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: THE WATSONYOUTH who wrote (67779)8/6/1999 12:15:00 AM
From: Charles R  Read Replies (2) of 1575424
 
THE WATSONYOUTH,

<<Given the context, you seem to be implying that G4 is Process issue. Is that right?>
You are right. It may not be. My guess is that it is a combination
of several issues (design/.18um Cu BEOL/performance/etc) However,
21 months is a long time. Seems like a minimum of 5 redesigns could have been completed by now. I don't know the complexity of the ALTIVEC
instruction set. (Can anyone comment?) >

Since you have said you do not "know" it is process problem, let me tell you a story.

It was sometime in 1992 when my boss came to my cube in a half excited state and asked me to review a bunch of documents about a hot silicon valley startup that was planning a 500MHz processor. He wanted to know if pumping money into it was a worthwhile cause. It was intriguing concept but the more I looked at this company's business aspects the more certain I was that these people were going to crash and burn. So, my recommendation was not to invest in this startup. That company was MicroUnity.

Now, I have lost touch with what happened to their neat plans but I take it that Altivec is what Mot got out of their MicroUnity investments. A fancy instruction set that, from what I gather, created a bunch of implementation problems for MicroUnity and now for Mot.

The Mot processor guys I know no longer work there so I can't confirm the problem side but it seems likely to me that they most likely have problems with the design.

Isn't it true that Mot is having good success with SRAM chips on the same process? If so, I would be hesitant to conclude that they are having process problems, especially when AMD is also confirming the good yields on that process.

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