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) -- Ignore unavailable to you. Want to Upgrade?


To: Joe NYC who wrote (16010)10/26/2000 1:12:18 AM
From: kash johalRead Replies (3) | Respond to of 275872
 
Joe,

Yes bin splits will be like a normal distribution.

Lets say midpoint of distribution of PIV is 1.5Ghz for arguments sake - ie equal parts above and below the 1.5Ghz speeds.

Individual parts will yield at all sorts of speeds - some at 1.4, some at 1.42 ... some at 1.49 ....some at 1.6 etc

Lets assume a very tight distribution os speeds with majority of parts between 1.4 to 1.6Ghz (actually unlikely).

The testers are set to a certain test spec to ensure it meets a minimum spec.

Say parts are to be sold at 1.4 and 1.5Ghz for marketing reasons.

With the above tester specs 50% of parts meet the 1.5Ghz spec. 100% of the parts will meet 1.4Ghz spec. So the company will will sell 50% of the parts as 1.4Ghz and balance of 50% will be saleable as 1.5Ghz parts.

Now the spread of MHZ is directly related to excellence in mfg control etc. The reason for very wide range of PIII speed yields has been postulated to be related to the use of notched gates.

It is quite likely that yield mid point is well above 1.5Ghz for PIV. The reasoning is that it would be extremely unlikely to have a +/- of 100Mhz on a 1.5Ghz part.

I would expect at least +/- of 200Mhz even with decent process control.

This implies a yield range of 1.4Ghz to 1.8Ghz assuming that intel isn't gonna throw a bunch of PIV parts away.

I hope that this explanation helps.

regards,

Kash