K8V SE Deluxe - Cool and Quiet issues --> PFN_LIST_CORRUPT

[Addy]

Hi ,

after flashing the BIOS of my K8V to Rev. 1003, which I hoped would solve my
problems with getting CnQ to work, actually made my problems worse. Without
CnQ, using QFan the MoBo runs quite well, but I´d like to get CnQ to work as
the PC runs day and night and electricity isn´t as cheap as it used to be.

But to the problem. When I switch on CnQ in the bios (and do all other
things required in the manual) my pc starts up and shows a BSOD saying
PFN_LIST_CORRUPT. If I turn CnQ off, there is no such problem. When I start
up WinXP in safe mode, there is no BSOD either. The PFN_LIST_CORRUPT seems
to be linked to bad RAM or bad driver behaviour.

Anybody had the same problem ?

Addy

My AMD64-PC

Asus K8V SE Deluxe, Athlon64 3200, 2x512MB Infineon DDR 400, Radeon 9600XT
ultimate, 1x160 GB Samsung SATA, Benq 822A DVDR.

 

[Paul]

Hi ,

after flashing the BIOS of my K8V to Rev. 1003, which I hoped would solve my
problems with getting CnQ to work, actually made my problems worse. Without
CnQ, using QFan the MoBo runs quite well, but I´d like to get CnQ to work as
the PC runs day and night and electricity isn´t as cheap as it used to be.

But to the problem. When I switch on CnQ in the bios (and do all other
things required in the manual) my pc starts up and shows a BSOD saying
PFN_LIST_CORRUPT. If I turn CnQ off, there is no such problem. When I start
up WinXP in safe mode, there is no BSOD either. The PFN_LIST_CORRUPT seems
to be linked to bad RAM or bad driver behaviour.

Anybody had the same problem ?

Addy

My AMD64-PC

Asus K8V SE Deluxe, Athlon64 3200, 2x512MB Infineon DDR 400, Radeon 9600XT
ultimate, 1x160 GB Samsung SATA, Benq 822A DVDR.

Have you tested the memory with memtest86 ?

Does the problem occur if there is only one DIMM in the system ?

I searched on abxzone.com (because I cannot find how to search on
forums.pcper.com) and one post suggested that with CnQ enabled,
the BIOS was overclocking the FSB and running 228x9, and the user
had to enter the BIOS and set it back to 200x10 to get it back to
normal.

This post has some suggestions for what to enable to get CnQ working.
As well, if you get complaints about "," is not a valid character,
you need to change the definition of the decimal point character
in your localization of Windows (i.e. the CnQ is "American" and
expects ".", while Europeans might use ",").

http://abxzone.com/forums/showpost.php?s=e18e7b69e2307e19f891ffbf9c742c1d&p=769420&postcount=13

I tried searching on "cool quiet k8v" at:
http://abxzone.com/forums/search.php

and there are a couple of other threads.

There were some similar issues with some people's P4 systems, and
they ended up doing a warranty return to Intel. It seems their
processors were defective, and at least one of the errors experienced
was the PFN thing. In the case of the Athlon64, it could be a BIOS
misprogramming issue.

CnQ is a bit of a departure, in terms of how it changes the
requirements for testing the processor at the factory. Standard
practice is to only test the processor at one set of conditions.
Some minimal tests might be done at other than stock speeds, but
that doesn't constitute a good enough test to guarantee those
conditions.

The invention of CnQ, means that the processor _should_ be tested
under all possible combinations of adjustable FID/VID supported
on the processor. That could increase test time by a couple of
orders of magnitude, and isn't likely to be how it is done. I'm
curious as to if/how AMD guarantees this feature doesn't affect
the integrity of the calculations the processor is doing. Maybe
they only test the two extremes or something - like max FID/VID
and min FID/VID. I wonder if there is a white paper on
this somewhere.

In any case, I would want _some_ independent kind of utility
that can query the current FSB clock, FID/VID settings etc.,
so a user can verify exactly how the system is being set up
at any given point in time. There are enough reports of CnQ
"flapping in the breeze" to make one wonder about the
provided software.

Paul

 

[Addy]

Have you tested the memory with memtest86 ?

Does the problem occur if there is only one DIMM in the system ?

I searched on abxzone.com (because I cannot find how to search on
forums.pcper.com) and one post suggested that with CnQ enabled,
the BIOS was overclocking the FSB and running 228x9, and the user
had to enter the BIOS and set it back to 200x10 to get it back to
normal.

This post has some suggestions for what to enable to get CnQ working.
As well, if you get complaints about "," is not a valid character,
you need to change the definition of the decimal point character
in your localization of Windows (i.e. the CnQ is "American" and
expects ".", while Europeans might use ",").

http://abxzone.com/forums/showpost.php?s=e18e7b69e2307e19f891ffbf9c742c1d&p=769420&postcount=13

I tried searching on "cool quiet k8v" at:
http://abxzone.com/forums/search.php

and there are a couple of other threads.

There were some similar issues with some people's P4 systems, and
they ended up doing a warranty return to Intel. It seems their
processors were defective, and at least one of the errors experienced
was the PFN thing. In the case of the Athlon64, it could be a BIOS
misprogramming issue.

CnQ is a bit of a departure, in terms of how it changes the
requirements for testing the processor at the factory. Standard
practice is to only test the processor at one set of conditions.
Some minimal tests might be done at other than stock speeds, but
that doesn't constitute a good enough test to guarantee those
conditions.

The invention of CnQ, means that the processor _should_ be tested
under all possible combinations of adjustable FID/VID supported
on the processor. That could increase test time by a couple of
orders of magnitude, and isn't likely to be how it is done. I'm
curious as to if/how AMD guarantees this feature doesn't affect
the integrity of the calculations the processor is doing. Maybe
they only test the two extremes or something - like max FID/VID
and min FID/VID. I wonder if there is a white paper on
this somewhere.

In any case, I would want _some_ independent kind of utility
that can query the current FSB clock, FID/VID settings etc.,
so a user can verify exactly how the system is being set up
at any given point in time. There are enough reports of CnQ
"flapping in the breeze" to make one wonder about the
provided software.

Paul

Hi Paul,

thanks for your intesive search. I followed your links and got some new
insights. But I couldn´t find a post where they talk about FSB overclocking
through CnQ. I had a similar idea as after one or twor trys with XP in Safe
Mode I restarted the PC an got a message from the board, that I had to reset
BIOS to standard because af a failed overclocking attempt. But the board
showed nothing.

I´m now in the process of testing my RAM with memtest86+ but I don´t seem to
get any error messages so far.

If you could point me to the article about the Overclocking thing, I´d be
grateful.

Addy

 

[Paul]

Hi Paul,

thanks for your intesive search. I followed your links and got some new
insights. But I couldn´t find a post where they talk about FSB overclocking
through CnQ. I had a similar idea as after one or twor trys with XP in Safe
Mode I restarted the PC an got a message from the board, that I had to reset
BIOS to standard because af a failed overclocking attempt. But the board
showed nothing.

I´m now in the process of testing my RAM with memtest86+ but I don´t seem to
get any error messages so far.

If you could point me to the article about the Overclocking thing, I´d be
grateful.

Addy

The overclocking comment is here:
http://abxzone.com/forums/showpost.php?p=768049&postcount=7

This thread mentions the ingredients of a working CnQ:
http://abxzone.com/forums/showthread.php?t=67457&highlight=cool+quiet+k8v

No idea if this is the proper order of execution or not I
added item 4 to the list and added comments.

1. download and install the cool n quiet driver and software from
the ASUS website. The AMD driver is here, as well as on the CD.
http://www.amd.com/us-en/assets/content_type/utilities/Cpudriver.ZIP
2. enable "Cool n Quiet" and the ACPI 2.0 in the BIOS
3. set the power setting to minimal management
(In XP under Power Management select "minimal" as the scheme)
4. If it doesn't work, upgrade to 1005.011 BIOS or later, as the people
in the thread found it works OK with that version.

From the AMD Cpudriver.ZIP readme file:

REQUIREMENTS:
AMD Athlon(tm) 64 systems with ACPI 2.0 performance control
objects (_PCT) on Windows XP <-----
DESCRIPTION:
This driver supports processor performance control via the
processor objects defined in the ACPI 2.0 specification.
KNOWN LIMITATIONS:
1) The processor performance state may not be restored to the
maximum state if the CPU runs at 100 percent. This problem occurs
if the computer is 100 percent busy when the power policy changes.
http://support.microsoft.com/?kbid=330512

2) "System Properties" reports wrong processor speed.
http://support.microsoft.com/default.aspx?scid=kb;en-us;303602

3) The Processor speed may be reported incorrectly in Windows XP.
http://support.microsoft.com/default.aspx?scid=kb;en-us;316965

HTH,
Paul

 

[Addy]

The overclocking comment is here:
http://abxzone.com/forums/showpost.php?p=768049&postcount=7

This thread mentions the ingredients of a working CnQ:
http://abxzone.com/forums/showthread.php?t=67457&highlight=cool+quiet+k8v

No idea if this is the proper order of execution or not I
added item 4 to the list and added comments.

1. download and install the cool n quiet driver and software from
the ASUS website. The AMD driver is here, as well as on the CD.
http://www.amd.com/us-en/assets/content_type/utilities/Cpudriver.ZIP
2. enable "Cool n Quiet" and the ACPI 2.0 in the BIOS
3. set the power setting to minimal management
(In XP under Power Management select "minimal" as the scheme)
4. If it doesn't work, upgrade to 1005.011 BIOS or later, as the people
in the thread found it works OK with that version.

From the AMD Cpudriver.ZIP readme file:

REQUIREMENTS:
AMD Athlon(tm) 64 systems with ACPI 2.0 performance control
objects (_PCT) on Windows XP <-----
DESCRIPTION:
This driver supports processor performance control via the
processor objects defined in the ACPI 2.0 specification.
KNOWN LIMITATIONS:
1) The processor performance state may not be restored to the
maximum state if the CPU runs at 100 percent. This problem occurs
if the computer is 100 percent busy when the power policy changes.
http://support.microsoft.com/?kbid=330512

2) "System Properties" reports wrong processor speed.
http://support.microsoft.com/default.aspx?scid=kb;en-us;303602

3) The Processor speed may be reported incorrectly in Windows XP.
http://support.microsoft.com/default.aspx?scid=kb;en-us;316965

HTH,
Paul

Thanks again for the intesive search. It seems the Overclocking phenomenon
is not my problem. Entering Bios It showed the right FSB settings.

The BIOS is not for my version of the board so it doesn´t help either.

I keep searching.

Addy

 

[Drew R.]

I have this board and ran into this problem. I did not find that it
was related to the CnQ and am still trying to track it down. I got
this error shortly after installing 1003. I also got CRC errors when
extracting files from archives (regarless of where or how I obtained
them). I believe this may a memory problem. I get memory errors with
both DIMMS installed (512MB x2) but not with one. The last thing I
found was my memory may be over-clocked - although it is set to
default timings. I will likely slow my memory down and re-test
tonight.