Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

FWD>>Pentium "Warranty"

1 view
Skip to first unread message

MA Crist

unread,
Nov 30, 1994, 11:22:40 AM11/30/94
to
Mail*Link(r) SMTP FWD>>Pentium "Warranty"

--------------------------------------
Intel Apologizes For Pentium Error, Issues "Warranty"

SANTA CLARA, CALIFORNIA, U.S.A., 1994 NOV 28 (NB) -- In a letter
now posted on the Internet, Intel President Andy Grove has
apologized for a computational error in the Pentium chip, and has
issued a "life-time warranty" on the chip to users "engaged in work
involving heavy-duty scientific/floating point calculations."

Intel has also opened up an 800 number, typically accessed mainly
by engineering customers, to all PC users with concerns about the
error, which received initial attention in an Internet newsgroup
before being reported by Electronic Engineering Times (EE Times)
and then parlayed into the general press by Cable News Network
(CNN).

In the letter on the Internet, Grove starts out by saying: "I am
truly sorry for the anxiety created among you by our floating point
issue. I read through some of the postings and it's clear that many of
you have done a lot of work around it and some of you are very
angry at us."

The Intel president then explains that the error went undetected in
pre-release testing, and was only discovered in further testing by
Intel last summer.

In response, Intel started a separate project, which included
mathematicians and scientists who work with Intel, to examine the
problem and its impact, according to Grove.

"The group concluded after months of work that (1) an error is only
likely to occur at a frequency of the order of once in nine billion
floating point divides, and that (2) this many divides in all the
programs they evaluated (which included many scientific programs)
would require elapsed times of use that would be longer than the
mean time to failure of the physical computer subsystems," he
writes.

"In other words, the error rate a user might see due to the
floating point problem would be swamped by other known computer
failure mechanisms. This explained why nobody -- not us, not our
OEM (original equipment manufacturer) customers, not the software
vendors we worked with, and not the many individual users -- had run
into it."

Grove goes on to pledge that Intel "would like to find all users of
the Pentium processor who are engaged in work involving heavy duty
scientific/floating point calculations," and that the company will
"resolve their problem in the most appropriate fashion including,
if necessary, by replacing their chips with new ones."

Intel, he adds, will "stand behind these chips for the life of your
computer."

The error was originally discovered by Thomas Nicely, a mathematics
professor at Lynchburg College in Virginia, according to John
Raftrey, an Intel spokesperson. Nicely determined that the Pentium
can return erroneous values for certain division operations.

Intel has since fixed the design, and has recently started to
produce corrected chips. The error, however, is present in all of
the estimated two million Pentium chips based on the original
design.

Grove's letter is posted in comp.sys.intel, the same newsgroup
where users have been discussing the flaw, Raftrey told Newsbytes.

Callers to the special "800" number (800-628-8686) are answered
by a "live person," he added. Each call is then routed to a staff
member with "the appropriate level of expertise" to handle the
particular inquiry.

Raftrey declined to say exactly how many calls Intel has received
on the special 800 number. "But calls are up," Newsbytes was told.

(Jacqueline Emigh/19941128/Reader Contact: 800-628-8686;
Press Contact: John Raftrey, Intel, 408-765-6007)

0 new messages