| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 5276.1 | | TLE::REAGAN | All of this chaos makes perfect sense | Mon May 05 1997 18:32 | 7 |
| Well, I certainly wouldn't put this in the same category as the
previous floating point bug. For those who don't want to look, the
bug is with Pentium II and Pentium Pro processors failing to set a
flag in the floating point status register when a float-to-int
conversion won't fit in a small integer.
-John
|
| 5276.2 | | QUARK::LIONEL | Free advice is worth every cent | Mon May 05 1997 18:47 | 4 |
| It would be inappropriate for Digital to try to "take advantage" of this,
especially when we are about to introduce our own Pentium II systems.
Steve
|
| 5276.3 | | LGP30::FLEISCHER | without vision the people perish (DTN 381-0426 ZKO1-1) | Mon May 05 1997 19:19 | 7 |
| > It would be inappropriate for Digital to try to "take advantage" of this,
> especially when we are about to introduce our own Pentium II systems.
We could always introduce them as our line of "small,
inaccurate machines".
Bob
|
| 5276.4 | | COOKIE::FROEHLIN | VMS...riding into the setting sun! | Mon May 05 1997 20:37 | 9 |
| > bug is with Pentium II and Pentium Pro processors failing to set a
> flag in the floating point status register when a float-to-int
> conversion won't fit in a small integer.
Wasn't that the bug area where the last Arianne Euro-rocket blew up?
Maybe they where an early beta site for these new processors. And Intel
made us believe it was a software bug...
Guenther
|
| 5276.5 | | DPE1::ARMSTRONG | | Mon May 05 1997 20:55 | 8 |
| > Wasn't that the bug area where the last Arianne Euro-rocket blew up?
> Maybe they where an early beta site for these new processors. And Intel
> made us believe it was a software bug...
The 'IntelSecrets' Web site mentions this...although the problem
in that case was with the software and not the hardware. The software
was not checking this flag....so dont trivialize this error.
bob
|
| 5276.6 | | TLE::REAGAN | All of this chaos makes perfect sense | Tue May 06 1997 15:13 | 14 |
| re:.4 No.
The bug was the routine in question returned a error string (in ASCII)
saying that a floating error occurred. The calling routine simply
took the ASCII characters as valid data (never thinking that an
error could occur).
Given the previous public trashing Intel took with the Pentium FP
problem, I would hope to see an official public statement from Intel.
Until then, all speculation will just spin out of control until we
end up some conspiracy theory about space aliens controlling the
government with advanced mind control.
-John
|
| 5276.7 | | STAR::PARKE | Sometimes pigeon, Sometimes statue | Tue May 06 1997 16:58 | 5 |
| On CNN last night it was mentioned the "Intel had found a glitch..."
"The problem is being investigated..." and "No recalls will be done"
|
| 5276.8 | | COOKIE::FROEHLIN | VMS...riding into the setting sun! | Tue May 06 1997 17:53 | 8 |
| .6> The bug was the routine in question returned a error string (in ASCII)
John, you're an easy believer...I think the Arianne was THE beta test
site for this chip and when the hardware bug showed up, the European
Space Agency agreed to cover this up as a software bug and Intel
replaces the rocket. ;-)
Guenther
|
| 5276.9 | | BBQ::WOODWARDC | ...but words can break my heart | Wed May 07 1997 01:02 | 8 |
|
>Until then, all speculation will just spin out of control until we
>end up some conspiracy theory about space aliens controlling the
>government with advanced mind control.
you heard about that too? ;') :') ;') :')
H
|
| 5276.10 | | MKTCRV::MANNERINGS | | Wed May 07 1997 08:20 | 1 |
| It is the space aliens who are sending out the empty envelopes, really.
|
| 5276.11 | a kind of ethic? | ATZIS3::SCHATZMANN_H | | Wed May 07 1997 08:50 | 14 |
| I think the best advantage about this new INTEL bug would be.
###############################################
# Make a BIG Press-Propaganda as a first one. #
###############################################
----> Tell that we will NOT take advantage out of this.
-Then do the same like last time.
.help Intel to swap tons of Pentiums-xy and earn money on this way.
.hope Alpha-Chip would never have the same Problem.
helmut
|