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 -- Ignore unavailable to you. Want to Upgrade?


To: kash johal who wrote (86019)1/8/2000 2:45:00 PM
From: Elmer  Respond to of 1572953
 
Re: "Overall seems like a large availability of Willamettes in Q4 seems a stretch and will require flawless execution. If a six month slip happens it could be big trouble."

It may not be as difficult as some make it out. First off, the .18u process will be more mature then and in high volume. Second, this is not a new architecture as was Merced/Itanium so the validation suite should be in place. Third, the chipset is not introducing a new memory architecture either, assuming Intel uses RamBus. Nevertheless it will require very good execution and please Intel, no more major screwups.

EP



To: kash johal who wrote (86019)1/8/2000 2:57:00 PM
From: Charles R  Read Replies (2) | Respond to of 1572953
 
<It seems to me that there are several risks with Willamette:>

Here are couple of other risks:

- Wilamette may need to be in 0.13 to scale well (mostly a power, yield issues) [By the way I got my first negative report on 0.13 recently - one of my sources is telling me a slip to Q4 is possible. If that happens AMD has much better chance of MHz parity or leadership going into Christmas]

- Wilamette has a lot of interesting circuits that are going to require DV that may put some interesting blocks along the way

- And finally, Wilamette definitely needs 0.13 to attain volumes. So, eve if everything goes absolutely perfectly, real volumes (tens of Mu) may not occur until 2001.