brooke/dither4
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
dither4/dither4.s

92 lines
1.4 KiB
ArmAsm
Raw Normal View History

initial atari tests (not with image) and dither improvements palette selection now biases to matches on the bright side of the RGB color cube, since black is always available for dithering does better but still not perfect atari binary just writes some test data to text framebuffer but builds and runs
2022-11-20 23:43:34 +00:00
SAVMSC = $58
work in progress
2022-11-25 19:51:27 +00:00
SDMCTL = $22F
SDLSTL = $230
SDLSTH = $231
COLPF0 = $D016
COLPF1 = $D017
COLPF2 = $D018
COLPF3 = $D019
COLBK = $D01A
WSYNC = $D40A
VCOUNT = $D40B
NMIEN = $D40E
.data
.import palette1
.import palette2
.import palette3
.import framebuffer
displaylist:
.repeat 3
.byte $70 ; 8 blank lines
.endrep
.byte $4e ; ANTIC mode e (160px 2bpp, 1 scan line per line)
.addr framebuffer
.repeat 191
.byte $0e
.endrep
.byte $41 ; jump and blank
.addr displaylist
initial atari tests (not with image) and dither improvements palette selection now biases to matches on the bright side of the RGB color cube, since black is always available for dithering does better but still not perfect atari binary just writes some test data to text framebuffer but builds and runs
2022-11-20 23:43:34 +00:00
.code
.export start
work in progress
2022-11-25 19:51:27 +00:00
; Loop through scan lines updating the palette
.proc update_palette
lda palette1,y
pha
ldx palette2,y
lda palette3,y
tay
pla
; Wait for horizontal blank
sta WSYNC
; Update color registers as fast as possible
sta COLPF0
stx COLPF1
sty COLPF2
partly producing output
2022-11-26 01:35:10 +00:00
rts
work in progress
2022-11-25 19:51:27 +00:00
.endproc
initial atari tests (not with image) and dither improvements palette selection now biases to matches on the bright side of the RGB color cube, since black is always available for dithering does better but still not perfect atari binary just writes some test data to text framebuffer but builds and runs
2022-11-20 23:43:34 +00:00
.proc start
work in progress
2022-11-25 19:51:27 +00:00
; Disable display DMA
lda #$00
sta SDMCTL
; Set up the display list
partly producing output
2022-11-26 01:35:10 +00:00
lda #.lobyte(displaylist)
work in progress
2022-11-25 19:51:27 +00:00
sta SDLSTL
partly producing output
2022-11-26 01:35:10 +00:00
lda #.hibyte(displaylist)
work in progress
2022-11-25 19:51:27 +00:00
sta SDLSTH
; Turn off interrupts except reset button
lda #$20
sta NMIEN
; Manually wait for first scan line
wait_vblank:
lda VCOUNT
partly producing output
2022-11-26 01:35:10 +00:00
bne wait_vblank
work in progress
2022-11-25 19:51:27 +00:00
; Re-enable display DMA
lda #$22
sta SDMCTL
even_scanline:
ldy VCOUNT
jsr update_palette
odd_scanline:
ldy VCOUNT
initial atari tests (not with image) and dither improvements palette selection now biases to matches on the bright side of the RGB color cube, since black is always available for dithering does better but still not perfect atari binary just writes some test data to text framebuffer but builds and runs
2022-11-20 23:43:34 +00:00
iny
work in progress
2022-11-25 19:51:27 +00:00
jsr update_palette
jmp even_scanline
initial atari tests (not with image) and dither improvements palette selection now biases to matches on the bright side of the RGB color cube, since black is always available for dithering does better but still not perfect atari binary just writes some test data to text framebuffer but builds and runs
2022-11-20 23:43:34 +00:00
.endproc
Reference in a new issue Copy permalink