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: andreas_wonisch who wrote (43451)6/8/2001 8:20:14 PM
From: Charles RRespond to of 275872
 
Andreas,

<However, it doesn't look like that these problems are terrible (like e.g. K6-III), since there are already 1.2 GHz Palominos available. >

Yeah, the K6 days are over. AMD has to slip hammer for another year for AMD to get back to K6 days.

<However, the 20% reduction in power consumption apparently doesn't allow an equal increase in clock-speed. That's probably why Palomino won't go above 1.6 GHz this year according to unofficial AMD roadmaps. IMO it would have been a better decision for AMD to focus on clock-speed instead of performance. The average customer isn't interested in high IPC. And cheap 2 GHz Pentium 4s with SDR memory couldn't be a compelling buying argument for those (although they will perform terrible, of course).>

Though I feel it is pretty likely that there is a bin split problem, I don't see any information to point to the potential root cause. If there is a bin-split problem it could be due to many things - it could be the additional speed paths from SSE, problems with PowerNOW, problems with relayed out cells or something completely different. Tough to tell where the problem may be.

Power is tightly coupled to frequency ramp in the sense that higher power makes the platform difficult to design and more expensive. I am not sure if the trade-off between power and frequency ramp is that straight-forward.

Chuck