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: combjelly who wrote (27989)	2/10/2001 3:58:32 PM
From: Paul Engel	Read Replies (1) \| Respond to of 275872

Re: C"While the P4 probably has significantly more process refinement ahead of it, it doesn't look like it has all that much more headroom than the Tbirds. " Message 15325634 {==================================} C't interview with P4 chief designer David Sagar See the report at heise.de (in German) David Sager confirmed that the P4 2 GHz will indeed be introduced this fall. He said that the architecture of the P4 will have almost infinite scalability. "The P4 will reach its performance limit only when Intel decides to give the processor a new name." He said that the P4 is the beginning of a new processor family. To further increase the performance there will be some minor and bigger design modifications, like bigger caches. The general architecture, however, will stay the same. "In summer, we will make the transition to 0.13µm." Mr. Sagar didn't say very much about SOI. However, since there was a lecture by Intel at the ISSCC about the performance gains of SOI on 0.18µm and 0.13µm process c't believes that SOI will come soon, too. Hmmm, together with the recent reports about a massive ramp of RAMBUS modules it seems that H2 will not be easy. Peter {====================================}

To: combjelly who wrote (27989)	2/10/2001 4:45:32 PM
From: fyodor_	Read Replies (2) \| Respond to of 275872

combjelly: An important point here. Currently both the Tbird and the Athlon can be over-clocked to about the same speed using liquid nitrogen. I'm assuming you mean both the TBird and the P4... If so: These LN2 overclocking comparisons are dangerous. The transistors can switch insanely fast at those temperatures and [some types of] noise issues are reduced significantly. This has a variety of implications that make comparisons almost completely invalid: - All heat issues are essentially nullified. Thus it doesn't matter if one core dissipates more heat than the other. With more conventional cooling, this is most certainly not the case, since transistors switch slower at higher temperatures. - With the transistors switching extremely fast, completely new limiting speed-paths can easily be brought to light. These may be very simple to solve or never arise, even with a finer process (i.e. a shrink). - Noise characteristics are changed. Needless to say, thermal noise is much lower at -196°C. Who knows what effect this might have? (I'm sure someone does, I just don't). I'm quite certain, however, that this effect is not accurately modeled by using a finer process. I'm sure there are plenty more, but suffice it to say that I don't really see what purpose is served by a comparison of the overclocking capabilities of the Athlon and the P4 at -196°C. (Not to mention the fact that the number of samples is way too small to be statistically significant.) -fyo