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 : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: TechieGuy-alt who wrote (156634)	4/17/2005 9:38:51 AM
From: aleph0	Respond to of 275872

I repeat ( # 156626 ), since PG visisted CT, I've noticed they've now become very "un"critical of Intel . I therefore cancelled my lost-standing abbonement. Need I say more ? CT is for me no longer the "bible" as it used to be regarded - but has joined the ranks of the commercial popular press IMO.

To: TechieGuy-alt who wrote (156634)	4/17/2005 11:12:20 AM
From: combjelly	Respond to of 275872

"Where did they get this info?" It appears to be guesswork based on public announcements, hints and slips. so you definitely should wash your hands after handling those performance figures... So the power envelope of the "K9" seems to be based on a power bump to the MII socket chips and the assumption being that is to prepare systems for "K9". The Intel results probably come from some hints he got somewhere on performance of Conroe/Merom and the rest was extrapolated. Given the number of fudge factors, i.e. how much an on-chip memory controller and any microarchitectural improvements would help means this is a pure exercise in speculation. Likewise with the "K9". It apparently assume no microarchitectural changes to the K8 core. Which isn't an irrational assumption, that is certainly a reasonable, low-cost way for AMD to do it. At 65nm, they could license IBM's cache technology, extend the crossbar and pack 4 cores on a chip the size of the current dual core chips but with roughly twice the L2 cache per core. But then what are they doing with all the design engineers they've been hiring, not to menton the ones they already have? Seems a waste of resources. I suspect that AMD has something more ambitious planned, although the hypothesized quad core may very well be in the works, it certainly would be a workable fall back if the "something more ambitious" plan runs into potholes along the road to release. Given that the next generation is going to be a critical one for AMD, a cautious, belt and suspenders approach would be smart.