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: Elmer who wrote (26028)11/16/1997 5:50:00 PM
From: Bill Jackson  Read Replies (1) | Respond to of 1583484
 
Elmer; You are right about the non factor of the bug. Way back when the first bug appeared in the 386/486? we got about one customer in 100 who got the free replacement. Large networks went with the MSI and either all or none did it depending on his ear to ground etc. All the nerd speedsters did it, and took delight in demonstrating the flaw.

I have heard about field programmable internal code?, is that the core microcode that can do that, or is there an external section that can be accessed to watch for the bug and insert the pathc on the fly, albeit far slower. This in rewriteable EEPROMs that I did not think could match CPU speeds.? So you get a ten cycle+? 'deep think' as this comes in to play.?

Bill



To: Elmer who wrote (26028)11/16/1997 5:58:00 PM
From: Ali Chen  Respond to of 1583484
 
Elmer, <because hardly anyone will bother with the patch. > But for those who will, who is going to pay for this?

As it was pointed out, this was a bug #81 in Pentium processor FAMILY. But you are right. Who will bother to develop any new patches? Customers have no choice currently, they have to buy these buggy Pentiums and live with it. This is, according to Paul Engel, "the processor of choice". This is a truly special meaning of "choice".