 registers S SAVE - Save a section of memory to a disk file T TRANSFER - Transfer memory to another location V VERIFY - Compare a section of memory with a disk file E EXIT - Exit Monitor mode . - Assembles a line of 4502 code > - Modifies memory ; - Modifies register contents @ - Display disk status $ - Display hex, decimal, octal, and binary value + & % The MONITOR accepts binary, octal, decimal and hexadecimal values for any numeric field. Numbers prefixed by one of the characters $ + & % are interpreted as base 16, 10, 8, or 2 values respectively. In the absence of a prefix, the base defaults to hexadecimal always. The assembler will use the base page form of an instruction wherever possible unless the address field is preceded by extra zeros to force the absolute form (except binary notation). The most significant byte of a 24-bit (3-byte) address field specifies the memory BANK to implement at the time the given command is executed. BANK bytes with the MSB set (i.e., banks greater than $7F) mean "use the current system configuration", which always includes the I/O area. If a BANK is not specified, BANK 0 is assumed. BANK 00 internal RAM bank 0 (System, BASIC program) BANK 01 internal RAM hank 1 (DOS, BASIC vars, color bytes) BANK 02 internal ROM bank 0 (DOS, C64 mode, CHRSETS) BANK 03 internal ROM bank 1 (Monitor, C65 mode) BANK 04-07 reserved for future expansion BANK 08-7F expansion RAM (graphic screens, RAM disk, etc.) BANK 80-FF MSB set means current config & I/O The monitor supports the editor autoscroll feature for memory dumps (forwards and backwards) and disassemblies (forward disassembly only). To send dump output to a printer, from BASIC open a CMD channel to the printer and enter the monitor (OPEN 4,4: CMD4: MONITOR). Give the dump command desired; output will be to the printer. 3.2.3. MONITOR COMMAND DESCRIPTIONS COMMAND: A PURPOSE: Enter a line of assembly code. SYNTAX: A
A number indicating the location in memory to place the assembled binary code. A 4502 assembly language mnemonic, eg., LDA The operand, when required, can be of any of the legal addressing modes. A is used to indicate the end of the assembly line. If are any errors on the line, a question mark is displayed to an error, and the cursor moves to the next line. The screen can be used to correct the error(s) on that line. As each line is entered, the machine code is written to the specified address and the line is automatically disassembled. Base page and relative addresses are calculated for you, and the appropriate word or byte relative mode selected automatically. To force an absolute addressing mode, supply leading zeros if necessary. .A 1800 LDX #$00 .A 1802 NOTE: A period (.) is equal to the ASSEMBLE command. . 1900 LDA #$23 COMMAND: C PURPOSE: Compare two areas of memory SYNTAX: C A number indicating the start of the area of memory to compare against. A number indicating the end of the area of memory to compare against. A number indicating the start of the other area of memory to compare with. The following example compares $8000-$9FFF in bank 0 with $8000-$9FFF in bank 1. Addresses of data that does not match are printed on the screen. C 8000 9FFF 18000 COMMAND: D PURPOSE: Disassemble machine code SYNTAX: D [address_1 [address_2] ] A number setting the address to start the disassembly. An optional ending address of code to be disassembled. The output of the disassembly is the same as that of an assembly, only preceded by a comma instead of an A or period. The object code is also displayed. Relative addresses in the disassembly are displayed as the 16-bit destination. A disassembly listing can be modified using the screen editor. Any changes to the mnemonic or operand on the screen, then hit the . This enters the line and calls the assembler for instructions. The object code cannot be modified this way. A disassembly can be paged. Typing a D causes the next of disassembly to be displayed. The autoscroll feature works in forward mode only, because backwards disassembly is not possible because all 256 opcodes are defined in the 4502 processor. The following example disassembles from ROM bank 3: D 3F000 3F005 . 03F000 A9 09 LDA #$09 . 03F002 A0 FF LDY #$FF . 03F004 18 CLC . 03F005 86 C2 STX $C2 Note that banks wrap to the next higher bank number. COMMAND: F PURPOSE: Fill a range of locations with a specified byte. SYNTAX: F The first location to fill with the . The last location to fill with the . The byte to fill with This command is useful for initializing data structures or any other RAM area. F 00600 007FF 00 Fills memory locations from $0600 to $07FF (RAM-0) with $00. Note that banks wrap to the next higher bank number. The maximum area that can be filled at one time is 64K, limited by the DMA device. COMMAND: G PURPOSE: Perform a JMP to a specified address SYNTAX: G
The address where execution is to start. When the address is not specified, execution begins at the current PC. (The current PC can be viewed or changed with the R command.) The GO command loads the processor's registers (displayable by the R command) and performs a JMP to the specified starting address. Caution is recommended in using the GO command. To return to MONITOR mode after performing a GO command, a BRK instruction must end the called routine. Also, the BANK specified must be able to handle interrupts (note that BANK bytes less than $80 do NOT include the operating system or I/O space). G FFC800 JuMPs to address $C800 in bank $FF (system configuration). COMMAND: H PURPOSE: Hunt through memory within a specified range for all occurences of a set of bytes. SYNTAX: H Address to start at Last address Data to search for. May be a number, sequence of numbers, or a PETSCII string. H 02000 0FFFF 46 52 45 44 Hunts for the series of bytes $46, $52, $45, $44 in memory bank 0 beginning at address $2000 and ending at $FFFF. The addresses of matches is displayed. H 0200 0FFFF 'FRED Hunts for the PETSCII string following an apostrophe. Note that banks wrap to the next higher bank number. COMMAND: L PURPOSE: Load a file from disk. SYNTAX: L <"filename"> [,device [,load_address] ] <"filename"> Is a filename in quotes. [device] Is a number indicating the device to load from. [load_address] Optional load address. If not given, the file is loaded into memory at the 16-bit address stored on disk (always RAM bank 0). The LOAD command causes a file to be loaded into memory. If the load address (including BANK) is given, the data is placed there. Otherwise the file is loaded into RAM bank 0 at the 16-bit load address specified by the first two bytes read from the PRG (program) type file. An error occurs if a load overflow the specified bank. L "filename" Loads "filename" from default system drive into RAM bank 0 at the address read from the file. L "filename",+10,80000 Loads "filename" from drive 10 (notice you must specify decimal for the drive number, or use hex equivalent) into expansion memory bank 8 at address $0000. Note that spaces between parameters after the filename are not permitted. COMMAND: M PURPOSE: Dump a section of memory in hex and PETSCII. SYNTAX: M [address_1 [address_2] ] [address_1] Starting address of memory dump. If omitted, one page is displayed starting from the last address used. [address_2] Ending address of memory dump. If omitted, one page is displayed starting at address_1. Memory dump width is sized to 40 or 80 columns, depending upon the text screen width. All data is displayed in hexadecimal and followed by a PETSCII interpretation of the data in reverse field (non-printing characters appear as periods). The autoscroll keys will scroll the dump forwards or backwards. Paging is also possible by typing M. The hex field of dump can be edited, and memory will be updated after a is typed on the edited line. M 29000 2900C >029000 3C 66 6E 6E 60 62 3C 00 :029008 46 41 49 54 20 4C 55 58 :FAIT LUX COMMAND: R PURPOSE: Display "shadow" 4502 registers. The PC (address), SR (status), A,X,Y,Z registers, and SP (stack pointer) are displayed. SYNTAX: R R PC SR AC XR YR SP ; BA1234 00 00 00 00 FB The address field contains the 8-bit bank plus the 16-bit segment address. The register dump can be edited by changing any field and pressing return. The data is used by the G (JMP) and J (JSR) commands. COMMAND: S PURPOSE: Save a section of memory in a disk file. SYNTAX: S <"filename">,,, <"filename"> Is a filename in quotes. Starting address of memory to be saved. Ending address PLUS ONE of memory to be saved. The SAVE command creates a PRG (program) type file and copies data into it from the specified memory area. All parameters are required. S "filename",8,A0000,AFFFF Saves expansion bank A in "filename" on drive 8 (you must specify decimal for the drive number, or use hex equivalent). The last byte at $FFFF will not be saved. Note that spaces between parameters after the filename are not permitted. The 16-bit segment address is saved as the first two bytes of the file, but the BANK address is not saved. The BANK wraps automatically to the next higher bank number, but note that LOAD is restricted to one bank, 64K bytes maximum. COMMAND: T PURPOSE: Transfer (copy) memory from one memory area to another SYNTAX: T Starting address of data to be copied. Ending address of data to be copied. Starting address of new location to copy data to. Data can be copied forwards or backwards to any location, even within the source range (eg., shift data up or down one byte) without any problem. An automatic compare is performed for each byte, and mismatches displayed on the screen. Because of the compare feature, it's not recommended you use the T command to copy data into write-only registers (the palette, for example). It works, but all the compares will fail. T 32000 3BFFF 82000 Copies BASIC ROM area to expansion RAM. COMMAND: V PURPOSE: Verify (compare) a disk file with the memory contents. SYNTAX: V <"filename"> [,device [,load_address] ] <"filename"> Is a filename in quotes. [device] Is a number indicating the device the file is on. [load_address] Optional load address. If not given, the file is compared to memory at the 16-bit address stored on disk (always RAM bank 0). The Verify command causes a file to be read and compared to memory. If the load address (including BANK) is given, the data read is compared to data there. Otherwise the data read is compared to RAM bank 0 at the 16-bit load address specified by the first two bytes of the PRG (program) type file. If there is a mismatch, the message 'VERIFYING ERROR' is displayed. If the data matches, nothing is displayed. An error occurs if the compare address overflows the specified bank. V "filename" Compares "filename" from the default system drive to RAM bank 0 at the address read from the file. V "filename",+10,80000 Compares "filename" from drive 10 (notice you must specify decimal for the drive number, or use hex equivalent) to expansion memory bank 8 at address $0000. Note that spaces between parameters after the filename are not permitted. COMMAND: X PURPOSE: Exit to BASIC SYNTAX: X COMMAND: > (greater than) PURPOSE: Pokes data (1 to 16 bytes) into memory SYNTAX: >
[byte]...
Address to start "poking" or displaying [byte] Data to be "poked". If not given, nothing is changed and the memory at that location is "peeked". Successive bytes are poked into successive locations. COMMAND: @ (at sign) PURPOSE: Disk operation: send command, display directory,status SYNTAX: @ [device] [,command] [device] Disk device number [command] Optional command (see DOS manual for specific commands) This command can be used to read a drive's status message, send a drive a DOS command, or display a disk directory. @ displays status of default system drive @9 displays status of drive 9 @+10 or @A displays status of drive 10 @,$ displays directory of default drive @9,$ displays status of drive 9 @,S0:*=SEQ displays all SEQ type files @,S0:FILE sends command to delete file "FILE" 3.3. EDITOR 3.3.1. EDITOR ESCAPE SEQUENCES This section contains a definition of the escape sequences that are present in the C64DX and a brief description of what each does. ESCape sequences are given by hitting the key and then another key. In PRINT strings, escape sequences are given by printing the escape character CHR$(27) followed by another character. In either case, the "other" character is defined as one of the following: KEY FUNCTION --- ---------------------------------------------------- @ Clear from cursor to end of screen A Enable auto-insert mode B Set bottom of screen window at cursor position C Disable auto-insert mode (set overwrite mode) D Delete current line E Set cursor to non-flashing mode F Set cursor to flashing mode G Enable bell (control-G) H Disable bell I Insert line J Move to start of current line K Move to end of current line L Enable scrolling M Disable scrolling N Normal screen fields [not implemented on C64DX] O Cancel insert, quote, reverse, underline & flash modes P Erase from cursor to start of current line Q Erase from cursor to end of current line R Set screen to reverse video [not implemented on C64DX] S Set bold attribute (VIC-III colors 16-31) T Set top of screen window at cursor position U Unset bold attibute V Scroll up W Scroll down X Swap 40/80 column display output device Y Set default tab stops (8 spaces) Z Clear all tab stops [ Set monochrome display (disable attributes) / Cancel insert, quote, rvs, ul & flash modes ] Set color display (enable attributes) 3.3.2. EDITOR CONTROL CODES This section contains a definition of the control codes that are present in the C64DX and a brief description of what each does. Control codes are given by pressing the key at the same time as another key. In PRINT strings, control codes are given by printing the control character with the CHRS() function. Control codes appear within quoted strings as reverse field characters. In any case, the control characters are: CHR$ KEYBOARD VALUE CONTROL FUNCTION ----- -------- ---------------------------------------------- 2 B Underline on 7 G Bell tone 9 I Forward TAB 10 J Line feed 11 K Disable case change C= key (was code 9) 12 L Enable case change C= key (was code 8) 14 N Set display upper/lower case mode 15 O Flash on 17 Q Cursor down 18 R Reverse on 19 S Home cursor 20 T Delete previous character 21 U Backup word 23 W Advance word 24 X Tab set/clear 26 Z Backup TAB 27 [ Escape character 29 ] Cursor right Shifted codes --------------------------------------------------------------------- 130 Underline off 142 Set uppercase/graphic mode 143 Flash off 145 Cursor up 146 Reverse mode off 147 Clear screen 148 Insert one character 157 Cursor left Color codes --------------------------------------------------------------------- 5 white 28 red 30 green 31 blue 129 orange 144 black 149 brown 150 light red 151 light gray 152 medium gray 153 light green 154 light blue 155 dark gray 156 purple 158 yellow 159 cyan Function keys ---------------------------------------------------------------- 3 Stop 16 F9 21 F10 22 F11 23 F12 25 F13 26 F14 131 Run 132 Help 133 F1 134 F3 135 F5 136 F7 137 F2 138 F4 139 F6 140 F8 3.4. KEKNEL 3.4.1. C64DX KERNEL ENTRY POINTS [*** THE FOLLOWING VECTORS AND JUMP TABLES ARE NOT FINAL ***] Where the default indirect vectors point to: FF09 nirq ;IRQ handler FF0B monitor_brk ;BRK handler (Monitor) FF0D nnmi ;NMI handler FF0F nopen ;open FF11 nclose ;close FF13 nchkin ;chkin FF15 nckout ;ckout FF17 nclrch ;clrch FF19 nbasin ;basin FF1B nbsout ;bsout FF1D nstop ;stop key scan FF1F ngetin ;getin FF21 nclall ;clall FF23 monitor_parser ;monitor command parser FF25 nload ;load FF27 nsave ;save FF29 talk ;Low level serial bus routines FF2B listen FF2D talksa FF2F second FF31 acptr FF33 ciout FF35 untalk FF37 unlisten FF39 DOS_talk ;newDOS routines FF3B DOS_listen FF3D DOS_talksa FF3F DOS_second FF41 DOS_acptr FF43 DOS_ciout FF45 DOS_untalk FF47 DOS_unlisten FF49 Get_DOS FF4B Leave_DOS FF4D jmp spin_spout ;setup fast serial port for input or output FF50 jmp close_all ;close all logical files for a given device FF53 jmp c64mode ;reconfigure system as a c/64 (no return!) FF56 jmp monitor_call ;map in Monitor & call it FF59 jmp bootsys ;boot alternate system from disk FF5C jmp phoenix ;call cold start routines, disk boot loader FF5F jmp lkupla ;search tables for given la FF62 jmp lkupsa ;search tables for given sa FF65 jmp swapper ;swap to alternate display device FF68 jmp pfkey ;program function key FF6B jmp setbnk ;set bank for load/save/verify/open FF6E jmp jsr_far ;JSR to any bank, RTS to calling bank FF71 jmp jmp_far ;JMP to any bank FF74 jmp lda_far ;LDA (X),Y from bank Z FF77 jmp sta_far :STA (X),Y to bank Z FF7A jmp cmp_far ;CMP (X),Y to bank Z FF7D jmp primm ;print immediate (always JSR to this routine!) FF80 ;release number of C65 Kernel ($FF=not released) FF81 jmp cint ;init screen editor & display chips FF84 jmp ioinit ;init I/O devices (ports, timers, etc.) FF87 jmp ramtas ;initialize RAM for system FF8A jmp restor ;restore vectors to initial system FF8D jmp vector ;change vectors for user FF90 jmp setmsg ;control OS messages FF93 jmp (isecond) ;send sa after listen FF96 jmp (italksa) ;send sa after talk FF99 jmp memtop ;set/read top of memory FF9C jmp membot ;set/read bottom of memory FF9F jmp key ;scan keyboard FFA2 jmp settmo ;old IEEE set timeout value FFA5 jmp (iacptr) ;read a byte from active serial bus talker FFA8 jmp (iciout) ;send a byte to active serial bus listener FFAB jmp (iuntalk) ;command serial bus device to stop talking FFAE jmp (iunlisten) ;command serial bus device to stop listening FFB1 jmp (ilisten) ;command serial bus device to listen FFB4 jmp (italk) ;command serial bus device to talk FFB7 jmp readss ;return I/O status byte FFBA jmp setlfs ;set la, fa, sa FFBD jmp setnam ;set length and fn adr FFC0 jmp (iopen) ;open logical file FFC3 jmp (iclose) ;close logical file FFC6 jmp (ichkin) ;open channel in FFC9 jmp (ickout) ;open channel out FFCC jmp (iclrch) ;close I/O channel FFCF jmp (ibasin) ;input from channel FFD2 jmp (ibsout) ;output to channel FFD5 jmp load ;load from file FFD8 jmp save ;save to file FFDB jmp Set Time ;set internal clock FFDE jmp Read Time :read internal clock FFE1 jmp (istop) ;scan stop key FFE4 jmp (igetin) ;get char from queue FFE7 jmp (iclall) ;clear all logical files (see close all) FFEA jmp ScanStopKey ;(was increment clock) & scan stop key FFED jmp scrorg ;return current screen window size FFF0 jmp plot ;read/set x,y coord FFF3 jmp iobase ;return I/O base FFF6 c65mode ;C64/C65 interface FFF8 c65mode FFFA nmi ;processor hardware vectors FFFC reset FFFE irq_kernel 3.4.2. C64DX EDITOR JUMP TABLE [*** THE FOLLOWING VECTORS AND JUMP TABLES ARE NOT FINAL ***] E000 cint ;initialize editor & screen E003 disply ;display character in .a, color in .x E006 lp2 ;get a key from IRQ buffer into .a E009 loopS ;get a chr from screen line into .a E00C print ;print character in .a E00F scrorg ;get size of window (rows,cols) in .x, .y E012 keyboard_scan ;scan keyboard subroutine E015 repeat ;repeat key logic & CKIT2 to store decoded key E018 plot ;read or set (.c) cursor position in .x, .y E01B mouse_cmd ;install/remove mouse driver E01E escape ;execute escape function using chr in .a E021 keyset ;redefine a programmable function key E024 editor_irq ;IRQ entry E027 palette_init ;initialize VIC palette E02A swap ;40/80 mode change E02D window ;set top left or bottom right (.c) of window E030 cursor ;turn on or off (.c) soft cursor 3.4.3. C64DX BASIC JUMP TABLE [*** THE FOLLOWING VECTORS AND JUMP TABLES ARE NOT FINAL ***] Format Conversions 7F00 ayint ;convert floating point to integer 7F03 givayf ;convert integer to floating point. 7F06 fout ;convert floating point to ASCII string 7F09 val_1 ;convert ASCII string to floating point 7F0C getadr ;convert floating point to an address 7F0F floatc ;convert address to floating point Math Functions 7F12 fsub ;MEM - FACC 7F15 fsubt ;ARG - FACC 7F18 fadd ;MEM + FACC 7F1B faddt ;ARG - FACC 7F1E fmult ;MEM * FACC 7F21 fmultt ;ARG * FACC 7F24 fdiv ;MEM / FACC 7F27 fdivt ;ARG / FACC 7F2A log ;compute natural log of FACC 7F2D int ;perform BASIC INT() on FACC 7F30 sqr ;compute square root of FACC 7F33 negop ;negate FACC 7F36 fpwr ;raise ARG to the MEM power 7F39 fpwrt ;raise ARG to the FACC power 7F3C exp ;compute EXP of FACC 7F3F cos ;compute COS of FACC 7F42 sin ;compute SIN of FACC 7F45 tan ;compute TAN of FACC 7F48 atn ;compute ATN of FACC 7F4B round ;round FACC 7F4E abs ;absolute value of FACC 7F51 sign ;test sign of FACC 7F54 fcomp ;compare FACC with MEM 7F57 rnd_0 ;generate random floating point number Movement 7F5A conupk ;move RAM MEM to ARG 7F5D romupk ;move ROM MEM to ARG 7F60 movfrm :move RAM MEM to FACC 7F63 movfm :move ROM MEM to FACC 7F66 movmf :move FACC to MEM 7F69 movfa ;move ARG to FACC 7F6C movaf ;move FACC to ARG 7F6F run 7F72 runc 7F75 clear 7F78 new 7F7B link_program 7F7E crunch 7F81 FindLine 7F84 newstt 7F87 eval 7F8A frmevl 7F8D run_a_program 7F90 setexc 7F93 linget 7F96 garba2 7F99 execute_a_line 7F9C chrget 7F9F chrgot 7FA2 chkcom 7FAS frmnum 7FA8 getadr 7FAB getnum 7FAE getbyt 7FB1 plsv Graphic Jump Table 8000 init ;Graphics BASIC init (same as command=0) 8002 parse ;Graphics BASIC command parser 8003 start ;0 commands 8006 screendef ;1 8008 screenopen ;2 800A screenclose ;3 800C screenclear ;4 800E screen ;5 8010 setpen ;6 8012 setpalette ;7 8014 setdmode ;8 8016 setdpat ;9 8018 line ;10 801A box ;11 801C circle ;12 801E polygon ;13 8020 ellipse ;14 8022 viewpclr ;15 8024 copy ;16 8026 cut ;17 8028 paste ;18 802A load ;19 802C char ;20 802E viewportdef ;21 3.4.4. C64DX SOFT VECTORS [*** THE FOLLOWING VECTORS AND JUMP TABLES ARE NOT FINAL ***] BASIC indirect vectors 02F7 jmp USR ;USR vector (must be set by application) 02FC esc_fn_vec ;Escape Function vector 02FE graphic_vector ;Graphic Kernel vector 0300 ierror ;indirect error (output error in .x) 0302 imain ;indirect main (system direct loop) 0304 icrnch ;indirect crunch (tokenization routine) 0306 iqplop ;indirect list (char list) 0308 igone ;indirect gone (char dispatch) 030A ieval ;indirect eval (symbol evaluation) 030C iesclk ;escape token crunch 030E iescpr ;escape token list 0310 iescex ;escape token execute Kernel indirect vectors 02FA iAutoScroll ;AutoScroll used by BASIC, Monitor, Editor 0312 itime ;(unused) 0314 iirq ;IRQ 0316 ibrk ;BRK 0318 inmi ;NMI 031A iopen 031C iclose 031E ichkin 0320 ickout 0322 iclrch 0324 ibasin 0326 ibsout 0328 istop 032A igetin 032C iclall 032E exmon ;Monitor command indirect 0330 iload 0332 isave Editor indirect vectors to routines & tables 0334 ctlvec ;'contrl' characters 0336 shfvec ;'shiftd' characters 0338 escvec ;'escape' characters 033A keyvec ;post keyscan, pre-evaluation of keys 033C keychk ;post-evaluation, pre-buffering of keys 033E decode ;vectors to 6 keyboard matrix decode tables 33E - Mode 1 --> normal keys 340 - Mode 2 --> keys 342 - Mode 3 --> keys 344 - Mode 4 --> keys 346 - Mode 5 --> keys 348 - Mode 6 --> keys 3.4.5. KERNEL DOCUMENTATION The KERNEL is the ROM resident operating system of the Commodore 64DX computer. All input, output, and memory management is controlled by the KERNEL. The KERNEL JUMP TABLE provides a standardized interface to many useful routines within the operating system. Application programmers are encouraged to utilize the JUMP TABLEs to simplify their operations and guarantee their functionality should hardware or, software modifications to the system become necessary. C64 STANDARD KERNEL CALLS The following system calls comprise the set of standard C64 system calls for the C64 class of machines, including the PLUS-4. Several of the calls, however function somewhat differently or may require slightly different setups. This was necessary to accommodate specific features of the system, notably the 40/80 column windowing Editor and banked memory facilities. As with all Kernel calls, the system configuration (BANK $FF) must be in context at the time of the call. C64DX KERNEL JUMP TABLE DESCRIPTIONS 1. $FF81 CINT ; initialize screen editor Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: .A used .X used .Y used Memory: init Editor RAM init Editor I/O Flags: none Example: SEI JSR $FF81 ; initialize screen editor CLI CINT is the Editor's initialization routine. Editor indirect vectors installed, programmable key definitions assigned, and the ASC/DIN key scanned for NATIONAL keyboard/charset determination. CINT sets the VIC bank, VIC nybble bank, enables the character ROM, resets SID volume, and clears the screen. The only thing it does not do that pertains to the Editor which is needed for IRQs (keyscan, VIC cursor blink, split screen modes), key lines, screen background colors, etc. (see IOINIT). Because CINT updates Editor indirect vectors that are used during IRQ processing, you should disable IRQs prior to calling it. CINT utilizes the status byte INIT STATUS as follows: $1104 bit 6 = 0 --> Full initialization. (Set INIT_STATUS bit 6) = 1 --> Partial initialization. (not keymatrix pointers) (not program key definitions) 2. $FF84 IOINIT ; init I/O devices Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: .A used .X used .Y used Memory: initialize I/O Flags: none Example: SEI JSR $FF84 ; initialize system I/O CLI IOINIT is perhaps the major function of the Reset handler. It initializes both CIA's (timers, keyboard serial port, user port), the 4510 port, the VIC chip, the UART and the DOS. It distinguishes a PAL system from an NTSC one and sets PALCNT if PAL. The system IRQ source, the VIC raster, is started (pending IRQs are cleared). IOINIT utilizes the status byte INIT STATUS as follows: $1104 bit 7 = 0 --> Full initialization. (set INIT STATUS bit 7) = 1 --> Partial initialization. You should be sure IRQs are disabled before calling IOINIT to avoid interrupts while the various I/O devices are being initialized. 3. $FF87 RAMTAS ; init RAM and buffers Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: .A used .X used .Y used Memory: initializes RAM Flags: none Example: JSR $FF87 ; initialize system RAM RAMTAS clears all base page RAM, allocates the sets pointers to the top and bottom of system RAM and points the SYSTEM_VECTOR to BASIC cold start. Lastly it sets a flag, DEJAVU, to indicate to other routines that system RAM has been initialized and that the SYSTEM_VECTOR is valid. It should be noted that the C64DX RAMTAS routine does NOT in any way test RAM. 4. $FF8A RESTOR ; init Kernel indirects Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: .A used .X used .Y used Memory: kernel indirects restored Flags: none Example: SEI JSR $FF8A ; restore kernel indirects CLI RESTOR restores the default values of all the Kernel indirect vectors from the Kernel ROM list. It does NOT affect any other vectors, such as those used by the Editor (see CINT) and BASIC. Because it is possible for an interrupt (IRQ or NMI) to occur during the updating of the interrupt indirect vectors, you should disable interrupts prior to calling RESTOR. See also the VECTOR call. 5. $FF8D VECTOR ; init or copy indirects Preparation: Registers: .X = adr (low) of user list .Y = adr (high) of user list Memory: system map Flags: .C = 0 --> load Kernel vectors .C = 1 --> copy Kernel vectors Calls: none Results: Registers: .A used .Y used Memory: as per call Flags: none Example: LDX #save_lo LDY #save_hi SEC JSR $FF87 ; copy indirects to 'save' VECTOR reads or writes the Kernel RAM indirect vectors. Calling VECTOR with the carry status set stores the current contents of the indirect vectors to the RAM address passed in the .X and .Y registers (to the current RAM bank). Calling VECTOR with the carry status clear updates the Kernel indirect vectors from the user list passed in the .X and .Y registers (from the current RAM bank). Interrupts (IRQ and NMI) should be disabled when updating the indirects. See also the RESTOR call. 6. $FF90 SETMSG ; kernel messages on/off Preparation: Registers: .A = message control Memory: system map Flags: none Calls: none Results: Registers: none Memory: MSGFLG updated Flags: none Example: LDA #0 JSR $FF90 ; turn OFF all Kernel messages SETMSG updates the Kernel message flag byte MSGFLG which determines whether system error and/or control messages will be displayed. BASIC normally disables error messages always and disables control messages in 'run' mode. Note that the Kernel error messages are not the verbose ones printed by BASIC, but simply the 'I/O ERROR #' message that you see when in the Monitor, for example. Examples of Kernel control messages are 'LOADING' and 'FOUND'. The MSGFLG control bits are: MSGFLG bit 7 = 1 --> enable CONTROL messages bit 6 = 1 --> enable ERROR messages 7. $FF93 SECND ; serial: send SA after LISTN Preparation: Registers: .A = SA (secondary address) Memory: system map Flags: none Calls: LISTN Results: Registers: .A used Memory: STATUS ($90) Flags: none Example: LDA #8 JSR $FFB1 ; LISTN device 8 LDA #15 JSR $FF93 ; pass it SA #15 SECND is a low-level serial routine used to send a secondary address (SA) to a LISTeNing device (see LISTN Kernel call). An SA is usually used to provide setup information to a device before the actual data I/O operation begins. Attention is released after a call to SECND. SECND is not used to send an SA to a TALKing device (see TKSA). (Most applications should use the higher level I/O routines: see OPEN and CKOUT). 8. $FF96 TKSA ; serial: send SA after TALK Preparation: Registers: .A = SA (secondary address) Memory: system map Flags: none Calls: TALK Results: Registers: .A used Memory: STATUS ($90) Flags: none Example: LDA #8 JSR $FFB4 ; TALK device 8 LDA #15 JSR $FF93 ; pass it SA #15 TKSA is a low-level serial routine used to send a secondary address (SA) to a device commanded to TALK (see TALK Kernel call). An SA is usually used to provide setup information to a device before the actual data I/O operation begins. (Most applications should use the higher level I/O routines: see OPEN and CHKIN). 9. $FF99 MEMTOP ; set/read top of system RAM Preparation: Registers: .X = lsb of MEMSIZ .Y = msb of MEMSIZ Memory: system map Flags: .C = 0 --> set top of memory .C = 1 --> read top of memory Calls: none Results: Registers: .X = lsb of MEMSIZ .Y = msb of MEMSIZ Memory: MEMSIZ Flags: none Example: SEC JSR $FF99 ; get top of user RAM DEY CLC JSR $FF99 ; lower it 1 block MEMTOP is used to read or set the top of system RAM, pointed to by MEMSIZ. This call is included in the C64DX for completeness, but neither the Kernel nor BASIC utilize MEMTOP as it has little meaning in the banked memory environment of the computer (even the RS-232 buffers are permanently allocated). None-the-less, set the carry status to load MEMSIZ into .X and .Y, and clear it to update the pointer from .X and .Y. Note that MEMSIZ references only system RAM. The Kernel initially sets MEMSIZ to $FF00. 10. $FF9C MEMBOT ; set/read bottom of system RAM Preparation: Registers: .X = lsb of MEMSTR .Y = msb of MEMSTR Memory: system map Flags: .C = 0 --> set bot of memory .C = 1 --> read bot of memory Calls: none Results: Registers: .X = lsb of MEMSTR .Y = msb of MEMSTR Memory: MEMSTR Flags: none Example: SEC JSR $FF9C ; get bottom of user RAM_0 INY CLC JSR $FF9C ; raise it 1 block MEMBOT is used to read or set the bottom of system RAM, pointed to by MEMSTR. This call is included in the C64DX for completeness, but neither the Kernel nor BASIC utilize MEMBOT as it has little meaning in the banked memory environment of the C64DX. None-the-less, set the carry status to load MEMSTR into .X and .Y, and clear it to update the pointer from .X and .Y. Note that MEMSTR references only system RAM. The Kernel initially sets MEMSTR to $2000 (BASIC text starts here). 11. $FF9F KEY ; scan keyboard Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: none Memory: keyboard buffer keyboard flags Flags: none Example: JSR $FF9F ; scan the keyboard KEY is an Editor routine which scans the entire keyboard. It distinguishes between shifted and unshifted keys, control keys, and programmable keys, setting keyboard status bytes and managing the keyboard buffer. After decoding the key, KEY will manage such features as toggling cases, pauses or delays, and key repeats. It is normally called by the operating system during the 60Hz IRQ processing. Upon conclusion, KEY leaves the keyboard hardware driving the key-line on which the STOP key is located. There are two indirect RAM jumps encountered during a keyscan: KEYVEC ($33A) and KEYCHK ($33C). KEYVEC (alias KEYLOG) is taken whenever a key depression is discovered, before the key in .A has been decoded. KEYCHK is taken after the key has been decoded, just before putting it into the key buffer. KEYCHK carries the ASCII character in .A, the keycode in .Y, and the shift-key status in .X. The keyboard decode matrices are addressed via indirect RAM vectors as well, located at DECODE. 12. $FFA2 SETTMO ; (reserved) Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: none Memory: TIMOUT Flags: none Example: LDA #value JSR $FFA2 ; update TIMOUT byte SETTMO is unused in the C64DX and is included for compatibility and completeness. It is used in the C64 by the IEEE communication cartridge to disable I/O timeouts. 13. $FFA5 ACPTR ; serial: byte input. Preparation: Registers: none Memory: system map Flags: none Calls: TALK TKSA (if necessary) Results: Registers: .A = data byte Memory: STATUS ($90) Flags: none Example: JSR $FFA5 ; input a byte from serial bus STA data ACPTR is a low-level serial I/O utility to accept a single byte from the current serial bus TALKer using full handshaking. To prepare for this routine a device must first have been established as a TALKer (see TALK) and passed a secondary address if necessary (see TKSA). The byte is returned in .A. (Most applications should use the higher level I/O routines: see BASIN and GETIN). 14. $FFA8 CIOUT ; serial: byte output Preparation: Registers: .A = data byte Memory: system map Flags: none Calls: LISTN SECND (if necessary) Results: Registers: .A used Memory: STATUS ($90) Flags: none Example: LDA data JSR $FFA8 ; send a byte via serial bus CIOUT is a low-level serial I/O utility to transmit a single byte to the current serial bus LISTNer using full handshaking. To prepare for this routine a device must first have been established as a LISTNer (see LISTN) and passed a secondary address if necessary (see SECND). The byte is passed in .A. Serial output data is buffered by one character, with the last character being transmitted with EOI after a call to UNLSN. (Most applications should use the higher level I/O routines; see BSOUT). 15. $FFAB UNTLK ; serial: send untalk Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: .A used Memory: STATUS ($90) Flags: none Example: JSR $FFAB ; UNTALK serial device UNTLK is a low-level Kernel serial bus routine that sends an UNTALK command to all serial bus devices. It commands all TALKing devices to stop sending data. (Most applications should use the higher level I/O routines; see CLRCH). 16. $FFAE UNLSN ; serial: send unlisten Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: .A used Memory: STATUS ($90) Flags: none Example: JSR $FFAE ; UNLISTEN serial device UNLSN is a low-level Kernel serial bus routine that sends an UNLISTEN command to all serial bus devices. It commands all LISTENing devices to stop reading data. (Most applications should use the higher level I/O routines; see CLRCH). 17. $FFB1 LISTN ; serial: send listen command Preparation: Registers: .A = device (0-31) Memory: system map Flags: none Calls: none Results: Registers: .A used Memory: STATUS ($90) Flags: none Example: JSR $FFB1 ; command device to LISTEN LISTN is a low-level Kernel serial bus routine that sends an LISTEN command to the serial bus device in .A. It commands the device to start reading data. (Most applications should use the higher level I/O routines; see CKOUT). 18. $FFB4 TALK ; serial: send talk command Preparation: Registers: .A = device (0-31) Memory: system map Flags: none Calls: none Results: Registers: .A used Memory: STATUS ($90) Flags: none Example: JSR $FFB4 ; command device to TALK TALK is a low-level Kernel serial bus routine that sends an TALK command to the serial bus device in .A. It commands the device to start sending data. (Most applications should use the higher level I/O routines; see CHKIN). 19. $FFB7 READSS ; read I/O status byte Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: .A = STATUS ($90 or $A6) Memory: STATUS cleared if RS-232 ($A6) Flags: none Example: JSR $FFB7 ; STATUS for last I/O READSS (alias READST) returns the status byte associated with the last I/O operation (serial or RS-232) performed. Serial bus and newDOS devices update STATUS ($90) and RS-232 I/O updates RSSTAT ($A6). Note that, to simulate an 6551, RSSTAT is cleared after it is read via READSS. The last I/O operation is determined by the contents of FA ($BA), thus applications which drive I/O devices using the lower-level Kernel calls should not use READSS. 20. $FFBA SETLFS ; set channel LA, FA, SA Preparation: Registers: .A = LA (logical #) .X = FA (device #) .Y = SA (secondary adr) Memory: system map Flags: none Calls: none Results: Registers: none Memory: LA, FA, SA updated Example: see OPEN SETLFS sets the logical file number (LA, $B8), device number (FA, $BA), and secondary address (SA, $B9) for the higher-level Kernel I/O routines. The LA must be unique among OPENed files and is used to identify specific files for I/O operations. The device number range is 0 to 31 and is used to target I/O. The SA is a command to be sent to the indicated device, usually to place it in a particular mode. If the SA is not needed, the .Y register should pass $FF. SETLFS is often used along with SETNAM and SETBNK calls prior to OPENs. See the Kernel OPEN, LOAD, and SAVE calls for examples. 21. $FFBD SETNAM ; set filename pointers Preparation: Registers: .A = string length .X = string adr_low .Y = string adr_high Memory: system map Flags: none Calls: SETBNK Results: Registers: none Memory: FNLEN, FNADR updated Flags: none Example: see OPEN SETNAM sets up the filename or command string for higher-level Kernel I/O calls such as OPEN, LOAD, and SAVE. The string (filename or command) length is passed in .A and updates FNLEN ($B7). The address of the string is passed in .X (low) and Y (high). See the companion call, SETBNK which specifies which RAM bank the string is found. If there is no string, SETNAM should still be called and a null ($00) length specified (the address does not matter). SETNAM is often used along with SETBNK and SETLFS calls prior to OPENs. See the Kernel OPEN, LOAD, and SAVE calls for examples. 22. $FFC0 OPEN ; open logical file Preparation: Registers: none Memory: system map Flags: none Calls: SETLFS, SETNAM, SETBNK Results: Registers: .A = error code (if any) .X used .Y used Memory: setup for I/O STATUS, RSSTAT updated Flags: .C = 1 --> error Example: OPEN 1,8,15,"I0" LDA #length ; fnlen LDX #filename JSR $FFBD ; SETNAM LDX #0 ; fnbank (RAM_0) JSR $FF68 ; SETBNK LDA #1 ; la LDX #8 ; fa LDY #15 ; sa JSR $FFBA ; SETLFS JSR $FFC0 ; OPEN BCS error filename .BYTE 'I0' length = 2 OPEN prepares a logical file for I/O operations. It creates a unique entry in the Kernel logical file tables LAT ($362), FAT ($36C), and SAT ($376) using its index LDTND ($98) and data supplied by the user via SETLFS. There can be up to ten logical files OPENed simultaneously. OPEN performs device specific opening tasks for serial, RS-232, keyboard & screen, devices, including clearing the previous status and transmitting any given filename or command string supplied by the user via SETNAM and SETBNK. The I/O status will be updated appropriately and can be read via READSS. The path to OPEN is through an indirect RAM vector at $31A. Applications may therefore provide their own OPEN procedures or supplement the system's by re-directing this vector to their own routine. 23. $FFC3 CLOSE ; close logical file Preparation: Registers: .A = LA (logical #) Memory: system map Flags: .C (see text below) Calls: none Results: Registers: .A = error code (if any) .X used .Y used Memory: logical tables updated STATUS, RSSTAT updated Flags: .C = 1 --> error Example: LDA #1 ; la JSR $FFC3 ; CLOSE BCS error CLOSE removes the logical file (LA) passed in .A from the logical file tables and performs device specific closing tasks. Keyboard, screen, and any unOPENed files pass through. RS-232 devices are not closed until all buffered data has been transmitted. Serial files are closed by transmitting a 'close' command (if an SA was given when it was opened), sending any, buffered character, and UNLiSTeNing the bus. There is a special provision incorporated into the CLOSE routine of systems featuring BASIC DOS command. If the following conditions are all TRUE, a full CLOSE is NOT performed: the table entry is removed but a 'close' command is NOT transmitted to the device. This allows the disk command channel to be properly OPENed and CLOSEd without the disk operating system closing ALL files on its end: .C = 1 --> indicates special CLOSE FA >=8 --> device is a disk SA =15 --> command channel The path to CLOSE is through an indirect RAM vector at $31C. Applications may therefore provide their own CLOSE procedures or supplement the system's by re-directing this vector to their own routine. 24. $FFC6 CHKIN ; set input channel Preparation: Registers: .X = LA (logical #) Memory: system map Flags: none Calls: OPEN Results: Registers: .A = error code (if any) .X used .Y used Memory: LA, FA, SA, DFLTN STATUS, RSSTAT updated Flags: .C = 1 --> error Example: LDX #1 ; la JSR $FFC6 ; CHKIN BCS error CHKIN establishes an input channel to the device associated with the logical address (LA) passed in .X, in preparation for a call to BASIN or GETIN. The Kernel variable DFLTN ($99) is updated to indicate the current input device and the variables LA, FA, and SA are updated with the file's parameters from its entry in the logical file tables (put there by OPEN). CHKIN performs certain device specific tasks: screen and keyboard channels pass through, and serial channels are sent a TALK command and the SA transmitted (if necessary). Call CLRCH to restore normal I/O channels. CHKIN is required for all input except the keyboard. If keyboard input is desired and no other input channel is established, you do not need to call CHKIN or OPEN. The keyboard is the default input device for BASIN and GETIN. The path to CHKIN is through an indirect RAM vector at $31E. Applications may therefore provide their own CHKIN procedures or supplement the system's by re-directing this vector to their own routine. 25. $FFC9 CKOUT ; set output channel Preparation: Registers: .X = LA (logical #) Memory: system map Flags: none Calls: OPEN Results: Registers: .A = error code (if any) .X used .Y used Memory: LA, FA, SA, DFLTO STATUS, RSSTAT updated Flags: .C = 1 --> error Example: LDX #1 ; la JSR $FFC9 ; CKOUT BCS error CKOUT establishes an output channel to the device associated with the logical address (LA) passed in .X, in preparation for a call to BSOUT. The Kernel variable DFLTO ($9A) is updated to indicate the current output device and the variables LA, FA, and SA are updated with the file's parameters from its entry in the logical file tables (put there by OPEN). CKOUT performs certain device specific tasks: keyboard channels are illegal, screen channels pass through, and serial channels are sent a LISTN command and the SA transmitted (if necessary). Call CLRCH to restore normal I/O channels. CKOUT is required for all output except the screen. If screen output is desired and no other output channel is established, you do not need to call CKOUT or OPEN. The screen is the default output device for BSOUT. The path to CKOUT is through an indirect RAM vector at $320. Applications may therefore provide their own CKOUT procedures or supplement the system's by re-directing this vector to their own routine. 26. $FFCC CLRCH ; restore default channels Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: .A used .X used Memory: DFLTI, DFLTO updated Flags: none Example: JSR $FFCC ; restore default I/O CLRCH (alias CLRCHN) is used to clear all open channels and restore the system default I/O channels after other channels have been established via CHKIN and/or CHKOUT. The keyboard is the default input device and the screen is the default output device. If the input channel was to a serial device, CLRCH first UNTLKs it. If the output channel was to a serial device, it is UNLiSteNed first. The path to CLRCH is through an indirect RAM vector at $322. Applications may therefore provide their own CLRCH procedures or supplement the system's by re-directing this vector to their own routine. 27. $FFCF BASIN ; input from channel Preparation: Registers: none Memory: system map Flags: none Calls: CHKIN (if necessary) Results: Registers: .A = character (or error code) Memory: STATUS, RSSTAT updated Flags: .C = 1 if error Example: LDY #0 ; index more JSR $FFCF ; input a character STA data,Y ; buffer it INY CMP #$0D ; carriage return? BNE more BASIN (alias CHRIN) reads a character from the current input device (DFLTN, $99) and returns it in .A. Input from devices other than the keyboard (the default input device) must be OPENed and CHKINed. The character is read from the input buffer associated with the current input channel: 1. RS-232 data is returned a character at a time from the RS-232 input buffer, waiting until a character is received if necessary. If RSSTAT is bad from a prior operation, input is skipped and null input (carriage return) is substituted. 2. Serial data is returned a character at a time directly from the serial bus, waiting until a character is sent if necessary. If STATUS ($90) is bad from a prior operation, input is skipped and null input (carriage return) is substituted. 3. Screen data is read from screen RAM starting at the current cursor position and ending with a faked carriage return at the end of the logical screen line. 4. Keyboard data is input by turning on the cursor reading characters from the keyboard buffer and echoing them on the screen until a carriage return is encountered. Characters are then returned one at a time from the screen until all characters input have been passed, including the carriage return. Any calls after the eol will start the process over again. The path to BASIN is through an indirect RAM vector at $324. Applications may therefore provide their own BASIN procedures or supplement the system's by re-directing this vector to their own routine. 28. $FFD2 BSOUT ; output to channel Preparation: Registers: .A = character Memory: system map Flags: none Calls: CKOUT (if necessary) Results: Registers: .A = error code (if any) Memory: STATUS, RSSTAT updated Flags: .C = 1 if error Example: LDA #character JSR $FFD2 ; output a character BSOUT (alias CHROUT) writes the character in .A to the current output device (DFLTO, $9A). Output to devices other than the screen (the default output device) must be OPENed and CKOUTed. The character is written to the output buffer associated with the current output channel: 1. RS-232 data is put a character at a time into the RS-232 output buffer, waiting until there is room if necessary. 2. Serial data is passed to CIOUT which buffers one character and sends the previous character. 3. Screen data is put into screen RAM at the current cursor position. 4. Keyboard output is illegal. The path to BSOUT is through an indirect RAM vector at $326. Applications may therefore provide their own BSOUT procedures or supplement the system's by re-directing this vector to their own routine. 29. $FFD5 LOAD ; load from file Preparation: Registers: .A = 0 --> LOAD .A > 0 --> VERIFY .X = load adr_lo (if SA=0) .Y = load adr_hi (if SA=0) Memory: system map Flags: none Calls: SETLFS, SETNAM, SETBNK Results: Registers: .A = error code (if any) .X = ending adr_lo .Y = ending adr_hi Memory: per command STATUS updated Flags: .C = 1 --> error Example: LOAD "program",8,1 LDA #length ; fnlen LDX #filename JSR $FFBD ; SETNAM LDA #0 ; load/verify bank (RAM_0) LDX #0 ; fnbank (RAM_0) JSR $FF68 ; SETBNK LDA #0 ; la (not used) LDX #8 ; fa LDY #$FF ; sa (SA>0 normal load) JSR $FFBA ; SETLFS LDA #0 ; load, not verify LDX #load_adr ; (used only if SA=0) JSR $FFD5 ; LOAD BCS error STX end_lo STY end_hi filename .BYTE 'program' length = 7 This routine LOADs data from an input device into memory. It can also be used to VERIFY that data in memory matches that in a file. LOAD performs device specific tasks for serial LOADs. You cannot LOAD from RS-232 devices, the screen, or the keyboard. While LOAD performs all the tasks of an OPEN, it does NOT create any logical files as an OPEN does. Also note that LOAD cannot 'wrap' memory banks. As with any I/O, the I/O status is updated appropriately and can be read via READSS. LOAD has two options that the user must select: 1. LOAD vs. VERIFY: the contents of .A passed at the call to LOAD determines which mode is in effect. If .A is zero, a LOAD operation will be performed and memory will be overwritten. If .A is non-zero, a VERIFY operation will be performed and the result passed via the error mechanism. 2. LOAD ADDRESS: the secondary address (SA) setup by the call to SETLFS determines where the LOAD starting address comes from. If the SA is zero, the user wants the address in .X and .Y at the time of the call to be used. If the SA is non-zero, the LOAD starting address is read from the file header itself and the file loaded to the same place from which it was SAVEd. The serial LOAD routine automatically attempts to access a newDOS drive, then attempts to BURST load a file, and resorts to the normal load mechanism (but still using the FAST serial routines) if the BURST handshake is not returned. The path to LOAD is through an indirect RAM vector at $330. Applications may therefore provide their own LOAD procedures or supplement the system's by re-directing this vector to their own routine. 30. $FFD8 SAVE ; save to file Preparation: Registers: .A = pointer to start adr .X = end_adr_lo .Y = end_adr_hi Memory: system map Flags: none Calls: SETLFS, SETNAM, SETBNK Results: Registers: .A = error code (if any) .X = used .Y = used Memory: STATUS updated Flags: .C = 1 --> error Example: SAVE "program",8 LDA #length ; fnlen LDX #filename JSR $FFBD ; SETNAM LDA #0 ; save from bank (RAM_0) LDX #0 ; fnbank (RAM_0) JSR $FF68 ; SETBNK LDA #0 ; la (not used) LDX #8 ; fa LDY #0 ; sa (cassette only) JSR $FFBA ; SETLFS LDA #start ; pointer to start address LDX end ; ending address lo LDY end+1 ; ending address hi JSR $FFD8 ; SAVE BCS error filename .BYTE 'program' length = 7 start .WORD address1 ; page_0 end .WORD address2 This routine SAVEs data from memory to an output device. SAVE performs device specific tasks for serial SAVEs. You cannot SAVE from RS-232 devices, the screen, or the keyboard. While SAVE performs all the tasks of an OPEN, it does NOT create any logical files as an OPEN does. The starting address of the area to be SAVEd must be placed in a base-page vector and the address of this vector passed to SAVE in .A at the time of the call. The address of the last byte to be SAVEd PLUS ONE is passed in .X and .Y at the same time. SAVE first attempts to access a newDOS drive. There is no BURST save: the normal FAST serial routines are used. As with any I/O, the I/O status will be updated appropriately and can be read via READSS. The path to SAVE is through an indirect RAM vector at $332. Applications may therefore provide their own SAVE procedures or supplement the system's by re-directing this vector to their own routine. 31. $FFDB SETTIM ; set internal clock Preparation: Registers: .A = hours .X = minutes .Y = seconds .Z = tenths Memory: system map Flags: none Calls: none Results: Registers: none Memory: TOD at CIA $DC00 updated Flags: none Example: LDA #0 ; reset clock TAX TAY TAZ JSR $FFDB ; SETTIM SETTIM sets the system CIA 24-hour TOD clock, which counts tenths of a second and automatically wraps at the 24-hour point. 32. $FFDE RDTIM ; read internal clock Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: .A = hours .X = minutes .Y = seconds .Z = tenths Memory: none Flags: none Example: JSR $FFDE ; RDTIM RDTIM reads the system CIA 24-hour TOD clock, which counts tenths of a second. The timer is automatically wrapped at the 24-hour point. 33. $FFE1 STOP ; scan stop key Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: .A = last keyboard row .X = used (if STOP key) Memory: none Flags: status valid Example: JSR $FFE1 ; scan STOP key BEQ stop ; branch if down STOP checks a Kernel variable STKEY ($91), which is updated by UDTIM during normal IRQ processing and contains the last scan of keyboard column C7. The STOP key is bit-7, which will be zero if the key is down. If it is, default I/O channels are restored via CLRCH and the keyboard queue is flushed by resetting NDX ($D0). The keys on keyboard line C7 are: bit: 7 6 5 4 3 2 1 0 key: STOP Q C= SPACE 2 CTRL <-- 1 The path to STOP is through an indirect RAM vector at $328. Applications may therefore provide their own STOP procedures or supplement the system's by re-directing this vector to their own routine. 34. $FFE4 GETIN ; read buffered data Preparation: Registers: none Memory: system map Flags: none Calls: CHKIN (if necessary) Results: Registers: .A = character (or error code) .X used .Y used Memory: STATUS, RSSTAT updated Flags: .C = 1 if error Example: wait JSR $FFE4 ; get any key BEQ wait STA character GETIN reads a character from the current input device (DFLTN $99) buffer and returns it in .A. Input from devices other than the keyboard (the default input device) must be OPENed and CHKINed. The character is read from the input buffer associated with the current input channel: 1. Keyboard input: a character is removed from the keyboard buffer and passed in .A. If the buffer is empty, a null ($00) is returned. 2. RS-232 input: a character is removed from the RS-232 input buffer and passed in .A. If the buffer is empty, a null ($00) is returned (use READSS to check validity). 3. Serial input: GETIN automatically jumps to BASIN. See BASIN serial I/O. 4. Screen input: GETIN automatically jumps to BASIN. See BASIN serial I/O. The path to GETIN is through an indirect RAM vector at $32A. Applications may therefore provide their own GETIN procedures or supplement the system's by re-directing this vector to their own routine. 35. $FFE7 CLALL ; close all files and channels Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: .A used .X used Memory: LDTND, DFLTN, DFLTO updated Flags: none Example: JSR $FFE7 ; close files CLALL deletes all logical file table entries by resetting the table index, LDTND ($98). It clears current serials channels (if any) and restores the default I/O channels via CLRCH. The path to CLALL is through an indirect RAM vector at $32C. Applications may therefore provide their own CLALL procedures or supplement the system's by re-directing this vector to their own routine. 36. $FFEA ScanStopKey (was UDTIM, which has no purpose on C64DX) Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: .A used .X used Memory: TIME, TIMER, STKEY updated Flags: none Example: JSR $FFEA ; ScanStopKey Scans key line C7, on which the STOP key lies, and stores the result in STKEY ($91). The Kernel routine STOP utilizes this variable. 37. $FFED SCRORG ; get current screen window size Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: .A = screen width .X = window width .Y = window height Memory: none Flags: none Example: JSR $FFED ; SCRORG SCRORG returns active window's size (maximum row & column #) & origin. entry: nothing required. exit: .C = max. screen width (0=80, 1=40) default = 0 .X = max. column number (# columns - 1) default = 79 .Y = max. line number (# lines minus 1) default = 24 .A = window address (home position), low default = $0800 .Z = window address, high 38. $FFF0 PLOT ; read/set cursor position Preparation: Registers: .X = cursor line .Y = cursor column Memory: system map Flags: .C = 0 --> set cursor position .C = 1 --> get cursor position Calls: none Results: Registers: .X = cursor line .Y = cursor column Memory: TBLX, PNTR updated Flags: .C = 1 --> error PLOT Reads or sets the cursor position within current window. Entry: .C = 1 Returns the cursor position (.Y=column, .X=line) relative to the current window origin (NOT screen origin). .C = 0 Sets the cursor position (.Y=column, .X=line) relative to the current window origin (NOT screen origin). Exit: When reading position, .X=line, .Y=column, .C=1 if wrapped line. When setting new position, .X=line, .Y=column, and .C = 0 Normal exit. The cursor has been moved to the position contained in .X & .Y relative to window origin (see SCRORG). .C = 1 Error exit. The requested position was outside the current window (see SCRORG). The cursor has not been moved. When called with the carry status set, PLOT returns the current cursor position relative to the current window origin (NOT screen origin). When called with the carry status clear, PLOT attempt to move the cursor to the indicated line and column relative to the current window origin (NOT screen origin). PLOT will return a clear carry status if the cursor was moved, and a set carry status if the requested position was outside the current window (NO CHANGE has been made). Editor variables that are useful: SCBOT - $E4 --> window bottom SCTOP - $E5 --> window top SCLF - $E6 --> window left side SCRT - $E7 --> window right side TBLX - $EC --> cursor line PNTR - $ED --> cursor column LINES - $EE --> maximum screen height COLUMNS $EF --> maximum screen width 39. $FFF3 IOBASE ; read base address of I/O block Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: .X = lsb of I/O block .Y = msb of I/O block Memory: none Flags: none Example: JSR $FFF3 ; find the I/O block IOBASE is unused in the C64DX and is included for compatibility and completeness. It returns the address of the I/O block in .X and .Y. NEW C64DX KERNEL CALLS The following system calls comprise a set of extensions to the standard CBM jump table. They are specifically for the C64DX machine and as such should not be considered as permanent additions to the standard jump table. With the exception of C64MODE they are all true subroutines and will terminate via RTSs. As with all Kernel calls, the system configuration (BANK $FF) must be in context at the time of the call. 1. $FF4D SPIN_SPOUT ;setup fast serial ports for I/O Preparation: Registers: none Memory: system map Flags: .C = 0 --> select SPINP .C = 1 --> select SPOUT Calls: none Results: Registers: .A used Memory: CIA_1, FSDIR register Flags: none Example: CLC JSR $FF4D ;setup for fast serial input The fast serial protocol utilizes CIA_1 (6526 at $DC00) and a special driver circuit controlled in part by the FSDIR register. SPINP and SPOUT are routines used by the system to set up the CIA and fast serial driver circuit for input or output. SPINP sets up CRA (CIA_1 register 14) and clears the FSDIR bit for input. SPOUT sets up CRA, ICR (CIA_1 register 13), timer A (CIA_l registers 4 & 5), and sets the FSDIR bit for output. Note the state of the TOD_IN bit of CRA is always preserved. These routines are required only by applications driving the fast serial bus themselves from the lowest level. 2. $FF50 CLOSE_ALL ;close all files on a device Preparation: Registers: .A --> device # (FA: 0-31) Memory: system map Flags: none Calls: none Results: Registers: .A used .X used .Y used Memory: none Flags: none Example: LDA #$08 JSR $FF50 ; close all files on device 8 The FAT is searched for the given FA. A proper CLOSE is performed for all matches. If one of the CLOSEd channels is the current I/O channel then the default channel is restored. This call is utilized, for example, by the BASIC command DCLOSE. 3. $FF53 C64MODE ;reconfigure system as a C64 Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: none Memory: none Flags: none Example: JMP $FF53 ;switch to C64 mode THERE IS NO RETURN FROM THIS ROUTINE. The system downloads code to RAM which reMAPs the system to put the C64 ROM in context, resets all VIC-III modes, and jumps to the C64 start routine. Return to C65 mode is by resetting the machine, although a program could do it very easily. A vector on the C64 side is provided to restart C64DX mode. 4. $FF56 MonitorCall ;enter Monitor mode Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: none Memory: none Flags: none Turns off BASIC receipt of IRQ, maps BASIC out, maps the Monitor in, and calls it. When the Monitor is exited, the system is restored, BASIC mapped in, and the system_vector taken (usually points to BASIC warm start entry). 5. $FF59 BOOT_SYS ;boot an alternate OS from disk Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: undefined Memory: undefined Flags: undefined Boot an alternate system. Reads the "home" sector of any diskette (physical track 0 sector 1, 512 bytes) into memory at $00400, turns off BASIC, and JMPs to it. Nothing done if disk not present. JMP not made if first byte is not $4C. It forces the "system" memory map, not user environment. No support for C128-style BOOT sector. Not related to BASIC 10.0 BOOT command, which RUNs a BASIC program called "AUTOBOOT.C65*" if found. 6. $FF5C PHOENIX ;???? C64DX diagnostics ???? Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: undefined Memory: undefined Flags: none Example: JSR $FF5C ;PHOENIX Not same thing as C128 Phoenix routine. In the C65 development system, this routine is called after BASIC inits and performs some system diagnostics, displaying results on the screen. 7. $FF5F LKUPLA ;search tables for given la 8. $FF62 LKUPSA ;search tables for given sa Preparation: Registers: .A = LA (logical file number) if LKUPLA .Y = SA (secondary address) if LKUPSA Memory: system map Flags: none Calls: none Results: Registers: .A = LA (only if found) .X = FA (only if found) .Y = SA (only if found) Memory: none Flags: .C = 0 if found .C = 1 if not found Example: LDY #$60 ;find an available SA again INY CPY #$6F BCS too_many ;too many files open JSR $FF62 ;LKUPSA BCC again ;get another if in use LKUPLA and LKUPSA are Kernel routines used primarily by BASIC DOS commands to work around a user's open disk channels. The Kernel requires unique logical device numbers (LAs) and the disk requires unique secondary addresses (SAs), therefore BASIC must find alternative unused values whenever it needs to establish a disk channel. 9. $FF65 SWAPPER ;switch between 40 & 80 column modes Preparation: Registers: none Memory: system map Flags: none Calls: none Results: Registers: .A used .X used .Y used Memory: screen cleared Flags: none Example: LDA $D7 ;check display mode BMI if_80 ;branch if 80 column JSR $FF5F ;switch from 40 to 80 MODE, location $D7, is toggled by SWAPPER to indicate the current display mode: $80 = 80-column, $00 = 40-column. Because they are both VIC screens, changing them requires clearing the screens since they share the same memory location. 10. $FF68 PFKEY ;program a function key Preparation: Registers: .A = pointer to string adr (lo/hi/bank) .Y = string length .X = key number (1-16) Memory: system map Flags: none Calls: none Results: Registers: .A used .X used .Y used Memory: PKYBUF, PKYDEF tables updated Flags: .C = 0 if successful .C = 1 if no room Example: LDA #$FA ;pointer to string address LDY #6 ;length LDX #15 ;key # ('HELP' key) JSR $FF68 ;install new key def'n BCS no_room >000FA 00 13 00 ;ptr to $1300 bank 0 >01300 53 54 52 49 4E 47 ;'string' PFKEY (alias KEYSET) is an Editor utility to replace a function key string with a user's string. Keys 1-14 are F1-F14, 15 is the HELP key, and 16 is the RUN string. The example above replaces the 'help' string assigned at system initialization to the HELP key with the string 'string'. Both the key length table, PKYBUF ($1000-$100F), and the definition area, PKYDEF ($1010-$10FF) are compressed and updated. The maximum length of all 16 strings is 240 characters. No change is made if there is insufficient room for a new definition. 11. $FF6B SETBNK ;set bank for I/O operations ;and filename Preparation: Registers: .A = BA, memory bank (0-FF) .X = FNBANK, filename bank Memory: system map Flags: none Calls: SETNAM Results: Registers: none Memory: BA, FNBANK updated Flags: none Example: see OPEN SETBNK is a prerequisite for any memory I/O operations and must be used along with SETLFS and SETNAM prior to OPENing files, etc. BA ($C6) sets the current 64KB memory bank for LOAD/SAVE/ VERIFY operations. FNBANK ($C7) indicates the bank in which the filename string is found. The Kernel routine SETBNK is often used along with SETNAM and SETLFS calls prior to OPENs. See the Kernel OPEN, LOAD, and SAVE calls for examples. 12. $FF6E JSRFAR ;gosub in another bank 13. $FF71 JMPFAR ;goto another bank Preparation: Registers: none Memory: system map, also: $02 --> bank (0-FF) $03 --> PC_high $04 --> PC_low $05 --> .S (status) $06 --> .A $07 --> .X $08 --> .Y $09 --> .Z Flags: none Calls: none Results: Registers: none Memory: as per call, also: $05 --> .S (status) $06 --> .A $07 --> .X $08 --> .Y $09 --> .Z Flags: none The two routines, JSRFAR and JMPFAR, enable code executing in the system bank of memory to call (or JMP to) a routine in any other bank. In the case of JSRFAR, the called routine must restore the system map before executing a return. JSRFAR calls JMPFAR. Both are RAM routines, located at $39C and $3B1 respectively. The user should take necessary precautions when calling a non-system bank that interrupts (IRQs & NMIs) will be handled properly (or disabled beforehand). 14. $FF74 LDA_FAR ;LDA (.X),Y from bank .Z Preparation: Registers: .X = pointer to base page pointer .Y = index .Z = bank (0-FF) Memory: setup indirect vector Flags: none Calls: none Results: Registers: .A = data Memory: DMA_LIST updated Flags: status valid LDA_FAR enables applications to read data from any other bank. It builds a DMA_LIST to fetch one byte, executes the DMA, and reads the byte. It's a ROM routine. 15. $FF77 STA_FAR ;STA (.X),Y from bank .Z Preparation: Registers: .A = data .X = pointer to base page pointer .Y = index .Z = bank (0-FF) Memory: setup indirect vector Flags: none Calls: none Results: Registers: .X used Memory: DMA_LIST updated Flags: status invalid STA_FAR enables applications to write data to any other bank. It builds a DMA_LIST to stash one byte, and executes the DMA. It's a ROM routine. 16. $FF7A CMP_FAR ;CMP (.X),Y from bank .Z Preparation: Registers: .A = data .X = pointer to a base page pointer .Y = index .Z = bank (0-FF) Memory: setup indirect vector Flags: none Calls: none Results: Registers: .X used Memory: none Flags: status valid CMP_FAR enables applications to compare data to any other bank. It builds calls LDA_FAR and compares the given byte with the byte fetched. It's a ROM routine. 17. $FF7D PRIMM ;print immediate utility Preparation: Registers: none Memory: none Flags: none Calls: none Results: Registers: none Memory: none Flags: none Example: JSR $FF7D ;display following text .BYTE 'message' .BYTE $00 ;terminator JMP continue ;execution resumes here PRIMM is a Kernel utility used to print (to the default output device) a PETSCII string which immediately follows the call. The string must be no longer than 255 characters and be terminated by a null ($00) character. It cannot contain any embedded null characters. Because PRIMM uses the system stack to find the string and a return address, you MUST NOT JMP to PRIMM. There must be a valid address on the stack. 3.4.6. BASIC 10.0 MATH PACKAGE This document details the many user-callable routines available in the C64DX BASIC 10.0 math package. FLOATING POINT MATH PACKAGE CONVENTIONS In BASIC memory the number is PACKED and looks like this: +--------+---------+--------+--------+-----+ | signed | B7=SGN | | | | | EXP +---------+ M A N T I S S A | LSB | | +$80 | MSB | | | | +--------+---------+--------+--------+-----+ Because the mantissa is normalized such that its msb is always 1, BASIC stores the SIGN of the mantissa here to save a byte of storage. It must be normalized when put in the FACC (see CONUPK). In the FACC the NORMALIZED number looks like this: $63 $64 $65 $66 $67 $68 FACEXP FACHO FACMOH FACMO FACLO FACSGN +--------+---------+--------+--------+-----+-------+ | signed | BIT 7=1 | | | | SIGN | | EXP +---------+ M A N T I S S A | LSB |+ = $00| | +$80 | MSB | | | |- = $00| +--------+---------+--------+--------+-----+-------+ Negative exponents are not stored 2's complement. The maximum exponent is 10^38 ($FF) and the minimum is 10^-39 ($01). A zero value for the exponent means the number is zero. Since the exponent is a power of 2, it can be described as the number of left (EXP>$80) or right (EXP<=$80) shifts to be performed on the normalized mantissa to create the binary representation of the value. There is a second floating accumulator called ARG which has the same layout. It is located at $6A through $6F. Throughout the math package the floating point format is: * the mantissa is 24 bits long. * the binary point is to the left of the msb. * the mantissa is always positive, and its msb is always 1. * number = mantissa * 2^exponent, sign in FACSGN. * the sign of the exponent is the msb of the exponent. * the exponent is stored in excess $80 (i.e., it is a signed 8-bit number with $80 added to it.) * an exponent of zero means the number is zero. (Note that the rest of the accumulator cannot be assumed to be zero.) * to keep the same number in the accumulator while shifting: right shifts --> increment exponent left shifts --> decrement exponent Arithmetic routine calling conventions: * For one argument functions: the argument is in the FACC. the result is left in the FACC. * For two argument operations: the first argument is in MEMORY (packed) or ARG (unpacked). the second argument is in the FACC. the result is left in the FACC. * Always call ROM routines with SYSTEM memory in context (BANK $FF). A note concerning precision. Since the mantissa is always normalized, the high order bit of the most significant byte is always one. This guarantees at least 40 bits (5 byte mantissa times 8 bits each) of precision, which is approximately 9 significant digits plus a few bits for rounding. In fact, there is a 'rounding' byte, FACOV ($71), which should, for the greatest degree of precision, be loaded whenever you load the FACC. The high order bit of FACOV is utilized in most of the math routines. While some of the 'movement' routines 'round' the loaded floating point number (i.e., FACOV = $00), others (such as CONUPK) do not - assuming the value of FACOV is the useful result of an operation in progress. In 99% of the cases you need not worry about it, as its significance is virtually nil. For the greatest degree of precision however, use it. A few examples of normalized (FACC) floating point numbers: VALUE EXP M A N T I S S A SIGN ------- ----- -------------------------- ---- 1E38 = FF 96 76 99 53 00 4E10 = A4 95 02 F9 00 00 2E10 = A3 95 02 F9 00 00 1E10 = A4 95 02 F9 00 00 10 = 84 A0 00 00 00 00 1 = 81 80 00 00 00 00 .5 = 80 80 00 00 00 00 .25 = 7F 80 00 00 00 00 .6 = 80 99 99 99 9A 00 1E-04 = 73 D1 B7 59 59 00 1E-37 = 06 88 1C EA 15 00 1E-38 = 02 D9 C7 DC EE 00 3E-39 = 01 82 AB 1E 2A 00 0 = 00 xx xx xx xx 00 -1 = 81 80 00 00 00 FF -5 = 83 A0 00 00 00 FF Now for a simple example of deriving the actual binary from the FACC: 5 = 83 A0 00 00 00 00 | \ | \ ($83-$80) ($A0) | which means: 2^3 * .10110000, or shift mantissa LEFT 3, which gives: 101.00000 (binary) or 5.0 (hex) =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: AYINT FUNCTION: CONVERT FLOATING POINT TO INTEGER PREPARATION: FACC contains floating point number (-32768<=n<=32767) RESULT: FACMO ($66) contains signed integer (msb) FACLO ($67) contains signed integer (lsb) ERROR: ?ILLEGAL QUANTITY ERROR if FACC too big. EXAMPLE: JSR AYINT ;INT(FACC) LDA $66 ;MSB LDA $67 ;LSB =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: GIVAYF FUNCTION: CONVERT INTEGER TO FLOATING POINT PREPARATION: .A contains signed integer (msb) .Y contains signed integer (lsb) RESULT: FACC contains floating point number EXAMPLE: LDA #>INTEGER LDY #POINTER STA INDEX1 ;SET POINTER TO STRING STY INDEX1+1 LDA #LENGTH ;SET STRING LENGTH JSR VAL_1 ;FACC = VAL(STRING) =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: GETADR FUNCTION: CONVERT FLOATING POINT TO ADDRESS PREPARATION: FACC contains floating point number (0<=n<=65535) RESULT: POKER ($16,$17) contains unsigned integer address ERROR: ?ILLEGAL QUANTITY ERROR if FACC too big. EXAMPLE: JSR GETADR ;ADR(FACC) LDA $16 ;LSB LDA $17 ;MSB =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: FLOATC FUNCTION: CONVERT ADDRESS TO FLOATING POINT PREPARATION: FACHO ($64) contains address (msb) FACMOH ($65) contains address (lsb) .X contains exponent ($90 always) .C=1 if positive (always) RESULT: FACC contains floating point number ERROR: ?OVERFLOW ERROR if FACC too big. EXAMPLE: LDA #
ADDRESS STA FACMOH ;SET ADDRESS STY FACHO LDY #$90 ;EXPONENT SEC ;POSITIVE JSR FLOATC ;FLOAT ADDRESS =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: FSUB FUNCTION: FACC = MEMORY - FACC PREPARATION: FACC contains floating point subtrahend .A = pointer (lsb) to packed floating point minuend .Y = pointer (msb) to packed floating point minuend SPECIAL NOTES: The minuend *MUST* be in VARBANK in packed format. FSUB calls CONUPK to normalize it. RESULT: FACC contains floating point difference ERROR: ?OVERFLOW ERROR if FACC too big. EXAMPLE: LDA #POINTER ;SET POINTER TO *PACKED* MINUEND JSR FSUB ;SUBTRACT MEM FROM FACC, DIFF IN FACC =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: FSUBT FUNCTION: FACC = ARG - FACC PREPARATION: FACC contains floating point subtrahend ARG contains floating point minuend SPECIAL NOTES: This routine is similar to FSUB. The only difference is the call to CONUPK. (FSUBT assumes you have already loaded ARG with unpacked minuend.) RESULT: FACC contains floating point difference ERROR: ?OVERFLOW ERROR if FACC too big. EXAMPLE: JSR FSUBT ;SUBTRACT ARG FROM FACC, DIFF IN FACC =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: FADD FUNCTION: FACC = MEMORY + FACC PREPARATION: FACC contains floating point addend .A = pointer (lsb) to packed floating point addend .Y = pointer (msb) to packed floating point addend SPECIAL NOTES: The second addend *MUST* be in VARBANK in packed format. FADD calls CONUPK to normalize it. RESULT: FACC contains floating point sum ERROR: ?OVERFLOW ERROR if result too big EXAMPLE: LDA #POINTER ;SET POINTER TO *PACKED* ADDEND JSR FADD ;ADD MEMORY TO FACC, SUM IN FACC =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: FADDT FUNCTION: FACC = ARG + FACC PREPARATION: FACC contains floating point addend ARG contains floating point addend ARISGN ($70) contains EOR(FACSGN,ARGSGN) .A contains FACEXP SPECIAL NOTES: This routine is similar to FADD. The only difference is the call to CONUPK. ********************************************* * You *MUST* put resultant sign in ARISGN. * * You *MUST* load FACEXP ($63) immediately * * before call so that status flags are set! * ********************************************* RESULT: FACC contains floating point sum ERROR: ?OVERFLOW ERROR if result too big EXAMPLE: LDA FACSGN EOR ARGSGN STA ARISGN ;SET RESULTANT SIGN LDA FACEXP ;SET STATUS FLAGS PER FACEXP JSR FADDT ;ADD ARG TO FACC, SUM IN FACC =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: FMULT FUNCTION: FACC = MEMORY * FACC PEPARATION: FACC contains floating point multiplier .A = pointer (lsb) to packed float. point multiplicand .Y = pointer (msb) to packed float. point multiplicand SPECIAL NOTES: The multiplicand *MUST* be in VARBANK in packed format. FMULT calls CONUPK to normalize it. RESULT: FACC contains floating point product ERROR: ?OVERFLOW ERROR if result too big EXAMPLE: LDA #POINTER ;SET POINTER TO *PACKED* MULTIPLICAND JSR FMULT ;MULTIPLY MEM BY FACC, PRODUCT IN FACC =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: FMULTT FUNCTION: FACC = ARG * FACC PREPARATION: FACC contains floating point multiplier ARG contains floating point muitiplicand SPECIAL NOTES: This routine is similar to FMULT. The only difference is the call to CONUPK. (FMULTT assumes you have already loaded ARG with unpacked multiplicand.) RESULT: FACC contains floating point product ERROR: ?OVERFLOW ERROR if result too big EXAMPLE: JSR FMULTT ;MULTIPLY ARG BY FACC, PRODUCT IN FACC =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: FDIV FUNCTION: FACC = MEMORY / FACC PREPARATION: FACC contains floating point divisor .A = pointer (lsb) to packed floating point dividend .Y = pointer (msb) to packed floating point dividend SPECIAL NOTES: The dividend *MUST* be in VARBANK in packed format. FDIV calls CONUPK to normalize it. RESULT: FACC contains floating point quotient ERROR: ?DIVISION BY ZERO ERROR if FACC zero EXAMPLE: LDA #POINTER ;SET POINTER TO *PACKED* DIVIDEND JSR FDIV ;DIVIDE MEM BY FACC, QUOTIENT IN FACC =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: FDIVT FUNCTION: FACC = ARG / FACC PREPARATION: FACC contains floating point divisor ARG contains floating point divldend ARISGN ($70) contains EOR(FACSGN,ARGSGN) .A contains FACEXP SPECIAL NOTES: This routine is similar to FDIV. The only difference is the call to CONUPK. (FDIVT assumes you have already loaded ARG with unpacked dividend.) ********************************************* * You *MUST* put resultant sign in ARISGN. * * You *MUST* load FACEXP ($63) immediately * * before call so that status flags are set! * ********************************************* RESULT: FACC contains floating point quotient ERROR: ?DIVISION BY ZERO ERROR if FACC zero EXAMPLE: LDA FACSGN EOR ARGSGN STA ARISGN ;SET RESULTANT SIGN LDA FACEXP ;SET STATUS FLAGS PER FACEXP JSR FDIVT ;DIVIDE ARG BY FACC, QUOTIENT IN FACC =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: LOG FUNCTION: FACC = LOG(FACC) natural logarithm (base e) PREPARATION: FACC contains floating point number RESULT: FACC contains floating point logarithm ERROR: ?ILLEGAL QUANTITY ERROR if FACC negative or zero EXAMPLE: JSR LOG ;FACC = LOG(FACC) =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: INT FUNCTION: FACC = INT(FACC) PREPARATION: FACC contains floating point number RESULT: FACC contains floating point greatest integer EXAMPLE: JSR INT ;FACC = INT(FACC) =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: SQR FUNCTION: FACC = SQR(FACC) PREPARATION: FACC contains floating point number RESULT: FACC contains floating point square root ERROR: ?ILLEGAL QUANTITY ERROR if FACC negative EXAMPLE: JSR SQR ;FACC = SQR(FACC) =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: NEGOP FUNCTION: FACC = -FACC (invert sign of FACC) PREPARATION: FACC contains floating point number RESULT: FACC contains floating point number with sign inverted EXAMPLE: JSR NEGOP ;FACC = -FACC =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: FPWR FUNCTION: FACC = ARG ^ MEMORY PREPARATION: ARG contains floating point number .A = pointer (lsb) to packed floating point power .Y = pointer (msb) to packed floating point power SPECIAL NOTES: The power *MUST* be in ROM or SYSTEM RAM in packed format as FPWR calls MOVFM to unpack it into FACC. RESULT: FACC contains floating point result ERROR: ?ILLEGAL QUANTITY ERROR if ARG negative ?OVERFLOW ERROR if result too big EXAMPLE: LDA #POINTER ;SET POINTER TO *PACKED* POWER JSR FPWR ;COMPUTE ARG ^ MEM, RESULT IN FACC =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: FPWRT FUNCTION: FACC = ARG ^ FACC PREPARATION: ARG contains floating point number FACC contains floating point power .A contains FACEXP SPECIAL NOTES: This routine is similar to FPWR. The only difference is the call to MOVFM. (FPWRT assumes you have already loaded FACC with unpacked power.) ********************************************* * You *MUST* load FACEXP ($63) immediately * * before call so that status flags are set! * ********************************************* RESULT: FACC contains floating point result ERROR: ?ILLEGAL QUANTITY ERROR if ARG negative ?OVERFLOW ERROR if result too big EXAMPLE: LDA FACEXP ;SET STATUS FLAGS PER FACEXP JSR FPWRT ;COMPUTE ARG ^ FACC, RESULT IN FACC =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: EXP (compute e ^ FACC) FUNCTION: FACC = EXP (FACC) PREPARATION: FACC contains floating point number RESULT: FACC contains fIoating point result ERROR: ?OVERFLOW ERROR if FACC too big EXAMPLE: JSR EXP ;FACC = EXP(FACC) =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: COS FUNCTION: FACC = COS(FACC) PREPARATION: FACC contains floating point number RESULT: FACC contains floating point cosine (in radians) EXAMPLE: JSR COS ;FACC = COS(FACC) =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: SIN FUNCTION: FACC = SIN(FACC) PREPARATION: FACC contains floating point number RESULT: FACC contains floating point sine (in radians) EXAMPLE: JSR SIN ;FACC = SIN(FACC) =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: TAN FUNCTION: FACC = TAN(FACC) PREPARATION: FACC contains floating point number RESULT: FACC contains floating point tangent (in radians) EXAMPLE: JSR TAN ;FACC = TAN(FACC) =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: ATN FUNCTION: FACC = ATN(FACC) PREPARATION: FACC contains floating point number RESULT: FACC contains floating point arctangent (in radians) EXAMPLE: JSR ATN ;FACC = ATN(FACC) =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: ROUND (round to 40 bits of precision) FUNCTION: FACC = FACC + FACOV(msb) PREPARATION: FACC contains floating point number FACOV(msb) contains 'extra' precision RESULT: none if FACC zero or FACOV(msb) zero one extra bit ADDED to FACC lsb if FACOV(msb) is set EXAMPLE: JSR ROUND ;ROUND FACC =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: ABS (make FACSGN(msb) = $00) FUNCTION: FACC = ABS(FACC) PREPARATION: FACC contains (SIGNED) floating point number RESULT: FACC contains (POSITIVE) floating point EXAMPLE: JSR ABS ;FACC = ABS(FACC) =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: SGN FUNCTION: .A = SGN(FACC) PREPARATION: FACC contains floating point number RESULT: .A --> $FF if FACC negative (FACC < 0) $00 if FACC zero (FACC = 0) $01 if FACC positive (FACC > 0) (status flags reflect contents of .A, carry invalid) EXAMPLE: JSR SGN ;SGN(FACC) ; BEQ will trap =0 ; BNE will trap <>0 ; BMI will trap <0 ; BPL will trap >=0 etc. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: FCOMP (compare FACC with MEMORY) FUNCTION: .A = FCOMP(FACC,MEMORY) PREPARATION: FACC contains floating point number .A = pointer (lsb) to packed floating point number .Y = pointer (msb) to packed floating point number SPECIAL NOTES: The number *MUST* be in ROM, or RAM currently in context below ROM, in PACKED format. *** FACOV is significant! RESULT: .A --> SFF if FACC < MEMORY $00 if FACC = MEMORY $01 if FACC > MEMORY (status flags reflect contents of .A, carry invalid) EXAMPLE: LDA #POINTER ;SET POINTER TO *PACKED* NUMBER JSR FCOMP ;COMPARE FACC WITH MEMORY ; BEQ will trap FACC = MEM ; BNE will trap FACC <> MEM ; BMI will trap FACC < MEM ; BPL will trap FACC >= MEM etc. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: RND0 FUNCTION: FACC = random floating point number (0 $00 to generate a 'true' random number $01 to generate next random number in sequence $FF to start a new sequence of random numbers based upon current contents of FACC. SPECIAL NOTES: *MUST* be called with the system bank in context. *MUST* load .A immediately before call so that status flags reflect contents of .A RESULT: FACC = floating point random number EXAMPLE: LDA #$FF ;START REPRODUCEABLE SEQUENCE BASED ON FACC JSR RND0 LDA #$01 JSR RND0 ;GENERATE (FIRST) RANDOM NUMBER IN SEQUENCE =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: CONUPK FUNCTION: ARG = UNPACK(RAM_CONSTANT) PREPARATION: .A = pointer (lsb) to packed floating point number .Y = pointer (msb) to packed floating point number SPECIAL NOTES: The number *MUST* be in VARBANK or SYSTEM RAM in packed format. RESULT: ARG loaded with normalized floating point number ARISGN ($6F) contains EOR(FACSGN,ARGSGN) .A contains FACEXP (status reflects contents of .A) EXAMPLE: LDA #POINTER ;SET POINTER TO *PACKED* NUMBER JSR CONUPK ;LOAD ARG ; BEQ traps ARG = $00 =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: ROMUPK FUNCTION: ARG = UNPACK(ROM_CONSTANT) PREPARATION: .A = pointer (lsb) to packed floating point number .Y = pointer (msb) to packed floating point number SPECIAL NOTES: The number *MUST* be in ROM or SYSTEM RAM currently in context (otherwise identical to CONUPK). RESULT: ARG loaded with normalized floating point number ARISGN ($6F) contains EOR(FACSGN,ARGSGN) .A contains FACEXP (status reflects contents of .A) EXAMPLE: LDA #POINTER ;SET POINTER TO *PACKED* NUMBER JSR ROMUPK ;LOAD ARG ; BEQ traps ARG = $00 =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: MOVFRM FUNCTION: FACC = UNPACK(RAM_CONSTANT) PREPARATION: .A = pointer (lsb) to packed floating point number .Y = pointer (msb) to packed floating point number SPECIAL NOTES: The number *MUST* be in VARBANK or SYSTEM RAM in packed format. RESULT: FACC loaded with normalized floating point number FACOV ($71) cleared EXAMPLE: LDA #POINTER ;SET POINTER TO *PACKED* NUMBER JSR MOVFRM ;LOAD FACC =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: MOVFA FUNCTION: FACC = ARG PREPARATION: ARG contains floating point number RESULT: FACC contains same number as ARG FACOV ($71) cleared .A contains FACEXP (but status invalid!) EXAMPLE: JSR MOVFA ;COPY ARG TO FACC =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- NAME: MOVAF FUNCTION: ARG = FACC PREPARATION: FACC contains floating point number RESULT: FACC will be ROUNDed and FACOV cleared. ARG contains same number as FACC .A contains FACEXP (but status invalid!) EXAMPLE: JSR MOVAF ;COPY FACC TO ARG 3.5. C65 DOS Documentation DIRECTORY HEADER DEFINITION ---------------------------------------------------------------------- BYTE DESCRIPTION ------ -------------------------------------------------------------- 0 TRACK mumber which points to the 1st dir. sector 1 SECTOR number which points to the 1st dir. sector 2 Disk format version number, which is currently 'D' 512 byte sectors 20 per track 20 Sectors per track 40 Tracks per side 2 sides (note they're inverted from normal MFM disk) 3 Must = 0 4 Bytes 4 thru 21 contain the volume name (label) 22 Bytes 22 and 23 contain the disk id (fake) 24 Must contain an $A0 25 DOS version number (CBMDOS = 1, 1581 = 3) 26 Format version number (currently = 'D' (fake)) 27 Bytes 27 thru 28 = $A0 29 NOT USED AT THIS TIME 30 NOT USED AT THIS TIME 31 NOT USED AT THIS TIME 32 NOT USED AT THIS TIME 33 NOT USED AT THIS TIME 34 Track number which points to this directory header 35 Sector number which points to this directory header 36 Bytes 36 thru 255 are not used at this time NOTE: If this is a subdirectory header then BYTES 32 and 33 contain the TRACK & SECTOR number of the DIRECTORY SECTOR that points to this DIRECTORY HEADER. See the partition command for a better description. If this is the ROOT header then they will contain a $00. BAM DEFINITION ---------------------------------------------------------------------- BYTE DESCRIPTION ------ -------------------------------------------------------------- 0 Track link for next bam sector, if last then end of BAMs 1 Sector link 2 Format type this disk was formatted under 3 Compliment version number of byte 2 above 4-5 Disk ID used when this disk was formatted 6 I/O byte used as follows; BIT 7- When set Verify is performed after each disk write BIT 6- Perform CRC check (not used by CBDOS) BIT 1- Huge relative files disabled 7 Auto loader flag (not used by CBDOS) 8-15 Not used at this time by any CBM DOS versions 16-255 BAM image BAM IMAGE ---------------------------------------------------------------------- BYTE DESCRIPTION ------ -------------------------------------------------------------- 0 Number of free sectors on this track 1 MSB flag for sector 7, LSB flag for sector 0 2 MSB flag for sector 15, LSB flag for sector 8 3 MSB flag for sector 23, LSB flag for sector 16 4 MSB flag for sector 31, LSB flag for sector 24 5 MSB flag for sector 39, LSB flag for sector 32 DIRECTORY SECTOR DEFINITION ---------------------------------------------------------------------- BYTE DESCRIPTION ------ -------------------------------------------------------------- 0 TRACK -- Points to the next directory track. 1 SECTOR -- Points to the next directory sector. (IF TRACK = 0 THEN THIS IS THE LAST DIRECTORY SECTOR) FILE ENTRY DESCRIPTION ---------------------------------------------------------------------- BYTE DESCRIPTION ------ -------------------------------------------------------------- 0 File status byte which is used as follows: BIT 7- Set indicates properly closed file BIT 6- File is locked (read only) BIT 5- Save with replace is CURRENTLY in effect, when file is closed this bit is deleted. BIT 4- NOT USED AT THIS TIME BIT 3- Bits 3 thru 0 are used to indicate the filetype: 0=DEL, 1=SEQ, 2=PRG, 3=USER, 4=REL, 5=CBM, 6=not used 7=used by dos to represent DIRECT type of file access 1 TRACK - link to the 1st sector of data for this file. 2 SECTOR - link to the 1st sector of data for this file. 3 Bytes 3-18 contain the filename in ASCII, padded with $A0 19 Side Sector TRACK link for relative files GEOS - Track number of GEOS file header 20 Side Sector SECTOR link for relative files GEOS - Sector number of GEOS file header 21 Record size for relative files GEOS - File structure type 0=SEQ, 1=VLIR 22 GEOS - FILE TYPES: 13 = Swap file 12 = System boot 11 = Disk device 10 = Input device 09 = Printer 08 = Font 07 = Appl. data 06 = Applications 05 = Desk Acc. 04 = System 03 = Basic data 02 = Assembly 00 = Not GEOS 23 Not used by CBM DOS previous to CBDOS GEOS - DATE: Year last modified (offset from 1990) CBDOS- Bits 7-4 contain the upper 4 bit's from the file type byte (see byte 0 above) for the UNNEW, UNSRATCH commands used by CBDOS 24 Not used by CBM DOS previous to CBDOS GEOS - DATE: Month last modified (1 thru 12) CBDOS- Bit's 7 thru 4 contain the lower 4 bit's from the file type byte (see byte 23 above) 25 GEOS - DATE: Day last modified (1 thru 31) 26 TRACK (from 1) for the save with replace file GEOS - DATE: Hour last modified (0 thru 23) 27 SECTOR (from 2) for the save with replace file GEOS - DATE: Minute last modified (0 thru 59) 28 LSB of the # of sectors used by this file 29 MSB of the # of sectors used by this file NOTE: Each sector in the directory contains 8 entries of 32 bytes each. SIDE SECTOR FORMAT DEFINITION ---------------------------------------------------------------------- BYTE DESCRIPTION ------ -------------------------------------------------------------- 0 Next Side Sector TRACK link ($FF if last) 1 Next Side Sector SECTOR link 2 Side Sector number If this is a SUPER SIDE SECTOR then this contains an $FE (see the description of the SUPER SIDE SECTOR below) 3 Record Size 4-5 TRACK & SECTOR link of Side Sector number 0 6-7 TRACK & SECTOR link of Side Sector number 1 8-9 TRACK & SECTOR link of Side Sector number 2 10-11 TRACK & SECTOR link of Side Sector number 3 12-13 TRACK & SECTOR link of Side Sector number 4 14-15 TRACK & SECTOR link of Side Sector number 5 16-17 TRACK & SECTOR link of the DATA BLOCK #0 18-19 TRACK & SECTOR link of the DATA BLOCK #l etc... NOTE: There are 91 groups to the largest file that this DOS can handle. SUPER SIDE SECTOR FORMAT DEFINITION ---------------------------------------------------------------------- BYTE DESCRIPTION ------ -------------------------------------------------------------- 0 Next Side Sector TRACK link (SFF if last) 1 Next Side Sector SECTOR 2 Contains an SFE to indicate this is a SUPER SIDED SECTOR 3-4 TRACK & SECTOR link of Side Sector number 0 5-6 TRACK & SECTOR link of Side Sector number 1 7-8 TRACK & SECTOR link of Side Sector number 2 9-10 TRACK & SECTOR link of Side Sector number 3 11-12 TRACK & SECTOR link of Side Sector number 4 13-14 TRACK & SECTOR link of Side Sector number 5 ..... ........................................... 253-254 TRACK & SECTOR link of Side Sector number 125 NOTE:There are 91 groups to the largest file that this DOS can handle. DATA SECTOR DEFINITION ---------------------------------------------------------------------- BYTE DESCRIPTION ------ -------------------------------------------------------------- 0-1 TRACK and SECTOR link to the next data block. If track=0 then sector contains the number of bytes used in this sector (which will always be at least 2 on the last block for the T&S link bytes). NOTE: Used by DEL, SEQ, PRG, REL (data blocks) and USR. ;*-------------------------------------------------------------------* ;* Format a track * ;* 10 sectors per track numbered 1-10, 512 byte sectors * ;*-------------------------------------------------------------------* ;* 12 Sync marks 00 * ;* 3 Header ID marks w/missing clock A1 * ;* 1 Header ID FE * ;* 4 Header bytes Track * ;* Side * ;* Sector * ;* Sector size * ;* 2 Header CRC bytes xx,xx * ;* 22 Data gap bytes 4E * ;* 12 Sync marks 00 * ;* 3 Data block ID marks w/missing clock A1 * ;* 1 Data block ID FB * ;* 512 Data block fill bytes 00 * ;* 2 Data block CRC bytes xx,xx * ;* 24 Sector gap bytes 4E * ;*-------------------------------------------------------------------* ;*-------------------------------------------------------------------* ;* Calculate the 2 bytes CRC for each sector header of an entire * ;* track of 10 sectors. AXYZ are trashed. * ;* * ;* This routine is based on the Cyclical Redundancy Check * ;* on the polynomial: A^16+A^12+A^5+1. * ;* * ;* HEADER contains TRACK,SIDE,SECTOR,2 [sector size] * ;* * ;* DO WHILE ne = 0 * ;* DO FOR each bit in the data byte (.a) [from lsb to msb] * ;* IF (LSB of crc) eor (LSB or data) * ;* THEN CRC = (CRC/2) EOR polynomial * ;* ELSE CRC = (CRC/2) * ;* END IF * ;* LOOP * ;* LOOP * ;*-------------------------------------------------------------------* ;*-------------------------------------------------------------------* ;* SIDE = (LogicalSector >= 20) AND 1 * ;* TRACK = LogicalTrack - 1 * ;* StartingSector = SIDE * 20 * ;* SECTOR = (LogicalSector - StartingSector) / 2 + 1 * ;* HALF = (LogicalSector - StartingSector) AND 1 * ;*-------------------------------------------------------------------* C65 Partition and Subdirectory Syntax This specification describes a _proposed_ C65 partition/subdirectory parser. OPEN la,fa,sa, "[#]/path/:filename" OPEN la,fa,15, "#/path/:[cmd_string]" where: # is an optional "drive" number, 0-3. /path/ is a partition or subdirectory name : delimits the path from the filename and: is a DOS command (such as I,N,S,C, etc.) (cmd_string) is an optional string required by some commands. The first example illustrates a typical filename specification, the second example illustrates a command channel instruction. OPEN la,fa,sa, "0/SUBDIR1/SUBDIR2/:FILE,S,W" Action taken Why ------------------------------------ ------------------------------- 1. Select the "root" 0 2. Find & enter two subdirectories /SUBDIR1/SUBDIR2/: (the trailing "/" is required to be compatible with CMD?) 3. Create & open file for writing FILE,S,W The "root" or "drive number", path, and ":" are all optional. If they are omitted, the file is opened in the current partition. Some similar, and legal, syntaxes are: OPEN la,fa,sa, "FILE,S,W" (create "FILE" in current part) OPEN la,fa,sa, ":FILE,S,W" (create "FILE" in current part) OPEN la,fa,sa, "0:FILE,S,W" (create "FILE" in current part) OPEN la,fa,sa, "/SUBDIR/ FILE,S,W" (from current partition, enter "SUBDIR" and create "FILE") OPEN la,fa,sa, "//SUBDIR/:FILE,S,W" (from Root partition, enter "SUBDIR" and create "FILE") OPEN la,fa,sa, "@0/SUBDIR/:FILE" (open "FILE" in "SUBDIR" for writing) Some questionable syntaxes, and their affect, are: OPEN la,fa,sa, "0FILE,S,W" (this would create file "0FILE") OPEN la,fa,sa, "/SUBDIR/FILE,S,W" (creates file "/SUBDIR/FILE" OPEN la,fa,sa, "@0:FILE,S,W" (open filr "FILE" in current partition for writing) OPEN la,fa,sa, "/0:FILE,S,W" (? should create file "0:file", note this is not the cmd chnl) Some legal commands: OPEN la,fa,sa,"I0" (initialize current partition) OPEN la,fa,sa,"I//" (initialize Root) OPEN la,fa,sa,"I0/SUBDIR/:" (enter "SUBDIR" and initialize) OPEN la,fa,sa,"N0/SUBDIR/:NAME,ID" (enter "SUBDIR" and "new" it) OPEN la,fa,sa,"S0/SUBDIR/:FILE" (delete "FILE" in "SUBDIR") OPEN la,fa,sa,"/0:SUBDIR" (1581 partition select, "/" in this context is a command itself) Some proposed general rules, designed to be compatible with both the 1581 subpartitioning syntax and CMD syntax: 1. The name of a subdirectory must always be separated from the filename by a colon (":"). 2. Each subdirectory name must be delimited by a slash ("/"). 3. To select Root directory (partiton), specify two slashes ("//"). This allows older applications specifying the drive number ("0:") to be run in a partition. CURRENT PARTITION ROUTINES Create Partition: "/0:PAR_NAME,"+(START-TRK)+(START-SECTOR)+(LO-BLKS)+(HI-BLKS) Select Partition: "/0:PAR_NAME" will select given filname as subdirectory "/0" will select root directory SELECT PARTITION This routine will allow the user to quickly select partition paths using the normal SA values other than 15. To use this new method the user opens the file using a normal SA and the filename MUST be structured as follows: "/:PATH_1/PATH_2/PATH_3..... ETC" If the dos does not find one of the filenames in the file path stream it will check to see if the file exists in the current directory and if it does it will open the file in the normal method as it does now. ;********************************************************************* ;* FILE_COMMANDS * ;* * ;* The following set of command channel routines were added to allow * ;* the user a graceful way of manipulating files: * ;* * ;* "F-L" Locate a file to prevent it from being scratched * ;* "F-U" Unlock a file and allow it to be scratched * ;* "F-R" Restore a file after it has been scratched * ;* * ;* Following each command above is the drive number, followed by a * ;* colon then followed by the filename(s). For example, to lock all * ;* the files on drive 0 you would send the following file command: * ;* * ;* OPENXX,XX,15,"F-L0:*" * ;* * ;* OPENXX,XX,15,"F-L0:FNAME,FNAME1,FNAME2,... ETC" * ;********************************************************************* ;********************************************************************* ;* BLOCK STATUS * ;* * ;* Syntax: "B-S:CHANNEL NUMBER, DRIVE NUMBER, TRACK, SECTOR" * ;* * ;* Then check error channel for normal errors then get one byte from * ;* from the channel number. If it is a 0 then the sector is free. * ;* 1 indicates the sector is in use. * ;* * ;* This command was added to enable an easy method of finding out if * ;* a given track or sector is currently marked as being used in a * ;* drive's BAM or not. * ;* * ;* CBDOS CHGUTIL * ;* * ;* COMMAND COMMENTS DRIVES USED ON * ;* "U0>B"+chr$(n) b = set fast/slow serial bus 1581 * ;* "U0>D"+chr$(n) d = set dirsecinc CBDOS * ;* "UO>H"+chr$(n) h = set head selection 0,1 1571 * ;* "U0>M"+chr$(n) m = set dos mode 1571 * ;* "U0>R"+chr$(n) r = set dos retries on errors 1571,1581 * ;* "UO>S"+chr$(n) s = set secinc 1571,1581,CBDOS * ;* "U0>V"+chr$(n) v = set verify on/off 1581,CBDOS * ;* "U0>?"+chr$(n) ? = set device number CBDOS * ;* "U0>L"+chr$(n) = set large rel files on/off CBDOS * ;* "U0>MR"+ xx = perform memory read 1581 * ;* "U0>MW"+ xx = perform memory write 1581 * ;* 12345 * ;* ^--------------- CMDSIZ points to end of starting string @1 * ;********************************************************************* FLOPPY DISK CONTROLLER ERRORS IP FDC DESCRIPTION -- --- ----------- 0 (0) no error 20 (2) can't find block header 23 (5) checksum error in data 25 (7) write-verify error 26 (8) write w/ write protect on 27 (9) crc error in header Information description ----------------------- 1 files scratched 2 selected partition 3 files locked 4 files unlocked 5 files restored Parameter errors ---------------- 30 general syntax 31 invalid command 32 long line 33 invalid filname 34 no filenames given Relative file errors -------------------- 50 record not present 51 overflow in record 52 file too large 53 big relative files disabled Open routine errors ------------------- 60 file open for write 61 file not open 62 file not found 63 file exists 64 file type mismatch Sector management errors ------------------------ 65 no block 66 illegal track or sector 67 illegal system t or s General channel/block errors ---------------------------- 02 channel selected 70 no channels available 71 bam corrupted error 72 disk full 73 cbdos v1.0 74 drive not ready 75 format error 76 controller error 77 selected partition illegal 78 directory full 79 file corrupted 3.6. C64DX RS-232 DRIVER 00A7 rs232_status - UART status byte 00A8 rs232_flags - open flag, xon/xoff status - b7: channel open (reset) - b6: flow control (1=x-line) - b5: duplex (1=half) - b1: XOFF received - b0: XOFF sent 00A9 rs232_jam - system character to xmit 00AA rs232_xon_char - XON character (null=disabled) 00AB rs232_xoff_char - XOFF character (null=disabled) 00B0 rs232_xmit_empty - xmit buffer empty flag (0=empty) 00B1 rs232_rcvr_buffer_lo - lowest page of input buffer 00B2 rs232_rcvr_buffer_hi - highest page of input buffer 00B3 rs232_xmit_buffer_lo - lowest page of output buffer 00B4 rs232_xmit_buffer_hi - highest page of output buffer 00B5 rs232_high_water - point at which receiver XOFFs 00B6 rs232_low_water - point at which receiver XONs 00C4 rs232_rcvr_head - pointer to end of buffer 00C6 rs232_rcvr_tail - pointer to start of buffer 00C8 rs232_xmit_head - pointer to end of buffer 00CA rs232_xmit_tail - pointer to start of buffer RS-232 interrupt-driven handler How it works: when an RS232 channel is OPENed, buffers are flushed, all flags and states are reset, and the receiver IRQ is enabled. When a byte is put into the xmit buffer by BSOUT, the xmit IRQ is enabled. The xmit IRQ is disabled whenever the xmit buffer is found to be empty or an XOFF is received (it is enabled whenever an XON is received). CLOSE will hang until the xmit buffer is empty, and BSOUT will hang when the xmit buffer is full. IRQs must be allowed by the user at all times (and especially during BSOUT calls) for proper operation. (The RS232 channel will work even if IRQs are disabled by the user, but thoughput will be reduced to the frame rate (normal system raster IRQ) and the system can hang forever should the xmit buffer become full and BSOUT is called with a byte to xmit). A successful CLOSE will disable all RS232 interrupts and re-init everything. Note that DOS calls disable both IRQ and NMI interrupts while the DOS code is in context. The remote should be XOFFed to avoid loss of data. Refer to the UART specification for register description & baud rate tables. Open an RS-232 channel This is different from the usual C64/C128 command string. 1 2 3 4 5 6 Command string bytes: baud|word|parity|stop(unused)|duplex|xline 4.0. C64DX DEVELOPMENT SOPPORT Please photocopy the attached 'C64DX PROBLEM REPORT' and use it to report any problems. If you have any requests or recommendations, please send a good description of it and explain why you want it. +----------------------------------------+-------------------------------------+ | C64 DX PROBLEM REPORT | Date | +----------------------------------------+-------------------------------------+ |Please complete this form as completely as possible and mail or express it to:| | | | Commodore Business Machines, Inc. Telephone: 215-431-9427 | | 1200 Wilson Drive Fax: 215-431-9156 | | West Chester, PA I9380 Email: fred@cbmvax.commodore.com | | | | Attention: Fred Bowen, Engineering | +------------------------------------------------------------------------------+ |Company Name | +------------------------------------------------------------------------------+ |Company Address | | | | | | | | | | | | | | | +----------------------------------------+-------------------------------------+ |Your Name |Your Phone | +----------------------------------------+-------------------------------------+ |Your system | | | | Serial No.________ PCB rev_________ Software ver________ ROM Cksum__________ | | | | 4510 rev__________ 4567 rev________ F011(DOS)___________ F018(DMA)__________ | | | | Peripherals: | +---------------------+--------------------------------------------------------+ |Your problem | Explain problem here and show how to cause it. Attach | | | sample program. | | ____ C64 mode | | | ____ C64DX mode | | | ____ Hardware | | | ____ Software | | | ____ Mechanical | | | ____ Documentation | | | ____ Compatibility | | | | It happens: | | | ____ all the time ____ frequently ____ occasionally | +---------------------+--------------------------------------------------------+ |In your opinion, how I bad is the problem? ____ Must fix, no workaround | | | | ____ I can work around it | | | | ____ Check here if you need to be contacted ____ Minor problem | +------------------------------------------------------------------------------+ |Please leave this space blank | | | | | | | | | | | |Number Received Contacted Completed | +------------------------------------------------------------------------------+ C64DX System Specification UPDATE * The Monitor parser now allows PETSCII input/conversion: 'A prints ASC() value of character >1800 'text puts text into memory LDA #'A * IRQ runs during graphics (Kernel finds its own base page). IRQ still does not run during DOS activity (not sure if they ever will). * The following Kernel Jump Table Entries have moved (and are still subject to further changes): FF05 nirq ;IRQ handler FF07 monitor_brk ;BRK handler (Monitor) FF09 nnmi ;NMI handler FF0B nopen ;open FF0D nclose ;close FF0E nchkin ;chkin FF11 nckout :ckout FF13 nclrch ;clrch FF15 nbasin :basin FF17 nbsout ;bsout FF19 nstop ;stop key scan FF1B ngetin ;getin FF1D nclall ;clall FF1F monitor_parser ;monitor command parser FF21 nload ;load FF23 nsave ;save FF25 talk FF27 listen FF29 talksa FF2B second FF2D acptr FF2F ciout FF31 untalk FF33 unlisten FF35 DOS_talk FF37 DOS_listen FF39 DOS_talksa FF3B DOS_second FF3D DOS_acptr FF3F DOS_ciout FF41 DOS_untalk FF43 DOS_unlisten FF45 Get_DOS FF47 Leave_DOS FF49 ColdStartDOS <<| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 0 | + á| - '| œ [|HOME|DEL | +----+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+-+----+ | TAB | | | | | | Z | | | | | š | | \ | RSTR | | | Q | W | E | R | T | Y | U | I | O | P | @ | * +| ^ ]| | +----+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+------+ |CTRL|SHFT| | | | | | | | | | ™ | Ž | ' | RETURN | | |LOCK| A | S | D | F | G | H | J | K | L | : [| ; ]| = #| | +----+----+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+-+--+-+----+----+----+ | C= | SHIFT | Y | | | | | | | < ;| > :| ? _| SHIFT| | | | | Z | X | C | V | B | N | M | , | . | / -| |  | +----+-------+-+--+----+----+----+----+----+----+----+----+-+--+-+----+----+----+ | | | | | | | | |  |  |  | +--------------------------------------------+ +----+----+----+ Notes: 1/ The operation of national keyboards is identical to C128 implementation. The ASCII/DIN key replaces the CAPS LOCK key, and can be toggled anytime to switch keyboard modes and automatically change the display. 2/ The national keyboard contains key legends for both national and ASCII modes. The national legends appear on the right top/bottom of the keys. 3/ The German keyboard has three (3) "deadkeys". They are accent d'aigue, accent grave, and accent circonflex. Pressing the "deadkey" followed by a valid vowel or accent character will 'build' the desired character: accent d'aigue: ‚ accent grave: …, Š, — accent circonflex: ƒ, ˆ, Œ, ", – 4/ National character ROM graphic characters differ from the C64 and ASCII (English) graphic character sets. PAINT x, y [,color] Working, but not completely to spec. Uses draw pen color and fills emptyness to any border. RND(0) Improved for better "randomness". Uses unused POT of second SID chip. PCB must allow lines to float. SET DISK # (without [TO #] parameter) allows user to clear DS$ message and specify which drive next DS$ comes from. SET VERIFY The new DOS65 defaults to verify-after-write OFF. This command works with 1581 drive, too. * Negative coordinates are now allowed for all graphics commands. Some commands require their arguments to be "onscreen" such as PAINT. * BASIC errors now force text mode, and TYPE, LIST, DISK, KEYLOAD, LOADIFF now catch all DOS errors. Autoboot filename=AUTOBOOT.C64DX.* * Opening an RS-232 channel, command string allows setting new features: 1 baud (0-16, where 16=MIDI rate) 2 word len 3 parity 4 stop bits (not used) 5 duplex 6 xline 7 xon char (0=incoming flow control disabled) 8 xoff char (0=outgoing flow control disabled) 9,10 input buffer pointer (page lo, hi) 11,12 output buffer pointer (page lo, hi) 13 high water mark (point at which xoff is xmitted) 14 low water mark (point at which xon is xmitted) For debug purposes, the border color will change if an RS232 buffer overflow occurs. To differentiate between a GET# of a null and a 'no data' null, test bit 3 of STatus (same as C64). * Support for latest DOS controller chip, F011D, includes error LED blink (border color still changes too, for now). Changes to improve FASTLOAD speed and improve SAVE speed. Will work with F011C chip, but error LED does not blink. Requires latest 'ELMER' PAL for disk LED to work correctly for either controller Chip. External drive LED will not work correctly until new PCB & F016 chip are designed. New DOS functions include COPY D0 TO D1, ability to change sector skews for files (U0>S#) and directory (U0>D#), and directory compress (i.e., empty trash) via "E" command. Physical interleave is now 7. * The DOS COPY/CONCAT bugs have been fixed, and COPY now allows forms such as COPY D0,"*.SRC" TO D1,"*" and COPY D0,"*=SEQ" TO D1, "*". Directory/partition paths not yet implemented, but will be. The following changes/updates/fixes have been made to the C64DX ROM code since the March 1, 1991 C64DX System Specification was printed. Please make note of them. Current ROM as of this update is 910501. CHAR Now works to spec. and supports the following imbedded control characters (although some are buggy; others are planned): ^F 6 flip ^I 9 invert ^O 15 overwrite ^R 18 reverse field on 146 reverse field off ^U 21 underline ^Y 25 tilt ^Z 26 mirror When specifying a character set from ROM, note that national versions of the C64DX will have the national character set at $39000 and the C64 character set at $3DC00. In US/English systems, the default C64DX-mode character set will be at $39000. CLR ERR$ Clears BASIC error stuff, useful after a TRAP. CURSOR [,] [column] [,row] [,style] where: column,row = x,y logical screen position style = flashing (0) or solid (1) ON, OFF = to turn the cursor on or off LINE x0, y0 [, [x1] [,y1]] ... where: (x1,y1)=(x0,y0) if not specified, drawing a dot. Additional coordinates (x2,y2), etc. draw a line from the previous point. LOADIFF "file" [,U#,D#] Loads an IFF picture from disk. Requires a suitable graphic screen to be already opened (this may change). The file must contain std IFF data in PRG file type. IFF pics can be ported directly from Amiga (eg., using XMODEM). Returns 'File Data Error' if it finds data it does not like. MOD (number, modulus) New function. MOUSE ON [,[port] [,[sprite] [,[hotspot] [,X/Yposition] ]]] MOUSE OFF where: port = (1...3) for joyport 1, 2, or either (both) sprite = (0...7) sprite pointer hotspot = x, y offset in sprite, default 0,0 position = normal, relative, or angular coordinates Defaults to sprite 0, port 2, last hotspot (0,0), and position. Kernel doesn't let hotspot leave the screen.