Assembly code snippets

Details

Title	Fast Alpha Blend (MMX)
Author	bitRAKE
Submitted by:	bitRAKE
Date added:	2002-02-17 21:46:06
Date modified:	2002-02-17 21:46:06

Comments

Under 7 cycles per pixel - too bad memory is so slow. For better speed: process your graphics in small bands, or integrate your other pixel processing with this code. That way the data will still be in the cache.

Snippet

AlphaBlend: ; 13056 ticks for 1000 quads on Athlon.
    mov eax,Source
    mov edx,Destination
    mov ecx,Count ; number of pixels/2
    dec ecx
    pxor mm7,mm7
    movq mm0,[eax+ecx*8]
    movq mm1,[edx+ecx*8]
@@: movq mm4,mm0
    movq mm2,mm0
    psrlw mm4,1
    movq mm3,mm1
    movq mm5,mm4
    punpcklbw mm0,mm7
    punpcklbw mm1,mm7
    punpckhbw mm2,mm7
    punpckhbw mm3,mm7
    psubsw mm0,mm1
    psubsw mm2,mm3
    punpcklwd mm4,mm4
    punpckhwd mm5,mm5
    punpckhdq mm4,mm4
    punpckhdq mm5,mm5
    psllw mm0,1
    psllw mm2,1
    pmulhw mm0,mm4
    pmulhw mm2,mm5
    paddsw mm0,mm1
    paddsw mm2,mm3
    packuswb mm0,mm2
    movq [edx+ecx*8],mm0
    dec ecx
    movq mm0,[eax+ecx*8]
    movq mm1,[edx+ecx*8]
    jns @B
    ret 12