SI
SI
discoversearch

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: Petz who wrote (197120)5/16/2006 4:12:20 PM
From: eracerRespond to of 275872
 
Re: For a large class of algorithms, including almost all encryption, decryption, audio processing and some video processing, the bottleneck is memory bandwidth. And as you know, Conroe's 1333 FSB can't even handle the memory bandwidth of two DDR2-800 channels, let alone a couple PCIe channels, gigabit ethernet and future coprocessor...

I believe you are greatly overestimating the impact of memory bandwidth here. The Core Duo already performs well in audio and video processing benchmarks even with a 667MHz FSB. While the lack of memory bandwidth will negatively impact performance in these type of benchmarks architecture still has a strong influence on performance.



To: Petz who wrote (197120)5/16/2006 4:17:05 PM
From: dougSF30Read Replies (1) | Respond to of 275872
 
Have you looked at NGMA scores on encoding/decoding benchmarks? I guess memory bandwidth is NOT the bottleneck...