Frank A. Stevenson wrote: > I like this idea (it would be great to make a practical implementation) > ... without giving to much thought about it, it seems that the shifting > can be avoided by unrolling the loop, and using the 64 registers as a > ring-buffer. I obviously did not think to hard about this :-( The amount of shifting is a problem, especially with the way ATI handles branches. (shifting only 2 around takes the same amount of time as shifting all 3) - I will need to think of something cleverer. :-)
Frank _______________________________________________ A51 mailing list [email protected] http://lists.lists.reflextor.com/cgi-bin/mailman/listinfo/a51
