Nvidia graphics driver bug

The nVidia graphics driver (96.86) distributed with Vista 32 RTM works
properly on my system. The driver released yesterday (97.46) as well as the
"RC2" driver (96.85) which was on nVidia's web site until yesterday both
fail when I resume from monitor "power-down." The power to the monitor is
restored, but the screen shows nothing but a light blue color with thin
white diagonal lines. It's impossible to recover from this condition and a
reboot is required. CTRL-ALT-DEL changes the color to light green.

This problem is recorded in the system log with five instances of this
error message: "IRQARB: ACPI BIOS does not contain an IRQ for the device in
PCI slot 22, function 0. Please contact your system vendor for technical
assistance" Each instance lists a different PCI slot.

Hardware: ASUS A32SLI-Deluxe motherboard, XFX 7900GT graphics card, AMD
4800+ dual core 64-bit processor. I have installed the latest BIOS (1303). I've
had no trouble with XP or Linux, both 32 and 64-bit versions.

I'll appreciate any help - and if anyone has a direct pipeline to Microsoft
or nVidia I'd appreciate having this bug passed on. I reported it to
Microsoft verbally about ten days ago, but the report seems to have gone to
the bit bucket in the sky.

Thanks to all.


formerprof

/formerprof/ said:
> The nVidia graphics driver (96.86) distributed with Vista 32 RTM works
> properly on my system. The driver released yesterday (97.46) as well as the
> "RC2" driver (96.85) which was on nVidia's web site until yesterday both
> fail when I resume from monitor "power-down." The power to the monitor is
> restored, but the screen shows nothing but a light blue color with thin
> white diagonal lines. It's impossible to recover from this condition and a
> reboot is required. CTRL-ALT-DEL changes the color to light green.
>
> This problem is recorded in the system log with five instances of this
> error message: "IRQARB: ACPI BIOS does not contain an IRQ for the device in
> PCI slot 22, function 0. Please contact your system vendor for technical
> assistance" Each instance lists a different PCI slot.
>
> Hardware: ASUS A32SLI-Deluxe motherboard, XFX 7900GT graphics card, AMD
> 4800+ dual core 64-bit processor. I have installed the latest BIOS (1303). I've
> had no trouble with XP or Linux, both 32 and 64-bit versions.
>
> I'll appreciate any help - and if anyone has a direct pipeline to Microsoft
> or nVidia I'd appreciate having this bug passed on. I reported it to
> Microsoft verbally about ten days ago, but the report seems to have gone to
> the bit bucket in the sky.

Perhaps your best bet on something of that nature...
http://forums.nvidia.com/

Supposedly info on drivers *is automatically passed on to MSFT via WER
(Windows Error Reporting) so you've been there and done that.

Robert McLaws: Windows Vista Edition
"I'm just an online pundit who's barely old enough to legally buy alcohol

Missing Drivers in Windows Vista
Microsoft's collecting much more information about device drivers than you
might think. Did you know that when Vista tries to install a driver, and it
can't find one (either out of the box or online), your Vista computer will
generate Windows Error Reporting error? It then submits that error report to
Microsoft. Microsoft can then prioritize how they interact with device
vendors based on real-world usability, and get the driver on Windows Update.
So if 10,000 users have some obscure piece of hardware, Microsoft can make
sure those users get the best experience possible. Cool, huh?"

CH



"dev" <dev@null.invalid> wrote in message
news:ugpE76dMHHA.320@TK2MSFTNGP06.phx.gbl...
> /formerprof/ said:
>> The nVidia graphics driver (96.86) distributed with Vista 32 RTM works
>> properly on my system. The driver released yesterday (97.46) as well as
>> the
>> "RC2" driver (96.85) which was on nVidia's web site until yesterday both
>> fail when I resume from monitor "power-down." The power to the monitor
>> is
>> restored, but the screen shows nothing but a light blue color with thin
>> white diagonal lines. It's impossible to recover from this condition and
>> a
>> reboot is required. CTRL-ALT-DEL changes the color to light green.
>>
>> This problem is recorded in the system log with five instances of this
>> error message: "IRQARB: ACPI BIOS does not contain an IRQ for the device
>> in
>> PCI slot 22, function 0. Please contact your system vendor for technical
>> assistance" Each instance lists a different PCI slot.
>>
>> Hardware: ASUS A32SLI-Deluxe motherboard, XFX 7900GT graphics card, AMD
>> 4800+ dual core 64-bit processor. I have installed the latest BIOS
>> (1303). I've
>> had no trouble with XP or Linux, both 32 and 64-bit versions.
>>
>> I'll appreciate any help - and if anyone has a direct pipeline to
>> Microsoft
>> or nVidia I'd appreciate having this bug passed on. I reported it to
>> Microsoft verbally about ten days ago, but the report seems to have gone
>> to
>> the bit bucket in the sky.
>
> Perhaps your best bet on something of that nature...
> http://forums.nvidia.com/

Dear Chad,

I do not thiink that a driver error is generated as one has to force a
hardware reset in order to get going again. The only report I've been able
to detect is the system log entry mentioned in my original post.

This problem may be unique to the 7900GT. I've now seen confirming reports
from four other users with the problem and all of them are running the
7900GT. Two workarounds have been suggested for use at the first corrupted
screen: ALT-F4 and CTRL-ALT-DEL/down arrow/ENTER/ENTER. I haven't had a
chance to try them out yet. Of course these are work-arounds, not
solutions.

So, alas, I fear that I haven't been there done that quite yet. All good
wishes.

formerprof



"Chad Harris" <fixvista-itneedsit.net> wrote in message
news:%23MaLglpMHHA.2456@TK2MSFTNGP06.phx.gbl...
> Supposedly info on drivers *is automatically passed on to MSFT via WER
> (Windows Error Reporting) so you've been there and done that.
>
> Robert McLaws: Windows Vista Edition
> "I'm just an online pundit who's barely old enough to legally buy alcohol
>
> Missing Drivers in Windows Vista
> Microsoft's collecting much more information about device drivers than you
> might think. Did you know that when Vista tries to install a driver, and
> it can't find one (either out of the box or online), your Vista computer
> will generate Windows Error Reporting error? It then submits that error
> report to Microsoft. Microsoft can then prioritize how they interact with
> device vendors based on real-world usability, and get the driver on
> Windows Update. So if 10,000 users have some obscure piece of hardware,
> Microsoft can make sure those users get the best experience possible.
> Cool, huh?"
>
> CH
>
>
>
> "dev" <dev@null.invalid> wrote in message
> news:ugpE76dMHHA.320@TK2MSFTNGP06.phx.gbl...
>> /formerprof/ said:
>>> The nVidia graphics driver (96.86) distributed with Vista 32 RTM works
>>> properly on my system. The driver released yesterday (97.46) as well as
>>> the
>>> "RC2" driver (96.85) which was on nVidia's web site until yesterday both
>>> fail when I resume from monitor "power-down." The power to the monitor
>>> is
>>> restored, but the screen shows nothing but a light blue color with thin
>>> white diagonal lines. It's impossible to recover from this condition and
>>> a
>>> reboot is required. CTRL-ALT-DEL changes the color to light green.
>>>
>>> This problem is recorded in the system log with five instances of this
>>> error message: "IRQARB: ACPI BIOS does not contain an IRQ for the device
>>> in
>>> PCI slot 22, function 0. Please contact your system vendor for technical
>>> assistance" Each instance lists a different PCI slot.
>>>
>>> Hardware: ASUS A32SLI-Deluxe motherboard, XFX 7900GT graphics card, AMD
>>> 4800+ dual core 64-bit processor. I have installed the latest BIOS
>>> (1303). I've
>>> had no trouble with XP or Linux, both 32 and 64-bit versions.
>>>
>>> I'll appreciate any help - and if anyone has a direct pipeline to
>>> Microsoft
>>> or nVidia I'd appreciate having this bug passed on. I reported it to
>>> Microsoft verbally about ten days ago, but the report seems to have gone
>>> to
>>> the bit bucket in the sky.
>>
>> Perhaps your best bet on something of that nature...
>> http://forums.nvidia.com/
>

formerprof schrieb:
>
> This problem is recorded in the system log with five instances of this
> error message: "IRQARB: ACPI BIOS does not contain an IRQ for the device
> in PCI slot 22, function 0. Please contact your system vendor for
> technical assistance" Each instance lists a different PCI slot.
>
> Hardware: ASUS A32SLI-Deluxe motherboard, XFX 7900GT graphics card, AMD
> 4800+ dual core 64-bit processor. I have installed the latest BIOS
> (1303). I've had no trouble with XP or Linux, both 32 and 64-bit versions.
>
Hello formerprof,

if there was a bug in the mentioned drivers concerning power management,
there would be much more problems of this kind. I suppose that the
problem lies - as mentioned in the log - in the bios of the mainboard or
the graphics card. Since Windows 98, Microsoft's demands on the correct
funtion of the BIOS and the ACPI-tables have risen continually, and
PM-failures have been the most common failures from bugs in the BIOS.

Speaking of myself, I have just needed to update the BIOS on my MSI-K9N
for PM to work properly.

Regards,

Detlev

Dear Detlev,

So far I've only heard of this problem occurring on nVidia 7900GT cards.
Several different main boards are reported, suggesting that there's
something about the 7900GT to which nVidia will have to attend. I hope it's
sooner rather than later. Many thanks and all good wishes.

formerprof



"Detlev Rackow" <detlev.rackow@gmx.de> wrote in message
news:eS%23vdd7MHHA.2028@TK2MSFTNGP03.phx.gbl...
> formerprof schrieb:
>>
>> This problem is recorded in the system log with five instances of this
>> error message: "IRQARB: ACPI BIOS does not contain an IRQ for the device
>> in PCI slot 22, function 0. Please contact your system vendor for
>> technical assistance" Each instance lists a different PCI slot.
>>
>> Hardware: ASUS A32SLI-Deluxe motherboard, XFX 7900GT graphics card, AMD
>> 4800+ dual core 64-bit processor. I have installed the latest BIOS
>> (1303). I've had no trouble with XP or Linux, both 32 and 64-bit
>> versions.
>>
> Hello formerprof,
>
> if there was a bug in the mentioned drivers concerning power management,
> there would be much more problems of this kind. I suppose that the problem
> lies - as mentioned in the log - in the bios of the mainboard or the
> graphics card. Since Windows 98, Microsoft's demands on the correct
> funtion of the BIOS and the ACPI-tables have risen continually, and
> PM-failures have been the most common failures from bugs in the BIOS.
>
> Speaking of myself, I have just needed to update the BIOS on my MSI-K9N
> for PM to work properly.
>
> Regards,
>
> Detlev