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.
Politics : Formerly About Advanced Micro Devices

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Process Boy who wrote (73484)9/29/1999 10:27:00 PM
From: THE WATSONYOUTH  Read Replies (3) of 1572901
 
Re:"Both, IMHO. By getting to a .18ish front end by this point in time has allowed them to scale MHz to the levels you see now. What's bad is AMd is de facto advertising this a .25 process, giving the impression of super MHz scalability with the design that may not in reality be there
going forward, disappointing some of those making the wildest speculations on MHz I've seen.
MOT is currently doing the same thing with the G4, only in reverse, having .18 gates and calling it .18, but with a .25 back end, probably so they won't be perceived as being behind on their Copper Process getting to .18."

I agree. It does appear that AMD is using a device design for the .25um Athlons that has thinner gate oxide and shorter channel lengths than Intel's .25um process. If so, some of the performance advantage observed to date is due to
process and not design. However, this will just make it more difficult (but not impossible) for AMD to further improve the device design for the .18um Athlons. We will see.
The Moto process, however, is in .25um groundrules (FEOL and BEOL) and the device (.15um Leff) is probably only slightly longer than what Intel is using nominally at .25um. So, If Moto actually delivers 500MHz G4s in .25um groundrules with the stated floating point and integer performance, it's not a bad result at all.

THE WATSONYOUTH
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext