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 : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Paul Engel who wrote (80639)	5/9/1999 9:52:00 AM
From: Scumbria	Read Replies (3) \| Respond to of 186894

Paul, Remember - this is a FULL CPU speed L2 cache - so for many applications where a 256K L2 cache is adequate, the faster cache access times - AND SCUMBRIAN LOW LATENCY - will show better performance than standard Pentium ///s with 512K 1/2 CPU speed L2 cache. After reading your comment I initially agreed with your statement, so I decided to try some calculations. I was surprised to find that the architectural performance difference between PIII and Coppermine is pretty close to zero. This is due to the smaller L2 on Coppermine. I used these values in my calculation: Coppermine: 96% L1 hit rate, 85% L2 Hit Rate, 0.50 Dram Page Hit Rate, 3 cycle L2 Miss penalty PIII: 96% L1 hit rate, 90% L2 Hit Rate, 0.40 Dram Page Hit Rate, 6 cycle L2 Miss penalty I assumed that Coppermine L2 latency is 3 clocks vs. 6 clocks on PIII, and that the DRAM page hit rate is better on Coppermine because of the better address locality coming out of the smaller L2 cache. The calculations are very sensitive to L2 hit rate, but I believe that the 90% PIII L2 Hit rate vs. the 85% Coppermine L2 Hit Rate is a reasonable approximation. Scumbria