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: Dan3 who wrote (14776)	10/18/2000 8:55:55 AM
From: minnow68	Respond to of 275872

Dan, You wrote "If P4 IPC really is as poor as early benchmarks have indicated, then Intel up the proverbial creek, with no paddle in sight." Current benchmarks show P4 to have about 27% lower IPC than an Athlon. I believe Mustang with DDR and 133 Mhz FSB will have an IPC at least 12% higher than the current Athlon. With all three at 1 Ghz, this gives us, P4=730Mhz effective, Mustang=1.12Ghz effective. This means the average Mustang IPC could be 53% higher than P4. Therefore, with linear scaling, a 2.5 Ghz P4 would only benchmark out the same as a 1.63 Ghz Mustang. But it gets even worse. As clock speed increases, scaling becomes worse with every increase in speed. Only on-die cache can keep it going. Mustang will have a considerable advantage there also. If the P4 benchmarks we have seen are accurate, then Intel would need to scale P4 well beyond 3 Ghz at 0.13 micron to be performance competitive with AMD. The bottom line is that you are quite correct. If P4 does not perform significantly better than the benchmarks we have seen, Intel will not come close to Mustang's performance. Mike

To: Dan3 who wrote (14776)	10/18/2000 4:20:36 PM
From: porn_start878	Read Replies (2) \| Respond to of 275872

P4 will be in volumes way before 2002, Intel will cut PIII's production to a celeron level by the end of Q3. They invested a lot on capacity, maybe it was to offset P4's die. I don't beleive anymore that they'll wait for .13mu to ramp the P4. If P4 IPC really is as poor as early benchmarks have indicated, then Intel up the proverbial creek, with no paddle in sight. I think everyone has just assumed that the final numbers will be much better. If they aren't, Intel won't be competitive for years. Willy will yield pretty well at 1.8-1.9GHz, if they need about 20-30% higher clockspeed to be on par to better than Mustang, which will yield at 1.33-1.66 max, they will do pretty fine. Where I could be wrong is the clockspeed delta needed... Mustang+DDR could improve Athlon's IPC by 20-25% so it's now 40-50% higher clockspeed that will be needed to be on par to better. I can't wait for both P4 and Mustang 266 benches... Max