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: wanna_bmw who wrote (57903)10/9/2001 2:25:58 PM
From: fyodor_Respond to of 275872
 
Wanna: I think that faster front side bus Northwoods will come after the first launch. There has been no word on 533MHz FSB chipsets, and usually there would be if they are set to launch in a few month's time.

DDR266 would be a good match for the 533MHz FSB and with rumors that Northwood has been pushed to Q1, I expect it will debut on the quad-pumped 133MHz FSB.

Second, DDR333 is going to be a low volume memory technology through most of 2002, and Intel doesn't like supporting anything that is low volume. Therefore, I expect support to come from others, like SiS and ALi, and even then, it won't be widely available.

The SiS645 chipset already supports DDR333. But, yes, volumes are going to be very low for all or most of Q1, but after that I think we'll see a fairly rapid ramp.

Third, either roadmaps that I've seen are conservative, or Intel doesn't feel like pushing frequencies too fast. I've seen 2.4GHz in Q2 and >2.5GHz in Q3, according to roadmaps from Register, Inquirer, etc.

That may well be. They'll likely choose (insofar as they have a choice) to stay only slightly ahead of AMD (in terms of performance or perception of same). Then they can devote greater amounts of .13&#181 resources to defend their dominance of the mobile market.

Lastly, the 193nm SVG equipment probably isn't going towards Northwood. Intel has said they plan to use it for the 100nm node, not 130nm.

I thought the hard phase mask was a temporary solution, since it's much more expensive and lowers overall throughput significantly.

But if you wish to argue these points in favor of Intel, I'd be happy to continue playing Devil's Advocate with you. ;-)

I wouldn't exactly call what I'm doing playing Devil's Advocate. What you are, perhaps, but I honestly believe what I am writing, make no mistake about that ;-). I honestly do think Intel's near-term roadmap is virtually unassailable by AMD and the only saving grace has been SVG's inability to deliver on their 193nm equipment.

(btw, I thought Intel had publicly announced that SVG had lost the order to deliver equipment for the 100nm node and that Intel would reevaluate products from all producers).

-fyo