Very Computer

Please continue to sign up and show we do care!

Machine              Quantity      Name and Email

 laptop (same as
 Midwest Micro)










no-name clone

: Please continue to sign up and show we do care!

: Machine              Quantity      Name and Email

:  laptop (same as
:  Midwest Micro)










: no-name clone

Machine              Quantity      Name and Email

 laptop (same as
 Midwest Micro)


Pentuim Clone 90       1           Andreas


Intel Plato-90         1           Kevin T. Van Maren






no-name clone



Escom P5/60            1           Rogier van Wanroij

Dell, when is the BIOS fix going to be ready?

David Dodell
---

Telephone: +1-602-860-1121
FAX      : +1-602-451-1165

PCI IDE Prefetch Buffers:
Enables/Disables PCI IDE Prefetch Buffers
Enabling this feature will significantly increase
PCI IDE Hard DIsk performance.

-- Plato Bios, rev 12.

Not only does Intel say that the performance "penalty" to disabling
the prefetch buffers is significant, it is a "feature" of their
motherboard.  As a feature, we have the right to expect it to
work, and NOT DESTROY OUR DATA.

Of course, once people realize that this is destroying their data,
Intel says that there is "NO PREFORMACE PENALTY" incurred by turning
it off.

"FIXED" drivers for OS/2 TURN-OFF the prefetch buffers; Linux, Free/Net
BSD, and others will probably follow this route, as the buffers cannot
be used reliably in a multitasking environment where the BIOS isn't used
for disk I/O.

Intel should REPLACE DEFECTIVE MOTHERBOARDS.

FREE of charge, and ON DEMAND.  As with the Pentium fiasco, proving that
you are affected by the DEFECT should not be a requirement.

I am willing to exchange my old motherboard so they can 'recycle' the cache
and other circuitry.  I would even be willing to get the same motherboard
back if a NON-DEFECTIVE IDE chip was used.  Surely Intel can remove a
soldered chip; they can certainly put enough of them ON the board.

The other option is a Triton-based motherboard.  This shouldn't be a
problem, since Intel is mass-producing them dirt cheap.

I disagree with Roedey about sending the motherboard in 6 weeks before
getting the new one, but I suppose it could be tolerable.  It is definatly
better than using the defective one, although the Pentium replacement
poilicy is a much better alternative for the consumers.

After all, Intel did promise "Satisfaction guaranteed".  And I am NOT
satisfied.

Kevin T. Van Maren

Quote:>Intel states in the BIOS:

>PCI IDE Prefetch Buffers:
>Enables/Disables PCI IDE Prefetch Buffers
>Enabling this feature will significantly increase
>PCI IDE Hard DIsk performance.

>-- Plato Bios, rev 12.

>Not only does Intel say that the performance "penalty" to disabling
>the prefetch buffers is significant, it is a "feature" of their
>motherboard.  As a feature, we have the right to expect it to
>work, and NOT DESTROY OUR DATA.

>Of course, once people realized that this is destroying their data,
>Intel says that there is "NO PREFORMACE PENALTY" incurred by turning
>it off.

Anyone willing to turn back on the prefetch buffers to test under
OS/2 with the old drivers?

I'd have to do a cpmplete backup, and then wipe my system after
I was done to prevent further corruption...anyone with a DAT drive
willing?

Quote:>Intel should REPLACE DEFECTIVE MOTHERBOARDS
>FREE of charge, and ON DEMAND.  As with the Pentium fiasco, proving that
>you are affected by the DEFECT should not be a requirement.

<...>
>After all, Intel did promise "Satisfaction guaranteed".  And I am NOT
>satisfied.

Kevin T. Van Maren

Quote:les) writes:

------------t-h-e---e-n-d---o-f---a-n-a-l-o-g-------------
 _  _          ::digital Entropy device::
|  |_||\ |.WWW:  http://phoenix.csc.calpoly.edu/~cnorman