WebSVN – Kolibri OS – Diff – /programs/demos/3DS/BUMP_TEX.INC


;CATMULL_SHIFT equ 8
;TEXTURE_SIZE EQU (TEX_X * TEX_Y)-1
;ROUND equ 8
;Ext = NON
;MMX = 1
;NON = 0
;------- Big thanks to Majuma (www.majuma.xt.pl) for absolutely great---
;------- DOS 13h mode demos --------------------------------------------
;------- Procedure draws bump triangle with texture, I use -------------
;--------Catmull Z-buffer algorithm- (Z coordinate interpolation)-------
;--------I calc texture pixel by this way: col1*col2/256 ---------------
bump_tex_triangle_z:
;------------------in - eax - x1 shl 16 + y1 -----------
;---------------------- ebx - x2 shl 16 + y2 -----------
;---------------------- ecx - x3 shl 16 + y3 -----------
;---------------------- edx - pointer to bump map-------
;---------------------- esi - pointer to env map--------
;---------------------- edi - pointer to screen buffer--
;---------------------- stack : bump coordinates--------
;----------------------         environment coordinates-
;----------------------         Z position coordinates--
;----------------------         pointer to Z buffer-----
;----------------------         pointer to texture------
;----------------------         texture coordinates-----
;-- Z-buffer - filled with coordinates as dword --------
;-- (Z coor. as word) shl CATMULL_SHIFT ----------------
.b_x1	equ ebp+4   ; procedure don't save registers !!!
.b_y1	equ ebp+6   ; each coordinate as word
.b_x2	equ ebp+8
.b_y2	equ ebp+10	 ; b - bump map coords
.b_x3	equ ebp+12	 ; e - env map coords
.b_y3	equ ebp+14
.e_x1	equ ebp+16
.e_y1	equ ebp+18
.e_x2	equ ebp+20
.e_y2	equ ebp+22
.e_x3	equ ebp+24
.e_y3	equ ebp+26
.z1	equ word[ebp+28]
.z2	equ word[ebp+30]
.z3	equ word[ebp+32]
.z_buff equ dword[ebp+34]	; pointer to Z-buffer
.tex_ptr equ dword[ebp+38]	; ptr to texture
.t_x1	equ ebp+42		; texture coords
.t_y1	equ ebp+44
.t_x2	equ ebp+46
.t_y2	equ ebp+48
.t_x3	equ ebp+50
.t_y3	equ ebp+52
 
 
 
.t_bmap equ dword[ebp-4]	; pointer to bump map
.t_emap equ dword[ebp-8]	; pointer to env map
.x1	equ word[ebp-10]
.y1	equ word[ebp-12]
.x2	equ word[ebp-14]
.y2	equ word[ebp-16]
.x3	equ word[ebp-18]
.y3	equ word[ebp-20]
 
.dx12	equ dword[ebp-24]
.dz12	equ	 [ebp-28]
.dbx12	equ dword[ebp-32]
.dby12	equ	 [ebp-36]
.dex12	equ dword[ebp-40]
.dey12	equ	 [ebp-44]
.dtx12	equ dword[ebp-48]
.dty12	equ	 [ebp-52]
 
.dx13  equ dword[ebp-52-4*1]
.dz13  equ	[ebp-52-4*2]
.dbx13 equ dword[ebp-52-4*3]
.dby13 equ	[ebp-52-4*4]
.dex13 equ dword[ebp-52-4*5]
.dey13 equ	[ebp-52-4*6]
.dtx13 equ dword[ebp-52-4*7]
.dty13 equ	[ebp-52-4*8]
 
 
.dx23  equ dword[ebp-(52+4*9)]
.dz23  equ	[ebp-(52+4*10)]
.dbx23 equ dword[ebp-(52+4*11)]
.dby23 equ	[ebp-(52+4*12)]
.dex23 equ dword[ebp-(52+4*13)]
.dey23 equ	[ebp-(52+4*14)]
.dtx23 equ dword[ebp-(52+4*15)]
.dty23 equ	[ebp-(52+4*16)]
 
.cx1   equ dword[ebp-(52+4*17)] 	; current variables
.cz1   equ	[ebp-(52+4*18)]
.cx2   equ dword[ebp-(52+4*19)]
.cz2   equ	[ebp-(52+4*20)]
.cbx1  equ dword[ebp-(52+4*21)]
.cby1  equ	[ebp-(52+4*22)]
.cbx2  equ dword[ebp-(52+4*23)]
.cby2  equ	[ebp-(52+4*24)]
.cex1  equ dword[ebp-(52+4*25)]
.cey1  equ	[ebp-(52+4*26)]
.cex2  equ dword[ebp-(52+4*27)]
.cey2  equ	[ebp-(52+4*28)]
 
.ctx1  equ dword[ebp-(52+4*29)]
.cty1  equ	[ebp-(52+4*30)]
.ctx2  equ dword[ebp-(52+4*31)]
.cty2  equ	[ebp-(52+4*32)]
 
       cld
       mov     ebp,esp
       push    edx	  ; store bump map
       push    esi	  ; store e. map
     ; sub     esp,120
 .sort3:		  ; sort triangle coordinates...
       cmp     ax,bx
       jle     .sort1
       xchg    eax,ebx
       mov     edx,dword[.b_x1]
       xchg    edx,dword[.b_x2]
       mov     dword[.b_x1],edx
       mov     edx,dword[.e_x1]
       xchg    edx,dword[.e_x2]
       mov     dword[.e_x1],edx
       mov     edx,dword[.t_x1]
       xchg    edx,dword[.t_x2]
       mov     dword[.t_x1],edx
       mov     dx,.z1
       xchg    dx,.z2
       mov     .z1,dx
 .sort1:
       cmp	bx,cx
       jle	.sort2
       xchg	ebx,ecx
       mov	edx,dword[.b_x2]
       xchg	edx,dword[.b_x3]
       mov	dword[.b_x2],edx
       mov	edx,dword[.e_x2]
       xchg	edx,dword[.e_x3]
       mov	dword[.e_x2],edx
       mov	edx,dword[.t_x2]
       xchg	edx,dword[.t_x3]
       mov	dword[.t_x2],edx
       mov     dx,.z2
       xchg    dx,.z3
       mov     .z2,dx
       jmp	.sort3
 .sort2:
       push	eax	; store triangle coords in variables
       push	ebx
       push	ecx
	 mov	  edx,80008000h  ; eax,ebx,ecx are ANDd together into edx which means that
	 and	  edx,ebx	 ; if *all* of them are negative a sign flag is raised
	 and	  edx,ecx
	 and	  edx,eax
	 test	  edx,80008000h  ; Check both X&Y at once
	 jne	  .loop23_done
    ;   mov     edx,eax         ; eax,ebx,ecx are ORd together into edx which means that
    ;   or      edx,ebx         ; if any *one* of them is negative a sign flag is raised
    ;   or      edx,ecx
    ;   test    edx,80000000h   ; Check only X
    ;   jne     .loop23_done
 
    ;   cmp     .x1,SIZE_X    ; {
    ;   jg      .loop23_done
    ;   cmp     .x2,SIZE_X     ; This can be optimized with effort
    ;   jg      .loop23_done
    ;   cmp     .x3,SIZE_X
    ;   jg      .loop23_done    ; {
 
 
       mov	bx,.y2	     ; calc delta 12
       sub	bx,.y1
       jnz	.bt_dx12_make
       mov	ecx,8
       xor	edx,edx
     @@:
       push	edx   ;dword 0
       loop	@b
       jmp	.bt_dx12_done
 .bt_dx12_make:
       movsx	ebx,bx
 
       mov	ax,.x2
       sub	ax,.x1
       cwde
       shl	eax,ROUND
 
       cdq
       idiv	ebx
 ;     mov      .dx12,eax
       push	 eax
 
 
 
 
 
 
 
 
 
if Ext>=SSE
 
       sub	 esp,28
   ;    mov       eax,256
       cvtsi2ss  xmm4,[i255d]
       cvtsi2ss  xmm3,ebx	     ;rcps
       divss	 xmm3,xmm4
       shufps	 xmm3,xmm3,0
 
       movd	 mm0,[.b_x1]
       movd	 mm1,[.b_x2]
       movd	 mm2,[.e_x1]
       movd	 mm3,[.e_x2]
 
       pxor	  mm4,mm4
       punpcklwd  mm0,mm4
       punpcklwd  mm1,mm4
       punpcklwd  mm2,mm4
       punpcklwd  mm3,mm4
 
       psubd	  mm1,mm0
       psubd	  mm3,mm2
 
       cvtpi2ps  xmm1,mm1
       movlhps	 xmm1,xmm1
       cvtpi2ps  xmm1,mm3
 
       divps	 xmm1,xmm3   ;xmm1--> | dby | dbx | dey | dex |
 
       shufps	 xmm1,xmm1,10110001b
			     ;xmm1--> | dbx | dby | dex | dey |
       cvtps2pi  mm0,xmm1 ;mm0,xmm1          ; mm0 -> 2 delta dwords
       movhlps	 xmm1,xmm1
       cvtps2pi  mm1,xmm1 ;mm1,xmm1
       movq	 .dey12,mm0
       movq	 .dby12,mm1
;-------------
  ;    pxor      mm0,mm0
  ;    pxor      mm1,mm1
   ;/   pinsrw    mm0,.z1,1
   ;/   pinsrw    mm0,.x1,0
   ;/   pinsrw    mm1,.z2,1
   ;/   pinsrw    mm1,.x2,0
       mov	 ax,.z2
       sub	 ax,.z1
       cwde
   ;/    movd      mm1,eax
 
   ;/    punpcklwd  mm0,mm4
   ;/    punpcklwd  mm1,mm4
 
  ;     cvtpi2ps   xmm1,mm1
  ;     cvtpi2ps   xmm2,mm0
  ;     subps      xmm1,xmm2
 
   ;/   psubd      mm1,mm0
 
       movd	  mm2,[.t_x1]
       movd	  mm3,[.t_x2]
 
       punpcklwd  mm2,mm4
       punpcklwd  mm3,mm4
       psubd	  mm3,mm2
 
   ;/  cvtpi2ps  xmm1,mm1
       cvtsi2ss  xmm1,eax
       movlhps	 xmm1,xmm1
       cvtpi2ps  xmm1,mm3
 
       divps	 xmm1,xmm3   ; xmm1--> | xxx | dz | dty | dtx |
 
       shufps	 xmm1,xmm1,11100001b
			     ; xmm1--> | xxx | dz | dtx | dty |
       cvtps2pi  mm0,xmm1    ; mm0 -> 2 delta dwords  | dtx | dty |
       movhlps	 xmm1,xmm1
       cvtss2si  eax,xmm1    ; eax --> 1 delta dword | dz |
       movq	 .dty12,mm0
       mov	 .dz12,eax
;----
;       mov       ax,.z2
;       sub       ax,.z1
;       cwde
;       mov       bx,.x2
;       sub       bx,.x1
;       movsx     ebx,bx
;       movd      mm1,eax
;       psllq     mm1,32
;       movd      mm1,ebx
 
;;       push      ebx
;;       push      eax
;;       movq      mm1,[esp]
;;       add       esp,8
;;;       mov       ax,.z1
;;;       mov       bx,.z2
;;;       shl       eax,16
;;;       shl       ebx,16
;;;       mov       ax,.x1
;;;       mov       bx,.x2
;       movd       mm2,[.t_x1]
;       movd       mm3,[.t_x2]
;;       movd      mm0,eax
;;       movd      mm1,ebx
 
;       pxor       mm4,mm4
;;       punpcklwd  mm0,mm4
;;       punpcklwd  mm1,mm4
;       punpcklwd  mm2,mm4
;       punpcklwd  mm3,mm4
 
;;       psubd    mm1,mm0
;       psubd      mm3,mm2
 
 
;       cvtpi2ps  xmm1,mm1
;       movlhps   xmm1,xmm1
;       cvtpi2ps  xmm1,mm3
 
;       divps     xmm1,xmm3   ; xmm1--> | dz | dx | dty | dtx |
 
;       shufps    xmm1,xmm1,10110001b
			     ; xmm1--> | dx | dz | dtx | dty |
;       cvtps2pi  mm0,xmm1    ; mm0 -> 2 delta dwords  | dtx | dty |
;       movhlps   xmm1,xmm1
;       cvtps2pi  mm1,xmm1    ; mm1 --> 2 delta dwords | dx | dz |
;       movq      .dty12,mm0
;       movq      .dz12,mm1
else
       mov     ax,.z2
       sub     ax,.z1
       cwde
       shl     eax,CATMULL_SHIFT
       cdq
       idiv    ebx
       push    eax
 
       mov	ax,word[.b_x2]
       sub	ax,word[.b_x1]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dbx12,eax
       push	 eax
 
       mov	ax,word[.b_y2]
       sub	ax,word[.b_y1]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dby12,eax
       push	 eax
 
       mov	ax,word[.e_x2]
       sub	ax,word[.e_x1]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dex12,eax
       push	 eax
 
       mov	ax,word[.e_y2]
       sub	ax,word[.e_y1]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dey12,eax
       push	 eax
 
;end if
 
       mov	ax,word[.t_x2]
       sub	ax,word[.t_x1]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dtx12,eax
       push	 eax
 
       mov	ax,word[.t_y2]
       sub	ax,word[.t_y1]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dty12,eax
       push	 eax
end if
   .bt_dx12_done:
 
       mov	bx,.y3	     ; calc delta13
       sub	bx,.y1
       jnz	.bt_dx13_make
       mov	ecx,8
       xor	edx,edx
     @@:
       push	edx   ;dword 0
       loop	@b
       jmp	.bt_dx13_done
 .bt_dx13_make:
       mov	ax,.x3
       sub	ax,.x1
       cwde
       movsx	ebx,bx
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dx13,eax
       push	 eax
 
 
 
 
 
 
 
 
 
 
if Ext>=SSE
 
       sub	 esp,28
   ;    mov       eax,255
       cvtsi2ss  xmm4,[i255d]
       cvtsi2ss  xmm3,ebx	     ;rcps
       divss	 xmm3,xmm4
       shufps	 xmm3,xmm3,0
 
       movd	 mm0,[.b_x1]
       movd	 mm1,[.b_x3]
       movd	 mm2,[.e_x1]
       movd	 mm3,[.e_x3]
 
       pxor	  mm4,mm4
       punpcklwd  mm0,mm4
       punpcklwd  mm1,mm4
       punpcklwd  mm2,mm4
       punpcklwd  mm3,mm4
 
       psubd	  mm1,mm0
       psubd	  mm3,mm2
 
       cvtpi2ps  xmm1,mm1
       movlhps	 xmm1,xmm1
       cvtpi2ps  xmm1,mm3
 
       divps	 xmm1,xmm3   ;xmm1--> | dby | dbx | dey | dex |
 
       shufps	 xmm1,xmm1,10110001b
			     ;xmm1--> | dbx | dby | dex | dey |
       cvtps2pi  mm0,xmm1	   ; mm0 -> 2 delta dwords
       movhlps	 xmm1,xmm1
       cvtps2pi  mm1,xmm1
       movq	 .dey13,mm0
       movq	 .dby13,mm1
;----------
       mov	 ax,.z3
       sub	 ax,.z1
       cwde
 
       movd	  mm2,[.t_x1]
       movd	  mm3,[.t_x3]
 
       punpcklwd  mm2,mm4
       punpcklwd  mm3,mm4
       psubd	  mm3,mm2
 
       cvtsi2ss  xmm1,eax
       movlhps	 xmm1,xmm1
       cvtpi2ps  xmm1,mm3
 
       divps	 xmm1,xmm3   ; xmm1--> | xxx | dz | dty | dtx |
 
       shufps	 xmm1,xmm1,11100001b
			     ; xmm1--> | xxx | dz | dtx | dty |
       cvtps2pi  mm0,xmm1    ; mm0 -> 2 delta dwords  | dtx | dty |
       movhlps	 xmm1,xmm1
       cvtss2si  eax,xmm1    ; eax --> 1 delta dword | dz |
       movq	 .dty13,mm0
       mov	 .dz13,eax
else
       mov     ax,.z3
       sub     ax,.z1
       cwde
       shl     eax,CATMULL_SHIFT
       cdq
       idiv    ebx
  ;    mov    .dz13,eax
       push    eax
 
 
       mov	ax,word[.b_x3]
       sub	ax,word[.b_x1]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dbx13,eax
       push	 eax
 
       mov	ax,word[.b_y3]
       sub	ax,word[.b_y1]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dby13,eax
       push	 eax
 
       mov	ax,word[.e_x3]
       sub	ax,word[.e_x1]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dex13,eax
       push	 eax
 
       mov	ax,word[.e_y3]
       sub	ax,word[.e_y1]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dey13,eax
       push	 eax
 
 
       mov	ax,word[.t_x3]
       sub	ax,word[.t_x1]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dtx13,eax
       push	 eax
 
       mov	ax,word[.t_y3]
       sub	ax,word[.t_y1]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dty13,eax
       push	 eax
end if
   .bt_dx13_done:
 
       mov	bx,.y3	     ; calc delta23
       sub	bx,.y2
       jnz	.bt_dx23_make
       mov	ecx,8
       xor	edx,edx
     @@:
       push	edx   ;dword 0
       loop	@b
       jmp	.bt_dx23_done
 .bt_dx23_make:
       mov	ax,.x3
       sub	ax,.x2
       cwde
       movsx	ebx,bx
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dx23,eax
       push	 eax
 
 
 
 
 
 
 
 
 
 
if Ext>=SSE
 
       sub	 esp,28
    ;   mov       eax,255
       cvtsi2ss  xmm4,[i255d]
       cvtsi2ss  xmm3,ebx	     ;rcps
       divss	 xmm3,xmm4
       shufps	 xmm3,xmm3,0
 
       movd	 mm0,[.b_x2]
       movd	 mm1,[.b_x3]
       movd	 mm2,[.e_x2]
       movd	 mm3,[.e_x3]
 
       pxor	  mm4,mm4
       punpcklwd  mm0,mm4
       punpcklwd  mm1,mm4
       punpcklwd  mm2,mm4
       punpcklwd  mm3,mm4
 
       psubd	  mm1,mm0
       psubd	  mm3,mm2
 
       cvtpi2ps  xmm1,mm1
       movlhps	 xmm1,xmm1
       cvtpi2ps  xmm1,mm3
 
       divps	 xmm1,xmm3   ;xmm1--> | dby | dbx | dey | dex |
 
       shufps	 xmm1,xmm1,10110001b
			     ;xmm1--> | dbx | dby | dex | dey |
       cvtps2pi  mm0,xmm1	   ; mm0 -> 2 delta dwords
       movhlps	 xmm1,xmm1
       cvtps2pi  mm1,xmm1
       movq	 .dey23,mm0
       movq	 .dby23,mm1
;----------
       mov	 ax,.z3
       sub	 ax,.z2
       cwde
 
       movd	  mm2,[.t_x2]
       movd	  mm3,[.t_x3]
 
       punpcklwd  mm2,mm4
       punpcklwd  mm3,mm4
       psubd	  mm3,mm2
 
       cvtsi2ss  xmm1,eax
       movlhps	 xmm1,xmm1
       cvtpi2ps  xmm1,mm3
 
       divps	 xmm1,xmm3   ; xmm1--> | xxx | dz | dty | dtx |
 
       shufps	 xmm1,xmm1,11100001b
			     ; xmm1--> | xxx | dz | dtx | dty |
       cvtps2pi  mm0,xmm1    ; mm0 -> 2 delta dwords  | dtx | dty |
       movhlps	 xmm1,xmm1
       cvtss2si  eax,xmm1    ; eax --> 1 delta dword | dz |
       movq	 .dty23,mm0
       mov	 .dz23,eax
else
 
       mov     ax,.z3
       sub     ax,.z2
       cwde
       shl     eax,CATMULL_SHIFT
       cdq
       idiv    ebx
     ; mov     .dz23,eax
       push    eax
 
       mov	ax,word[.b_x3]
       sub	ax,word[.b_x2]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dbx23,eax
       push	 eax
 
       mov	ax,word[.b_y3]
       sub	ax,word[.b_y2]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dby23,eax
       push	 eax
 
       mov	ax,word[.e_x3]
       sub	ax,word[.e_x2]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dex23,eax
       push	 eax
 
       mov	ax,word[.e_y3]
       sub	ax,word[.e_y2]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dey23,eax
       push	 eax
 
 
       mov	ax,word[.t_x3]
       sub	ax,word[.t_x2]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dtx23,eax
       push	 eax
 
       mov	ax,word[.t_y3]
       sub	ax,word[.t_y2]
       cwde
       shl	eax,ROUND
       cdq
       idiv	ebx
 ;     mov      .dty23,eax
       push	 eax
end if
      ;  sub     esp,40
   .bt_dx23_done:
       sub	 esp,64
       movsx	eax,.x1
       shl	eax,ROUND
       mov	.cx1,eax
       mov	.cx2,eax
  ;     push     eax
  ;     push     eax
 
       movsx	ebx,word[.b_x1]
       shl	ebx,ROUND
       mov	.cbx1,ebx
       mov	.cbx2,ebx
      ; push     ebx
      ; push     ebx
 
       movsx	ecx,word[.b_y1]
       shl	ecx,ROUND
       mov	.cby1,ecx
       mov	.cby2,ecx
      ; push     ecx
      ; push     ecx
 
       movsx	edx,word[.e_x1]
       shl	edx,ROUND
       mov	.cex1,edx
       mov	.cex2,edx
    ;   push     edx
    ;   push     edx
 
       movsx	eax,word[.e_y1]
       shl	eax,ROUND
       mov	.cey1,eax
       mov	.cey2,eax
    ;   push     eax
    ;   push     eax
 
       movsx	ebx,.z1
       shl	ebx,CATMULL_SHIFT
       mov	.cz1,ebx
       mov	.cz2,ebx
   ;    push     ebx
   ;    push     ebx
 
      ; sub      esp,16
       movsx	ecx,word[.t_x1]
       shl	ecx,ROUND
       mov	.ctx1,ecx
       mov	.ctx2,ecx
       ;push     ecx
       ;push     ecx
 
       movsx	edx,word[.t_y1]
       shl	edx,ROUND
       mov	.cty1,edx
       mov	.cty2,edx
      ; push     edx
      ; push     edx
 
 
       movsx	ecx,.y1
       cmp	cx,.y2
       jge	.loop12_done
  .loop12:
       call	.call_line
 
if Ext >= MMX
       movq	mm0,.cby2
       movq	mm1,.cby1
       movq	mm2,.cey2
       movq	mm3,.cey1
       movq	mm4,.cty1
       movq	mm5,.cty2
       movq	mm6,.cz1
       movq	mm7,.cz2
       paddd	mm0,.dby12
       paddd	mm1,.dby13
       paddd	mm2,.dey12
       paddd	mm3,.dey13
       paddd	mm4,.dty13
       paddd	mm5,.dty12
       paddd	mm6,.dz13
       paddd	mm7,.dz12
       movq	.cby2,mm0
       movq	.cby1,mm1
       movq	.cey1,mm3
       movq	.cey2,mm2
       movq	.cty1,mm4
       movq	.cty2,mm5
       movq	.cz1,mm6
       movq	.cz2,mm7
else
       mov	edx,.dbx13
       add	.cbx1,edx
       mov	eax,.dbx12
       add	.cbx2,eax
       mov	ebx,.dby13
       add	.cby1,ebx
       mov	edx,.dby12
       add	.cby2,edx
 
       mov	eax,.dex13
       add	.cex1,eax
       mov	ebx,.dex12
       add	.cex2,ebx
       mov	edx,.dey13
       add	.cey1,edx
       mov	eax,.dey12
       add	.cey2,eax
 
       mov	eax,.dtx13
       add	.ctx1,eax
       mov	ebx,.dtx12
       add	.ctx2,ebx
       mov	edx,.dty13
       add	.cty1,edx
       mov	eax,.dty12
       add	.cty2,eax
 
       mov	eax,.dx13
       add	.cx1,eax
       mov	ebx,.dx12
       add	.cx2,ebx
       mov	ebx,.dz13
       add	.cz1,ebx
       mov	edx,.dz12
       add	.cz2,edx
end if
       inc	ecx
       cmp	cx,.y2
       jl	.loop12
    .loop12_done:
 
       movsx	ecx,.y2
       cmp	cx,.y3
       jge	.loop23_done
 
       movsx	eax,.z2
       shl	eax,CATMULL_SHIFT
       mov	.cz2,eax
 
       movsx	ebx,.x2
       shl	ebx,ROUND
       mov	.cx2,ebx
 
       movzx	edx,word[.b_x2]
       shl	edx,ROUND
       mov	.cbx2,edx
 
       movzx	eax,word[.b_y2]
       shl	eax,ROUND
       mov	.cby2,eax
 
       movzx	ebx,word[.e_x2]
       shl	ebx,ROUND
       mov	.cex2,ebx
 
       movzx	edx,word[.e_y2]
       shl	edx,ROUND
       mov	.cey2,edx
 
       movzx	eax,word[.t_x2]
       shl	eax,ROUND
       mov	.ctx2,eax
 
       movzx	ebx,word[.t_y2]
       shl	ebx,ROUND
       mov	.cty2,ebx
 
     .loop23:
       call	.call_line
 
if Ext >= MMX
       movq	mm0,.cby2
       movq	mm1,.cby1
       movq	mm2,.cey2
       movq	mm3,.cey1
       movq	mm4,.cty1
       movq	mm5,.cty2
       movq	mm6,.cz1
       movq	mm7,.cz2
       paddd	mm0,.dby23
       paddd	mm1,.dby13
       paddd	mm2,.dey23
       paddd	mm3,.dey13
       paddd	mm4,.dty13
       paddd	mm5,.dty23
       paddd	mm6,.dz13
       paddd	mm7,.dz23
       movq	.cby2,mm0
       movq	.cby1,mm1
       movq	.cey2,mm2
       movq	.cey1,mm3
       movq	.cty1,mm4
       movq	.cty2,mm5
       movq	.cz1,mm6
       movq	.cz2,mm7
else
       mov	edx,.dbx13
       add	.cbx1,edx
       mov	eax,.dbx23
       add	.cbx2,eax
       mov	ebx,.dby13
       add	.cby1,ebx
       mov	edx,.dby23
       add	.cby2,edx
 
       mov	eax,.dex13
       add	.cex1,eax
       mov	ebx,.dex23
       add	.cex2,ebx
       mov	edx,.dey13
       add	.cey1,edx
       mov	eax,.dey23
       add	.cey2,eax
 
       mov	eax,.dx13
       add	.cx1,eax
       mov	ebx,.dx23
       add	.cx2,ebx
       mov	ebx,.dz13
       add	.cz1,ebx
       mov	edx,.dz23
       add	.cz2,edx
 
       mov	eax,.dtx13
       add	.ctx1,eax
       mov	ebx,.dtx23
       add	.ctx2,ebx
       mov	edx,.dty13
       add	.cty1,edx
       mov	eax,.dty23
       add	.cty2,eax
 
end if
       inc	ecx
       cmp	cx,.y3
       jl	.loop23
    .loop23_done:
 
       mov	esp,ebp
ret   50
 
.call_line:
 
       pushad
       push	.tex_ptr
       push	dword .cty2
       push	.ctx2
       push	dword .cty1
       push	.ctx1
       push	dword .cz1
       push	dword .cz2
       push	.z_buff
       push	.t_emap
       push	.t_bmap
       push	dword .cey2
       push	.cex2
       push	dword .cey1
       push	.cex1
       push	dword .cby2
       push	.cbx2
       push	dword .cby1
       push	.cbx1
       push	ecx
 
       mov	eax,.cx1
       sar	eax,ROUND
       mov	ebx,.cx2
       sar	ebx,ROUND
 
       call	bump_tex_line_z
 
       popad
ret
bump_tex_line_z:
;--------------in: eax - x1
;--------------    ebx - x2
;--------------    edi - pointer to screen buffer
;stack - another parameters :
.y	equ dword [ebp+4]
.bx1q	equ	  [ebp+8]
.bx2q	equ	  [ebp+16]
.ex1q	equ	  [ebp+24]
.ex2q	equ	  [ebp+32]
.tx1q	equ	  [ebp+60]
.tx2q	equ	  [ebp+68]
;.bx1q   equ       [ebp+8]
;.bx2q   equ       [ebp+16]
;.ex1q   equ       [ebp+24]
;.exyq   equ       [ebp+32]
.bx1	equ dword [ebp+8]   ;   ---
.by1	equ dword [ebp+12]  ;       |
.bx2	equ dword [ebp+16]  ;       |
.by2	equ dword [ebp+20]  ;       |>   b. map and e. map coords
.ex1	equ dword [ebp+24]  ;       |>   shifted shl ROUND
.ey1	equ dword [ebp+28]  ;       |
.ex2	equ dword [ebp+32]  ;       |
.ey2	equ dword [ebp+36]  ;   ---
.bmap	equ dword [ebp+40]  ; bump map offset
.emap	equ dword [ebp+44]  ; env map offset
.z_buff equ dword [ebp+48]
.z2	equ dword [ebp+52]  ;   -- |>   z coords shifted
.z1	equ dword [ebp+56]  ;   --       shl  CATMULL_SHIFT
 
.tx1	equ dword [ebp+60]  ;  -----
.ty1	equ dword [ebp+64]  ;       |>   shifted shl ROUND
.tx2	equ dword [ebp+68]  ;       |
.ty2	equ dword [ebp+72]  ;   ---
.tex_map equ dword [ebp+76]  ; texture offset  ( pointer )
 
 
.x1	equ dword [ebp-4]
.x2	equ dword [ebp-8]
.dbx	equ dword [ebp-12]
.dby	equ dword [ebp-16]
.dbyq	equ qword [ebp-16]  ; - new
.dex	equ dword [ebp-20]
.dey	equ dword [ebp-24]
.deyq	equ qword [ebp-24]  ; - new
.dz	equ dword [ebp-28]
.dtx	equ dword [ebp-32]
.dty	equ dword [ebp-36]
.dtyq	equ qword [ebp-36]
 
.cbx	equ dword [ebp-40]
.cby	equ dword [ebp-44]
.cbyq	equ qword [ebp-44]  ; - new
.cex	equ dword [ebp-48]
.cey	equ dword [ebp-52]
.ceyq	equ qword [ebp-52]  ; - new
.cz	equ dword [ebp-56]
.czbuff equ dword [ebp-60]
.ctx	equ dword [ebp-64]
.cty	equ dword [ebp-68]
.ctyq	equ qword [ebp-68]
.c_scr	equ dword [ebp-72]
 
.temp1	equ	   ebp-80
.temp2	equ	   ebp-88
.temp3	equ	   ebp-76
.temp4	equ	   ebp-84
.temp5	equ	   ebp-92
 
	mov	ebp,esp
 
 
 
	mov	ecx,.y
	or	ecx,ecx
	jl	.bl_end
	cmp	ecx,SIZE_Y
	jge	.bl_end
 
	cmp	eax,ebx
	jl	.bl_ok
	je	.bl_end
 
	xchg	eax,ebx
if Ext=NON
	mov	edx,.bx1
	xchg	edx,.bx2
	mov	.bx1,edx
	mov	edx,.by1
	xchg	edx,.by2
	mov	.by1,edx
 
	mov	edx,.ex1
	xchg	edx,.ex2
	mov	.ex1,edx
	mov	edx,.ey1
	xchg	edx,.ey2
	mov	.ey1,edx
 
	mov	edx,.tx1
	xchg	edx,.tx2
	mov	.tx1,edx
	mov	edx,.ty1
	xchg	edx,.ty2
	mov	.ty1,edx
else
	movq	mm0,.bx1q
	movq	mm1,.bx2q
	movq	mm2,.ex1q
	movq	mm3,.ex2q
	movq	mm4,.tx1q
	movq	mm5,.tx2q
	movq	.bx2q,mm0
	movq	.bx1q,mm1
	movq	.ex1q,mm3
	movq	.ex2q,mm2
	movq	.tx1q,mm5
	movq	.tx2q,mm4
end if
 
	mov	edx,.z1
	xchg	edx,.z2
	mov	.z1,edx
  .bl_ok:
	push	eax
	push	ebx	      ;store x1, x2
	cmp	.x1,SIZE_X
	jge	.bl_end
	cmp	.x2,0
	jle	.bl_end
 
	mov	ebx,.x2
	sub	ebx,.x1
 
if Ext>=SSE
 
       sub	 esp,28
       cvtsi2ss  xmm3,ebx	     ;rcps
       shufps	 xmm3,xmm3,0
 
       cvtpi2ps  xmm0,.bx1q ;mm0
       movlhps	 xmm0,xmm0
       cvtpi2ps  xmm0,.ex1q ;mm2
       cvtpi2ps  xmm1,.bx2q ;mm1
       movlhps	 xmm1,xmm1
       cvtpi2ps  xmm1,.ex2q ;mm3
       subps	 xmm1,xmm0
 
       divps	 xmm1,xmm3
 
       shufps	 xmm1,xmm1,10110001b
       cvtps2pi  mm0,xmm1	   ; mm0 -> 2 delta dwords
       movhlps	 xmm1,xmm1
       cvtps2pi  mm1,xmm1
       movq	 .deyq,mm0
       movq	 .dbyq,mm1
 
       movd	 mm2,.z1
       movd	 mm3,.z2
 
       cvtpi2ps  xmm0,.tx1q ;mm0
       movlhps	 xmm0,xmm0
       cvtpi2ps  xmm0,mm2
       cvtpi2ps  xmm1,.tx2q ;mm1
       movlhps	 xmm1,xmm1
       cvtpi2ps  xmm1,mm3
       subps	 xmm1,xmm0
 
       divps	 xmm1,xmm3
 
       shufps	 xmm1,xmm1,10110100b
       cvtps2pi  mm0,xmm1	   ; mm0 -> 2 delta dwords
       movhlps	 xmm1,xmm1
       cvtps2pi  mm1,xmm1
       movd	 .dz,mm0
       movq	 .dtyq,mm1
 
else
 
	mov	eax,.bx2       ; calc .dbx
	sub	eax,.bx1
	cdq
	idiv	ebx
	push	eax
 
	mov	eax,.by2       ; calc .dby
	sub	eax,.by1
	cdq
	idiv	ebx
	push	eax
 
	mov	eax,.ex2       ; calc .dex
	sub	eax,.ex1
	cdq
	idiv	ebx
	push	eax
 
	mov	eax,.ey2       ; calc .dey
	sub	eax,.ey1
	cdq
	idiv	ebx
	push	eax
 
 
	mov	eax,.z2        ; calc .dz
	sub	eax,.z1
	cdq
	idiv	ebx
	push	eax
 
	mov	eax,.tx2       ; calc .dtx
	sub	eax,.tx1
	cdq
	idiv	ebx
	push	eax
 
	mov	eax,.ty2       ; calc .dty
	sub	eax,.ty1
	cdq
	idiv	ebx
	push	eax
 
end if
	cmp	.x1,0	      ; set correctly begin variable
	jge	@f	      ; CLIPPING ON FUNCTION
			      ; cutting triangle exceedes screen
	mov	ebx,.x1
	neg	ebx
	imul	ebx	      ; eax = .dz * abs(.x1)
	add	.z1,eax
	mov	.x1,0
 
	mov	eax,.dbx
	imul	ebx
	add    .bx1,eax
 
	mov	eax,.dby
	imul	ebx
	add	.by1,eax
 
	mov	eax,.dex
	imul	ebx
	add	.ex1,eax
 
	mov	eax,.dey
	imul	ebx
	add	.ey1,eax
 
	mov	eax,.dtx
	imul	ebx
	add	.tx1,eax
 
	mov	eax,.dty
	imul	ebx
	add	.ty1,eax
 
      @@:
	cmp	.x2,SIZE_X
	jl	@f
	mov	.x2,SIZE_X
      @@:
	mov	eax,SIZE_X	 ;calc memory begin in buffers
	mul	.y
	add	eax,.x1
	lea	esi,[4*eax]
	add	esi,.z_buff	  ; z-buffer filled with dd variables
	lea	eax,[eax*3]
	add	edi,eax
 
 
	mov	ecx,.x2
	sub	ecx,.x1
	; init current variables
	push	.bx1   ; current b, e and t shifted shl ROUND   .cbx
	push	.by1					     ;  .cby
	push	.ex1					     ;  .cex
	push	.ey1					     ;  .cey
 
	push	.z1    ; current z shl CATMULL_SHIFT         ; .cz
	push	esi					     ; .czbuff
 
	push	.tx1	  ;         .ctx
	push	.ty1	  ;         .cty
	push	edi	  ;         .c_scr
if Ext>=MMX
	movq	mm7,.ctyq
	movq	mm6,.cbyq
	movq	mm5,.ceyq
;        movq    mm4,.dtyq
;        movq    mm3,.dbyq
end if
 
     .draw:
    ; if TEX = SHIFTING   ;bump drawing only in shifting mode
	mov	esi,.czbuff	 ; .czbuff current address in buffer
	mov	ebx,.cz 	 ; .cz - cur z position
	cmp	ebx,dword[esi]
	jge	.skip
if Ext=NON
	mov	eax,.cby
	shr	eax,ROUND
	mov	esi,.cbx
	shr	esi,ROUND
else
	movq	mm1,mm6
	psrld	mm1,ROUND
	movd	eax,mm1
	psrlq	mm1,32
	movd	esi,mm1
end if
 
 
 
	shl	eax,TEX_SHIFT
	add	esi,eax 	;-  ; esi - current bump map index
 
	mov	ebx,esi
	dec	ebx
	and	ebx,TEXTURE_SIZE
	add	ebx,.bmap
	movzx	eax,byte [ebx]
 
	mov	ebx,esi
	inc	ebx
	and	ebx,TEXTURE_SIZE
	add	ebx,.bmap
	movzx	ebx,byte [ebx]
	sub	eax,ebx
 
	mov	ebx,esi
	sub	ebx,TEX_X
	and	ebx,TEXTURE_SIZE
	add	ebx,.bmap
	movzx	edx,byte [ebx]
 
	mov	ebx,esi
	add	ebx,TEX_X
	and	ebx,TEXTURE_SIZE
	add	ebx,.bmap
	movzx	ebx,byte [ebx]
	sub	edx,ebx
 
     ;  eax - horizontal sub    modificated x coord
     ;  edx - vertical   sub    modificated y coord
if Ext=NON
	mov	ebx,.cex       ;.cex - current env map X
	shr	ebx,ROUND
	add	eax,ebx
 
 
	mov	ebx,.cey       ;.cey - current  env map y
	shr	ebx,ROUND
	add	edx,ebx
 
else
	movq	mm1,mm5        ; mm5 - copy of cur env coords
	psrld	mm1,ROUND
	movd	ebx,mm1
	psrlq	mm1,32
	add	eax,ebx
	movd	ebx,mm1
	add	edx,ebx
;        movq    qword[.temp1],mm3
;        add     eax,dword [.temp1]
;        add     edx,dword [.temp1+4]
end if
 
	or	eax,eax
	jl	.black
	cmp	eax,TEX_X
	jg	.black
	or	edx,edx
	jl	.black
	cmp	edx,TEX_Y
	jg	.black
 
	shl	edx,TEX_SHIFT	; zaburzenie w emapie = zaburzenie w teksturze
	add	edx,eax 	; proponuje nie stawiac czarnego pixela tylko
	lea	esi,[edx*3]	; niezaburzony.
	add	esi,.emap	;
	lodsd
 
if Ext=NON
	mov	edx,.cty
	shr	edx,ROUND  ; sar
 
	mov	edi,.ctx
	shr	edi,ROUND  ; sar
else
	movq	mm1,mm7
	psrld	mm1,ROUND
	movd	edx,mm1
	psrlq	mm1,32
	movd	edi,mm1
 
end if
 
	shl	edx,TEX_SHIFT
	add	edi,edx
	and	edi,TEXTURE_SIZE
	lea	esi,[edi*3]
	add	esi,.tex_map
 
if Ext=NON
	mov	edx,eax
	lodsd
	push	ax
	mul	dl
	mov	dl,ah
	pop	ax
	shr	ax,8
	mul	dh
	mov	al,dl
	mov	edi,.c_scr
	stosw
	shr	edx,16
	shr	eax,16
	mul	dl
	shr	ax,8
	stosb
else
	movd	   mm0,eax
	pxor	   mm1,mm1
	punpcklbw  mm0,mm1
	movd	   mm2,[esi]
	punpcklbw  mm2,mm1
	pmullw	   mm0,mm2
	psrlw	   mm0,8
	packuswb   mm0,mm1
	mov	   edi,.c_scr
	movd	   [edi],mm0
 
end if
 
	jmp	.actual_zbuff	; actualize z buffer
     @@:
     .black:
	xor	eax,eax
	mov	edi,.c_scr
	stosd
     .actual_zbuff:
	mov	eax,.cz
	mov	edi,.czbuff
	stosd
 
      .skip:
	add	.czbuff,4
	add	.c_scr,3
 
if Ext=NON
	mov	eax,.dbx
	add	.cbx,eax
	mov	ebx,.dby
	add	.cby,ebx
 
	mov	edx,.dex
	add	.cex,edx
	mov	eax,.dey
	add	.cey,eax
 
	mov	ebx,.dtx
	add	.ctx,ebx
	mov	edx,.dty
	add	.cty,edx
 
else
	paddd	mm7,.dtyq
	paddd	mm6,.dbyq
	paddd	mm5,.deyq
end if
	mov	eax,.dz
	add	.cz,eax
 
	dec	ecx
	jnz	.draw
 
  .bl_end:
	mov	esp,ebp
ret 76
;Ext = MMX
 
;     else
;        movq    mm5, qword[.temp1]  ;-
;        paddd   mm5, qword[.temp5]  ; .temp5 == low dword = TEX_X, high dword = -TEX_X
;        pand    mm5, qword[.temp3]  ; .temp3 == low = high dword = TEX_SIZE
;        paddd   mm5, qword[.temp4]  ; .temp4 == low = high dword = offset .bmap
;        movd    ebx,mm5
;        psrlq   mm5,32
;     end if

Rev 1245	Rev 1776
1	;CATMULL_SHIFT equ 8	1	;CATMULL_SHIFT equ 8
2	;TEXTURE_SIZE EQU (TEX_X * TEX_Y)-1	2	;TEXTURE_SIZE EQU (TEX_X * TEX_Y)-1
3	;ROUND equ 8	3	;ROUND equ 8
4	;Ext = NON	4	;Ext = NON
5	;MMX = 1	5	;MMX = 1
6	;NON = 0	6	;NON = 0
7	;------- Big thanks to Majuma (www.majuma.xt.pl) for absolutely great---	7	;------- Big thanks to Majuma (www.majuma.xt.pl) for absolutely great---
8	;------- DOS 13h mode demos --------------------------------------------	8	;------- DOS 13h mode demos --------------------------------------------
9	;------- Procedure draws bump triangle with texture, I use -------------	9	;------- Procedure draws bump triangle with texture, I use -------------
10	;--------Catmull Z-buffer algorithm- (Z coordinate interpolation)-------	10	;--------Catmull Z-buffer algorithm- (Z coordinate interpolation)-------
11	;--------I calc texture pixel by this way: col1*col2/256 ---------------	11	;--------I calc texture pixel by this way: col1*col2/256 ---------------
12	bump_tex_triangle_z:	12	bump_tex_triangle_z:
13	;------------------in - eax - x1 shl 16 + y1 -----------	13	;------------------in - eax - x1 shl 16 + y1 -----------
14	;---------------------- ebx - x2 shl 16 + y2 -----------	14	;---------------------- ebx - x2 shl 16 + y2 -----------
15	;---------------------- ecx - x3 shl 16 + y3 -----------	15	;---------------------- ecx - x3 shl 16 + y3 -----------
16	;---------------------- edx - pointer to bump map-------	16	;---------------------- edx - pointer to bump map-------
17	;---------------------- esi - pointer to env map--------	17	;---------------------- esi - pointer to env map--------
18	;---------------------- edi - pointer to screen buffer--	18	;---------------------- edi - pointer to screen buffer--
19	;---------------------- stack : bump coordinates--------	19	;---------------------- stack : bump coordinates--------
20	;---------------------- environment coordinates-	20	;---------------------- environment coordinates-
21	;---------------------- Z position coordinates--	21	;---------------------- Z position coordinates--
22	;---------------------- pointer to Z buffer-----	22	;---------------------- pointer to Z buffer-----
23	;---------------------- pointer to texture------	23	;---------------------- pointer to texture------
24	;---------------------- texture coordinates-----	24	;---------------------- texture coordinates-----
25	;-- Z-buffer - filled with coordinates as dword --------	25	;-- Z-buffer - filled with coordinates as dword --------
26	;-- (Z coor. as word) shl CATMULL_SHIFT ----------------	26	;-- (Z coor. as word) shl CATMULL_SHIFT ----------------
27	.b_x1 equ ebp+4 ; procedure don't save registers !!!	27	.b_x1 equ ebp+4 ; procedure don't save registers !!!
28	.b_y1 equ ebp+6 ; each coordinate as word	28	.b_y1 equ ebp+6 ; each coordinate as word
29	.b_x2 equ ebp+8	29	.b_x2 equ ebp+8
30	.b_y2 equ ebp+10 ; b - bump map coords	30	.b_y2 equ ebp+10 ; b - bump map coords
31	.b_x3 equ ebp+12 ; e - env map coords	31	.b_x3 equ ebp+12 ; e - env map coords
32	.b_y3 equ ebp+14	32	.b_y3 equ ebp+14
33	.e_x1 equ ebp+16	33	.e_x1 equ ebp+16
34	.e_y1 equ ebp+18	34	.e_y1 equ ebp+18
35	.e_x2 equ ebp+20	35	.e_x2 equ ebp+20
36	.e_y2 equ ebp+22	36	.e_y2 equ ebp+22
37	.e_x3 equ ebp+24	37	.e_x3 equ ebp+24
38	.e_y3 equ ebp+26	38	.e_y3 equ ebp+26
39	.z1 equ word[ebp+28]	39	.z1 equ word[ebp+28]
40	.z2 equ word[ebp+30]	40	.z2 equ word[ebp+30]
41	.z3 equ word[ebp+32]	41	.z3 equ word[ebp+32]
42	.z_buff equ dword[ebp+34] ; pointer to Z-buffer	42	.z_buff equ dword[ebp+34] ; pointer to Z-buffer
43	.tex_ptr equ dword[ebp+38] ; ptr to texture	43	.tex_ptr equ dword[ebp+38] ; ptr to texture
44	.t_x1 equ ebp+42 ; texture coords	44	.t_x1 equ ebp+42 ; texture coords
45	.t_y1 equ ebp+44	45	.t_y1 equ ebp+44
46	.t_x2 equ ebp+46	46	.t_x2 equ ebp+46
47	.t_y2 equ ebp+48	47	.t_y2 equ ebp+48
48	.t_x3 equ ebp+50	48	.t_x3 equ ebp+50
49	.t_y3 equ ebp+52	49	.t_y3 equ ebp+52
50		50
51		51
52		52
53	.t_bmap equ dword[ebp-4] ; pointer to bump map	53	.t_bmap equ dword[ebp-4] ; pointer to bump map
54	.t_emap equ dword[ebp-8] ; pointer to env map	54	.t_emap equ dword[ebp-8] ; pointer to env map
55	.x1 equ word[ebp-10]	55	.x1 equ word[ebp-10]
56	.y1 equ word[ebp-12]	56	.y1 equ word[ebp-12]
57	.x2 equ word[ebp-14]	57	.x2 equ word[ebp-14]
58	.y2 equ word[ebp-16]	58	.y2 equ word[ebp-16]
59	.x3 equ word[ebp-18]	59	.x3 equ word[ebp-18]
60	.y3 equ word[ebp-20]	60	.y3 equ word[ebp-20]
61		61
62	.dx12 equ dword[ebp-24]	62	.dx12 equ dword[ebp-24]
63	.dz12 equ [ebp-28]	63	.dz12 equ [ebp-28]
64	.dbx12 equ dword[ebp-32]	64	.dbx12 equ dword[ebp-32]
65	.dby12 equ [ebp-36]	65	.dby12 equ [ebp-36]
66	.dex12 equ dword[ebp-40]	66	.dex12 equ dword[ebp-40]
67	.dey12 equ [ebp-44]	67	.dey12 equ [ebp-44]
68	.dtx12 equ dword[ebp-48]	68	.dtx12 equ dword[ebp-48]
69	.dty12 equ [ebp-52]	69	.dty12 equ [ebp-52]
70		70
71	.dx13 equ dword[ebp-52-4*1]	71	.dx13 equ dword[ebp-52-4*1]
72	.dz13 equ [ebp-52-4*2]	72	.dz13 equ [ebp-52-4*2]
73	.dbx13 equ dword[ebp-52-4*3]	73	.dbx13 equ dword[ebp-52-4*3]
74	.dby13 equ [ebp-52-4*4]	74	.dby13 equ [ebp-52-4*4]
75	.dex13 equ dword[ebp-52-4*5]	75	.dex13 equ dword[ebp-52-4*5]
76	.dey13 equ [ebp-52-4*6]	76	.dey13 equ [ebp-52-4*6]
77	.dtx13 equ dword[ebp-52-4*7]	77	.dtx13 equ dword[ebp-52-4*7]
78	.dty13 equ [ebp-52-4*8]	78	.dty13 equ [ebp-52-4*8]
79		79
80		80
81	.dx23 equ dword[ebp-(52+4*9)]	81	.dx23 equ dword[ebp-(52+4*9)]
82	.dz23 equ [ebp-(52+4*10)]	82	.dz23 equ [ebp-(52+4*10)]
83	.dbx23 equ dword[ebp-(52+4*11)]	83	.dbx23 equ dword[ebp-(52+4*11)]
84	.dby23 equ [ebp-(52+4*12)]	84	.dby23 equ [ebp-(52+4*12)]
85	.dex23 equ dword[ebp-(52+4*13)]	85	.dex23 equ dword[ebp-(52+4*13)]
86	.dey23 equ [ebp-(52+4*14)]	86	.dey23 equ [ebp-(52+4*14)]
87	.dtx23 equ dword[ebp-(52+4*15)]	87	.dtx23 equ dword[ebp-(52+4*15)]
88	.dty23 equ [ebp-(52+4*16)]	88	.dty23 equ [ebp-(52+4*16)]
89		89
90	.cx1 equ dword[ebp-(52+4*17)] ; current variables	90	.cx1 equ dword[ebp-(52+4*17)] ; current variables
91	.cz1 equ [ebp-(52+4*18)]	91	.cz1 equ [ebp-(52+4*18)]
92	.cx2 equ dword[ebp-(52+4*19)]	92	.cx2 equ dword[ebp-(52+4*19)]
93	.cz2 equ [ebp-(52+4*20)]	93	.cz2 equ [ebp-(52+4*20)]
94	.cbx1 equ dword[ebp-(52+4*21)]	94	.cbx1 equ dword[ebp-(52+4*21)]
95	.cby1 equ [ebp-(52+4*22)]	95	.cby1 equ [ebp-(52+4*22)]
96	.cbx2 equ dword[ebp-(52+4*23)]	96	.cbx2 equ dword[ebp-(52+4*23)]
97	.cby2 equ [ebp-(52+4*24)]	97	.cby2 equ [ebp-(52+4*24)]
98	.cex1 equ dword[ebp-(52+4*25)]	98	.cex1 equ dword[ebp-(52+4*25)]
99	.cey1 equ [ebp-(52+4*26)]	99	.cey1 equ [ebp-(52+4*26)]
100	.cex2 equ dword[ebp-(52+4*27)]	100	.cex2 equ dword[ebp-(52+4*27)]
101	.cey2 equ [ebp-(52+4*28)]	101	.cey2 equ [ebp-(52+4*28)]
102		102
103	.ctx1 equ dword[ebp-(52+4*29)]	103	.ctx1 equ dword[ebp-(52+4*29)]
104	.cty1 equ [ebp-(52+4*30)]	104	.cty1 equ [ebp-(52+4*30)]
105	.ctx2 equ dword[ebp-(52+4*31)]	105	.ctx2 equ dword[ebp-(52+4*31)]
106	.cty2 equ [ebp-(52+4*32)]	106	.cty2 equ [ebp-(52+4*32)]
107		107
108	cld	108	cld
109	mov ebp,esp	109	mov ebp,esp
110	push edx ; store bump map	110	push edx ; store bump map
111	push esi ; store e. map	111	push esi ; store e. map
112	; sub esp,120	112	; sub esp,120
113	.sort3: ; sort triangle coordinates...	113	.sort3: ; sort triangle coordinates...
114	cmp ax,bx	114	cmp ax,bx
115	jle .sort1	115	jle .sort1
116	xchg eax,ebx	116	xchg eax,ebx
117	mov edx,dword[.b_x1]	117	mov edx,dword[.b_x1]
118	xchg edx,dword[.b_x2]	118	xchg edx,dword[.b_x2]
119	mov dword[.b_x1],edx	119	mov dword[.b_x1],edx
120	mov edx,dword[.e_x1]	120	mov edx,dword[.e_x1]
121	xchg edx,dword[.e_x2]	121	xchg edx,dword[.e_x2]
122	mov dword[.e_x1],edx	122	mov dword[.e_x1],edx
123	mov edx,dword[.t_x1]	123	mov edx,dword[.t_x1]
124	xchg edx,dword[.t_x2]	124	xchg edx,dword[.t_x2]
125	mov dword[.t_x1],edx	125	mov dword[.t_x1],edx
126	mov dx,.z1	126	mov dx,.z1
127	xchg dx,.z2	127	xchg dx,.z2
128	mov .z1,dx	128	mov .z1,dx
129	.sort1:	129	.sort1:
130	cmp bx,cx	130	cmp bx,cx
131	jle .sort2	131	jle .sort2
132	xchg ebx,ecx	132	xchg ebx,ecx
133	mov edx,dword[.b_x2]	133	mov edx,dword[.b_x2]
134	xchg edx,dword[.b_x3]	134	xchg edx,dword[.b_x3]
135	mov dword[.b_x2],edx	135	mov dword[.b_x2],edx
136	mov edx,dword[.e_x2]	136	mov edx,dword[.e_x2]
137	xchg edx,dword[.e_x3]	137	xchg edx,dword[.e_x3]
138	mov dword[.e_x2],edx	138	mov dword[.e_x2],edx
139	mov edx,dword[.t_x2]	139	mov edx,dword[.t_x2]
140	xchg edx,dword[.t_x3]	140	xchg edx,dword[.t_x3]
141	mov dword[.t_x2],edx	141	mov dword[.t_x2],edx
142	mov dx,.z2	142	mov dx,.z2
143	xchg dx,.z3	143	xchg dx,.z3
144	mov .z2,dx	144	mov .z2,dx
145	jmp .sort3	145	jmp .sort3
146	.sort2:	146	.sort2:
147	push eax ; store triangle coords in variables	147	push eax ; store triangle coords in variables
148	push ebx	148	push ebx
149	push ecx	149	push ecx
150	mov edx,80008000h ; eax,ebx,ecx are ANDd together into edx which means that	150	mov edx,80008000h ; eax,ebx,ecx are ANDd together into edx which means that
151	and edx,ebx ; if all of them are negative a sign flag is raised	151	and edx,ebx ; if all of them are negative a sign flag is raised
152	and edx,ecx	152	and edx,ecx
153	and edx,eax	153	and edx,eax
154	test edx,80008000h ; Check both X&Y at once	154	test edx,80008000h ; Check both X&Y at once
155	jne .loop23_done	155	jne .loop23_done
156	; mov edx,eax ; eax,ebx,ecx are ORd together into edx which means that	156	; mov edx,eax ; eax,ebx,ecx are ORd together into edx which means that
157	; or edx,ebx ; if any one of them is negative a sign flag is raised	157	; or edx,ebx ; if any one of them is negative a sign flag is raised
158	; or edx,ecx	158	; or edx,ecx
159	; test edx,80000000h ; Check only X	159	; test edx,80000000h ; Check only X
160	; jne .loop23_done	160	; jne .loop23_done
161		161
162	; cmp .x1,SIZE_X ; {	162	; cmp .x1,SIZE_X ; {
163	; jg .loop23_done	163	; jg .loop23_done
164	; cmp .x2,SIZE_X ; This can be optimized with effort	164	; cmp .x2,SIZE_X ; This can be optimized with effort
165	; jg .loop23_done	165	; jg .loop23_done
166	; cmp .x3,SIZE_X	166	; cmp .x3,SIZE_X
167	; jg .loop23_done ; {	167	; jg .loop23_done ; {
168		168
169		169
170	mov bx,.y2 ; calc delta 12	170	mov bx,.y2 ; calc delta 12
171	sub bx,.y1	171	sub bx,.y1
172	jnz .bt_dx12_make	172	jnz .bt_dx12_make
173	mov ecx,8	173	mov ecx,8
174	xor edx,edx	174	xor edx,edx
175	@@:	175	@@:
176	push edx ;dword 0	176	push edx ;dword 0
177	loop @b	177	loop @b
178	jmp .bt_dx12_done	178	jmp .bt_dx12_done
179	.bt_dx12_make:	179	.bt_dx12_make:
180		180	movsx ebx,bx
-		181
181	mov ax,.x2	182	mov ax,.x2
182	sub ax,.x1	183	sub ax,.x1
183	cwde	184	cwde
184	movsx ebx,bx	185	shl eax,ROUND
185	shl eax,ROUND	-
186	cdq	186	cdq
187	idiv ebx	187	idiv ebx
188	; mov .dx12,eax	188	; mov .dx12,eax
189	push eax	189	push eax
190		190
191	mov ax,.z2	-
192	sub ax,.z1	-
193	cwde	-
194	shl eax,CATMULL_SHIFT	-
195	cdq	-
196	idiv ebx	-
197	push eax	-
198		-
199	if Ext>=SSE	191	if Ext>=SSE
200		192
201	sub esp,16	193	sub esp,28
202	; mov eax,256	194	; mov eax,256
203	cvtsi2ss xmm4,[i255d]	195	cvtsi2ss xmm4,[i255d]
204	cvtsi2ss xmm3,ebx ;rcps	196	cvtsi2ss xmm3,ebx ;rcps
205	divss xmm3,xmm4	197	divss xmm3,xmm4
206	shufps xmm3,xmm3,0	198	shufps xmm3,xmm3,0
207		199
208	movd mm0,[.b_x1]	200	movd mm0,[.b_x1]
209	movd mm1,[.b_x2]	201	movd mm1,[.b_x2]
210	movd mm2,[.e_x1]	202	movd mm2,[.e_x1]
211	movd mm3,[.e_x2]	203	movd mm3,[.e_x2]
212		204
213	pxor mm4,mm4	205	pxor mm4,mm4
214	punpcklwd mm0,mm4	206	punpcklwd mm0,mm4
215	punpcklwd mm1,mm4	207	punpcklwd mm1,mm4
216	punpcklwd mm2,mm4	208	punpcklwd mm2,mm4
217	punpcklwd mm3,mm4	209	punpcklwd mm3,mm4
218		210
219	psubd mm1,mm0	211	psubd mm1,mm0
220	psubd mm3,mm2	212	psubd mm3,mm2
221		213
222	cvtpi2ps xmm1,mm1	214	cvtpi2ps xmm1,mm1
223	movlhps xmm1,xmm1	215	movlhps xmm1,xmm1
224	cvtpi2ps xmm1,mm3	216	cvtpi2ps xmm1,mm3
225		217
226	divps xmm1,xmm3 ;xmm1--> \| dby \| dbx \| dey \| dex \|	218	divps xmm1,xmm3 ;xmm1--> \| dby \| dbx \| dey \| dex \|
227		219
228	shufps xmm1,xmm1,10110001b	220	shufps xmm1,xmm1,10110001b
229	;xmm1--> \| dbx \| dby \| dex \| dey \|	221	;xmm1--> \| dbx \| dby \| dex \| dey \|
230	cvtps2pi mm0,xmm1 ; mm0 -> 2 delta dwords	222	cvtps2pi mm0,xmm1 ;mm0,xmm1 ; mm0 -> 2 delta dwords
231	movhlps xmm1,xmm1	223	movhlps xmm1,xmm1
232	cvtps2pi mm1,xmm1	224	cvtps2pi mm1,xmm1 ;mm1,xmm1
233	movq .dey12,mm0	225	movq .dey12,mm0
234	movq .dby12,mm1	226	movq .dby12,mm1
235	;-------------	227	;-------------
236	; mov ax,.z2	228	; pxor mm0,mm0
-		229	; pxor mm1,mm1
-		230	;/ pinsrw mm0,.z1,1
-		231	;/ pinsrw mm0,.x1,0
-		232	;/ pinsrw mm1,.z2,1
-		233	;/ pinsrw mm1,.x2,0
-		234	mov ax,.z2
-		235	sub ax,.z1
-		236	cwde
-		237	;/ movd mm1,eax
-		238
-		239	;/ punpcklwd mm0,mm4
-		240	;/ punpcklwd mm1,mm4
-		241
-		242	; cvtpi2ps xmm1,mm1
-		243	; cvtpi2ps xmm2,mm0
-		244	; subps xmm1,xmm2
-		245
-		246	;/ psubd mm1,mm0
-		247
-		248	movd mm2,[.t_x1]
-		249	movd mm3,[.t_x2]
-		250
-		251	punpcklwd mm2,mm4
-		252	punpcklwd mm3,mm4
-		253	psubd mm3,mm2
-		254
-		255	;/ cvtpi2ps xmm1,mm1
-		256	cvtsi2ss xmm1,eax
-		257	movlhps xmm1,xmm1
-		258	cvtpi2ps xmm1,mm3
-		259
-		260	divps xmm1,xmm3 ; xmm1--> \| xxx \| dz \| dty \| dtx \|
-		261
-		262	shufps xmm1,xmm1,11100001b
-		263	; xmm1--> \| xxx \| dz \| dtx \| dty \|
-		264	cvtps2pi mm0,xmm1 ; mm0 -> 2 delta dwords \| dtx \| dty \|
-		265	movhlps xmm1,xmm1
-		266	cvtss2si eax,xmm1 ; eax --> 1 delta dword \| dz \|
-		267	movq .dty12,mm0
-		268	mov .dz12,eax
-		269	;----
-		270	; mov ax,.z2
237	; sub ax,.z1	271	; sub ax,.z1
238	; cwde	272	; cwde
239	; mov bx,.x2	273	; mov bx,.x2
240	; sub bx,.x1	274	; sub bx,.x1
241	; movsx ebx,bx	275	; movsx ebx,bx
242	; movd mm1,eax	276	; movd mm1,eax
243	; psllq mm1,32	277	; psllq mm1,32
244	; movd mm1,ebx	278	; movd mm1,ebx
245	;; push ebx	279
-		280	;; push ebx
246	;; push eax	281	;; push eax
247	;; movq mm1,[esp]	282	;; movq mm1,[esp]
248	;; add esp,8	283	;; add esp,8
249	;;; mov ax,.z1	284	;;; mov ax,.z1
250	;;; mov bx,.z2	285	;;; mov bx,.z2
251	;;; shl eax,16	286	;;; shl eax,16
252	;;; shl ebx,16	287	;;; shl ebx,16
253	;;; mov ax,.x1	288	;;; mov ax,.x1
254	;;; mov bx,.x2	289	;;; mov bx,.x2
255	; movd mm2,[.t_x1]	290	; movd mm2,[.t_x1]
256	; movd mm3,[.t_x2]	291	; movd mm3,[.t_x2]
257	;; movd mm0,eax	292	;; movd mm0,eax
258	;; movd mm1,ebx	293	;; movd mm1,ebx
259		294
260	; pxor mm4,mm4	295	; pxor mm4,mm4
261	;; punpcklwd mm0,mm4	296	;; punpcklwd mm0,mm4
262	;; punpcklwd mm1,mm4	297	;; punpcklwd mm1,mm4
263	; punpcklwd mm2,mm4	298	; punpcklwd mm2,mm4
264	; punpcklwd mm3,mm4	299	; punpcklwd mm3,mm4
265		300
266	;; psubd mm1,mm0	301	;; psubd mm1,mm0
267	; psubd mm3,mm2	302	; psubd mm3,mm2
268		303
269		304
270	; cvtpi2ps xmm1,mm1	305	; cvtpi2ps xmm1,mm1
271	; movlhps xmm1,xmm1	306	; movlhps xmm1,xmm1
272	; cvtpi2ps xmm1,mm3	307	; cvtpi2ps xmm1,mm3
273		308
274	; divps xmm1,xmm3 ; xmm1--> \| dz \| dx \| dty \| dtx \|	309	; divps xmm1,xmm3 ; xmm1--> \| dz \| dx \| dty \| dtx \|
275		310
276	; shufps xmm1,xmm1,10110001b	311	; shufps xmm1,xmm1,10110001b
277	; xmm1--> \| dx \| dz \| dtx \| dty \|	312	; xmm1--> \| dx \| dz \| dtx \| dty \|
278	; cvtps2pi mm0,xmm1 ; mm0 -> 2 delta dwords \| dtx \| dty \|	313	; cvtps2pi mm0,xmm1 ; mm0 -> 2 delta dwords \| dtx \| dty \|
279	; movhlps xmm1,xmm1	314	; movhlps xmm1,xmm1
280	; cvtps2pi mm1,xmm1 ; mm1 --> 2 delta dwords \| dx \| dz \|	315	; cvtps2pi mm1,xmm1 ; mm1 --> 2 delta dwords \| dx \| dz \|
281	; movq .dty12,mm0	316	; movq .dty12,mm0
282	; movq .dz12,mm1	317	; movq .dz12,mm1
283	else	318	else
284		319	mov ax,.z2
-		320	sub ax,.z1
-		321	cwde
-		322	shl eax,CATMULL_SHIFT
-		323	cdq
-		324	idiv ebx
-		325	push eax
-		326
285	mov ax,word[.b_x2]	327	mov ax,word[.b_x2]
286	sub ax,word[.b_x1]	328	sub ax,word[.b_x1]
287	cwde	329	cwde
288	shl eax,ROUND	330	shl eax,ROUND
289	cdq	331	cdq
290	idiv ebx	332	idiv ebx
291	; mov .dbx12,eax	333	; mov .dbx12,eax
292	push eax	334	push eax
293		335
294	mov ax,word[.b_y2]	336	mov ax,word[.b_y2]
295	sub ax,word[.b_y1]	337	sub ax,word[.b_y1]
296	cwde	338	cwde
297	shl eax,ROUND	339	shl eax,ROUND
298	cdq	340	cdq
299	idiv ebx	341	idiv ebx
300	; mov .dby12,eax	342	; mov .dby12,eax
301	push eax	343	push eax
302		344
303	mov ax,word[.e_x2]	345	mov ax,word[.e_x2]
304	sub ax,word[.e_x1]	346	sub ax,word[.e_x1]
305	cwde	347	cwde
306	shl eax,ROUND	348	shl eax,ROUND
307	cdq	349	cdq
308	idiv ebx	350	idiv ebx
309	; mov .dex12,eax	351	; mov .dex12,eax
310	push eax	352	push eax
311		353
312	mov ax,word[.e_y2]	354	mov ax,word[.e_y2]
313	sub ax,word[.e_y1]	355	sub ax,word[.e_y1]
314	cwde	356	cwde
315	shl eax,ROUND	357	shl eax,ROUND
316	cdq	358	cdq
317	idiv ebx	359	idiv ebx
318	; mov .dey12,eax	360	; mov .dey12,eax
319	push eax	361	push eax
320		362
321	end if	363	;end if
322		364
323	mov ax,word[.t_x2]	365	mov ax,word[.t_x2]
324	sub ax,word[.t_x1]	366	sub ax,word[.t_x1]
325	cwde	367	cwde
326	shl eax,ROUND	368	shl eax,ROUND
327	cdq	369	cdq
328	idiv ebx	370	idiv ebx
329	; mov .dtx12,eax	371	; mov .dtx12,eax
330	push eax	372	push eax
331		373
332	mov ax,word[.t_y2]	374	mov ax,word[.t_y2]
333	sub ax,word[.t_y1]	375	sub ax,word[.t_y1]
334	cwde	376	cwde
335	shl eax,ROUND	377	shl eax,ROUND
336	cdq	378	cdq
337	idiv ebx	379	idiv ebx
338	; mov .dty12,eax	380	; mov .dty12,eax
339	push eax	381	push eax
340		382	end if
341	.bt_dx12_done:	383	.bt_dx12_done:
342		384
343	mov bx,.y3 ; calc delta13	385	mov bx,.y3 ; calc delta13
344	sub bx,.y1	386	sub bx,.y1
345	jnz .bt_dx13_make	387	jnz .bt_dx13_make
346	mov ecx,8	388	mov ecx,8
347	xor edx,edx	389	xor edx,edx
348	@@:	390	@@:
349	push edx ;dword 0	391	push edx ;dword 0
350	loop @b	392	loop @b
351	jmp .bt_dx13_done	393	jmp .bt_dx13_done
352	.bt_dx13_make:	394	.bt_dx13_make:
353	mov ax,.x3	395	mov ax,.x3
354	sub ax,.x1	396	sub ax,.x1
355	cwde	397	cwde
356	movsx ebx,bx	398	movsx ebx,bx
357	shl eax,ROUND	399	shl eax,ROUND
358	cdq	400	cdq
359	idiv ebx	401	idiv ebx
360	; mov .dx13,eax	402	; mov .dx13,eax
361	push eax	403	push eax
362		404
363	mov ax,.z3	-
364	sub ax,.z1	-
365	cwde	-
366	shl eax,CATMULL_SHIFT	-
367	cdq	-
368	idiv ebx	-
369	; mov .dz13,eax	-
370	push eax	-
371		-
372	if Ext>=SSE	405	if Ext>=SSE
373		406
374	sub esp,16	407	sub esp,28
375	; mov eax,255	408	; mov eax,255
376	cvtsi2ss xmm4,[i255d]	409	cvtsi2ss xmm4,[i255d]
377	cvtsi2ss xmm3,ebx ;rcps	410	cvtsi2ss xmm3,ebx ;rcps
378	divss xmm3,xmm4	411	divss xmm3,xmm4
379	shufps xmm3,xmm3,0	412	shufps xmm3,xmm3,0
380		413
381	movd mm0,[.b_x1]	414	movd mm0,[.b_x1]
382	movd mm1,[.b_x3]	415	movd mm1,[.b_x3]
383	movd mm2,[.e_x1]	416	movd mm2,[.e_x1]
384	movd mm3,[.e_x3]	417	movd mm3,[.e_x3]
385		418
386	pxor mm4,mm4	419	pxor mm4,mm4
387	punpcklwd mm0,mm4	420	punpcklwd mm0,mm4
388	punpcklwd mm1,mm4	421	punpcklwd mm1,mm4
389	punpcklwd mm2,mm4	422	punpcklwd mm2,mm4
390	punpcklwd mm3,mm4	423	punpcklwd mm3,mm4
391		424
392	psubd mm1,mm0	425	psubd mm1,mm0
393	psubd mm3,mm2	426	psubd mm3,mm2
394		427
395	cvtpi2ps xmm1,mm1	428	cvtpi2ps xmm1,mm1
396	movlhps xmm1,xmm1	429	movlhps xmm1,xmm1
397	cvtpi2ps xmm1,mm3	430	cvtpi2ps xmm1,mm3
398		431
399	divps xmm1,xmm3 ;xmm1--> \| dby \| dbx \| dey \| dex \|	432	divps xmm1,xmm3 ;xmm1--> \| dby \| dbx \| dey \| dex \|
400		433
401	shufps xmm1,xmm1,10110001b	434	shufps xmm1,xmm1,10110001b
402	;xmm1--> \| dbx \| dby \| dex \| dey \|	435	;xmm1--> \| dbx \| dby \| dex \| dey \|
403	cvtps2pi mm0,xmm1 ; mm0 -> 2 delta dwords	436	cvtps2pi mm0,xmm1 ; mm0 -> 2 delta dwords
404	movhlps xmm1,xmm1	437	movhlps xmm1,xmm1
405	cvtps2pi mm1,xmm1	438	cvtps2pi mm1,xmm1
406	movq .dey13,mm0	439	movq .dey13,mm0
407	movq .dby13,mm1	440	movq .dby13,mm1
408	else	441	;----------
-		442	mov ax,.z3
-		443	sub ax,.z1
-		444	cwde
-		445
-		446	movd mm2,[.t_x1]
-		447	movd mm3,[.t_x3]
-		448
-		449	punpcklwd mm2,mm4
-		450	punpcklwd mm3,mm4
-		451	psubd mm3,mm2
-		452
-		453	cvtsi2ss xmm1,eax
-		454	movlhps xmm1,xmm1
-		455	cvtpi2ps xmm1,mm3
-		456
-		457	divps xmm1,xmm3 ; xmm1--> \| xxx \| dz \| dty \| dtx \|
-		458
-		459	shufps xmm1,xmm1,11100001b
-		460	; xmm1--> \| xxx \| dz \| dtx \| dty \|
-		461	cvtps2pi mm0,xmm1 ; mm0 -> 2 delta dwords \| dtx \| dty \|
-		462	movhlps xmm1,xmm1
-		463	cvtss2si eax,xmm1 ; eax --> 1 delta dword \| dz \|
-		464	movq .dty13,mm0
-		465	mov .dz13,eax
-		466	else
409	mov ax,word[.b_x3]	467	mov ax,.z3
-		468	sub ax,.z1
-		469	cwde
-		470	shl eax,CATMULL_SHIFT
-		471	cdq
-		472	idiv ebx
-		473	; mov .dz13,eax
-		474	push eax
-		475
-		476
-		477	mov ax,word[.b_x3]
410	sub ax,word[.b_x1]	478	sub ax,word[.b_x1]
411	cwde	479	cwde
412	shl eax,ROUND	480	shl eax,ROUND
413	cdq	481	cdq
414	idiv ebx	482	idiv ebx

Subversion Repositories Kolibri OS

(root)/programs/demos/3DS/BUMP_TEX.INC @ 1776 – Rev 1245 → 1776