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.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: Ali Chen who wrote (39279)	10/14/1998 1:20:00 AM
From: Joe NYC	Respond to of 1573685

Ali, You may want to compare Xeon-512-low latency with P-II-512-high latency, with the same results. Both PII and Xeon have a high latency L2 (Xeon slightly lower), while Mendocino has low latency L2. Some info on latencies from Cyrix presentation at WinHEC ftp://www.national.com/cyrix/developr/whpres98.pdf Joe

To: Ali Chen who wrote (39279)	10/14/1998 4:00:00 AM
From: Tenchusatsu	Read Replies (1) \| Respond to of 1573685

Ali, <Your conclusion is not quite correct. You may want to compare Xeon-512-low latency with P-II-512-high latency, with the same results. > Xeon's full-speed off-chip cache is still higher in latency than Mendocino's on-chip cache. Look, if you're saying that latency doesn't matter, then why would Mendocino be running on par with a Pentium II with four times the L2 cache? If you think that no one needs more than 128K of L2 cache, then why is Sharptooth going with 256K? If 128K is enough, then why would K7 need an L2 cache given that its L1 is going to be 128K? And I'd sure like to know how your dual independent bus argument relates to the L2 cache latency problem. Tenchusatsu