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: pgerassi who wrote (195104)4/26/2006 12:30:12 AM
From: eracerRead Replies (1) | Respond to of 275872
 
Re: ...Conroe will underperform A64 X2 at same clock when hit with these average users...

...The primary bottleneck for these here will be the latency to memory and to a lesser extent, the BW. Here Woodcrest will fall down wrt Opteron. And the amount of the drop wrt Opteron will get worse as the socket count goes up. And I think that this will be far more typical of the workloads that server purchasers will use than yours.


Then AMD should definitely make Athlon 64 FX and Opterons with no L2 cache at all. If power desktop users and servers are cruching so much code that 4MB is practically useless then just imagine how useless 1MBx2 L2 is on Athlon 64 FX and Opteron. I say give the Semprons 1MB L2 cache for single core single-threaded use and give the 128KB L2 cache to the Opterons and Athlon 64 FXs.



To: pgerassi who wrote (195104)4/26/2006 7:43:17 AM
From: fastpathguruRead Replies (1) | Respond to of 275872
 
Not to mention that adding cache ain't exactly rocket science for AMD either...

fpg