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: Scumbria who wrote (46755)	1/20/1999 4:36:00 PM
From: Kevin K. Spurway	Respond to of 1572369

Re: "Are we forgetting the 128K L1 on K7? The L2 latency becomes decreasingly important as the L1 miss rate decreases." I don't think Paul is forgetting it. I think he never knew it in the first place! Kevin

To: Scumbria who wrote (46755)	1/20/1999 4:48:00 PM
From: Paul Engel	Read Replies (3) \| Respond to of 1572369

Scumby - Re: "Are we forgetting the 128K L1 on K7? " No, WE ARE NOT FORGETTING IT. Remember - this is a SPLIT cache - 64K Data and 64K Instruction. For most applications, this will require going OFF CHIP to the L2 cache quite frequently - for EITHER DATA OR INSTRUCTIONS OR BOTH. The performance penalty for going off chip to a 166 or 200 MHz L2 cache - including all the latencies - will be detrimental to the K7's performance. The K6-3, on the other hand, will have 262 K of COMBINED L1 and L2 cache - all of it ON CHIP running at full CPU speed. Paul Paul