Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: Charles R who wrote (19210)	11/15/2000 2:04:46 PM
From: jcholewa	Read Replies (2) \| Respond to of 275872

> > <If you optimized 100% of the double precision floating point code in current 3D games, > > you would get nearly a 0% boost in performance. > > Did I say anything about the optimizations being only > about double precision FP? You said SSE2 optimizations. That means either double precision vector floating point (which doesn't usually apply to the games) or vector integer, which is unimpressive in the P4 (it's equal to regular MMX, but only if you make your code more explicitly parallel). So, what do you think exists in SSE2 and not in SSE that can help 3D games? > <This isn't just my opinion. > > Yes, it is. Of course it isn't. It is that, plus additional information which followed that statement. > <It's more of a logical deduction derived from the fact that there is > no such code in 3D games (except for really rare circumstances).> > I am no authority on software code for 3D games. If you think you understand > software architecture for games well and feel strongly about it, that's cool. I don't feel strongly about it at all. I merely cannot think of anything that SSE2 offers that can help 3D games. > I have heard from folks who are pretty good in the graphics area that SSE2 allows for some neat optimizations. > Time will tell. Please ask them for details. They may be speaking about CAD-type operations, which could be immensely assisted by SSE2's 4-way double precision vector floating-point capability.     -JC PS: Tag, you're it! :)