mirror of
https://github.com/deater/dos33fsprogs.git
synced 2025-04-16 20:40:18 +00:00
starpath: more optimization
This commit is contained in:
Vince Weaver
parent
61a50b4e9f
commit
d4232185d4
@ -13,6 +13,8 @@ SETGR = $FB40
|
||||
|
||||
FULLGR = $C052
|
||||
|
||||
COLOR = $30
|
||||
|
||||
FRAME = $F0
|
||||
YPOS = $F1
|
||||
XPOS = $F2
|
||||
@ -46,10 +48,51 @@ starpath:
|
||||
; initialize
|
||||
;=============================
|
||||
|
||||
jsr init_multiply_tables
|
||||
|
||||
lda #0
|
||||
sta FRAME
|
||||
init_multiply_tables:
|
||||
|
||||
; Build the add tables
|
||||
|
||||
ldx #$00
|
||||
txa
|
||||
.byte $c9 ; CMP #immediate - skip TYA and clear carry flag
|
||||
lb1: tya
|
||||
adc #$00 ; 0
|
||||
ml1: sta square1_hi,x ; square1_hi[0]=0
|
||||
tay ; y=0
|
||||
cmp #$40 ; subtract 64 and update flags (c=0)
|
||||
txa ; a=0
|
||||
ror ; rotate
|
||||
ml9: adc #$00 ; add 0
|
||||
sta ml9+1 ; update add value
|
||||
inx ; x=1
|
||||
ml0: sta square1_lo,x ; square1_lo[0]=1
|
||||
bne lb1 ; if not zero, loop
|
||||
inc ml0+2 ; increment values
|
||||
inc ml1+2 ; increment values
|
||||
clc ; c=0
|
||||
iny ; y=1
|
||||
bne lb1 ; loop
|
||||
|
||||
; Build the subtract tables based on the existing one
|
||||
|
||||
ldx #$00
|
||||
ldy #$ff
|
||||
second_table:
|
||||
lda square1_hi+1,x
|
||||
sta square2_hi+$100,x
|
||||
lda square1_hi,x
|
||||
sta square2_hi,y
|
||||
lda square1_lo+1,x
|
||||
sta square2_lo+$100,x
|
||||
lda square1_lo,x
|
||||
sta square2_lo,y
|
||||
dey
|
||||
inx
|
||||
bne second_table
|
||||
|
||||
; lda #0
|
||||
stx FRAME
|
||||
|
||||
next_frame:
|
||||
lda #0 ; start with YPOS=0
|
||||
@ -58,10 +101,10 @@ yloop:
|
||||
lda #0 ; start with XPOS=0
|
||||
sta XPOS
|
||||
xloop:
|
||||
ldx #14
|
||||
ldx #14 ; start Depth at 14
|
||||
|
||||
depth_loop:
|
||||
stx DEPTH ; start Depth at 14
|
||||
stx DEPTH
|
||||
|
||||
|
||||
;===============
|
||||
@ -78,8 +121,6 @@ depth_loop:
|
||||
jsr multiply_u8x8 ; 8-bit unsigned multiply
|
||||
|
||||
sta YPH ; store out to YPH:YPL
|
||||
;lda PRODLO
|
||||
;sta YPL
|
||||
|
||||
;========================
|
||||
; XP=(X*6)-DEPTH
|
||||
@ -202,7 +243,9 @@ plot_pixel:
|
||||
;=====================
|
||||
; set color
|
||||
|
||||
jsr SETCOL ; Set COLOR with ROM routine (mul*17)
|
||||
sta COLOR
|
||||
|
||||
; jsr SETCOL ; Set COLOR with ROM routine (mul*17)
|
||||
|
||||
;=====================
|
||||
; plot point
|
||||
@ -237,7 +280,7 @@ yloop_done:
|
||||
jmp next_frame
|
||||
|
||||
color_lookup:
|
||||
.byte 0,5,10,5,10,7,15,15,2,1,3,9,13,12
|
||||
.byte $00,$55,$AA,$55,$AA,$77,$FF,$FF,$22,$11,$33,$99,$DD,$CC
|
||||
|
||||
|
||||
; Fast mutiply
|
||||
@ -271,24 +314,18 @@ color_lookup:
|
||||
|
||||
|
||||
; Fast 8x8 bit unsigned multiplication, 16-bit result
|
||||
; Input: M1xM2
|
||||
; Result: M2:M1
|
||||
|
||||
; input AxY
|
||||
; Result: M2,A:M1
|
||||
;
|
||||
|
||||
; input A/Y
|
||||
|
||||
multiply_u8x8:
|
||||
sta M1
|
||||
sty M2
|
||||
|
||||
lda M1
|
||||
sta sm1a+1 ; 3
|
||||
sta sm3a+1 ; 3
|
||||
eor #$ff ; invert the bits for subtracting ; 2
|
||||
sta sm2a+1 ; 3
|
||||
sta sm4a+1 ; 3
|
||||
|
||||
ldy M2
|
||||
sec
|
||||
sm1a:
|
||||
lda square1_lo,Y
|
||||
@ -299,42 +336,13 @@ sm3a:
|
||||
lda square1_hi,Y
|
||||
sm4a:
|
||||
sbc square2_hi,Y
|
||||
sta M2
|
||||
; sta M2
|
||||
|
||||
rts
|
||||
|
||||
|
||||
; Fast mutiply -- setup tables
|
||||
|
||||
|
||||
; Note for our purposes we only care about 8.8 x 8.8 fixed point
|
||||
; with 8.8 result, which means we only care about the middle two bytes
|
||||
; of the 32 bit result. So we disable generation of the high and low byte
|
||||
; to save some cycles.
|
||||
|
||||
;
|
||||
; The old routine took around 700 cycles for a 16bitx16bit=32bit mutiply
|
||||
; This routine, at an expense of 2kB of looku tables, takes around 250
|
||||
; If you reuse a term the next time this drops closer to 200
|
||||
|
||||
; This routine was described by Stephen Judd and found
|
||||
; in The Fridge and in the C=Hacking magazine
|
||||
; http://codebase64.org/doku.php?id=base:seriously_fast_multiplication
|
||||
|
||||
; The key thing to note is that
|
||||
; (a+b)^2 (a-b)^2
|
||||
; a*b = ------- - --------
|
||||
; 4 4
|
||||
; So if you have tables of the squares of 0..511 you can lookup and subtract
|
||||
; instead of multiplying.
|
||||
|
||||
; Table generation: I:0..511
|
||||
; square1_lo = <((I*I)/4)
|
||||
; square1_hi = >((I*I)/4)
|
||||
; square2_lo = <(((I-255)*(I-255))/4)
|
||||
; square2_hi = >(((I-255)*(I-255))/4)
|
||||
|
||||
|
||||
.ifndef square1_lo
|
||||
square1_lo = $2000
|
||||
square1_hi = $2200
|
||||
@ -349,48 +357,5 @@ square2_hi = $2600
|
||||
; square2_hi[i]=(( ((i-255)*(i-255))/4)>>8)&0xff;
|
||||
; }
|
||||
|
||||
init_multiply_tables:
|
||||
|
||||
; Build the add tables
|
||||
|
||||
ldx #$00
|
||||
txa
|
||||
.byte $c9 ; CMP #immediate - skip TYA and clear carry flag
|
||||
lb1: tya
|
||||
adc #$00 ; 0
|
||||
ml1: sta square1_hi,x ; square1_hi[0]=0
|
||||
tay ; y=0
|
||||
cmp #$40 ; subtract 64 and update flags (c=0)
|
||||
txa ; a=0
|
||||
ror ; rotate
|
||||
ml9: adc #$00 ; add 0
|
||||
sta ml9+1 ; update add value
|
||||
inx ; x=1
|
||||
ml0: sta square1_lo,x ; square1_lo[0]=1
|
||||
bne lb1 ; if not zero, loop
|
||||
inc ml0+2 ; increment values
|
||||
inc ml1+2 ; increment values
|
||||
clc ; c=0
|
||||
iny ; y=1
|
||||
bne lb1 ; loop
|
||||
|
||||
; Build the subtract tables based on the existing one
|
||||
|
||||
ldx #$00
|
||||
ldy #$ff
|
||||
second_table:
|
||||
lda square1_hi+1,x
|
||||
sta square2_hi+$100,x
|
||||
lda square1_hi,x
|
||||
sta square2_hi,y
|
||||
lda square1_lo+1,x
|
||||
sta square2_lo+$100,x
|
||||
lda square1_lo,x
|
||||
sta square2_lo,y
|
||||
dey
|
||||
inx
|
||||
bne second_table
|
||||
|
||||
|
||||
rts
|
||||
|
||||
|
||||
361
graphics/gr/starpath/starpath_256.s
Normal file
361
graphics/gr/starpath/starpath_256.s
Normal file
@ -0,0 +1,361 @@
|
||||
; An Apple II lores version of Hellmood's amazing 64B DOS Star Path Demo
|
||||
;
|
||||
; See https://hellmood.111mb.de//starpath_is_55_bytes.html
|
||||
;
|
||||
; deater -- Vince Weaver -- vince@deater.net -- 25 February 2025
|
||||
|
||||
|
||||
PLOT = $F800 ; PLOT AT Y,A (A colors output, Y preserved)
|
||||
PLOT1 = $F80E ; PLOT at (GBASL),Y (need MASK to be $0f or $f0)
|
||||
|
||||
SETCOL = $F864 ; COLOR=A
|
||||
SETGR = $FB40
|
||||
|
||||
FULLGR = $C052
|
||||
|
||||
COLOR = $30
|
||||
|
||||
FRAME = $F0
|
||||
YPOS = $F1
|
||||
XPOS = $F2
|
||||
DEPTH = $F3
|
||||
C = $F4
|
||||
AL = $F5
|
||||
M1 = $F6
|
||||
M2 = $F7
|
||||
TEMP = $F8
|
||||
YPL = $F9
|
||||
YPH = $FA
|
||||
XPL = $FB
|
||||
XPH = $FC
|
||||
Q = $FD
|
||||
|
||||
;=============================
|
||||
;=============================
|
||||
; star path
|
||||
;=============================
|
||||
;=============================
|
||||
|
||||
starpath:
|
||||
;=============================
|
||||
; setup graphics
|
||||
;=============================
|
||||
|
||||
jsr SETGR ; set graphics
|
||||
bit $C052 ; set full-screen graphics
|
||||
|
||||
;=============================
|
||||
; initialize
|
||||
;=============================
|
||||
|
||||
|
||||
init_multiply_tables:
|
||||
|
||||
; Build the add tables
|
||||
|
||||
ldx #$00
|
||||
txa
|
||||
.byte $c9 ; CMP #immediate - skip TYA and clear carry flag
|
||||
lb1: tya
|
||||
adc #$00 ; 0
|
||||
ml1: sta square1_hi,x ; square1_hi[0]=0
|
||||
tay ; y=0
|
||||
cmp #$40 ; subtract 64 and update flags (c=0)
|
||||
txa ; a=0
|
||||
ror ; rotate
|
||||
ml9: adc #$00 ; add 0
|
||||
sta ml9+1 ; update add value
|
||||
inx ; x=1
|
||||
ml0: sta square1_lo,x ; square1_lo[0]=1
|
||||
bne lb1 ; if not zero, loop
|
||||
inc ml0+2 ; increment values
|
||||
inc ml1+2 ; increment values
|
||||
clc ; c=0
|
||||
iny ; y=1
|
||||
bne lb1 ; loop
|
||||
|
||||
; Build the subtract tables based on the existing one
|
||||
|
||||
ldx #$00
|
||||
ldy #$ff
|
||||
second_table:
|
||||
lda square1_hi+1,x
|
||||
sta square2_hi+$100,x
|
||||
lda square1_hi,x
|
||||
sta square2_hi,y
|
||||
lda square1_lo+1,x
|
||||
sta square2_lo+$100,x
|
||||
lda square1_lo,x
|
||||
sta square2_lo,y
|
||||
dey
|
||||
inx
|
||||
bne second_table
|
||||
|
||||
; lda #0
|
||||
stx FRAME
|
||||
|
||||
next_frame:
|
||||
lda #0 ; start with YPOS=0
|
||||
sta YPOS
|
||||
yloop:
|
||||
lda #0 ; start with XPOS=0
|
||||
sta XPOS
|
||||
xloop:
|
||||
ldx #14 ; start Depth at 14
|
||||
|
||||
depth_loop:
|
||||
stx DEPTH
|
||||
|
||||
|
||||
;===============
|
||||
; YP = Y*4*DEPTH
|
||||
;===============
|
||||
|
||||
lda YPOS ;
|
||||
asl
|
||||
asl ; A is YPOS*4
|
||||
|
||||
tay ; multiply Y*4*DEPTH
|
||||
; lda DEPTH
|
||||
txa
|
||||
jsr multiply_u8x8 ; 8-bit unsigned multiply
|
||||
|
||||
sta YPH ; store out to YPH:YPL
|
||||
|
||||
;========================
|
||||
; XP=(X*6)-DEPTH
|
||||
; curve X by depth
|
||||
;=========================
|
||||
|
||||
lda XPOS ; load XPOS
|
||||
asl
|
||||
sta XPL
|
||||
asl
|
||||
;clc ; carry always 0 as x never more than 40?
|
||||
adc XPL
|
||||
sta XPL ; XPL=XPOS*6
|
||||
|
||||
sta AL ; AL also is XPOS*6
|
||||
|
||||
sec ; Subtract DEPTH
|
||||
sbc DEPTH
|
||||
sta XPL ; XP=(XPOS*6)-DEPTH
|
||||
|
||||
; if carry set means not negative
|
||||
; and draw path
|
||||
; otherwise we draw the sky
|
||||
bcs draw_path
|
||||
|
||||
;========================
|
||||
; draw the sky
|
||||
;========================
|
||||
draw_sky:
|
||||
|
||||
;================================
|
||||
; set color to white for star?
|
||||
|
||||
lda #31 ; C=31
|
||||
sta C
|
||||
|
||||
;=====================
|
||||
; calc A=(XPOS*6)+YP
|
||||
|
||||
; ??? used to see if star
|
||||
|
||||
clc ; A=X*6+YP
|
||||
lda AL
|
||||
adc M1 ; YPL from previous multiply
|
||||
|
||||
;==============
|
||||
; see if star
|
||||
|
||||
cmp #6 ; if A&0xFF < 6 then skip, we are star
|
||||
bcc plot_pixel
|
||||
|
||||
;==============
|
||||
; not star, sky
|
||||
|
||||
lda YPOS ; C=Y/4+32
|
||||
lsr
|
||||
lsr
|
||||
clc
|
||||
adc #32
|
||||
sta C
|
||||
|
||||
bne plot_pixel ; bra
|
||||
|
||||
;====================
|
||||
; draw path
|
||||
;====================
|
||||
|
||||
draw_path:
|
||||
;=================================
|
||||
; calc XP*DEPTH and get high byte
|
||||
|
||||
ldy XPL
|
||||
; lda DEPTH
|
||||
txa
|
||||
jsr multiply_u8x8 ; 8-bit unsigned multiply
|
||||
|
||||
;===================================
|
||||
; calc Q= (XP*DEPTH)/256 | (YP/256)
|
||||
; for texture pattern
|
||||
|
||||
ora YPH ; Q=(XP*DEPTH)/256 | YP/256
|
||||
sta Q
|
||||
|
||||
;==============================
|
||||
; calc C = Q & (Depth + Frame)
|
||||
; mask geometry by time shifted depth
|
||||
|
||||
clc
|
||||
; lda DEPTH
|
||||
txa
|
||||
adc FRAME ; add depth plus frame D+F
|
||||
|
||||
and Q ; C = Q & (D+FRAME)
|
||||
sta C
|
||||
|
||||
;=========================
|
||||
; increment depth
|
||||
|
||||
; inc DEPTH ; DEPTH=DEPTH+1
|
||||
inx
|
||||
|
||||
;==========================
|
||||
; to create gaps
|
||||
|
||||
cmp #16 ; IF C<16 THEN 3
|
||||
bcc depth_loop
|
||||
|
||||
;===========================
|
||||
; plot pixel
|
||||
; XPOS,YPOS COLOR=LOOKUP(C/2-8)
|
||||
plot_pixel:
|
||||
lda C
|
||||
lsr
|
||||
sec
|
||||
sbc #8 ; A is C/2-8
|
||||
|
||||
tay
|
||||
lda color_lookup,Y ; Lookup in color table
|
||||
|
||||
;=====================
|
||||
; set color
|
||||
|
||||
sta COLOR
|
||||
|
||||
; jsr SETCOL ; Set COLOR with ROM routine (mul*17)
|
||||
|
||||
;=====================
|
||||
; plot point
|
||||
|
||||
ldy XPOS
|
||||
lda YPOS
|
||||
jsr PLOT ; PLOT AT Y,A (Y preserved)
|
||||
|
||||
;===================
|
||||
; increment xloop
|
||||
|
||||
inc XPOS
|
||||
lda XPOS
|
||||
cmp #40
|
||||
bne xloop
|
||||
; beq xloop_done
|
||||
; jmp xloop
|
||||
xloop_done:
|
||||
|
||||
;===================
|
||||
; increment yloop
|
||||
|
||||
inc YPOS
|
||||
lda YPOS
|
||||
cmp #48
|
||||
bne yloop
|
||||
; beq yloop_done
|
||||
; jmp yloop
|
||||
yloop_done:
|
||||
inc FRAME
|
||||
|
||||
jmp next_frame
|
||||
|
||||
color_lookup:
|
||||
.byte $00,$55,$AA,$55,$AA,$77,$FF,$FF,$22,$11,$33,$99,$DD,$CC
|
||||
|
||||
|
||||
; Fast mutiply
|
||||
|
||||
; Note for our purposes we only care about 8.8 x 8.8 fixed point
|
||||
; with 8.8 result, which means we only care about the middle two bytes
|
||||
; of the 32 bit result. So we disable generation of the high and low byte
|
||||
; to save some cycles.
|
||||
|
||||
;
|
||||
; The old routine took around 700 cycles for a 16bitx16bit=32bit mutiply
|
||||
; This routine, at an expense of 2kB of looku tables, takes around 250
|
||||
; If you reuse a term the next time this drops closer to 200
|
||||
|
||||
; This routine was described by Stephen Judd and found
|
||||
; in The Fridge and in the C=Hacking magazine
|
||||
; http://codebase64.org/doku.php?id=base:seriously_fast_multiplication
|
||||
|
||||
; The key thing to note is that
|
||||
; (a+b)^2 (a-b)^2
|
||||
; a*b = ------- - --------
|
||||
; 4 4
|
||||
; So if you have tables of the squares of 0..511 you can lookup and subtract
|
||||
; instead of multiplying.
|
||||
|
||||
; Table generation: I:0..511
|
||||
; square1_lo = <((I*I)/4)
|
||||
; square1_hi = >((I*I)/4)
|
||||
; square2_lo = <(((I-255)*(I-255))/4)
|
||||
; square2_hi = >(((I-255)*(I-255))/4)
|
||||
|
||||
|
||||
; Fast 8x8 bit unsigned multiplication, 16-bit result
|
||||
|
||||
; input AxY
|
||||
; Result: M2,A:M1
|
||||
;
|
||||
|
||||
multiply_u8x8:
|
||||
sta sm1a+1 ; 3
|
||||
sta sm3a+1 ; 3
|
||||
eor #$ff ; invert the bits for subtracting ; 2
|
||||
sta sm2a+1 ; 3
|
||||
sta sm4a+1 ; 3
|
||||
|
||||
sec
|
||||
sm1a:
|
||||
lda square1_lo,Y
|
||||
sm2a:
|
||||
sbc square2_lo,Y
|
||||
sta M1
|
||||
sm3a:
|
||||
lda square1_hi,Y
|
||||
sm4a:
|
||||
sbc square2_hi,Y
|
||||
; sta M2
|
||||
|
||||
rts
|
||||
|
||||
|
||||
; Fast mutiply -- setup tables
|
||||
|
||||
.ifndef square1_lo
|
||||
square1_lo = $2000
|
||||
square1_hi = $2200
|
||||
square2_lo = $2400
|
||||
square2_hi = $2600
|
||||
.endif
|
||||
|
||||
; for(i=0;i<512;i++) {
|
||||
; square1_lo[i]=((i*i)/4)&0xff;
|
||||
; square1_hi[i]=(((i*i)/4)>>8)&0xff;
|
||||
; square2_lo[i]=( ((i-255)*(i-255))/4)&0xff;
|
||||
; square2_hi[i]=(( ((i-255)*(i-255))/4)>>8)&0xff;
|
||||
; }
|
||||
|
||||
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user