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 : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?


To: fyodor_ who wrote (133951)5/1/2001 11:20:25 PM
From: dale_laroy  Read Replies (1) | Respond to of 186894
 
"I seem to recall that UMC's (or maybe it was TSMC's) .13µ process would feature a gate length of 100nm. That's where Intel and AMD are on their .18µ processes. Sure, gate lengths aren't everything - and I'm not claiming that Intel/AMD are a full generation ahead - but I just don't see how UMC (or any other non-IBM foundry) is going to pull off production volumes of a modern x86 chip operating at ~ 1GHz."

It probably was TSMC, but the same would almost certainly apply to UMC. Nevertheless, assuming that UMC's 0.13-micron process technology were equivalent to AMD's 0.18-micron process technology, this could still enable them to produce 1.333+ GHz Athlon/Duron processors. If UMC were to produce Appaloosa processors at Palomino, or even TBird, speed grades in 2002, these would be very competitive with a desktop Tualatin.



To: fyodor_ who wrote (133951)5/1/2001 11:39:54 PM
From: Elmer  Read Replies (1) | Respond to of 186894
 
IMHO, you are completely correct. UMC is a foundry and, as such, needs very high yields on an extremely wide variety of chips. Their process just isn't going to be optimized for clock speed

Exactly. UMC has to yield well across a whole host of products from many different customers on standard cell, custom and even gate arrays. They can not optimize for 1 customer. That's why I think this must be for the K6-2 which is no longer leading edge.

EP