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: spyhunter who wrote (99550)3/22/2000 11:57:00 PM
From: Elmer  Read Replies (1) | Respond to of 1571271
 
Re: "The 486SX was definitely originally with the FPU on the die and disabled. I don't know if they came out with a different version. But the first versions WERE that way. We always laughed about it. Remember your history..."

You are the one who needs a history lesson. I was much closer to that project than I can say here. The first engineering samples were 486DX units with the FPU disabled, but only samples. Production was a different die. This is one of the great urban legends of the semiconductor industry. It was never true but many people repeat it as though it was gospel.

2 main problems.

#1 The FPU on the 486 was a significant percentage of the die area and it would have been a great waste of silicon to simply disable it.

#2 Using die with a defective FPU would have never provided significant quantities. Assume the yield was 66% on a 486DX. Also assume the FPU was 25% of the die area. That means only ~25% of the ~33% defective die were FPU failures. That means the 486DX fallout would have provided only about 8% the volume of the 486DX yet at one point the 486SX was the volume leader. Even this number is too optimistic because not all failures are the result of point defects but are systemic and not local to one area of the die.

There are other reasons that are show stoppers as well but they are not worth going into here.

EP