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: fyodor_ who wrote (20292)	11/23/2000 1:32:47 AM
From: Jim McMannis	Read Replies (2) \| Respond to of 275872

Fyodor, You make some good points about FPU, 3DNow, SSE... OTOH, The P4 seems an odd chip. It does a few things really well and a lot of things really bad. Certainly wasn't designed as an all 'round chip... It scales well, runs bandwidth extensive games like quake3 well (the speed comes from the bandwith), it's pretty good at MPEG4/MP3 encoding, about equal to a 1.2 Athlon (1.5 Ghz P4 comparison). On a lot of other benchmarks it loses miserably. It also seems to rely heavily on SSE2 optimization, not unlike the K6-2 relied on 3DNOW to achieve good gaming scores where as the P-II was the more well rounded chip with the stronger FPU... ---- What puzzles me, the Mhz not withstanding, is why Intel would design such a chip in the first place... Perhaps they saw it as a means to incorporate the Rambus bandwidth into a chip. I'm perplexed... Jim

To: fyodor_ who wrote (20292)	11/23/2000 9:40:42 AM
From: that_crazy_doug	Read Replies (2) \| Respond to of 275872

<< Sure, the benchmarks migrated in that direction, but I believe that has more to do with a real change in the programs users run. 3D games, MP3 decoding and now DVD / DivX / MPEG encoding. >> So now the question is what is the p4 good at, the answer is memory bandwidth and double precision floating point (with the correct optimizations). Are these 2 things going to overshadow everything else moving forward? (similar to how fp overshadowed int during the p5 -> p6 transition) The memory bandwidth issue will be less of a factor because AMD could always come up with a similar platform. I'm not sure how big the double precision floating point will be, but we'll find out.