Debugging

[britlion]

Is a pain.

Right now I've got a program that:

Code:

FOR n=40 to 50 FastPlot (n,50) next n

draws pixels in all the wrong places So my fastplot is wrong?

Code:

FOR n=40 to 50 FastPlot (n,50) pause 0 next n

Ah! That fixed it. Now it plots all the pixels, with pauses, in all the right places.

*headscratch*

Presumably fastplot is wrong. I think. But...huh? How does that make it work better?!?

Code:

SUB FastPlot (x as uByte, y as uByte) ASM ld HL,FastPlotYValue+1 LD E,(IX+7) ; y LD (HL),E call FastPlotASMStart END ASM return ASM FastPlotASMStart: ld hl,BLtabpow2 ld (FastPlotSaveX+1),A and 7 ; x mod 8 ;ld b,0 ;ld c,a ;add hl,bc add a,l ld l,a jr nc, FastPlotSkip1 inc h FastPlotSkip1: ;ld a,(hl) ; optimize - load e with (hl) directly, not (hl) -> a -> e. ld e,(hl) ; e contains one bit set ld hl,BLTabLinIdx END ASM FastPlotYValue: ASM FastPlotYValue: ld a,0 ; Self modding code that puts y value here ;ld b,0 ;ld c,a ;add hl,bc add a,l ld l,a jr nc, FastPlotSkip2 inc h FastPlotSkip2: ld a,(hl) ; table lookup ;ld h,0 ;ld l,a ;add hl,hl ;add hl,hl ;add hl,hl ;add hl,hl ;add hl,hl ; x32 (16 bits) (this multiply routine is 66 t states) rrca rrca rrca ld h,a and %11100000 xor h or %01000000 ld h,a ;- this is faster at 35 T states to do hl=a*32 ;set 6,h ; adds the screen start address (16384) (moved up to previous and %010 instruction) FastPlotSaveX: ld a, 255 ; X Self Modifying - this is written into to save X earlier. ;srl a ;srl a ;srl a ; x/8 24 T states rrca rrca rrca and %00011111 ;x / 8 19 T states or l ld l,a ; + x/8. ld a,(hl) or e ; or = superposition mode. ld (hl),a ; set the pixel. ret END ASM #ifndef BLTabLineIdxPresent #define BLTabLineIdxPresent ASM ;; small screen lines lookup table BLTabLinIdx: defb 0,8,16,24,32,40,48,56,1,9,17,25,33,41,49,57 defb 2,10,18,26,34,42,50,58,3,11,19,27,35,43,51,59 defb 4,12,20,28,36,44,52,60,5,13,21,29,37,45,53,61 defb 6,14,22,30,38,46,54,62,7,15,23,31,39,47,55,63 defb 64,72,80,88,96,104,112,120,65,73,81,89,97,105,113,121 defb 66,74,82,90,98,106,114,122,67,75,83,91,99,107,115,123 defb 68,76,84,92,100,108,116,124,69,77,85,93,101,109,117,125 defb 70,78,86,94,102,110,118,126,71,79,87,95,103,111,119,127 defb 128,136,144,152,160,168,176,184,129,137,145,153,161,169,177,185 defb 130,138,146,154,162,170,178,186,131,139,147,155,163,171,179,187 defb 132,140,148,156,164,172,180,188,133,141,149,157,165,173,181,189 defb 134,142,150,158,166,174,182,190,135,143,151,159,167,175,183,191 BLtabpow2: ;; lookup table with powers of 2 defb 128,64,32,16,8,4,2,1 END ASM #endif END SUB FOR n=40 to 50 FastPlot (n,60) next n FOR n=40 to 50 FastPlot (n,50) pause 0 next n

Ideas accepted

(Not sure any are in the right places, but the non pause version had them non-contiguous, which was very weird).

[boriel]

Hi, Britlion

Sorry for the delay! (as always)
Are you using this FastPlot implementation?
https://zxbasic.readthedocs.io/en/docs/l.../fastplot/

Ok. You're using the 2nd implementation (the Wiki one does not include the lookup table?)

Update: In the original code, even with PAUSE 0, the code plotted in the wrong region of the screen for me...
Update: I've fastcalled your routine. Now it does not hang, but still the plots are a bit odd

Code:

SUB Fastcall FastPlot (x as uByte, y as uByte) ASM pop hl ; hl = RETURN ADDRESS ex (sp), hl ; Callee H = y ex de, hl ; d = y ld HL,FastPlotYValue+1 ld (hl), d FastPlotASMStart: ld hl,BLtabpow2 ld (FastPlotSaveX+1),A and 7 ; x mod 8 add a,l ld l,a jr nc, FastPlotSkip1 inc h FastPlotSkip1: ld e,(hl) ; e contains one bit set ld hl,BLTabLinIdx FastPlotYValue: ld a,0 ; Self modding code that puts y value here add a,l ld l,a jr nc, FastPlotSkip2 inc h FastPlotSkip2: ld a,(hl) ; table lookup ; OR a <= CLEAR CARRY FLAG missing?? rrca rrca rrca ld h,a and %11100000 xor h or %01000000 ld h,a ;- this is faster at 35 T states to do hl=a*32 FastPlotSaveX: ld a, 255 ; X Self Modifying - this is written into to save X earlier. rrca rrca rrca and %00011111 ;x / 8 19 T states or l ld l,a ; + x/8. ld a,(hl) or e ; or = superposition mode. ld (hl),a ; set the pixel. END ASM return #ifndef BLTabLineIdxPresent #define BLTabLineIdxPresent ASM ;; small screen lines lookup table BLTabLinIdx: defb 0,8,16,24,32,40,48,56,1,9,17,25,33,41,49,57 defb 2,10,18,26,34,42,50,58,3,11,19,27,35,43,51,59 defb 4,12,20,28,36,44,52,60,5,13,21,29,37,45,53,61 defb 6,14,22,30,38,46,54,62,7,15,23,31,39,47,55,63 defb 64,72,80,88,96,104,112,120,65,73,81,89,97,105,113,121 defb 66,74,82,90,98,106,114,122,67,75,83,91,99,107,115,123 defb 68,76,84,92,100,108,116,124,69,77,85,93,101,109,117,125 defb 70,78,86,94,102,110,118,126,71,79,87,95,103,111,119,127 defb 128,136,144,152,160,168,176,184,129,137,145,153,161,169,177,185 defb 130,138,146,154,162,170,178,186,131,139,147,155,163,171,179,187 defb 132,140,148,156,164,172,180,188,133,141,149,157,165,173,181,189 defb 134,142,150,158,166,174,182,190,135,143,151,159,167,175,183,191 BLtabpow2: ;; lookup table with powers of 2 defb 128,64,32,16,8,4,2,1 END ASM #endif END SUB FOR n=40 to 50 FastPlot (n,60) next n FOR n=40 to 50 FastPlot (n,50) pause 0 next n

[boriel]

More updates: I've located a possible bug (signaled in the code above):

• Did you forgot to clear the Carry Flag (with an OR a) before doing RRCA (see the mark above, after the label FastPlotSkip2.
  
• I think you also forgot to subtract the y value from 191 as in the original plot routines, so the Y position is reversed or eve n out of the screen.
  

Can you check that? 

[britlion]

I can.

I rarely can get here, since the site refuses to allow me to access it with my VPN running, which is a pain.

Thanks for looking - I'll go dig on that.

[britlion]

Ok. You're using the 2nd implementation (the Wiki one does not include the lookup table?)

The wiki should probably link. It does say "HRPrintFast also uses the same table - it's important to only include it once" - however, the screen tables are enormous in full - so even HRPrintFast links to the saved file in this forum... - I couldn't see a way to attach a file to the wiki.