Author Topic: Texture drawing (Read 34503 times)

AngelFish · « **Reply #60 on:** July 01, 2011, 06:01:23 pm »

Awww, you're not a Brandon. As a general rule, people whose first names are Brandon are always supper uber-1337 coderz who can do pretty much anything they want with calculators.

ben_g · « **Reply #61 on:** July 01, 2011, 06:03:18 pm »

Your previous post lets me think your name is Brandon. Am I correct?

AngelFish · « **Reply #62 on:** July 01, 2011, 06:04:07 pm »

O.o

No, but thanks for the compliment.

Calc84maniac, on the other hand...

ben_g · « **Reply #63 on:** July 01, 2011, 06:07:44 pm »

But anyway, Why would you think I'm an uber 1337 coder when I'm just a newb?

* ben_g turns of his computer and will be gone for 10 days.

ben_g · « **Reply #64 on:** July 14, 2011, 11:06:00 am »

I have ported half of the code to asm. This code should draw the top third of a textured quad. My problem is that it doesn't draw anything. It gets stuck in an endless loop, and I don't know why.

Spoiler For The code:

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; ;TexQuad ;in: x1, y1, u1, v1, x2, y2, u2, v2, x3, y3, u3, v3, x4, y4, u4, v4 ;out: textured quad is drawn to the buffer TexQuad: ld a, (y1) ld b, a ld a, (y2) sub b ld (temp), a ld a, (x1) ld b, a ld a, (x2) sub b ld h, a ld l, 0 ld de, (temp) ld e, 0 call DivFP ld (dx1), hl ld a, (y1) ld b, a ld a, (y3) sub b ld (temp), a ld a, (x1) ld b, a ld a, (x3) sub b ld h, a ld l, 0 ld de, (temp) ld e, 0 call DivFP ld (dx2), hl ld a, (y2) ld b, a ld a, (y4) sub b ld (temp), a ld a, (x2) ld b, a ld a, (x4) sub b ld h, a ld l, 0 ld de, (temp) ld e, 0 call DivFP ld (dx3), hl ld a, (y3) ld b, a ld a, (y4) sub b ld (temp), a ld a, (x3) ld b, a ld a, (x4) sub b ld h, a ld l, 0 ld de, (temp) ld e, 0 call DivFP ld (dx4), hl ld a, (y1) ld b, a ld a, (y2) sub b ld (temp), a ld a, (u1) ld b, a ld a, (u2) sub b ld h, a ld l, 0 ld de, (temp) ld e, 0 call DivFP ld (du1), hl ld a, (y1) ld b, a ld a, (y3) sub b ld (temp), a ld a, (u1) ld b, a ld a, (u3) sub b ld h, a ld l, 0 ld de, (temp) ld e, 0 call DivFP ld (du2), hl ld a, (y2) ld b, a ld a, (y4) sub b ld (temp), a ld a, (u2) ld b, a ld a, (u4) sub b ld h, a ld l, 0 ld de, (temp) ld e, 0 call DivFP ld (du3), hl ld a, (y3) ld b, a ld a, (y4) sub b ld (temp), a ld a, (u3) ld b, a ld a, (u4) sub b ld h, a ld l, 0 ld de, (temp) ld e, 0 call DivFP ld (du4), hl ld a, (y1) ld b, a ld a, (y2) sub b ld (temp), a ld a, (v1) ld b, a ld a, (v2) sub b ld h, a ld l, 0 ld de, (temp) ld e, 0 call DivFP ld (dv1), hl ld a, (y1) ld b, a ld a, (y3) sub b ld (temp), a ld a, (v1) ld b, a ld a, (v3) sub b ld h, a ld l, 0 ld de, (temp) ld e, 0 call DivFP ld (dv2), hl ld a, (y2) ld b, a ld a, (y4) sub b ld (temp), a ld a, (v2) ld b, a ld a, (v4) sub b ld h, a ld l, 0 ld de, (temp) ld e, 0 call DivFP ld (dv3), hl ld a, (y3) ld b, a ld a, (y4) sub b ld (temp), a ld a, (v3) ld b, a ld a, (v4) sub b ld h, a ld l, 0 ld de, (temp) ld e, 0 call DivFP ld (dv4), hl ld hl, (x1) ld l, 0 ld (tx1), hl ld hl, (x2) ld l, 0 ld (tx2), hl ld hl, (u1) ld l, 0 ld (tu1), hl ld hl, (v1) ld l, 0 ld (tv1), hl ld hl, (u2) ld l, 0 ld (tu2), hl ld hl, (v2) ld l, 0 ld (tv2), hl ld hl, (y1) ld l, 0 ld (_ty), hl QuadDrawLoop1: ld a, (tx1) ld (temp), a ;i=temp QuadPlotLoop1: ld a, (tx1) ld b, a ld a, (tx2) sub b ld c, a ld a, (tx1) ld d, a ld e, 0 ld a, (temp) sub b ld h, a ld l, 0 call DivFP ld a, h ld (temp+1), a ;tmp=temp+1 ld b, a ld a, 1 sub b ld d, a ld e, 0 ld a, (tu1) ld b, a ld c, 0 call MulFP push hl ld de, (temp+1) ld e, 0 ld bc, (tu2) ld c, 0 call MulFP pop de add hl, de ld a, h ld (temp2), a ;u=temp2 ld a, (temp+1) ld b, a ld a, 1 sub b ld d, a ld e, 0 ld a, (tv1) ld b, a ld c, 0 call MulFP push hl ld de, (temp+1) ld e, 0 ld bc, (tv2) ld c, 0 call MulFP pop de add hl, de ld a, h ld (temp2+1), a ;v=temp2+1 ;Get the pixel at (u,v) from the texture ld a, (temp2+1) ;load the v coordinate in a ld hl, texture ld d, 0 ld e, a add hl, de ld a, (hl) ;and load the byte of the v coordinate in a ld c, a ld a, (temp2) ld b, 1 cp b jr c, BitShiftDone ;no shifting if u=0 ld b, a BitShiftLoop: sll c djnz BitShiftLoop BitShiftDone: push bc ;1 ld a, (_ty) ld l, a ld a, (temp) getPixel: ld h, 0 ld d, h ld e, l add hl, hl add hl, de add hl, hl add hl, hl ld e, a srl e srl e srl e add hl, de ld de, PlotSScreen add hl, de and 7 ld b, a ld a, $80 ret z rrca djnz $-1 pop bc ;0 bit 7, c jr z, ResPixel SetPixel: or (hl) ld (hl), a jr EndPlot ResPixel: cpl and (hl) ld (hl), a EndPlot: ld hl, (tx1) ld de, (dx1) add hl, de ld (tx1), hl ld hl, (tx2) ld de, (dx2) add hl, de ld (tx2), hl ld hl, (tu1) ld de, (du1) add hl, de ld (tu1), hl ld hl, (tu2) ld de, (du2) add hl, de ld (tu2), hl ld hl, (tv1) ld de, (dv1) add hl, de ld (tv1), hl ld hl, (tv2) ld de, (dv2) add hl, de ld (tv2), hl ld a, (_ty) inc a ld (_ty), a ld c, a ld a, (y2) ld h, a ld a, c ld c, h cp c jp nz, QuadDrawLoop1 ret ;voorlopig

All of the math routines are for 8.8 signed floating point math. These are all fully tested so they don't couse the cash.

thepenguin77 · « **Reply #65 on:** July 14, 2011, 01:13:56 pm »

Now that is one of the scariest pieces of code I have ever seen.

And this	should definitely be this
Code: [Select] `ld a, (_ty) inc a ld (_ty), a ld c, a ld a, (y2) ld h, a ld a, c ld c, h cp c jp nz, QuadDrawLoop1`	Code: [Select] `ld hl, _ty inc (hl) ld a, (y2) cp (hl) jp nz, QuadDrawLoop1`

But, as to why it's not actually working. Your loop should theoretically work as long as you are giving it valid numbers. (It will actually finish no matter what you give it though) So that leads me to believe that the problem is with your pixel plotting. Most likely, you are trying to plot pixels that are off screen and that is what is causing all of your problems. You end up plotting pixels in your program code, which causes crashes.

If you are really unlucky, you might even be plotting pixels at _ty.

ben_g · « **Reply #66 on:** July 14, 2011, 05:18:07 pm »

Quote from: thepenguin77 on July 14, 2011, 01:13:56 pm

Now that is one of the scariest pieces of code I have ever seen.

How do you mean?

Quote from: thepenguin77 on July 14, 2011, 01:13:56 pm

But, as to why it's not actually working. Your loop should theoretically work as long as you are giving it valid numbers. (It will actually finish no matter what you give it though) So that leads me to believe that the problem is with your pixel plotting. Most likely, you are trying to plot pixels that are off screen and that is what is causing all of your problems. You end up plotting pixels in your program code, which causes crashes.

If you are really unlucky, you might even be plotting pixels at _ty.

I tried jumping over the plotting part, and it still crashes. By the way, this is done in an app, and ass variables defined inside the app itself won't work, then isn't the program protected against SMC?

By the way: thanks for the optimization.

thepenguin77 · « **Reply #67 on:** July 14, 2011, 05:43:19 pm »

It's scary because it's so massive and very repetitive (though I've seen worse; much, much worse). I'm sure with the proper optimizations you could probably cut about 25% or more off.

Anyways, I believe I found the error. You inlined your GetPixel which I think is rather funny, but there was an extra RET Z in it that you missed. This normally wouldn't cause crashes, but you did a PUSH BC around it, so you are returning into BC, which is definitely not a place where you want to go.

And yes, being in an app protects you accidental SMC.

ben_g · « **Reply #68 on:** July 14, 2011, 06:42:07 pm »

The ret z dit couse the crash.

But it still doesn't work. It draws a line instaed while it should draw the top third of the quad in solid black (the texture is an 8x8 solid rectangle).

ben_g · « **Reply #69 on:** July 16, 2011, 07:44:20 pm »

many thanks to thepenguin77 and calc84 for helping me through omnomIRC.

But it still doesn't work. It gets stuck in a endless loop again.

I included the most recent scource code.

Spoiler For source code:

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; ;TexQuad ;in: x1, y1, u1, v1, x2, y2, u2, v2, x3, y3, u3, v3, x4, y4, u4, v4 ;out: textured quad is drawn to the buffer TexQuad: ld a, (y1) ld b, a ld a, (y2) sub b ld (temp), a ld a, (x1) ld b, a ld a, (x2) sub b ld h, a ld l, 0 ld de, (temp) ld e, 0 call DivFP ld (dx1), hl ld a, (y1) ld b, a ld a, (y3) sub b ld (temp), a ld a, (x1) ld b, a ld a, (x3) sub b ld h, a ld l, 0 ld a, (temp) ld d, a ld e, 0 call DivFP ld (dx2), hl ld a, (y2) ld b, a ld a, (y4) sub b ld (temp), a ld a, (x2) ld b, a ld a, (x4) sub b ld h, a ld l, 0 ld a, (temp) ld d, a ld e, 0 call DivFP ld (dx3), hl ld a, (y3) ld b, a ld a, (y4) sub b ld (temp), a ld a, (x3) ld b, a ld a, (x4) sub b ld h, a ld l, 0 ld a, (temp) ld d, a ld e, 0 call DivFP ld (dx4), hl ld a, (y1) ld b, a ld a, (y2) sub b ld (temp), a ld a, (u1) ld b, a ld a, (u2) sub b ld h, a ld l, 0 ld a, (temp) ld d, a ld e, 0 call DivFP ld (du1), hl ld a, (y1) ld b, a ld a, (y3) sub b ld (temp), a ld a, (u1) ld b, a ld a, (u3) sub b ld h, a ld l, 0 ld a, (temp) ld d, a ld e, 0 call DivFP ld (du2), hl ld a, (y2) ld b, a ld a, (y4) sub b ld (temp), a ld a, (u2) ld b, a ld a, (u4) sub b ld h, a ld l, 0 ld a, (temp) ld d, a ld e, 0 call DivFP ld (du3), hl ld a, (y3) ld b, a ld a, (y4) sub b ld (temp), a ld a, (u3) ld b, a ld a, (u4) sub b ld h, a ld l, 0 ld a, (temp) ld d, a ld e, 0 call DivFP ld (du4), hl ld a, (y1) ld b, a ld a, (y2) sub b ld (temp), a ld a, (v1) ld b, a ld a, (v2) sub b ld h, a ld l, 0 ld a, (temp) ld d, a ld e, 0 call DivFP ld (dv1), hl ld a, (y1) ld b, a ld a, (y3) sub b ld (temp), a ld a, (v1) ld b, a ld a, (v3) sub b ld h, a ld l, 0 ld a, (temp) ld d, a ld e, 0 call DivFP ld (dv2), hl ld a, (y2) ld b, a ld a, (y4) sub b ld (temp), a ld a, (v2) ld b, a ld a, (v4) sub b ld h, a ld l, 0 ld a, (temp) ld d, a ld e, 0 call DivFP ld (dv3), hl ld a, (y3) ld b, a ld a, (y4) sub b ld (temp), a ld a, (v3) ld b, a ld a, (v4) sub b ld h, a ld l, 0 ld a, (temp) ld d, a ld e, 0 call DivFP ld (dv4), hl ld a, (x1) ld h, a ld l, 0 ld (tx1), hl ld a, (x2) ld h, a ld l, 0 ld (tx2), hl ld a, (u1) ld h, a ld l, 0 ld (tu1), hl ld a, (v1) ld h, a ld l, 0 ld (tv1), hl ld a, (u2) ld h, a ld l, 0 ld (tu2), hl ld a, (v2) ld h, a ld l, 0 ld (tv2), hl ld a, (y1) ld (_ty), a QuadDrawLoop1: ld a, (tx1) ld (temp), a ;i=temp QuadPlotLoop1: ld a, (tx1) ld b, a ld a, (tx2) sub b ld c, a ld a, (tx1) ld d, a ld e, 0 ld a, (temp) sub b ld h, a ld l, 0 call DivFP ld a, h ld (temp+1), a ;tmp=temp+1 ld b, a ld a, 1 sub b ld d, a ld e, 0 ld a, (tu1) ld b, a ld c, 0 call MulFP push hl ld a, (temp+1) ld d, a ld e, 0 ld a, (tu2) ld b, a ld c, 0 call MulFP pop de add hl, de ld a, h ld (temp2), a ;u=temp2 ld a, (temp+1) ld b, a ld a, 1 sub b ld d, a ld e, 0 ld a, (tv1) ld b, a ld c, 0 call MulFP push hl ld a, (temp+1) ld d, a ld e, 0 ld a, (tv2) ld b, a ld c, 0 call MulFP pop de add hl, de ld a, h ld (temp2+1), a ;v=temp2+1 ;Get the pixel at (u,v) from the texture ld hl, (temp2+1) ld h, 0 ;1 byte smaller ld de, texture add hl, de ld a, (temp2) ld b, a inc b ld a, (hl) shiftLoop: rl c djnz shiftLoop ;the extra shift gets rid of the jr ;and pulls the bit out of carry BitShiftDone: push af ;1 ld a, (_ty) ld l, a ld a, (temp) ld b, 94 cp b jr nc, ContDrawLoop1 getPixel: ld h, 0 ld d, h ld e, l add hl, hl add hl, de add hl, hl add hl, hl ld e, a srl e srl e srl e add hl, de ld de, PlotSScreen add hl, de and 7 ld b, a ld a, $80 jr z, PlotPixel rrca djnz $-1 PlotPixel: pop af ;0 jr nc, ResPixel SetPixel: or (hl) ld (hl), a jr EndPlot ResPixel: cpl and (hl) ld (hl), a EndPlot: ld a, (temp) ld b, 255 cp b jp z, ContDrawLoop1 ld hl, temp inc (hl) ld a, (tx2) cp (hl) jp nc, QuadPlotLoop1 ContDrawLoop1: ld hl, (tx1) ld de, (dx1) add hl, de ld (tx1), hl ld hl, (tx2) ld de, (dx2) add hl, de ld (tx2), hl ld hl, (tu1) ld de, (du1) add hl, de ld (tu1), hl ld hl, (tu2) ld de, (du2) add hl, de ld (tu2), hl ld hl, (tv1) ld de, (dv1) add hl, de ld (tv1), hl ld hl, (tv2) ld de, (dv2) add hl, de ld (tv2), hl ld hl, _ty inc (hl) ld a, (y2) cp (hl) jp nz, QuadDrawLoop1 ld a, (y2) ld (_ty), a QuadDrawLoop2: ld a, (tx1) ld (temp), a ;i=temp QuadPlotLoop2: ld a, (tx1) ld b, a ld a, (tx2) sub b ld c, a ld a, (tx1) ld d, a ld e, 0 ld a, (temp) sub b ld h, a ld l, 0 call DivFP ld a, h ld (temp+1), a ;tmp=temp+1 ld b, a ld a, 1 sub b ld d, a ld e, 0 ld a, (tu1) ld b, a ld c, 0 call MulFP push hl ld a, (temp+1) ld d, a ld e, 0 ld a, (tu2) ld b, a ld c, 0 call MulFP pop de add hl, de ld a, h ld (temp2), a ;u=temp2 ld a, (temp+1) ld b, a ld a, 1 sub b ld d, a ld e, 0 ld a, (tv1) ld b, a ld c, 0 call MulFP push hl ld a, (temp+1) ld d, a ld e, 0 ld a, (tv2) ld b, a ld c, 0 call MulFP pop de add hl, de ld a, h ld (temp2+1), a ;v=temp2+1 ;Get the pixel at (u,v) from the texture ld hl, (temp2+1) ld h, 0 ;1 byte smaller ld de, texture add hl, de ld a, (temp2) ld b, a inc b ld a, (hl) shiftLoop2: rl c djnz shiftLoop2 ;the extra shift gets rid of the jr ;and pulls the bit out of carry BitShiftDone2: push af ;1 ld a, (_ty) ld l, a ld a, (temp) getPixel2: ld h, 0 ld d, h ld e, l add hl, hl add hl, de add hl, hl add hl, hl ld e, a srl e srl e srl e add hl, de ld de, PlotSScreen add hl, de and 7 ld b, a ld a, $80 jr z, PlotPixel2 rrca djnz $-1 PlotPixel2: pop af ;0 jr nc, ResPixel2 SetPixel2: or (hl) ld (hl), a jr EndPlot2 ResPixel2: cpl and (hl) ld (hl), a EndPlot2: ld hl, temp inc (hl) ld a, (tx2) cp (hl) jp nc, QuadPlotLoop2 ld hl, (tx1) ld de, (dx3) add hl, de ld (tx1), hl ld hl, (tx2) ld de, (dx2) add hl, de ld (tx2), hl ld hl, (tu1) ld de, (du3) add hl, de ld (tu1), hl ld hl, (tu2) ld de, (du2) add hl, de ld (tu2), hl ld hl, (tv1) ld de, (dv3) add hl, de ld (tv1), hl ld hl, (tv2) ld de, (dv2) add hl, de ld (tv2), hl ld hl, _ty inc (hl) ld a, (y2) cp (hl) jp nz, QuadDrawLoop2 QuadDrawLoop3: ld a, (tx1) ld (temp), a ;i=temp QuadPlotLoop3: ld a, (tx1) ld b, a ld a, (tx2) sub b ld c, a ld a, (tx1) ld d, a ld e, 0 ld a, (temp) sub b ld h, a ld l, 0 call DivFP ld a, h ld (temp+1), a ;tmp=temp+1 ld b, a ld a, 1 sub b ld d, a ld e, 0 ld a, (tu1) ld b, a ld c, 0 call MulFP push hl ld de, (temp+1) ld e, 0 ld bc, (tu2) ld c, 0 call MulFP pop de add hl, de ld a, h ld (temp2), a ;u=temp2 ld a, (temp+1) ld b, a ld a, 1 sub b ld d, a ld e, 0 ld a, (tv1) ld b, a ld c, 0 call MulFP push hl ld a, (temp+1) ld d, a ld e, 0 ld a, (tv2) ld b, a ld c, 0 call MulFP pop de add hl, de ld a, h ld (temp2+1), a ;v=temp2+1 ;Get the pixel at (u,v) from the texture ld hl, (temp2+1) ld h, 0 ;1 byte smaller ld de, texture add hl, de ld a, (temp2) ld b, a inc b ld a, (hl) shiftLoop3: rl c djnz shiftLoop3 ;the extra shift gets rid of the jr ;and pulls the bit out of carry BitShiftDone3: push af ;1 ld a, (_ty) ld l, a ld a, (temp) getPixel3: ld h, 0 ld d, h ld e, l add hl, hl add hl, de add hl, hl add hl, hl ld e, a srl e srl e srl e add hl, de ld de, PlotSScreen add hl, de and 7 ld b, a ld a, $80 jr z, PlotPixel3 rrca djnz $-1 PlotPixel3: pop af ;0 jr nc, ResPixel3 SetPixel3: or (hl) ld (hl), a jr EndPlot3 ResPixel3: cpl and (hl) ld (hl), a EndPlot3: ld hl, temp inc (hl) ld a, (tx2) cp (hl) jp nc, QuadPlotLoop3 ld hl, (tx1) ld de, (dx3) add hl, de ld (tx1), hl ld hl, (tx2) ld de, (dx4) add hl, de ld (tx2), hl ld hl, (tu1) ld de, (du3) add hl, de ld (tu1), hl ld hl, (tu2) ld de, (du4) add hl, de ld (tu2), hl ld hl, (tv1) ld de, (dv3) add hl, de ld (tv1), hl ld hl, (tv2) ld de, (dv4) add hl, de ld (tv2), hl ld hl, _ty inc (hl) ld a, (y2) cp (hl) jp nz, QuadDrawLoop3 ret

ben_g · « **Reply #70 on:** July 17, 2011, 06:41:27 pm »

After placing ret's and jumps at some positions in the code and counting the bytes in the memory editor of wabbitemu, I found out that thi routine does anything but drawing a textured quad. Amost nothing of it is doing what it should do, so I decided to put this on hold, make an other game, and try again when i'm more used to asm.

So, does anybody have a good idea for a good 2D or wireframe 3D game?

Edit: I just noticed i tripple posted, but there is more than 6 hours between those posts, so it's ok, right?

ztrumpet · « **Reply #71 on:** July 17, 2011, 07:14:02 pm »

Quote from: ben_g on July 17, 2011, 06:41:27 pm

Edit: I just noticed i tripple posted, but there is more than 6 hours between those posts, so it's ok, right?

Yup, it's fine.

Good luck with your new project.

thepenguin77 · « **Reply #72 on:** July 18, 2011, 09:37:20 am »

Yeah, the triple post was justified because if you hadn't of triple posted, I probably wouldn't have checked the pages and I would have missed your edits. (I always look at all the asm posts, just so you know. So if I don't answer, it either means I have no idea, or it would be annoying for me do whatever the request was (write a whole routine set for someone).)

ben_g · « **Reply #73 on:** July 20, 2011, 02:49:38 pm »

It doesn't look that bad on a calculator:

For some reason, the pixels on the edge between two texels are sometimes wrong. Does anybody knows how that can be fixed?

This is in basic, btw.

EDIT:

How could i have forgot the scource?

anyway, here it is. Sorry, it's the program itself, not just te code. I tried opening it with notepad, but all i got was a bunch of strange characters. So i hope you have an emulator to open it

btw: still no texture correction, and there are some bugs. If you try to change the coordinates: keep in mind that it only works when Y1 < Y2 < Y3 < Y4 , and there is also a bug when you try to draw a trapezoid with the paralel lines more vertical than horizontal.

EDIT2:

I found a converter. Here's the basic source code:

Code: [Select]

AxesOff
ClrDraw
[[1,1,1,1,1,1,1,1][1,0,1,0,1,0,1,1][1,1,0,1,0,1,0,1][1,0,1,0,1,0,1,1][1,1,0,1,0,1,0,1][1,0,1,0,1,0,1,1][1,1,0,1,0,1,0,1][1,1,1,1,1,1,1,1->[A]
4->dim(LDX
4->dim(LDU
4->dim(LDV
2->dim(LTX
2->dim(LTY
2->dim(LTU
2->dim(LTV
{50,40,80,10->LX
{10,11,60,61->LY
{1,8,1,8->LU
{1,1,8,8->LV
(LX(2)-LX(1))/(LY(2)-LY(1->LDX(1
(LX(3)-LX(1))/(LY(3)-LY(1->LDX(2
(LX(4)-LX(2))/(LY(4)-LY(2->LDX(3
(LX(4)-LX(3))/(LY(4)-LY(3->LDX(4
(LU(2)-LU(1))/(LY(2)-LY(1->LDU(1
(LV(2)-LV(1))/(LY(2)-LY(1->LDV(1
(LU(3)-LU(1))/(LY(3)-LY(1->LDU(2
(LV(3)-LV(1))/(LY(3)-LY(1->LDV(2
(LU(4)-LU(2))/(LY(4)-LY(2->LDU(3
(LV(4)-LV(2))/(LY(4)-LY(2->LDV(3
(LU(4)-LU(3))/(LY(4)-LY(3->LDU(4
(LV(4)-LV(3))/(LY(4)-LY(3->LDV(4
LX(1->LTX(1
LX(1->LTX(2
LU(1->LTU(1
LV(1->LTV(1
LU(1->LTU(2
LV(1->LTV(2
LY(1->Y
If [A](LU(1),LV(1
Then
Pxl-On(Y,LX(1
Else
Pxl-Off(Y,LX(1
End
For(Y,LY(1)+1,LY(2
LTX(1)+LDX(1->LTX(1
LTU(1)+LDU(1->LTU(1
LTV(1)+LDV(1->LTV(1
LTX(2)+LDX(2->LTX(2
LTU(2)+LDU(2->LTU(2
LTV(2)+LDV(2->LTV(2
For(X,LTX(1),LTX(2
(X-LTX(1))/(LTX(2)-LTX(1->T
round((1-T)*LTU(1)+T*LTU(2),0)->U
round((1-T)*LTV(1)+T*LTV(2),0)->V
If [A](U,V)
Then
Pxl-On(Y,round(X,0
Else
Pxl-Off(Y,round(X,0
End
End
End
For(Y,LY(2)+1,LY(3
LTX(1)+LDX(3->LTX(1
LTX(2)+LDX(2->LTX(2
LTU(1)+LDU(3->LTU(1
LTV(1)+LDV(3->LTV(1
LTU(2)+LDU(2->LTU(2
LTV(2)+LDV(2->LTV(2
For(X,LTX(1),LTX(2
(X-LTX(1))/(LTX(2)-LTX(1->T
round((1-T)*LTU(1)+T*LTU(2),0->U
round((1-T)*LTV(1)+T*LTV(2),0->V
If [A](U,V
Then
Pxl-On(Y,round(X,0
Else
Pxl-Off(Y,round(X,0
End
End
End
For(Y,LY(3)+1,LY(4)-1
LTX(1)+LDX(3->LTX(1
LTU(1)+LDU(3->LTU(1
LTV(1)+LDV(3->LTV(1
LTX(2)+LDX(4->LTX(2
LTU(2)+LDU(4->LTU(2
LTV(2)+LDV(4->LTV(2
For(X,LTX(1),LTX(2
(X-LTX(1))/(LTX(2)-LTX(1->T
round((1-T)*LTU(1)+T*LTU(2),0->U
round((1-T)*LTV(1)+T*LTV(2),0->V
If [A](U,V
Then
Pxl-On(Y,round(X,0
Else
Pxl-Off(Y,round(X,0
End
End
End
If [A](LU(4),LV(4
Then
Pxl-On(LY(4),LX(4
Else
Pxl-Off(LY(4),LX(4
End
DelVar LX
DelVar LY
DelVar LU
DelVar LV
DelVar [A]
DelVar LDU
DelVar LDV
DelVar LDX
DelVar LTU
DelVar LTV
DelVar LTX
DelVar LTY

Does anybody see what causes the wrong pixels in this program?

thepenguin77 · « **Reply #74 on:** July 20, 2011, 05:11:17 pm »

Of course we would have to see the source, but I assume this has something to do with rounding. Most likely the pixels that look bad are split between two adjacent pixels and every now and then go the wrong way.

Author Topic: Texture drawing (Read 34503 times)

AngelFish

Re: Texture drawing

ben_g

Re: Texture drawing

AngelFish

Re: Texture drawing

ben_g

Re: Texture drawing

ben_g

Re: Texture drawing

thepenguin77

Re: Texture drawing

ben_g

Re: Texture drawing

thepenguin77

Re: Texture drawing

ben_g

Re: Texture drawing

ben_g

Re: Texture drawing

ben_g

Re: Texture drawing

ztrumpet

Re: Texture drawing

thepenguin77

Re: Texture drawing

ben_g

Re: Texture drawing

thepenguin77

Re: Texture drawing