12 Jun
2006
12 Jun
'06
6:16 a.m.
Hi, 2006/6/12, Michael Niedermayer <michaelni@gmx.at>:
Hi
no objections ... btw, if you interrested, then that code can be optiized further ... like replacing srca[x] || srca[x+1] by *(unit16_t*)&srca[x] or changing the for loops to asm loops
Seems that all functions have already been vectorized except vo_draw_alpha_uyvy. How frequently is this piece of code used? Is an SSE2 version necessary here? -- Zuxy Beauty is truth, While truth is beauty. PGP KeyID: E8555ED6