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: fyo who wrote (67058)	7/30/1999 4:29:00 PM
From: Dan3	Respond to of 1574422

This is where the rambus latency could become an issue. At 1GHZ, a ns is a processor clock cycle. If coppermine and K7 are both shipping at 1GHZ, with similar per clock capability, and both are getting a cache hit rate of 99%, then wouldn't a 10 to 20 ns loss per cache miss due to Rambus translates into 5 to 10% poorer performance on typical benchmarks? If DDRDRAM reverses DRDRAMS streaming advantage, or PC166 is used (already shipping from many of the companies that will begin shipping DRDRAM next quarter) the delta could be a little bigger. So if coppermine and k7 are essentially equal performers, but Intel continues to insist on using rambus, coppermine could end up looking like the CYRIX MII - and get a rap for not being a "real" 1GHZ chip.