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 (74787)	3/17/2002 7:22:00 PM
From: Elmer	Respond to of 275872

All in all, I'd have to say things look ugly for AMD - unless Tbred brings serious enhancements to the table. Intel will have higher bandwidth and lower latency with dual DDR chipsets later this year. Both are going to significantly undermine AMD's QuantiSpeed rating. If AMD wants to keep it a "True Performance Initiative", they'd better either have something serious to offer, or adjust the QuantiSpeed formula for new speeds appropriately. Thanks for the analysis. You're going to upset a lot of people here who are sure AMD is taking off for the Mooooon. The NorthWood advantage will only increase with the introduction of the 533MHz FSB and with AMD gaining little from .13u (even if they can eventually correct their yield problems), it looks bleak. I can't help but laugh at those who keep claiming the P4 is a bad design. It isn't even starting to hit it's stride. EP

To: fyodor_ who wrote (74787)	3/17/2002 8:42:43 PM
From: Dan3	Read Replies (2) \| Respond to of 275872

Re: Athlon vs P4 - How Do They Scale? How could you work so hard, and then so profoundly misunderstand the data you'd collected? I posted why Athlon's cache would let it scale better, you basically proved my point, then interpreted AMD's quantispeed adjusted numbers as representing performance changes due to cache performance. AMD will have to change the rate at which it adjusts its model numbers in future chips - no doubt about it. But IPC for Athlon falls off more slowly as mhz increase - you've just proved in practice what I predicted in theory. Look at the measured Delta P over Delta f in your listings. You might also further note that my analysis predicts P4 hitting a wall as its prefetching of 128 byte cachelines saturates memory bandwidth, rather than expecting a more rapid reduction in performance scaling (compared to Athlon) as clock speeds increase. But you've just shown that happening too. If your point is that AMD will have to provide more mhz per quantispeed increase, then I agree 100%, but that's rather obvious and expected: A model speed increase 50% higher than the frequency increase won't be sustained. A 1500+ is 11% faster than its 1333 clock, while a 2100+ is 17% faster than its 1733 clock - I'm amazed that you failed to understand the impact of that variable, and used it to compare Delta P / Delta f. AMD will almost certainly modify their ratings for Thoroughbred - in fact, an upcoming change in the model numbers for Thoroughbred may be why AMD is being so tight lipped about Thoroughbred. Furthermore, I think that the numbers you've posted make it clear that Athlons's advanced cache design will result in Athlon's architecture providing higher performance on any given memory technology, which is very important if we assume that both P4 and Athlon will keep scaling in clockspeed to the point that memory becomes a bottleneck. Note that you've been seeing a 400mhz fsb connected to dual memory channels trying to compete with a 266mhz fsb connected to a single memory channel - and, even burdened with that disadvantage in memory bandwidth, the 2100+ Athlon's advanced cache lets it score higher than the 2200 P4 in most tests. Thanks for your hard work, and you've certainly proven that AMD has to adjust the rate at which it is scaling model numbers when it ships Thoroughbred. Regards, Dan

To: fyodor_ who wrote (74787)	3/18/2002 11:24:51 AM
From: milo_morai	Read Replies (1) \| Respond to of 275872

You should be proud now. Tony Viola aka Whitecliff writes ragingbull.lycos.com He seems more worried about AMD then ever. How much time did it take again? :o)

To: fyodor_ who wrote (74787)	3/18/2002 12:54:12 PM
From: AK2004	Read Replies (1) \| Respond to of 275872

Joe <edited> could you you run me through your numbers because I, frankly, do not see how you got to your conclusion. we can equally compare dP/df to dP/df or dP/dQf to dP/df Let's be conservative (against AMD) and use the 2nd. sandra 2002 MM yields no difference considering how much of a difference especially considering what a bios upgrade made ( dP/df to dP/df in amd's favor ) SiSoft Sandra 2002 Pro Multimedia Bench FLOAT AthlonXP Qf f(GHz) score DP/Df dP/df dP/dQf 2100+ 1.733 11123 6.42E+03 6.42E+03 4.30E+03 2000+ 1.666 10693 6.42E+03 6.47E+03 4.27E+03 1900+ 1.600 10266 6.42E+03 6.37E+03 4.27E+03 1800+ 1.533 9839 6.42E+03 6.34E+03 4.25E+03 1700+ 1.466 9414 6.42E+03 6.50E+03 4.29E+03 1600+ 1.400 8985 6.42E+03 6.43E+03 4.31E+03 1500+ 1.333 8554 6.42E+03 n/a n/a Northwood f(GHz) score DP/Df dP/df 2.200 10795 4.91E+03 4.89E+03 2.000 9818 4.91E+03 4.93E+03 1.800 8833 4.91E+03 4.87E+03 1.600 7859 4.91E+03 n/a Let's look at p4 to p4; NW is definitely faster but the scaling derivative is not a nicely defined function to begin with just look at: Willamette f(GHz) fps DP/Df dP/df 2.000 52.4 26.20 17.00 1.900 50.7 26.68 13.00 1.800 49.4 27.44 22.00 1.700 47.2 27.76 18.00 1.600 45.4 28.38 20.00 1.500 43.4 28.93 16.00 1.400 41.8 29.86 n/a Having said that, I say that my knowledge of processors is close to zero but from stand point of pure math I do not see how you arrived to your conclusion - I am sure I am missing something Regards -Albert ps sandra overall yields even less conclusive results Sysmark 2002 (overall score) AthlonXP Qf f(GHz) score DP/Df dP/df dP/dQf 2100+ 1.733 188 108.5 194.0 130.0 2000+ 1.666 175 105.0 15.2 10.0 1900+ 1.600 174 108.8 14.9 10.0 1800+ 1.533 173 112.9 89.6 60.0 1700+ 1.466 167 113.9 30.3 20.0 1600+ 1.400 165 117.9 59.7 40.0 1500+ 1.333 161 120.8 n/a n/a Northwood f(GHz) score DP/Df dP/df 2.200 216 98.2 55.0 2.000 205 102.5 105.0 1.800 184 102.2 45.0 1.600 175 109.4 Willamette f(GHz) score DP/Df dP/df 2.000 183 91.5 60.0 1.900 177 93.2 70.0 1.800 170 94.4 50.0 1.700 165 97.1 50.0 1.600 160 100.0 70.0 1.500 153 102.0 60.0 1.400 147 105.0 n/a <edited> what you showed does create an impression that NW scaling is better but from what I see it is far less than conclusive ( willing to be convinced otherwise )