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: Dan3 who wrote (142765)	9/5/2001 10:20:05 AM
From: wanna_bmw	Read Replies (1) \| Respond to of 186894

Dan, Re: Your example of Truespace. Goodness, I'm sure it took you a while, but congratulations on finding an example where a 1GHz Pentium III beats a 1.5GHz Pentium 4. However, there are quite a few problems with your example. Come on, you knew I would present a counter-argument. Don't sigh on me. ;-) The big one is that the version of Truespace used in the test was version 4.2, while Truespace 5.1 has been available for some time. Now, I'm all for admitting it when you prove to me that the Pentium 4 has a weakness in popular "old" code, but you'll also have to prove that the version it was tested on would still be the one that most people would be using. However, you seem to think that a rendering studio or business would spend the money to upgrade one or several of their workstations to Pentium 4 based ones, but they wouldn't spend the $250 to upgrade to the latest version of their software. Of course, the $250 isn't just to get adequate speed, but also to get all the new features that a new release normally has, such as bugfixes, new rendering styles or plugins, etc. Most definitely, rendering studios keep up with the times, and are always ready to have the latest version of the software they use for their work. caligari.com Second, different testing situations produce different results. Look at Anand's results from testing the same application, and you'll see the same 1.5GHz Pentium 4 coming very close to beating the 1.4GHz Athlon. How can that be that the same processor was falling behind the 1GHz Pentium III? Different testing conditions. There are many, and it's nearly impossible to test all of them. Very probably, there are many more conditions where the Pentium 4 does very well with respect to the Athlon. anandtech.com Third, your blanket statement before was that the 1.8GHz Pentium 4 under "older" code would perform equivalently like a 900MHz Pentium III or Athlon. So far, you have shown one testing condition under one application that already has a newer version available, in which a 1.5GHz Pentium 4 scores below a 1GHz Pentium III, but above an 800MHz Pentium III. Congratulations, but you hardly are able to make your point. What I had assumed you would use for proof was a benchmark called PovRay. I have also done some research in the area, and found that the Pentium 4 used to do particularly horrible in this enthusiast rendering application. I have a link with "old" PovRay binaries, and a recompiled version here. tech-report.com As you can see, with the original binaries, there is a test where the 2.0GHz Pentium 4 can barely outperform that 1.2GHz Pentium III (Tualatin based). Surely, under the same circumstances, the slower Pentium 4 chips would be in the range of the 900MHz Pentium III. Normally, this would prove your point (if by only one example), if it weren't for the recompiled binaries in the same test. This test shows that with a simple recompile without any change in the code, a Pentium 4 2.0GHz can go from scoring almost 20% below the 1.4GHz Athlon and equal to the 1.2GHz Pentium III, all the way to scoring within a percentage point of the 1.4GHz Athlon, and 32% ahead of the 1.2GHz Pentium III. Nevermind what changes in the code could produce, if ever a newer version is released. That just goes to show that the few "old" binaries that used to show the Pentium 4 in bad light have already been updated by recompiling the code. Other legacy benchmarks show the Pentium 4 able to keep up with the fastest Athlon, provided you allow for the frequency advantage. The "newer" benchmarks that you don't like because the Athlon loses, show a Pentium 4 able to outperform the Athlon, with only a nominal frequency advantage necessary. Benchmarks like Quake III, WebMark 2001, and SysMark 2001 all show a 1.5GHz Pentium 4 able to keep up with or exceed the 1.4GHz Athlon. You'll discount these, but you will count those "older" benchmarks that already have newer versions available that people use, and that perform much better on the Pentium 4. That's just another one of your double standards that is all too common on this board. I would just love you to admit that you were wrong about a 1.8GHz Pentium 4 getting close to the performance of a 900MHz Pentium III in "old" software, but I'm sure you'll find a way of arguing your way out of it. wanna_bmw