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 (20509)	11/25/2000 5:05:31 PM
From: niceguy767	Respond to of 275872

fyodor: "And the P4 will scale a heck of a lot better from 1.5GHz than the P3 will from 1GHz, even you know that ;)" I don't know that...What I do know is that PWeeiii has hit a wall at 1 gig...and P4 has demonstrated a huge bellyflop at 1.4...Beyond those facts, I cannot project...but as indicators, I'm not anticipating a great deal in store for either architecture down the road...I would hope for INTC's sake that the P4 at 1.5 scales better than the P3 at 1.0, as it looks like the P4 needs to get to about 2 gigs before it'll compete on a performance basis with the 1.2 gig Athy/DDR combo...

To: fyodor_ who wrote (20509)	11/25/2000 5:50:13 PM
From: Gopher Broke	Read Replies (1) \| Respond to of 275872

And the P4 will scale a heck of a lot better from 1.5GHz than the P3 will from 1GHz, even you know that Given the P3 has not managed to scale one iota above 1GHz, one would hope the P4 does a lot better. The processor may scale OK but it will be interesting to see how Intel can scale the heatsink. Fins protruding outside the case with the next version perhaps? Or will it be back to the blowhole-in-the case approach? With all the hints Intel was dropping at the last CC I think we may find P4 does not scale too well until 3Q 2001 and .13 process arrives. And there is plenty of opportunity for another cockup there. A few bouts of copper contamination would not be amusing for Intel.

To: fyodor_ who wrote (20509)	11/25/2000 6:32:09 PM
From: hmaly	Read Replies (1) \| Respond to of 275872

fyodor Re..But not even you are blind enough to think that the P3 is a better PROCESSOR than the P4. If you are talking about a 6 yr. old design against a brand spanking new P4, yes the P4 is a better processor. But, the PII and then PIII were monsters when they were introduced. P4 doesn't inspire anyone except the adherents to someday one yr from now. Besides AMD is on a roll. Intel needed a chip with some wow to stop the momentum now; not a yr. from now. Re Absolute performance matters, sure, but so does scalability etc. etc. And the P4 will scale a heck of a lot better from 1.5GHz than the P3 will from 1GHz, even you know that ;) Scalability without performance will be useless. Balance also counts. Just ask Scumbria on that. Add to that ,Intel has been boasting about 3ghz double pumped cpu with a trace cache,20 stage pipeline etc. but in the end the chip doesn't perform any better than the PIII even with another 400 mhz, . What happened to all of those wiz bang features anyway. Add to that the new features of the P4 that, when supported, can increase the performance significantly. The problem with that statement is that AMD will also have compatibility with some portions of SSE2, by then; so the applications will also help AMD. If Dirk was right at the Nov. CC; and the Hammer performance is 3X T-bird, P4 would have to scale up to 4.5 ghz just to beat it on performance. Not impossible but unlikely by 2002.

To: fyodor_ who wrote (20509)	11/25/2000 10:39:47 PM
From: combjelly	Read Replies (1) \| Respond to of 275872

"And the P4 will scale a heck of a lot better from 1.5GHz than the P3 will from 1GHz, " While I will agree with this, that is only because it seems like 1GHz is all that the PIII can do in 0.18 micron. The amount that the P4 can scale beyond 1.5 or 1.6GHz is in some doubt in my mind. Consider the thermal requirements of keeping the Tj at 85 degrees or less. Now currently the P4 is consuming somewhere in the range of 50 watts on the average, but the peak seems to be somewhere in the 70+ watt range. Now there is not enough information to guess what the thermal load at 2GHz would be, but 70 or so average and 100+ watts peak would not be unexpected. Based on this, 1.7GHz may be it until 0.13 micron and/or Northwood. And that is not particularly impressive.