+RXB Posted February 25 Share Posted February 25 7 hours ago, Retrospect said: hey @RXB I liked the video. Also you gotta love the sound of a proper TI keyboard. I miss that so much. I've a request for RXB if it's at all possible? The CALL ROLLUP, ROLLDOWN command is pretty awesome , is there any way this can be extended with something like this; CALL BOX(1,1,8,32) .... what that could do is specify an area of the screen for ROLL to play with, so ROLL only rolls that specific area of the screen , like in my example there it would be from row 1, column 1, down to row 8 column 32 ... or the user could specify whole screen with CALL BOX(1,1,24,32) ... my idea is we could specify a portion of the screen to ROLL for example with a starfield , leaving the rest of the screen for the scoreboard to stay still. This would enable coders to create pretty good games for just the console. My very first package before RXB was WINDYXB (my daughters name) that had CALL WINDOW(row,column,height,width,string-variable) ! 255 max string Now I could substitute string-variable for VDP address so you could actually save up to 768 bytes. 2 Quote Link to comment Share on other sites More sharing options...
+Retrospect Posted February 25 Share Posted February 25 4 minutes ago, RXB said: My very first package before RXB was WINDYXB (my daughters name) that had CALL WINDOW(row,column,height,width,string-variable) ! 255 max string Now I could substitute string-variable for VDP address so you could actually save up to 768 bytes. Cool ... could you set up a call window or call box that lets call roll just roll what's in the window ? Quote Link to comment Share on other sites More sharing options...
+RXB Posted February 25 Share Posted February 25 11 hours ago, Retrospect said: Cool ... could you set up a call window or call box that lets call roll just roll what's in the window ? That seems pretty specialized to just a very few applications. On other hand: CALL WINDOW(row,column,height,width,string-variable) ! 255 max This could be used to replace many programs using DISPLAY AT or CALL HCHAR to draw a box or rectangle with contents there in one command. Almost all commands have to be very universal or as replacements for other commands, this is how RXB has evolved over last 30 years. The idea is to find common sets of commands being used and replace them with a single command like CALL JOYST/MOTION/KEY and IF something THEN line number is RXB command CALL JOYMOTION or like ON/KEY/GOTO in a single command like RXB command CALL ONKEY 1 Quote Link to comment Share on other sites More sharing options...
+Vorticon Posted February 27 Share Posted February 27 Quick question: does RXB use any of the low memory of the 32K expansion or is it available for user machine language routines? Quote Link to comment Share on other sites More sharing options...
SteveB Posted February 27 Share Posted February 27 As Rich always emphasize, RXB runs on the unexpanded console ... so I guess it does not use the lower 8k at all when present. I tested some of my games with RXB, mainly my wordsearch which stores the vocabulary there. 1 Quote Link to comment Share on other sites More sharing options...
+Retrospect Posted February 27 Share Posted February 27 It might do ... he does have it set up for 32K expansion when present. Check out the SIZE command. It comes up on screen like a full screen of details. (if expansions are present) I've never tried RXB using just 16K console. Hey that's a fair point, I don't know if my RXB Cosmic Hero game will work on 16K console? Quote Link to comment Share on other sites More sharing options...
+TheBF Posted February 27 Share Posted February 27 6 hours ago, Vorticon said: Quick question: does RXB use any of the low memory of the 32K expansion or is it available for user machine language routines? A quick test can use the Heatmap in Classic99. If you don't see any "red" lines in the low RAM area it's not being used, at least not for the program you are running. 2 1 Quote Link to comment Share on other sites More sharing options...
+Retrospect Posted March 1 Share Posted March 1 A quick video of Cosmic Hero working on a 16K configuration. Glad that works, it was meant for a stock console with only RXB. js99er-20230301002116.webm 2 1 Quote Link to comment Share on other sites More sharing options...
Asmusr Posted March 1 Share Posted March 1 How come there are 8192 assembly bytes free on an unexpanded console? And what does >A040 Program end address mean when the program is in VDP RAM? Quote Link to comment Share on other sites More sharing options...
+Retrospect Posted March 1 Share Posted March 1 5 hours ago, Asmusr said: How come there are 8192 assembly bytes free on an unexpanded console? That is a very interesting question. There literally should only be 11840 bytes free (a little more if disk drive turned off) .... wtf ... @RXB have you seen this? Quote Link to comment Share on other sites More sharing options...
+RXB Posted March 1 Share Posted March 1 5 hours ago, Asmusr said: How come there are 8192 assembly bytes free on an unexpanded console? And what does >A040 Program end address mean when the program is in VDP RAM? Well I have fixed this I defaulted to show it but this was before I move to console only and now it works properly. It is in RXB 2023A that is not out yet hopefully Tursi will include it in next RXB release. 1 Quote Link to comment Share on other sites More sharing options...
+RXB Posted March 3 Share Posted March 3 In order to gain some space in XB ROMs I have seen how to eliminate some insanely useless Assembly in XB ROM 1 that is no faster then GPL. In this case it is GRSUB1 in XB GROM 3 that calls the XML GREAD is changed here: *********************************************************** * SUBROUTINE TO READ 2 BYTES OF DATA FROM VDP OR ERAM * (use GREAD) *********************************************************** GRSUB1 FETCH @FAC4 Fetch the source addr on ERAM DST *FAC4,@FAC2 Put it in @FAC2 CZ @RAMTOP If ERAM present BS G6823 DST 2,@FAC4 @FAC4 : Byte count XML GREAD Read data from ERAM * @FAC6 : Destination addr on CP BR G6827 ERAM not exists G6823 DST V*FAC2,@FAC6 Read data from VDP G6827 RTN *********************************************************** RXB NEW PATCH CODE *********************************************************** GRSUB1 FETCH @FAC4 Fetch the source addr on ERAM DST *FAC4,@FAC2 Put it in @FAC2 CZ @RAMTOP If ERAM present BS G6823 DST 2,@FAC4 @FAC4 : Byte count DST *FAC2,@FAC6 Read data from ERAM * @FAC6 : Destination addr on CPU BR G6827 ERAM not exists G6823 DST V*FAC2,@FAC6 Read data from VDP G6827 RTN *********************************************************** Now this is the Assembly in XB ROM 1 that is to be removed as it is not any faster than DST *FAC2,@FAC6 vs XML GREAD 3943 7EA6 AORG >7EA6 99/4 ASSEMBLER GREADS PAGE 0091 3945 3946 * (RAM to RAM) 3947 * Read data from ERAM 3948 * @GSRC : Source address on ERAM 3949 * @DEST : Destination address in CPU 3950 * Where the data stored after read from ERAM 3951 * @BCNT3 : byte count 3952 7EA6 0203 GREAD1 LI R3,BCNT3 # of bytes to move 7EA8 8356 3953 7EAA 0202 LI R2,GSRC Source in ERAM 7EAC 8354 3954 7EAE 0201 LI R1,DEST Destination in CPU 7EB0 8358 3955 7EB2 1006 JMP GRZ1 Jump to common routine 3956 * Read data from ERAM to CPU 3957 * @ADDR1 : Source address on ERAM 3958 * @ADDR2 : Destination address in CPU 3959 * Where the data stored after read from ERAM 3960 * @BCNT1 : byte count 3961 7EB4 0203 GREAD LI R3,BCNT1 # of bytes to move 7EB6 834E 3962 7EB8 0202 LI R2,ADDR1 Source in ERAM 7EBA 834C 3963 7EBC 0201 LI R1,ADDR2 Destination in CPU 7EBE 8350 3964 * Common ERAM to CPU transfer routine 3965 7EC0 C112 GRZ1 MOV *R2,R4 3966 7EC2 DC74 GRZ2 MOVB *R4+,*R1+ Move byte from ERAM to CPU 3967 7EC4 0613 DEC *R3 One less to move, done? 3968 7EC6 16FD JNE GRZ2 No, copy the rest 3969 7EC8 045B RT 3970 ************************************************************ There are a few other Assembly routines in XB ROM 1 that are no faster than GPL commands or same speed but take up way more to set up vs GPL. I am looking at maybe as much as almost 1K of assembly in XB ROM 1 that is much less efficient than GPL CALL MOVE command that only takes 7 bytes. But is no slower than the above assembly routine due to all the time to set up just to use assembly will benefit more overall. As example in GPL CALL MOVE @BCNT1,*ADDR1,*ADDR2 only takes 7 bytes and does not require reloading variables to make a call to make it work. So 7 bytes vs 34 bytes but they do the same thing at same amount of time. I still have to do some timing checks but I do need room to add INTEGERs to XB to go with Floating Point too. 2 Quote Link to comment Share on other sites More sharing options...
+Lee Stewart Posted March 3 Share Posted March 3 15 hours ago, RXB said: In order to gain some space in XB ROMs I have seen how to eliminate some insanely useless Assembly in XB ROM 1 that is no faster then GPL. In this case it is GRSUB1 in XB GROM 3 that calls the XML GREAD is changed here: *********************************************************** * SUBROUTINE TO READ 2 BYTES OF DATA FROM VDP OR ERAM * (use GREAD) *********************************************************** GRSUB1 FETCH @FAC4 Fetch the source addr on ERAM DST *FAC4,@FAC2 Put it in @FAC2 CZ @RAMTOP If ERAM present BS G6823 DST 2,@FAC4 @FAC4 : Byte count XML GREAD Read data from ERAM * @FAC6 : Destination addr on CP BR G6827 ERAM not exists G6823 DST V*FAC2,@FAC6 Read data from VDP G6827 RTN *********************************************************** RXB NEW PATCH CODE *********************************************************** GRSUB1 FETCH @FAC4 Fetch the source addr on ERAM DST *FAC4,@FAC2 Put it in @FAC2 CZ @RAMTOP If ERAM present BS G6823 DST 2,@FAC4 @FAC4 : Byte count DST *FAC2,@FAC6 Read data from ERAM * @FAC6 : Destination addr on CPU BR G6827 ERAM not exists G6823 DST V*FAC2,@FAC6 Read data from VDP G6827 RTN *********************************************************** Unless some later routine (other than the now unused GREAD) depends on the byte-count being in FAC4, I would eliminate DST 2,@FAC4 @FAC4 : Byte count ...lee Quote Link to comment Share on other sites More sharing options...
apersson850 Posted March 10 Share Posted March 10 On 2/25/2023 at 8:32 PM, RXB said: CALL WINDOW(row,column,height,width,string-variable) ! 255 max This could be used to replace many programs using DISPLAY AT or CALL HCHAR to draw a box or rectangle with contents there in one command. I've done exactly that as a unit for the p-system. Could be used to present pop-up error messages or whatever. The p-system uses a simulated 80-column screen, but I designed my window thing so it would always show on the same physical place on the screen, regardless of which logical section you were viewing. 1 Quote Link to comment Share on other sites More sharing options...
+InsaneMultitasker Posted March 12 Share Posted March 12 Does RXB (and XB) have source code to the crunch buffer, and if so, how does it handle converting variables from lower case to upper case? I looked for the source but couldn't find the code for this. Quote Link to comment Share on other sites More sharing options...
HOME AUTOMATION Posted March 12 Share Posted March 12 Maybe this... ******************************************************************************** AORG >7B88 TITL 'CRUNCHS' QUOTE EQU >22 COMMA EQU >2C LISTZ EQU >02 OLDZ EQU >05 SAVEZ EQU >07 MERGEZ EQU >08 RETURZ EQU >88 UNBRKZ EQU >8F DATAZ EQU >93 RESTOZ EQU >94 REMZ EQU >9A CALLZ EQU >9D IMAGEZ EQU >A3 RUNZ EQU >A9 COLONZ EQU >B5 QUOTEZ EQU >C7 UNQSTZ EQU >C8 USINGZ EQU >ED MAXKEY EQU 10 * * CRUNCH copies a line (normally in LINBUF) to CRNBUF in the * process, it turns the line number (if any) binary, and * converts all reserved words to tokens. CALL is a GPL XML * followed by a single byte which indicates the type of * crunch to be done. Possible types include: * >00 - Normal crunch * >01 - crunch as a data statement (input stmt) * REGISGERS: * R0 - R1 Scratch * R2 - R3 Scratch * R4 Points to R8LB * R5 Points to length byte of string/numeric * R6 Indicates numeric copy mode (numeric/line #) * R7 Mode of copy (strings, names, REMs, etc) * R8 Character buffer * R9 Points to name during keyword scan * R11 - R12 Links * R13 GROM read data pointer * R15 VDP write address pointer * CRUNCH MOV R11,R12 Save return link MOVB *R13,R3 Read call code BL @PUTSTK Save GROM address CLR @FAC Assume no line number LI R4,R8LB Set up W/S low-byte pointer CLR R8 Initialize character buffer BL @GETNB Scan line for 1st good char MOVB R1,*R4 Save character JEQ CRU28 If empty line, return * Now check crunch call mode, normal or input statement SRL R3,8 Normal curnch call? JEQ CRU01 Yes, crunch the statement * Initialize for input statement crunch LI R2,CRU84 No, must be crunch input stmt LI R10,CRU83 so set up move indicators LI R7,CRU80 JMP CRU10 And jump into it * Initialize for normal line crunch CRU01 INC @BUFLEV Indicate CRNBUF is destroyed CLR @ARG4 Assume no symbol MOVB R8,@PRGFLG Clear program flag BL @GETINT Try to read a line number MOV R0,@FAC Put line number into final JEQ CRU02 If no line number BL @GETNB Skip all leading spaces MOVB R1,*R4 Save character in R8LB JEQ CRU28 If nothing left in line CRU02 LI R7,CRU16 Set normal scan move LI R6,CRU96 Set normal numeric scan mode JMP CRU10 Merge into normal scan code * Main loop of the input copy routine. Sets R8LB to next * character, R0 to its character property byte * R7 indicates dispatch mode. CRU04 LI R6,CRU96 Set normal numeric mode CRU05 LI R7,CRU16 Set normal scan mode CRU06 BL @PUTCHR Copy into crunch buffer CRU08 BL @GETCHR Get next input character CLR R0 Assume nil property MOVB R1,*R4 Copy to crunch buffer JEQ CRU12 Finish up if we reach a null *-----------------------------------------------------------* * Replace following line for adding lowercase character * * set to 99/4A 5/12/81 * * * CRU10 MOVB @CPTBL(R8),R0 Fetch char's prop table vec * CRU10 CB *R4,@ENDPRO Higher then "z" * JHE CRU09 Yes, give CPNIL property * MOVB @CPTBL(R8),R0 Fetch char's prop table value * B *R7 Dispatch to appropriate code * CRU09 MOVB CPNIL,R0 Don't go to CPT, just take * * CPNIL prop * *-----------------------------------------------------------* CRU12 B *R7 Dispatch to appropriate code CRU14 MOV R8,R8 End of line? JNE CRU06 Not yet CRU15 MOV @RAMPTR,R3 Now check for trailing spaces DEC R3 Backup to read last character BL @GETV1 Go read it CB R1,@CBH20 Last character a space? JNE CRU28 No, so end of line, exit DEC @RAMPTR Yes, backup pointer to delete JMP CRU15 And test new last character *-----------------------------------------------------------* * The following two lines are added for adding lowercase * * character set for 99/4A 5/13/81 * ENDPRO BYTE >7B ASCII code for char after "z" * EVEN * *-----------------------------------------------------------* * * Normal scan mode -- figures out what to do with this char CRU16 MOVB *R4,*R4 At end of line? JEQ CRU28 Yes, clean up and return MOVB R0,R0 Set condition on char prop JLT CRU08 Ignore separators (spaces) MOV @RAMPTR,R9 Save crunch pointer SLA R0,2 Scan property bits 1 and 2 JOC CRU32 Break chars are 1 char tokens JLT CRU18 Alpha, prepare to pack name SLA R0,2 Scan property bits 3 and 4 JNC CRU20 Jump if not multi-char oper BL @GETCHR Check next char to see if we SRL R1,8 have a 2 char operator JEQ CRU32 If read end of line-single oper BL @BACKUP Backup read pointer CB @CPTBL(R1),@LBCPMO Next char also a multi-oper? JNE CRU32 No, want single-char oper BL @PUTCHR Copy in first char to oper JMP CRU36 And scan keyword table * Set name copy mode CRU18 LI R7,CRU76 Alphabetic: set name copy mode *-----------------------------------------------------------* * Insert following 2 lines for adding lowercase character * * set in 99/4A 5/12/81 * SRL R0,2 Adjust R0 for LOWUP routine * BL @LOWUP Translate lowercase to upper * * if necessary * *-----------------------------------------------------------* JMP CRU06 And resume copy * Handle single character operators CRU20 JLT CRU32 Bit 4: single character oper SLA R0,2 Scan property bits 5 and 6 JOC CRU24 If numeric JLT CRU26 If digit only CI R8,QUOTE Is it a string quote? JNE ERRIVN No, unknown char so error MOV R7,R10 Yes, save current mode CRU22 LI R8,QUOTEZ Convert char to quote token BL @PUTCHR Put in token LI R7,CRU68 Set string, copy mode MOV @RAMPTR,R5 Save pointer to length byte JMP CRU06 Continue copy w/quote token CRU24 CI R8,'.' A decimal point JNE CRU26 No, decode as numeric/line # LI R6,CRU96 Yes, decode as numeric CRU26 B *R6 Handle numeric or line # BERRSY B @CERSYN Long distance SYNTAX ERROR CRU27 BL @PUTCHR Put out last char before end INC @VARW Skip last character * Here for successful completion of scan CRU28 SWPB R8 Mark end of line with a null BL @PUTCHR Put the end of line in CRNADD EQU $+2 LI R0,CRNBUF Get start of crunch buffer NEG R0 Negate for backwards add A @RAMPTR,R0 Calculate line length MOVB @R0LB,@CHAT Save length for GPL BL @GETSTK Restore GROM address B *R12 Return with pointer beyond null * Keyword table scanning routine. Name has already been * copied into crunch area starting at R9; RAMPTR point just * beyond name in input line. * R3 is name length, R1 indexes into the table CRU32 BL @BACKUP Fix pointer for copy(next line) CRU36 BL @GETCHR Read last character MOVB R1,*R4 Put into output buffer BL @PUTCHR Copy into crunch buffer CRU38 MOV @RAMPTR,R3 Get end pointer S R9,R3 Sub start to get length of name CI R3,MAXKEY Is longer than any keyword? JH CRU61 Yes, can't be a keyword MOV R3,R2 Get name length and DEC R2 corremt 0 length name indexing SLA R2,1 Turn it into an index AI R2,KEYTAB Add in address of table list MOVB R2,@GRMWAX(R13) Load address to GROM SWPB R2 MOVB R2,@GRMWAX(R13) MOVB *R13,R2 Read address of correct table MOVB *R13,@R2LB Both bytes * R2 now contains the address of the correct table CRU40 MOVB R2,@GRMWAX(R13) Load address of table MOV R3,R0 Copy of length for compare MOVB @R2LB,@GRMWAX(R13) MOVB @R9LB,*R15 Source is in VDP A R3,R2 Address of next keyword in table MOVB R9,*R15 INC R2 Skip token value CRU42 CB @XVDPRD,*R13 Compare the character JL CRU61A If no match possible JNE CRU40 No match, but match possible DEC R0 Compared all? JNE CRU42 No, check next one MOV R9,@RAMPTR Name matched so throw out name MOVB *R13,*R4 Read the token value CLR @ARG4 Indicate keyword found * Check for specially crunched statements LI R7,CRU14 Assume a REM statement LI R0,SPECTB-1 Now check for special cases *********************************************************** < * For GRAM KRACKER XB or RichGKXB or SXB substitute with: * < * CI R8,>000B * < *********************************************************** < CI R8,MERGEZ Is this a command? < JH CRU47 No, continue on MOV @FAC,R3 Yes, attempt to put in program? JNE ERRCIP Yes, *COMMAND ILLEGAL IN PROGRAM* CI R9,CRNBUF Command 1st token in line? JNE BERRSY No, *SYNTAX ERROR* CRU47 INC R0 Skip offset value CB *R4,*R0+ In special table? JEQ CRU53A Yes, handle it JH CRU47 If still possible match *********************************************************** < * For GRAM KRACKER XB or RichGKXB or SXB substitute with: * < * CI R8,>000C * < *********************************************************** < CI R8,MERGEZ A specially scanned command? < JL CRU27 Yes, exit crunch LI R0,LNTAB Now check for line number CRU48 CB *R4,*R0+ In table? JEQ CRU52 Yes, change to line # crunch JH CRU48 May still be in table CI R8,COMMAZ Just crunch a comma? JEQ CRU50 Yes, so retain current numeric CI R8,TOZ Just crunch a TO? JNE CRU53 No, so reset to normal numeric CRU50 B @CRU05 Yes, resume normal copy CRU52 LI R6,CRU100 Set line number scan mode JMP CRU50 Set normal scan mode ERRIVN INC @ERRCOD *ILLEGAL VARIABLE NAME ERRCIP INC @ERRCOD *COMMAND ILLEGAL IN PROGRAM ERRNQT INC @ERRCOD *NONTERMINATED QUOTED STING CBH20 EQU $+1 ERRNTL A @C4,@ERRCOD *NAME TO LONG JMP CRU28 Exit back to GPL OFFSET EQU $ CRU53 B @CRU04 Stmt sep resets to normal scan CRU53A MOVB *R0,R1 Pick up offset from table SRL R1,8 Make into offset B @OFFSET(R1) Goto special case handler * Process a LIST statement CRU57 BL @PUTCHR Put the list token in BL @GETNB Get next character CI R1,QUOTE*256 Device name available? JNE CRU28 No, no more to crunch, exit LI R10,CRU106 Yes, set after string scan mode B @CRU22 Crunch the device name * Process an IMAGE statement CRU54 LI R10,CRU83B Image after, string copy mode JMP CRU59 Handle similar to data stmt * Process a DATA statement CRU58 LI R10,CRU83 After-datum skip spaces CRU59 C @RAMPTR,@CRNADD Image & data must be 1st on line JNE JNESY1 If not, error LI R2,CRU84 (non)quote string copy mode CRU60 LI R7,CRU80 Now set check-for-quote mode CRU74 B @CRU06 And copyin statement token * Here when don't find something in the keyword table CRU61 CI R3,15 Is it longer than name can be? JH ERRNTL Yes, name to long CRU61A MOV @ARG4,R0 Symbol name last time too? JNE JNESY1 Yes, can't have 2 in a row DEC @ARG4 Indicate symbol noe CRU62 LI R7,CRU16 No keyword,; leave in CRNBUF LI R6,CRU96 Assume normal numeric scan CRU64 B @CRU08 And continue to scan line * Process a SUB statement CRU65 MOV @RAMPTR,R3 Get the current crunch pointer DEC R3 Point at last character put in BL @GETV1 Read it CB R1,@GOZTOK Was it a GO? JEQ CRU52 Yes, SUB is part of GO SUB * Process a CALL SUB statement CRU66 LI R7,CRU93 Set name copy JMP CRU74 And get next character CRU32L B @CRU32 * Now the various mode copy routines; string, names, image, * and data statements CRU68 MOV R8,R8 Premature end of line? JEQ ERRNQT Yes, *NONTERMINATED QUOTED STRING CI R8,QUOTE Reach end of string? JNE CRU74 No, continue copying BL @GETCHR Get next character MOVB R1,R1 Read end of line? JEQ CRU70 Yes, can't be double quote CI R1,QUOTE*256 Is it two quotes in a row? JEQ CRU74 Yes, copy in a normal quote BL @BACKUP No, backup & rtn to normal scan CRU70 MOV R10,R7 Needed for image/data stmts CRU72 BL @LENGTH Calculate length of string JMP CRU64 Resume scan * Names *-----------------------------------------------------------* * Replace following two lines for adding lowercase * * character set in 99/4A 5/12/81 * * CRU76 ANDI R0,CPALNM*256 Is this char alpha or digit * * JEQ CRU74 Yes, continue packing * CRU76 ANDI R0,CPULNM*256 Is this char alpha (both are * * upper and lower) or a digit? * JNE CRU78 Yes, continue packing * *-----------------------------------------------------------* * No, finish w/name packing CI R8,'$' Does name end with a $? JEQ CRU32L Yes, include it in name MOVB *R4,*R4 At an end of line? JEQ CRU79 Yes, don't back up pointer BL @BACKUP Backup for next char CRU79 B @CRU38 Jump to name/keyword check CRU82 B @CRU22 *-----------------------------------------------------------* * Add following 2 lines for adding lowercase character set * * for 99/4A 5/12/81 * CRU78 BL @LOWUP Translate lower to upper if * * necessary * JMP CRU74 Continue packing * *-----------------------------------------------------------* * DATA: Scan spaces after a quoted string datum CRU83 CI R8,COMMA Hit a comma? JEQ CRU85A Yes, get back into scan * IMAGE: Scan spaces after a quoted string datum CRU83B MOVB R0,R0 At a space? JLT CRU64 Yes, ignore it MOV R8,R8 At end of line? JEQ CRU62 Yes, exit scan JNESY1 JMP JNESYN No, unknown character * DATA: Scan imbedded blanks and check trailing blanks CRU83A MOV @VARW,@ARG2 Save input pointer BL @GETNB Look for next non-blank MOVB R1,R1 At end of line? JEQ CRU92 Yes, end string and exit CI R10,CRU83B Scanning an image? JEQ CRU83C Yes, commas are not significant CI R1,COMMA*256 Hit a comma? JEQ CRU85 Yes, ignore trailing spaces CRU83C MOV @ARG2,@VARW No, restore input pointer JMP CRU74 and include imbedded space * DATA: Scan unquoted strings CRU84 JLT CRU83A If hit a space-end of string MOV R8,R8 At end-of-line? JEQ CRU92 Yes, put in length and exit CI R8,COMMA Reached a comma? JNE CRU74 No, scan unquoted string CI R10,CRU83B Scanning an IMAGE stmt? JEQ CRU74 Commas are not significant CRU85 BL @LENGTH Yes, end the string CRU85A LI R8,COMMAZ Load a comma token INC @VAR5 Count comma for input stmt JMP CRU60 And resume in string mode * IMAGE/DATA: Check for leading quote mark CRU80 JLT CRU64 Ignore leading separators CI R8,QUOTE Quotoed string? JEQ CRU82 Yes, like any string, R10 ok MOV R8,R8 End of line? JEQ BCRU28 Yes, end it CI R10,CRU83B Scanning an IMAGE? JEQ CRU88 Yes, ignore commas CI R8,COMMA At a comma? JEQ CRU85A Yes, put it in directly CRU88 MOV R2,R7 No, set unquote string copy mode * IMAGE & DATA: Scan unquoted strings CRU86 LI R8,UNQSTZ Load unquoted string token BL @PUTCHR Put the token in MOV @RAMPTR,R5 Save current crunch pointer BL @BACKUP Back up to scan again CRU87 JMP CRU74 Resume scan * CALL and SUB statements *-----------------------------------------------------------* * Replace following 2 lines for adding lowercase character * * set for 99/4A 5/12/81 * * CRU94 ANDI R0,CPALNM*256 Still an alpha-numeric * * JNE CRU74 Yes, include in name * CRU94 ANDI R0,CPULNM*256 Still an alpha(U & L)-numeric * JNE CRU91 Yes, transfer L to U, then * * include in name * *-----------------------------------------------------------* MOV R8,R8 At end of line? JEQ CRU92 Yes, get out now CRU90 BL @BACKUP No, reset read pointer CRU92 LI R7,CRU16 Normal scanning mode JMP CRU72 Calculate & put in string length *-----------------------------------------------------------* * Add following lines for adding lowercase character set * * for 99/4A 5/12/81 * CRU91 BL @LOWUP Transfer lowercase char to * * uppercase char if necessary * B @CRU74 Include in name * *-----------------------------------------------------------* * CALL and SUB statements before hit name CRU93 JLT CRU64 If a space, ignore it MOV R0,R0 Premature EOL or NIL char, prop? JEQ CERSYN Yes, *SYNTAX ERROR *-----------------------------------------------------------* * Replace following line for adding lowercase character set * * for 99/4A 5/12/81 * * ANDI R0,CPALPH*256 An alphabetic to start name?* ANDI R0,CPUL*256 An alphabetic (both U & L) to * * start name? * *-----------------------------------------------------------* JEQ CERSYN No, syntax error LI R7,CRU94 Set up to copy name JMP CRU86 Put in the unqst token * Numerics CRU96 LI R7,CRU98 Set after-initialize scan CLR @ARG Clear the 'E' flag JMP CRU86 Set up for the numeric CRU98 MOV R8,R8 At end of line? JEQ CRU92 Yes end the number SLA R0,2 Scan property bit 2 JLT CRU99A If alpha, might ge 'E' SLA R0,3 Scan property bits 4 and 5 JNC CRU99 Bit 4=oper, if not oper, jmp MOV @ARG,R0 If operator, follow an 'E'? CRU99 CLR @ARG Previous char no longer an 'E' JLT CRU87 If still numeric JMP CRU90 No longer numeric CRU99A CI R8,'E' 'E' to indicate an exponent? JNE CRU90 No, so end the numeric MOV @ARG,R0 An 'E' already encountered? JNESYN JNE CERSYN Yes, so error SETO @ARG No, indicated 1 encountered now JMP CRU87 And include it in the number * Line numbers CRU100 MOV R8,R8 At end of line? JEQ BCRU28 Yes, exit crunch BL @GETINT Try to get a line number MOV R0,R0 Get a line number? JEQ CRU105 No, back to normal numeric mode LI R8,LNZ Load a line number token BL @PUTCHR Put it out MOV R0,R8 Set up to put out binary # SWPB R8 Swap to put MSB of # 1st BL @PUTCHR Put out 1st byte of line # SRL R8,8 Bare the 2nd byte of line # JMP CRU87 Jump back into it CRU105 B @CRU04 Back to normal numeric mode * Handle a list statement CRU106 JLT CRU93 If space, ignore it MOV R8,R8 At end of line? JEQ BCRU28 Yes, exit crunch CI R8,':' Get a colon? JNE CERSYN No, *SYNTAX ERROR LI R8,COLONZ Need to put colon in B @CRU27 And exit crunch * Error handling routine ERRLTL INC @ERRCOD * LINE TO LONG 3 DECT @RAMPTR Backup so can exit to GPL ERRBLN INC @ERRCOD * BAD LINE NUMBER 2 CERSYN INC @ERRCOD * SYNTAX ERROR 1 BCRU28 B @CRU28 Exit back to GPL * Back up pointer in input line to rescan last character BACKUP DEC @VARW Back up the pointer MOVB @VARW1,*R15 Write LSB of address NOP MOVB @VARW,*R15 Write MSB of address LI R0,>7F00 >7F is an edge character <<<<<<<<<< SB @XVDPRD,R0 At an edge chracter? JEQ BACKUP Yes, back up one more RT And return to caller * Put a character into the crunch buffer PUTCHR MOV @RAMPTR,R1 Fetch the current pointer CI R1,CRNEND At end of buffer? JH ERRLTL Yes, LINE TO LONG MOVB @R1LB,*R15 Put out LSB of address ORI R1,WRVDP Enable VDP write MOVB R1,*R15 Put out MSB of address INC @RAMPTR Increment the pointer MOVB *R4,@XVDPWD Write out the byte RT And return *-----------------------------------------------------------* * Move LENGTH to GETNB, becuase CRUNCH is running out of * * space, 1/21/81 * * Calculate and put length of string/number into length * * byte * * LENGTH MOV R11,R3 Save return address * * MOV @RAMPTR,R0 Save current crunch pointer * * MOV R0,R8 Put into R8 for PUTCHR below* * S R5,R8 Calculate length of string * * DEC R8 RAMPTR is post-incremented * * MOV R5,@RAMPTR Address of length byte * * BL @PUTCHR Put the length in * * MOV R0,@RAMPTR Restore crunch pointer * * B *R3 And return * *-----------------------------------------------------------* * * Get a small non-negative integer * CALL: VARW - TEXT POINTER, points to second character * R8 - First character in low byte * BL @GETINT * R0 - NUMBER * VARW - Text pointer, if there is a number, points to * character after number. If there is not a * number, unchanged. * R8 - 0 in high byte * DESTROYS: R1, R2 GETINT MOV R11,R3 Save return address MOV R8,R0 Get possible digit LI R2,10 Get radix in register for speed AI R0,-'0' Convert from ASCII to binary C R0,R2 Is the character a digit? JL GETI02 Yes, there is a number! CLR R0 No, indicate no number B *R3 Done, no number GETI01 MPY R2,R0 Multiply previous by radix MOV R0,R0 Overflow? JNE ERRBLN Yes, bad line number MOV R1,R0 Get low order word of product A R8,R0 Add in next digit JLT ERRBLN If number went negative, error GETI02 BL @GETCHR Get next character MOVB R1,*R4 Put into normal position JEQ GETI03 If read end of line AI R8,-'0' Convert from ASCII to binary C R8,R2 Is this character a digit? JL GETI01 Yes, try to pack it in DEC @VARW No point to 1st char after number GETI03 CLR R8 Clean up our mess MOV R0,R0 Hit a natural zero? JEQ ERRBLN Yes, its an error B *R3 And return * The LINE NUMER TABLE * All tokens which appear in the table must have numerics * which follow them crunched as line numbers. LNTAB BYTE ELSEZ GOZTOK BYTE GOZ <<<<<<<<<< BYTE GOTOZ BYTE GOSUBZ BYTE RETURZ BYTE BREAKZ BYTE UNBRKZ BYTE RESTOZ BYTE ERRORZ BYTE RUNZ BYTE THENZ BYTE USINGZ BYTE >FF Indicate end of table EVEN ************************************************************* * Table of specially crunched statements * * 2 bytes - special token * * Byte 1 - token value * * Byte 2 - "address" of special handler * * Offset from label OFFSET in this assembly of * * the special case handler * ************************************************************* SPECTB BYTE LISTZ,CRU57-OFFSET BYTE OLDZ,CRU58-OFFSET BYTE SAVEZ,CRU58-OFFSET BYTE MERGEZ,CRU58-OFFSET BYTE SSEPZ,CRU53-OFFSET BYTE TREMZ,CRU74-OFFSET BYTE DATAZ,CRU58-OFFSET BYTE REMZ,CRU74-OFFSET BYTE CALLZ,CRU66-OFFSET BYTE SUBZ,CRU65-OFFSET BYTE IMAGEZ,CRU54-OFFSET BYTE >FF EVEN * * TRANSFER LOWERCASE CHARACTER TO UPPERCASE CHARACTER * R0 - Last digit indicates whether this character is a * lowercase character LOWUP ANDI R0,CPLOW*256 Is lowercase prop set? JEQ LU01 No, just return SB @CBH20,*R4 Change lower to upper LU01 RT AORG >7FFE DATA >C68C ******************************************************************************** 1 1 Quote Link to comment Share on other sites More sharing options...
+InsaneMultitasker Posted March 12 Share Posted March 12 1 hour ago, HOME AUTOMATION said: Maybe this... ******************************************************************************** AORG >7B88 TITL 'CRUNCHS' That's the file! Is this file and related files somewhere on WHT and/or another repository? I compared this file to the Geneve's ABASIC source, and the latter is definitely built upon the XB source. It even answered a few questions. Thank you, very cool. 4 Quote Link to comment Share on other sites More sharing options...
HOME AUTOMATION Posted March 12 Share Posted March 12 Got it here... 2 1 Quote Link to comment Share on other sites More sharing options...
+Retrospect Posted March 15 Share Posted March 15 I did a quick RXB demo showing bit-shifted scrolling, twinkling stars using RXB instead of ordinary XB. These stars will scroll a little bit faster than normal XB due to CALL HPUT being quite perky. I've provided the source code for anyone who wants to have a play around with it. Video shows program running under normal speed ( still quite slow, the routine was originally meant for compiled games! ) and running under CPU overdrive ( more acceptable for what it is )RXB BIT SHIFTED STARS.txt Note: The great thing about CALL HPUT is that it can address more .... there's a line in there that needs to delete a star and then re-display it 1 column to the left, HPUT can do this in one command and that saves time and bytes in comparison to normal XB which would have to use DISPLAY AT twice. Classic99 QI399.063 - CPU Overdrive 2023-03-15 00-36-09.mp4 3 Quote Link to comment Share on other sites More sharing options...
+Vorticon Posted March 20 Share Posted March 20 I was intrigued by the CALL USER functionality of RXB2023, so I tried a small test. I wrote a test program in Stevie, saved it to disk, and tried to load it using CALL USER, but it's coming up all on 1 line. The program displays fine in TI99Dir and each line is properly terminated by a carriage return. Any thoughts on the issue? Classic99 QI399.063 - CPU Overdrive 2023-03-20 16-51-34.mp4 Quote Link to comment Share on other sites More sharing options...
MikeV Posted March 21 Share Posted March 21 16 hours ago, Vorticon said: I was intrigued by the CALL USER functionality of RXB2023, so I tried a small test. I wrote a test program in Stevie, saved it to disk, and tried to load it using CALL USER, but it's coming up all on 1 line. The program displays fine in TI99Dir and each line is properly terminated by a carriage return. Any thoughts on the issue? I've only experimented with CALL USER to see how it works. While it works as advertised I have not found many uses for it to date. Entering the same short program in TI Writer works as it should, the E/A (no CR) will produce the same results you see. Fred's nifty adaptation of notepad in TIDIR displays both the TIW and E/A versions properly (it's one of those smart programs). I would check and see if your line entries all terminate with a ".", which is how Fred's program represents a CR. I unfortunately do not know anything about Stevie... yet! It is on the to-do list. Quote Link to comment Share on other sites More sharing options...
+Vorticon Posted March 21 Share Posted March 21 46 minutes ago, MikeV said: I've only experimented with CALL USER to see how it works. While it works as advertised I have not found many uses for it to date. Entering the same short program in TI Writer works as it should, the E/A (no CR) will produce the same results you see. Fred's nifty adaptation of notepad in TIDIR displays both the TIW and E/A versions properly (it's one of those smart programs). I would check and see if your line entries all terminate with a ".", which is how Fred's program represents a CR. I unfortunately do not know anything about Stevie... yet! It is on the to-do list. Ah! That may be the case. I'll check and see how Stevie terminates lines. The CALL USER functionality actually allows you to create and edit XB programs in a full screen proper editor then simply "paste" them into RXB without the need to use emulators or external PC programs. Few users nowadays use real hardware for initial development but there are special cases where programming on the real thing is necessary such as when working on interfacing projects as in my case. So this functionality will come in real handy! 1 Quote Link to comment Share on other sites More sharing options...
+Vorticon Posted March 24 Share Posted March 24 XB GEM has a DV2XB subprogram which does the same thing but a lot faster and does not require that pesky CR line termination. I really think RXB should could benefit from enhancing the CALL USER functionality and removing the CR limitation. Quote Link to comment Share on other sites More sharing options...
+RXB Posted May 25 Share Posted May 25 On 3/24/2023 at 5:24 AM, Vorticon said: XB GEM has a DV2XB subprogram which does the same thing but a lot faster and does not require that pesky CR line termination. I really think RXB should could benefit from enhancing the CALL USER functionality and removing the CR limitation. XB 256 cannot do half as much as RXB CALL USER, can it edit lines and run multiple XB programs and load Assembly support routines all from a single DV 80 file? I demoed in 2000 RXB CALL USER ran 52K of Assembly support from various XB Assembly programs. (I did not write any of them like STAR or XB40) So XB 256 is great but 20 years behind what I already did in 2000. Quote Link to comment Share on other sites More sharing options...
+RXB Posted May 25 Share Posted May 25 RXB 2023A now has a routine installed that was discussed on the XB 256 page. I think this came from CALL LOAD(8192,A,B,"",8198,C,D) where the "" skips to another address. I always thought this was more than a strange way to skip to another address or filename. The idea in XB 256 was to use CALL PEEK(8192,A,B,"",8198,C,D) The "" was to indicate a jump to another address in the PEEK routine. I thought this was a waste of space as double quotes is 2 characters along with the previous comma and ending comma before the next address. RXB 2023A has modified this with a single character token insted of the quote marks it uses the & (and) symbol token. So RXB 2023A has CALL PEEK(8192,A,B,&,8198,C,D) One less token used and when reading it makes a little more logical sense to read. 1 Quote Link to comment Share on other sites More sharing options...
Like us on Facebook
Follow us on Twitter
AtariAge on Instagram
Watch us on YouTube
Chat on Discord
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.