XP Setup hangs with dual PIII-550 on Asus P2B-DS v1.04

[Nir Gilboa]

Hi,

I'm trying to install XP (Sp2) on an Asus P2B-DS v1.04 with dual
PIII-550. The bus speed is set to 100 FSB, the cpus are both known to be
OK, and I have not made any hardware modifications to the board.

I use F5 to to select MPS Multiprocessor. The setup hangs somewhere
after the "Copying Files" part but never in the same place exactly. I
don't get a BSOD or anything at all - it just hangs.

Installing with the "Standard PC" HAL works fine on both of the CPUs.
Also I've been able to determine this has nothing to do with any
hardware attached after removing everything that was connected to the mb.

I think it's clear that it is an issue with using dual processors. Can
someone help ? any ideas are much appreciated.

Also I've tried to find concise information on how to fix hardware bugs
on the board (I guess ACPI and USB) but all i found was some
instructions in german.. If anyone cares to explain in a bit of detail
that would also be very much appreciated.

Thanks,
Nir Gilboa.

 

[Dave]

I have seen this before.
In my case it turned out to be the CDROM XP was installing from.
Assuming you have a fairly new drive and an original copy of XP - all
should go well.
If not, try a new CDROM + the original (not copied) XP cd.

D.L

 

[Paul]

I have seen this before.
In my case it turned out to be the CDROM XP was installing from.
Assuming you have a fairly new drive and an original copy of XP - all
should go well.
If not, try a new CDROM + the original (not copied) XP cd.

D.L

I tried a Google on "p2b-ds acpi bios" and the results I got, hint
that perhaps certain BIOS revisions don't work well with ACPI. Since
Asus doesn't keep the release notes for the old releases, it is
hard to say what the history of ACPI and SMP is.

http://groups.google.ca/group/maili..._frm/thread/88047c74854e25de/ba2a5184ebd6ad36

There is also a resistor mod, to get the reported CPU utilization
to work properly in Win2K. i don't know if the same CPU utilization
bug exists in WinXP or not.

http://groups.google.ca/group/alt.c..._frm/thread/65c3c59da8cfbc49/9466aa9102d61c7d

This is a good site for P2B-DS owners. If anyone knows the answer to
your questions, it is the guy at the email address at the very bottom
of this page

http://www.tipperlinne.com/p2bmod

HTH,
Paul

 

[Michael W. Ryder]

Hi,

I'm trying to install XP (Sp2) on an Asus P2B-DS v1.04 with dual
PIII-550. The bus speed is set to 100 FSB, the cpus are both known to be
OK, and I have not made any hardware modifications to the board.

I use F5 to to select MPS Multiprocessor. The setup hangs somewhere
after the "Copying Files" part but never in the same place exactly. I
don't get a BSOD or anything at all - it just hangs.

Installing with the "Standard PC" HAL works fine on both of the CPUs.
Also I've been able to determine this has nothing to do with any
hardware attached after removing everything that was connected to the mb.

I think it's clear that it is an issue with using dual processors. Can
someone help ? any ideas are much appreciated.

Also I've tried to find concise information on how to fix hardware bugs
on the board (I guess ACPI and USB) but all i found was some
instructions in german.. If anyone cares to explain in a bit of detail
that would also be very much appreciated.

Thanks,
Nir Gilboa.

Have you tried copying the XP install disk to the hard drive and
installing from there? Microsoft told me to do this after trying to
install XP on my P2B-S. I realize that my computer was not the dual CPU
version, but the problem may be with the SCSI controller.

 

[Nir Gilboa]

Hi

Thank you Dave, Paul and Michael.
I will try to install from the original XP cd, although what I've tried
from is a copy of my Volume License cd from work. I'll also try to copy
all the installation files to disk and post again if it doesn't work.

I'm still open to other ideas in cas this doesn't work...
Nir

 

[Stephan Grossklass]

Hi,

I'm trying to install XP (Sp2) on an Asus P2B-DS v1.04 with dual
PIII-550. The bus speed is set to 100 FSB, the cpus are both known to be
OK, and I have not made any hardware modifications to the board.

I use F5 to to select MPS Multiprocessor. The setup hangs somewhere
after the "Copying Files" part but never in the same place exactly. I
don't get a BSOD or anything at all - it just hangs.

Hmm. Do you have the latest BIOS (1014 beta 3)?

Also I've tried to find concise information on how to fix hardware bugs
on the board (I guess ACPI and USB) but all i found was some
instructions in german.. If anyone cares to explain in a bit of detail
that would also be very much appreciated.

To fix the ACPI problem, all you have to do is moving an SMD resistor
near the APIC. (Of course that's a gross understatement if you're not
routined in SMD soldering.) The FAQ page covering this issue is here:
http://rma.asus.de/support/faq/faq075_W2K_ACPI_dual-mb.htm
I don't think there ever was anything official for the USB issues. But
searching did turn up this:
http://www.hole.fi/~dogo/p2b-usb/
Apparently you need to replace R112 by a zero ohm resistor.
I think the older board revisions were also affected by the bus
termination voltage issue discussed here:
http://www.angelfire.com/electronic2/locksmith/asusreworkguide.html

Stephan

 

[P2B]

I tried a Google on "p2b-ds acpi bios" and the results I got, hint
that perhaps certain BIOS revisions don't work well with ACPI. Since
Asus doesn't keep the release notes for the old releases, it is
hard to say what the history of ACPI and SMP is.

http://groups.google.ca/group/maili..._frm/thread/88047c74854e25de/ba2a5184ebd6ad36

There is also a resistor mod, to get the reported CPU utilization
to work properly in Win2K. i don't know if the same CPU utilization
bug exists in WinXP or not.

http://groups.google.ca/group/alt.c..._frm/thread/65c3c59da8cfbc49/9466aa9102d61c7d

This is a good site for P2B-DS owners. If anyone knows the answer to
your questions, it is the guy at the email address at the very bottom
of this page

http://www.tipperlinne.com/p2bmod

HTH,
Paul

I got mail!

I've never seen this problem, but then I've never tried second guessing
XP's choice of HAL. Left alone, it will pick 'ACPI Multiprocessor PC'
for the P2B-DS, which IME works just fine in conjunction with BIOS
1014beta3. I suspect the OP's attempt to force 'MPS Multiprocessor' is
causing his grief. We'll see.

The CPU utilisation issue also affects XP, but as you suggest it's
merely a reporting bug. Doing the resistor mod to fix it has no effect
on performance.

P2B

 

[Daniel Mandic]

Hi Gilbert!


I always press F5 when installing NT5 or NT5.1.

I have found out that it is possible to press F5 and F6, at setup. So
you can add SCSI and turn on Standard.
You might need MPS and Standard-PC. I always prevent the setup from
loading the ACPI. ACPI is more for paranoid computer-user who wants to
save something..... You can also shut down and power off the machine,
when not in need.

I hope you will find a non-hardware solution to get the setup doing MPS
without ACPI :-(



Best Regards,

Daniel Mandic



P.S.: F5 is not documented... You have to press it while setup shows
"press F6 for add. Controller..". Press it sometimes (two or three- not
too much).... one time might be too little. Or, one good press should
do. As I said, you can press F6 too, as well....... ;-)

 

[Mercury]

I had / have a P2B-DS that was running NT4, W2k, then W2K3 24x7 for years
without a problem.
I did get the CPU reporting issue fixed and agree that it is cosmetic.
I never F5'd anything in my life and think this really is for the exceptions
rather than the rule.

XP SP2 is very much akin to W2K3.

HTH.

 

[Roger Hamlett]

I had / have a P2B-DS that was running NT4, W2k, then W2K3 24x7 for years
without a problem.
I did get the CPU reporting issue fixed and agree that it is cosmetic.
I never F5'd anything in my life and think this really is for the
exceptions rather than the rule.

XP SP2 is very much akin to W2K3.

HTH.

I have to agree. F5, is normally used if you have a DP motherboard and
want to install in single processor mode, or if it's dual nature is not
correctly recognised. The P2B-D, is normally OK in this regard.
A question to the original poster. What drives are installed?.
I have seen this 'hang' behaviour, on the -DS, when trying to install the
machine with no IDE drives present. Removing the hardware wouldn't help in
this situation!.

Best Wishes

 

[Nir Gilboa]

Hi

I've tried letting the Setup program determine the HAL, but what happens
is it hangs at "Setup is starting Windows..." even before entering the
graphical part.

I've also tried making xp setup disks and doing the same thing - letting
setup determine the HAL. Same thing happens - hangs at "Setup is
starting Windows...". The second time I tried this I got a BSOD that
says PAGE_FAULT_IN_NONPAGED_AREA with 0x0000050 - and I couldn't find
that error on MS's support site.

Also i'd like to mention I am using BIOS 1014.Beta003 with MPS 1.4
disabled and that the processors are exactly identical SL3F7 and SL3FJ.

Another thing I tried over the weekend is opening up the pc and cleaning
dust off the Slot 1's, the mb and the SEC cartridges, but no luck so far.

These results suggest something might be wrong with the motherboard -
but believe me it has been working just fine with XP Pro (SP2) and one
PIII550. Also I used to run W2K with dual PII350 without so much as a
hiccup.

What else can I possibly do to troubleshoot this problem ? Any help is
appreciated - I am totally stumped here.

Nir

 

[Roger Hamlett]

Hi

I've tried letting the Setup program determine the HAL, but what happens
is it hangs at "Setup is starting Windows..." even before entering the
graphical part.

I've also tried making xp setup disks and doing the same thing - letting
setup determine the HAL. Same thing happens - hangs at "Setup is
starting Windows...". The second time I tried this I got a BSOD that
says PAGE_FAULT_IN_NONPAGED_AREA with 0x0000050 - and I couldn't find
that error on MS's support site.

Also i'd like to mention I am using BIOS 1014.Beta003 with MPS 1.4
disabled and that the processors are exactly identical SL3F7 and SL3FJ.

Another thing I tried over the weekend is opening up the pc and cleaning
dust off the Slot 1's, the mb and the SEC cartridges, but no luck so
far.

These results suggest something might be wrong with the motherboard -
but believe me it has been working just fine with XP Pro (SP2) and one
PIII550. Also I used to run W2K with dual PII350 without so much as a
hiccup.

What else can I possibly do to troubleshoot this problem ? Any help is
appreciated - I am totally stumped here.

Nir

You haven't said what disk configuration you have?. As I have already
posted, I have seen this type of behaviour, when I tried to configure such
a system with only SCSI drives attached, or with only an IDE CD.
Seperately, has this board got the fix for the interrupt problem?. There
was a fault on these boards, which in an ACPI configuration, led to
non-stop interrupts being generated, hanging/killing the system. This was
fixed by moving one resistor on the board. The default XP install would
use ACPI, and since this appears to make the fault worse, it may be that
your board has not got the fix. There used to be board pictures of the
rework on the Asus site. This problem only showed in SMP configurations,
so might be related....

Best Wishes

 

[Daniel Mandic]

Hi

I've tried letting the Setup program determine the HAL, but what
happens is it hangs at "Setup is starting Windows..." even before
entering the graphical part.

I've also tried making xp setup disks and doing the same thing -
letting setup determine the HAL. Same thing happens - hangs at "Setup
is starting Windows...". The second time I tried this I got a BSOD
that says PAGE_FAULT_IN_NONPAGED_AREA with 0x0000050 - and I couldn't
find that error on MS's support site.

Also i'd like to mention I am using BIOS 1014.Beta003 with MPS 1.4
disabled and that the processors are exactly identical SL3F7 and
SL3FJ.

Another thing I tried over the weekend is opening up the pc and
cleaning dust off the Slot 1's, the mb and the SEC cartridges, but no
luck so far.

These results suggest something might be wrong with the motherboard -
but believe me it has been working just fine with XP Pro (SP2) and
one PIII550. Also I used to run W2K with dual PII350 without so much
as a hiccup.

What else can I possibly do to troubleshoot this problem ? Any help
is appreciated - I am totally stumped here.

Nir

--

 

[Daniel Mandic]

Nir Gilboa wrote:

port site.

Also i'd like to mention I am using BIOS 1014.Beta003 with MPS 1.4
disabled and that the processors are exactly identical SL3F7 and
SL3FJ.

Hi!



I also use a P3B-F. I cannot disable S.M.A.R.T., or the onboard USB -
Probably the 1008 beta bios.

I am going to change to an official BIOS, without P3 Tualatin Support.
My Tualatin is here in the P2B-F and the beta bios is working more than
properly, exactly as I need. I stuckked with Award 4.51PG, the best
IMO. Award 6 and above drives me crazy... why do the producer place
buttons in the BIOS when they don´t work.

Well,.. as you don´t have a Tualatin you might change the bios. The 550
PIII don´t need the beta.




Best Regards,

Daniel Mandic

 

[P2B]

Hi

I've tried letting the Setup program determine the HAL, but what happens
is it hangs at "Setup is starting Windows..." even before entering the
graphical part.

I've also tried making xp setup disks and doing the same thing - letting
setup determine the HAL. Same thing happens - hangs at "Setup is
starting Windows...". The second time I tried this I got a BSOD that
says PAGE_FAULT_IN_NONPAGED_AREA with 0x0000050 - and I couldn't find
that error on MS's support site.

It's there - you have to delete the leading zeros when searching stop
errors:

http://support.microsoft.com/default.aspx?scid=kb;en-us;171003

It says 0x50 during installation may be caused by faulty RAM. That
doesn't explain why it works with one processor, but the fact
installation attempts don't always fail in the same manner is consistent
with a RAM issue.

I suggest running memtest86 with all tests enabled overnight. IME if
memtest86 reports even a single error, the chances of OS installation
completing successfully are low.

Also i'd like to mention I am using BIOS 1014.Beta003 with MPS 1.4
disabled and that the processors are exactly identical SL3F7 and SL3FJ.

Another thing I tried over the weekend is opening up the pc and cleaning
dust off the Slot 1's, the mb and the SEC cartridges, but no luck so far.

These results suggest something might be wrong with the motherboard -
but believe me it has been working just fine with XP Pro (SP2) and one
PIII550. Also I used to run W2K with dual PII350 without so much as a
hiccup.

What else can I possibly do to troubleshoot this problem ? Any help is
appreciated - I am totally stumped here.

I suspect a processor seating issue - slot 1 processors can be a pain,
especially on older boards that may have seen many processor insertion
cycles. The SMP signals between processors are particularly sensitive,
and any contact resistance can cause your symptoms.

Try cleaning the processor edge connectors and processor slots with
isopropyl alcohol - 95%+ is best, but you can use 70% (rubbing alcohol)
provided you allow time for the 30% water to evaporate. The slots can be
cleaned with a piece of cardboard the same thickness as the processor
edge soaked in alcohol - if there's oxidation of the contacts, the
cardboard comes out with black marks on it.

Make sure your processor retention mechanisms are locked and holding the
processors securely. Sometimes it helps to fully insert the processor
into the slot, then withdraw it 1mm or so before locking the retention
mechanism.

You could also try underclocking during installation - if it completes
at 66Mhz FSB but not 100Mhz, you have hardware misbehaving at the higher
bus speed.

HTH

P2B

 

[Nir Gilboa]

Hi

Thank you very much to everyone who tried to help. You guys really are
at least as crazy as me for spending time on this

Roger: The mb does not have any modifications made to it. From what you
are saying it does sound like I should fix the ACPI issue. I found a few
places that make reference to moving the R79 resistor to R80. If you
find those board pictures somewhere i'd be happy to try it.

Robert: I have one IDE 80Gb drive attached to Primary IDE and one CD-ROM
on the secondary IDE.

P2B: It might be a memory issue, but in the one CPU config the
installation always works without a hitch, which leads me to believe the
memory issue if it exists is not the main problem.

Cleaning the slots and underclocking are also good ideas and i will
definitely try them. Do you happen to know if moving the R79 to R80 is
the correct fix for ACPI and what it actually does ?

---

I have been able to find a workaround - I install with one CPU and force
the MPS Multiprocessor HAL. The installation completes successfully. I
then reboot after re-inserting the second CPU. XP loads fine and when
opening the task manager i could see two working processors, and
accurate usage percentages. No 50% use bug or anything.

The second time i tried to open task manager the pc simply rebooted. No
BSOD, no error in the event log, NADA. As if I pressed the reset button

Another "feature" of this setup is that hangs. XP hangs usually after
about 2-3 minutes with only a reset helping.. There is an error in the
System log which says: "Error code 10000050, parameter1 ffbfaacc, param2
00000000, param3 bf830313, param4 0000000" Also I do see the "Windows
has recovered from a serious error" and all the BCCodes and hex numbers
but i have no idea how to use them to understand what is wrong.

Interesting, isn't it ? I think the next thing I will try is cleaning
the slots and underclocking. If that doesn't work, I think i'll just
have to sacrifice the mb and processors to big orange flames Or I
could settle with using just one CPU heh

Anyway thanks again to everyone - if you still have ideas i'd love to
hear them.

Nir

 

[P2B]

You haven't said what disk configuration you have?. As I have already
posted, I have seen this type of behaviour, when I tried to configure such
a system with only SCSI drives attached, or with only an IDE CD.

I've never connected an IDE device to any of my P2B-DS boards, and have
never had such problems. However, I agree knowing details of the OP's
disk configuration could well facilitate the diagnosis.

Seperately, has this board got the fix for the interrupt problem?. There
was a fault on these boards, which in an ACPI configuration, led to
non-stop interrupts being generated, hanging/killing the system. This was
fixed by moving one resistor on the board. The default XP install would
use ACPI, and since this appears to make the fault worse, it may be that
your board has not got the fix. There used to be board pictures of the
rework on the Asus site. This problem only showed in SMP configurations,
so might be related....

That was strictly a cosmetic issue, at least in W2K, according to Asus -
the CPU meter never dropped below 50%.

I once had a dozen or so P2B-DS 1.04 boards sent in at once for the ACPI
fix, so I did some testing with the first one. I installed W2K (no
problem), confirmed it had picked the ACPI Multiprocessor HAL, observed
the CPU meter pegged at 50% on an idle system, and ran benchmarks. After
moving the resistor, the CPU meter dropped as expected but there was no
significant change in the benchmark results.

I have not run an equivalent test using XP, but I would expect similar
results.

P2B

 

[Daniel Mandic]

Anyway thanks again to everyone - if you still have ideas i'd love to

hear them.

Nir

Try to install Microsoft Windows NT4 Workstation- or Server. Just to
see if this Error occurs, too. Or Win2000.

Good Luck.




Best Regards,

Daniel Mandic

 

[P2B]

Hi

Thank you very much to everyone who tried to help. You guys really are
at least as crazy as me for spending time on this

Roger: The mb does not have any modifications made to it. From what you
are saying it does sound like I should fix the ACPI issue. I found a few
places that make reference to moving the R79 resistor to R80. If you
find those board pictures somewhere i'd be happy to try it.

Robert: I have one IDE 80Gb drive attached to Primary IDE and one CD-ROM
on the secondary IDE.

P2B: It might be a memory issue, but in the one CPU config the
installation always works without a hitch, which leads me to believe the
memory issue if it exists is not the main problem.

I tend to agree, but running memtest86 overnight before installing the
OS is standard procedure here. If it hangs, crashes, or reports errors
there's a hardware issue (not always memory). If it's still running
clean by morning, chances are the hardware is good to go.

Cleaning the slots and underclocking are also good ideas and i will
definitely try them. Do you happen to know if moving the R79 to R80 is
the correct fix for ACPI and what it actually does ?

You have that backwards, the fix is moving R80 to R79, but I don't know
exactly why it fixes the problem. It sounds like your board already has
the fix, but I posted a picture of the location so you can check - this
one has been fixed:

http://tipperlinne.com/images/acpi-fix.jpg

I have been able to find a workaround - I install with one CPU and force
the MPS Multiprocessor HAL. The installation completes successfully. I
then reboot after re-inserting the second CPU. XP loads fine and when
opening the task manager i could see two working processors, and
accurate usage percentages. No 50% use bug or anything.

The second time i tried to open task manager the pc simply rebooted. No
BSOD, no error in the event log, NADA. As if I pressed the reset button

Classic symptom of dodgy connections between the processor and slot on
these boards.

P2B

 

[Roger Hamlett]

I've never connected an IDE device to any of my P2B-DS boards, and have
never had such problems. However, I agree knowing details of the OP's
disk configuration could well facilitate the diagnosis.

Have you seen it take a very long time booting into W2K/WXP?. If so, this
is caused by the same problem. It didn't happen on the first
boards/BIOS's, but some time latter, there was a BIOS change, that caused
this problem. I went through a dozen P2B variants, and several BIOS
versions, and all would display it in some configurations. It was however
very 'odd' in terms of the patterns that gave it. For example, I had a
system that ran W2K fine, once installed, but would always crash on the
install, if using an IDE CD, but copying the disk to the SCSI HD, and
installing from this worked fine. Another system installed perfectly the
first time, but changing to a latter CD drive, introduced this behaviour.
Five drives latter, from different manufacturers, it evntually worked
again. The problem particularly appeared when there was an IDE CD from
particular manufacturers, but only SCSI HD's.
There was also a seperate issue, that might apply in this case, relating
to the video card, and the SCSI BIOS. Some video cards caused a problem
with the SCSI BIOS on these machines during boot. There is an area in the
upper memory block, that is incorrectly flagged during bootup, and in some
cases leads to both the SCSI system, and the video, trying to map stuff to
the same address. Some families of video cards, never really worked...

That was strictly a cosmetic issue, at least in W2K, according to Asus -
the CPU meter never dropped below 50%.

Hardly 'cosmetic'. It was caused by the ACPI interrupt continuously
triggering, and the CPU handling these, basically sits 'flat out' doing
little else. It normally did not cause install problems on W2K, but I have
seen it give trouble on XP.

I once had a dozen or so P2B-DS 1.04 boards sent in at once for the ACPI
fix, so I did some testing with the first one. I installed W2K (no
problem), confirmed it had picked the ACPI Multiprocessor HAL, observed
the CPU meter pegged at 50% on an idle system, and ran benchmarks. After
moving the resistor, the CPU meter dropped as expected but there was no
significant change in the benchmark results.

I have not run an equivalent test using XP, but I would expect similar
results.

P2B

Depends on the benchmark. Most benchmark tests, only use one processor.
Try a compile that supports SMP, and the time needed reduces directly in
line with the change in processor useage.

Best Wishes