Advanced Micro Devices - Moderated (AMD) Message Board

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)

AMD 213.43

+6.2%

Dec 19 9:30 AM EST

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next

To: Petz who wrote (77264)	4/16/2002 4:34:41 PM
From: wanna_bmw	Read Replies (1) of 275872

Petz, Re: "I find Madison's floating point performance underwhelming. That's a theoretical peak of 5.9 gigaflops per CPU. An Athlon XP2100+ at 1.73 GHz has a theoretical peak of 3.46 Gflops (64 bit floating point)." Two things. First, there is a big difference between "theoretical" FLOPs and actual FLOPs. Show me a test that can show how many sustainable megaflops that an Athlon is capable of achieving. Second, there should be something to be said about scalability. You simply divided the number of CPUs by the total number of teraflops that the system was capable of producing. However, you should know that performance doesn't scale linearly with the number of CPUs, especially when the number is approaching more than 1000. There are certainly very few Athlon supercomputers out there, and those that exist are usually clustered dual processor systems, which is much more inefficient for scaling. If you have a comparable 1400 processor Athlon system that you can show produces the same floating point throughput, then we may have a discussion; otherwise, I think you are writing off the McKinley/Madison floating point performance prematurely. wbmw

Report TOU Violation

Share This Post

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next