Silicon Investor (SI) -- The First Internet Community

STOCKTALK

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 : AMD:News, Press Releases and Information Only! -- Ignore unavailable to you. Want to Upgrade?

To: Paul Engel who wrote (6449)	6/1/1998 12:33:00 PM
From: Ali Chen	Read Replies (1) \| Respond to of 6843

Paul, <no understanding of the FDIV problem!> if you really want (and can) to learn the details of the famous Pentium FDIV bug, I suggest you to read this: http.cs.berkeley.edu instead of relying on some superficial journalism. <The FDIV problem had nothing directly related to division by the integer 3> This is a rare correct remark from you, mainly due to the use of word "directly". Indirectly, my example was taken off-head from the above article in attempt to illustrate the difference between the uniform probabilistic reasoning and distinct speciality of buggy divisors (so-called bruised integers) that are frequently found in mathematical models used in all applications. - Ali

To: Paul Engel who wrote (6449)	6/1/1998 3:40:00 PM
From: Petz	Read Replies (1) \| Respond to of 6843

Paul, you know EP's claim about the AMD K6-classic extended precision multiply error is false. ANYONE could produce the Pentium FDIV bug by typing a simple formula into ANY spreadshett application. The truth is NO COMMERCIAL APPLICATION HAS EVER DEMONSTRATED A MATH ERROR USING THE K6 CHIP. Please tell me what I type into 1-2-3 OR Excel OR Quattro to demonstrate a math error on my K6. When you tell me, you've won the argument. Better yet, tell me ANY APPLICATION AT ALL that gives an error and how you demonstrate it. I'm waiting........ I wonder how many math errors there will be on the 400 mm^2 die Merced chip? Petz