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 : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: fp_scientist who wrote (133094)	4/21/2001 10:40:56 PM
From: Elmer	Read Replies (1) \| Respond to of 186894

Anyhow, I thought the SpecFP score for the P4 was "cooked". I now have to concede that it performs extremely well for my application (even without SSE2). My suspicion is that most of the advantage may be memory bandwidth but I don't know for sure. It will be interesting for me to find out how the P4 performs with DDR for my application when it becomes available. We shall see. Nice to hear you are happy with today's P4 running non optimized code. Maybe you should try Intel's Fortran compiler? EP

To: fp_scientist who wrote (133094)	4/22/2001 6:09:04 AM
From: fyodor_	Read Replies (1) \| Respond to of 186894

fp_scientist: My suspicion is that most of the advantage may be memory bandwidth but I don't know for sure. It will be interesting for me to find out how the P4 performs with DDR for my application when it becomes available. I have to agree with Elmer that you should try and get your hands on an Intel compiler. My own experiences with them aren't too hot, but if you can get some SSE2 usage in there, you might see a nice performance increase. Of course, it all depends ;-). Your code sounds like it might well be bandwidth limited, in which case you may not see much of an increase from SSE2. Supposedly, the use of SSE2 added only around 5% to the P4s SPECfp scores (but then SPECfp is _very_ bandwidth sensitive). Have you by any chance tried your stuff on a SDR Athlon? If so, how did that compare to the DDR Athlon? -fyo

To: fp_scientist who wrote (133094)	4/22/2001 8:00:06 AM
From: pgerassi	Read Replies (1) \| Respond to of 186894

Dear Fp_scientist: Of course on really large datasets, 1GB and up, P4 can not get enough RDRAM to run it. The i850 only has 4 RIMMs and the largest I have seen is 256MB. Thus anything requiring at least 1GB, page swapping starts and P4 will run like a dog. Xeon3, P3, and Athlons can get 3-4GB PC133 SDRAM systems. The problem being that P3 and Xeon run slower than Athlon on these datasets both at same clock and in overall performance. Pete