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: wanna_bmw who wrote (52888)	8/29/2001 12:21:43 AM
From: Dan3	Read Replies (2) \| Respond to of 275872

Wanna_bmw - do you know anything at all about how a cache works? You're message indicates that you don't. Consider for a moment, how would a CPU go about checking for a particular address being in cache in 2 clocks? That's the beginning of your journey along the path to understanding the wayness of cache, and why it matters. My earlier comment is absolutely correct: the cache in a $100 Athlon is effectively 325% larger than the cache in a $4,000 Itanium for a lot of code. Too bad Intel can't figure out how to make either victim or high set associative caches. Ultimately, unless they can correct this, it may end up losing them a lot of their server business. Perhaps I should clarify that I am comparing Itanium's main (L3) cache with Athlon's main (L2) cache. I think that's appropriate, too. Itanium's L1 and L2 combined (160K) are barely larger than Athlon's L1 cache (128K), so comparing Athlons's L2 with Itanium's L3 is reasonable. Itanium's 96K L2 is 6-way, while its 2 meg or 4 meg L3 is 4-way. Athlon's 256K L2 is 16-way.