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: Epinephrine who wrote (90316)1/28/2000 7:59:00 PM
From: Process Boy  Read Replies (1) of 1574005
 
E - <Now your turn, I have given you my opinion in good faith, please share yours.>

I believe Willamette micro-architecture enhancements will be significantly "better" than what Mustang can possibly do with the Athlon core by EOY. You'll have to rely on rumor as to what those might be, until Intel discloses some details.

As for the process enhancements, I CLEARLY believe Intel's transisotrs are about a generation ahead of AMD, and improving all the time. AMD will undoubtably be improving theirs also, but from what was disclosed at the IEDM with their 50nm paper no one at Intel, i.e., people smarter than me on the subject, is too concerned. Intel's transistors display the best drive currents in the industry. Intel believes that xtor drive current at current dimensions has the most direct payoff on device performance. Coupling these two factors, i.e., new dsign, with better process, is what leads me to my conclusions that Intel will be ahead. I certainly have the chance of being wrong.

As for Cumine vs. Athlon, I've psoted several times that AMD actually has the advantage with regard to design scalability, but I believe Intel has the advantage in process. I believe the race to 1GHz will be close, with worst case for Intel being 2nd by a matter of weeks at the most. I could certainly be wrong here also. If you are AMD-centric, you could certainly take solace in repeated rumors that AMD could release 1GHz "anytime it wants". I have no idea if these rumors are true or not. If they are, obviously my position is in deep doo-doo, but I would have to ask the question why they don't do it then. I've heard ideas on this that are pretty unconvincing.

BTW, an article just posted cites MDR as basically agreeing with the stance I've taken for several months. Not they have a great track record.

There is an ISSCC conference coming up in Feb, where Intel will disclose details on a 1GHz .18 Al interconnect CPU. Maybe there will be more detail to discuss at that time.

Bottomline: I believe Willie (new design) + Intel's process will be a hard combo to beat EOY. I have a friendly wager on it.

And despite a recent set of flames to the contrary, I could not even if I wanted to tell you conclusively what Willie speed will be at launch. I believe it to be a work in progress.

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