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 (19219)	11/15/2000 2:31:57 PM
From: jcholewa	Read Replies (1) \| Respond to of 275872

> I do not know how SSE2 is being used in game optimization. Kap seems to have answered how the SpecFP > optimizations are being done. As you see it has very little to do with prefetch and a lot to do with SSE2 optimizations. > If Kap is right and there is only a hardware prefetch, I wouldn't be surprised if the prefetch gains amount to much. > (Hardware prefetch from experience can even lead to performance decrease.) Kap is right about SSE2 being able to strogly affect specfp. I just wasn't aware of whether or not optimizations for it were really put in. His point it valid. Regarding prefetch ... I know that hardware prefetch is in Willamette, but I would be very surprised if Willamette did not support the full SSE instruction set, which includes by definition prefetch opcodes. Smart use of prefetch can do a much better job than hardware prefetch based on stride prediction in some cases. Okay, okay, fine. I will hold much of my specfp judgement until we can get more clarification on the issue. :)       > <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.> > I will try to see if I can get any meaningful details. Thank you. I apologize if it seems as if I'm busting your balls. :)     -JC

To: Charles R who wrote (19219)	11/15/2000 3:15:27 PM
From: fyodor_	Respond to of 275872

<Chuck: I do not know how SSE2 is being used in game optimization. Kap seems to have answered how the SpecFP optimizations are being done.> SPECfp uses double precision floating point (as I've just been informed ;)), so using SSE2 will boost the performance considerably. SSE2 does not include any instructions that are of particular benefit in games. Like JC said, games don't make use of double precision floating point and the integer SIMD (in the P4 implementation) is rather weak. -fyo