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)
AMD 214.25-3.1%3:59 PM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Scumbria who wrote (10595)9/28/2000 5:13:34 PM
From: Charles R of 275872
 
Scumbria,

<I distinctly remember you saying that you heard about serious problems on the Willamette design last year.>

It is true that there were serious problems with Willamette before tapeout (you may find a post or two from me on this topic around Q2 99, I believe). The tapout was pushed out after that and did not occur until very late Q4 (December, if my memory serves me right).

The chip came back in early January and after a sub GHz lab bring-up went to 1.3G very quickly. Since then my significant remaining concerns about Willamette have been heat dissipation and potential speed limitations because of double pumped ALU. I still wonder how well this beast is going to scale.

<So why is the 20 stage pipe after decode barely able to run faster than the 5-stage K7 post-decode pipe? Something is amiss with the Willy microarchitecture.>

Clearly, temperature is a BIG problem and as we discussed over this thread many times, the double-pumped ALU could be another. You have also brought up 2-cycle cache which clearly does not help. Other than that, all we can do is see how the clock speed battle will roll out next year - I bet it is going to be extremely interesting.

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