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 : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: dougSF30 who wrote (253045)	6/8/2008 7:21:58 PM
From: mas_	Read Replies (1) \| Respond to of 275872

Certain benchmarks disable prefetching from main memory due to the bandwidth constraints of the old FSB connection. If you do aggressive speculative prefetching without enough bandwidth, and the prefetchers are pulling enough false positives in, it costs you performance, because the unneeded data is plugging the narrow pipe to main memory, slowing the real data flow. That is what is happening but bandwidth is not the problem and cure, the latency of cache pollution caused by very aggressive prefetching and easing of is. You are the one who is confused.

To: dougSF30 who wrote (253045)	6/8/2008 7:52:06 PM
From: mas_	Read Replies (1) \| Respond to of 275872

because the unneeded data is plugging the narrow pipe to main memory, slowing the real data flow. Stop making stuff up. rethink.intel.com While most workloads experience higher performance because of the hardware prefetcher, some workloads can cause the prefetcher to guess incorrectly and populate the processor’s caches with unneeded data. For some workloads and data structures, hardware prefetching may actually lower performance. Cache pollution is the problem in Intel's own words.