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 : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Road Walker who wrote (169708)	8/21/2002 12:45:58 PM
From: Tenchusatsu	Respond to of 186894

John, I liked this part of the article: But the world is different now, and clock speeds don't matter as much, AMD says. Yep, always a case of sour grapes for AMD. Tenchusatsu

To: Road Walker who wrote (169708)	8/21/2002 12:46:16 PM
From: wanna_bmw	Read Replies (2) \| Respond to of 186894

John, Re: "It's been about a year since we detailed AMD's plans for the repositioning of its Athlon chips against Intel's Pentium 4 line. A key part of that strategy involves constantly debunking the "megahertz myth" which is to a large extent laudable." Eventually, I would expect some efforts by Intel to debunk the "QuantiSpeed" myth. Just look at the benchmarks from Anandtech. anandtech.com In Content Creation Winstone, the Athlon XP 2600+ gets outperformed by a Pentium 4 2.26GHz. The correct rating for the chip in this benchmark should be 2200+, since that's the fastest Pentium 4 chip it's able to get a 3% advantage on. So why bother with the QuantiSpeed at all? The chip already runs at 2.13GHz. Why not call it that, since it is close enough to how it performs. In Xmpeg 4.5/DiVX 5.02, it only manages to outperform the Pentium 4 2.2GHz by 1.5%. Again, 2200+ is more fitting for Content Creation and DiVX encoding. The 2.53GHz Pentium 4 outperforms it by 16.5%, so how can AMD claim a faster speed? In LAME MP3 encoding, they are perhaps justified. The 2600+ does outperform the Pentium 4 by about 5.5%. In 3D Studio MAX and Maya 4.0.1, the story is also different, since the Athlon can outperform the Pentium 4 2.53GHz by 4-21%, depending on the renderer used. Tom's Hardware seems to have a different version of 3D Studio MAX with Intel optimizations, since in his tests, the Pentium 4 leads. Here, it depends on which version you have, so again, the model number is ambiguous. Considering a different renderer, Lightwave 7.5, the Pentium 4 2.53GHz leads 16-62%. The 2.2GHz Pentium 4 even leads by 2.5-42% in this benchmark. AMD's model numbers only tell part of the story. In Unreal Tournament, the two CPUs essentially tie, with megahertz hardly making a difference. QuantiSpeed, therefore, has little meaning here. Jedi Knight 2 and Serious Sam 2 are similar. There is very little difference between the top CPUs. These tests are pretty much video card limited at this point. Comanche 4 is the only one where the Pentium 4 is able to get a significant enough lead, and at 2.4GHz, it can still outperform the Athlon XP 2600+. As for SPEC Viewperf, even though it uses some real applications, the tests seem to be very architecture dependent, with megahertz hardly making any difference at all. The Athlon wins all these, but by how much is different for each test component. No model number could accurately describe the behavior in this test. It also seems to go against all the other results in the review. Anand says: "We're quite skeptical of the viewperf results not only because they don't agree with any of our other numbers from this review but mainly because the 3ds max test does not agree with our real-world 3ds max tests from earlier in this review." So ignoring ViewPerf, Intel leads in 4 benchmarks, AMD leads in 3, and several games were a draw. At best, AMD deserves a rating similar to Intel's higher speeds, but no single number accurately resembles the results. It will either be too aggressive or too conservative. The "ModelHurts Myth" is just as bad as the "Megahertz Myth", because with QuantiSpeed, AMD can claim the world's fastest microprocessor, but clearly it is not true. The Pentium 4 wins the majority of tests, and all AMD has been able to do is catch up. The results speak for themselves, but most reviewers don't seem to notice, unfortunately. Yet, I see the time is approaching when QuantiSpeed will come to increasing pressure from the press. It simply does not accurately portray performance. Right now, AMD is having a tough time competing in a down market, so they have been reduced to deception to even the odds. It will catch up with them sooner or later, I'm guessing. Right now, I wouldn't go near the stock at its current price. I might consider it a better buy at <$7. wbmw