| Tuesday, March 12, 2002 |
09:50 - Benchmarking Fallout
http://www.theregister.co.uk/content/35/24378.html
|
(top)  |
Damn, The Register is hard to get to these days. I don't know if they're being constantly hammered by linkers-in from their coverage of the now-infamous C't benchmarking brouhaha, but whatever it is, they've certainly got the world keeping an eye on them. As some of the loudest online backers of Linux and Mac OS X and protesters against the Microsoft hegemony, they've got a torch to bear now, whether they like it or not.
They've just posted a page full of responses from users about their Mac benchmarking results article from yesterday. The first one in particular, by researcher Justin G. Cordesman, is particularly worth reading.
As for SSE2 vs. Altivec, SSE2 is a toy by comparison. Its architecture does not offer the range of generalized high precision capability that the altivec instruction set does. It is filled with bandwidth limitations, particularly its tiny number of harder to use registers that make it nearly impossible to keep the pipeline full, and it is capable of basically no parallelism whatsoever with the regular FP unit on the processor (which means it must start and stop each unit to switch back and forth, and the lack of generalization makes this an excruciating performance penalty). The small number of registers in particular makes the P3 a better scientific computing processor than the P4 for real world applications because the P4's pipe is too deep to keep it filled. This can be graphically demonstrated with fully optimized applications that force significant branching on real world data.
The rest of the responses on the page are less in-depth but also useful-- except the last one, which so beautifully exemplifies the pouty attitude of so many zealots in any community-- "You don't agree with me? You won't do everything I say? Fine! I'll take my ball and go home!" Promptly upon which the ball spontaneously deflates.
|
|
