Hi Timothy, On Monday 04 April 2005 15:34, Timothy Miller wrote: > In the case where we have run out of multipliers and need more, we > can either code our own (as you have done), or tell the synthesis > tool to infer one in random logic with however many stages of > pipeline that we want.
Nice! (The automagic part.) I was lucky enough to run into a good contact at Xilinx and now have a good opportunity to do some research on spartan 3 parts, particularly availability, power requirements and cost. The 3S2000 seems to be the smallest part that can fit an OGL 1.3 fixed function pipeline at decent speed. The 3S4000 on the other hand would allow more scope for optimization and enough headroom for some next-generation development. So it would be really nice to be able to use the 3S4000. But if the board ends up twice as expensive and needs active cooling, then the 3S200 would have to do. Regards, Daniel _______________________________________________ Open-graphics mailing list [email protected] http://lists.duskglow.com/mailman/listinfo/open-graphics List service provided by Duskglow Consulting, LLC (www.duskglow.com)
