ASUS P5N-D+Q6600 - S.O.S. - new system shuts down playing ANY game!

[agentdcooper]

hello all - please help!

I have a brand new system, I have given this 29 days of
troubleshooting, and it is my last day before I have to apply for an
RMA on my 30day newegg.com return policy for a couple products I
purchased... My system will post, boot into windows --- running vista
x64 ultimate SP1 and windows xp SP2/32bit in dual boot setup --- once
in windows I can run ANY application, video, etc without problem, for
days on end == but as soon as I load up a game [and that is _the_ main
reason I built this box], my system SHUTS down, out of nowhere! No
slow down, no warning, just *boom* power off!!$@#
I am an old-school troubleshooter, and know my shit pretty well, but
this has me stumped! someone please help a fellow out =)

....the problem is OS independant, RAM independant, _probably not heat
related_, just plain has me perplexed....

Games I can start up ANYTIME, and the problem occurs [actually every
game I've tried so far];
Call of Duty 4 - direct X 10 and DX9
Crysis [64bit and 32bit] - direct X 10 and DX9
Gears of War - direct X 10 and DX9
Bioshock - direct X 10 and DX9

- I've ran different OSs, same problem [vista x64 ultimate, now SP1
and windows xp SP2/32bit]
- I've swapped between, in and out trying each single stick, to using
all 4, 4x1G sticks of ram [CORSAIR 1Gx2 TWIN2x2049-6400 R, and
KINGSTON 1Gx2 KVR800D2N5K2/2G]
- tried every one of my 4x PCI-E 6pin connectors connecting one at a
time to my video card --- the power supply is 650watt hec zephyr SLI/
crossfire supported, I figured 4x+12V rails @ 17A/each should be fine
for a single 8800 GTS 512MB??
- tried adding 5 more case fans, positioning them and repositioning
them, to just currently putting the motherboard/video/ram and all
outside of the case, exposed to the cold air in my basement [62degree
room temp], on an anti-static sheet and foam, so to be sure its not a
grounding issue in the case or something...
- uninstalled/reinstalled all different versions of drivers nvidia,
and "tweaked" others - currently running stock nvidia 169.44 on vista
x64 ultimate downloaded from evga's site, and the current nvidia
drivers for XP from evga's site
- reinstalled OS multiple times, currently on a fresh minimal installs
of both OSs [installed indepently first, now dual booting]
- tried OC'n, stock, everything in between = ASUS overclock speed/FSB/
voltage settings... I've read you need to specify voltage config for
Vcore/HT/NB/SB/DIMMs with this motherboard and processor, I've tried
this using all stock settings from manufactures' specifications, as
well as starting from incrementing voltage in steps of .05 increments
where applicable
- specified memory timings, and tried AUTO
- disabled all unnecessary things in BIOS to try and strip it down,
and tried defaults, and custom settings....
- making sure all fan settings are set to performance, then disabled
-- many differnt combos!
:: yes, I crank up the nvidia 8800 onboard fan to 100% via ntune
every time I boot
:: CPU fan is _huge_ arctic cooling freezer 7 pro, and seems to do a
good job
:: again the whole machine is outside the case, exposed to the air,
so it says nice n cool, with a couple fans pointed at it...
- no add-on cards are in the system other then video card, everything
else is onboard, or just ram+DVD and hard drive
- I've enabled and checked all "event/error/application" logs vista
x64 has default, tried using the tool "everest" and enabled logging
every 1 second, playing a game, while pulling EVERYTHING, also
speedfan and nvidia monitor has event logging, whcih isn't telling me
anything either... I am open to trying any other program that may help
me diagnose/isolate the issue

I'm not sure what to think, is it the power supply, video card,
motherboard, CPU, or ram??? somebody please help ='(

is there anything out there that I could use to test the power supply
itself, and see at what point it shuts off?? it just seems like it
would have to be power, heat, or motherboard/cpu related but heat just
doesn't seem to be the answer - I boot up and idle around motherboard
temp @ 23C, CPU @ ~20-26C, video card @ ~22-26C and I seem to top out
at = motherboard @ 25C, CPU @ ~24-30C, video card @ ~28-33C --- this
isn't too hot is it?

;;for reference i have some files uploaded that may give ya an idea
what I am dealing with here;;
nvidia ntune system information output
http://nen-tftp.techiekb.com/cooppost/sysinfo.log

cpu-z html output (OC'd to 3.0GHz)
http://nen-tftp.techiekb.com/cooppost/cpuz.htm

example Everest log that captures the power down issue [last entry]
http://nen-tftp.techiekb.com/cooppost/everest-log_2008-03-20_02-21-40_log.htm

another example Everest log capture with power down issue [last entry]
http://nen-tftp.techiekb.com/cooppost/everest-log_2008-03-20_02-08-09_log.htm

SpeedFan log capture during power down issue
http://nen-tftp.techiekb.com/cooppost/SFLog20080317-0001.csv

another SpeedFan log capture during power down issue
http://nen-tftp.techiekb.com/cooppost/SFLog20080316.csv

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
;;my PC build as of 2/20/2008 ;;

P5N-D nForce 750i SLI Motherboard
http://www.newegg.com/Product/Product.aspx?Item=N82E16813131232ASUS

Intel Core 2 Quad Q6600 Kentsfield 2.4GHz CPU
http://www.newegg.com/Product/Product.aspx?Item=N82E16819115018

CORSAIR XMS2 2x1GB DDR2 800MHz - TWIN2X2048-6400 (1 came DOA)
http://www.newegg.com/Product/Product.aspx?Item=N82E16820145590

Kingston 2x1GB DDR2 800MHz PC6400 - KVR800D2N5K2/2G
http://www.enuinc.com/m-02g-080-kin-k2n5.html

EVGA GeForce 8800GTS (G92) 512MB 256-bit GDDR3 Video Card
http://www.newegg.com/Product/Product.aspx?Item=N82E16814130325

hec Zephyr 650 ATX12V 650W Power Supply
http://www.newegg.com/Product/Product.aspx?Item=N82E16817339003

COOLER MASTER Centurion 5 CAC-T05-UW Computer Case
http://www.newegg.com/Product/Product.aspx?Item=N82E16811119068

Western Digital Caviar SE SATA2 WD800AAJS Hard Drive
http://www.newegg.com/Product/Product.aspx?Item=N82E16822136195

SAMSUNG 20X DVD±RW w/LightScribe SATA Model SH-S203N
http://www.newegg.com/Product/Product.aspx?Item=N82E16827151154

1x - Antec 120mm TriCool Blue case fan
http://www.enuinc.com/fan-120-ant-blu.html

3x - Antec 80mm Ball Bearing case fans
http://www.enuinc.com/fan-080-ant-001-bb.html

 

[agentdcooper]

Thanks for the reply! Any help is good help, the more brains the
better =)

The way usenet works, and free time working around sleep and
work and family, etc, you could easily need more than one
day even if we posted something that solves a problem.

Get the RMA NOW. You don't have to actually RMA it now, but
because you are running out of time you should go ahead and
get it then if you get the problem fixed, just let the RMA
expire instead of sending them the parts (contact them to
cancel it later if necessary).

excellent suggestion, this is exactly what I did today! thank-you
newegg! RMA'n memory [1 was DOA, but have to replace @ same time], the
video card and PSU. I am going with replacements for RAM and video
card, but requesting refund on the PSU -- I am going to replace the
hec Zypher 650W PSU for this instead = http://www.newegg.com/Product/Product.aspx?Item=N82E16817341010
-- it's an OCZ StealthXStream OCZ600SXS 600W Power Supply, reviews are
good, sounds like well-built PSU, I was very hesitant in the first
place to buy the hec PSU, but we shall see if it was the source of the
trouble...

Run Memtest86+ for a few hours. Run Orthos for at least 60
minutes. If either of these show any error you need correct
that subsystem fault. More likely, when the game starts and
the video card enters 3D mode it consumes more power and
this is enough to overwhelm your PSU. I'd have thought an
HEC 650W would be enough but it may be a budgetized design
or have some flaw or failure. Check voltages with a
multimeter if possible.

Everest stability tests, ATI Tool stability tests, and ntune stability
tests have been run, nothing seems to knock the computer off-line
except starting up ANY game, sometimes I can play a game for 2mins,
then iit powers the system off. when I say "powers the system off", I
mean it just shuts down with no warning, to an OFF state - I have to
push the power button, the screen just goes black, all fans stop - no
blue screen or warning.

Also do the standard windows things like checking Event
Viewer.

Event viewer just says the game took over desktop manager, then ends
there -- not much use, I turned on ALL reporting/verbosity(sp?), with
1second updates

What exactly does "shuts down" mean? Does it:

A) Reset and hang
B) Reset and boot
C) No reset, just hang/crash, then is there a bluescreen?
D) Turn off, then the switch on case front can turn it on.
E) Turn off, then PSU AC power switch or cord must be
pulled before it'll turn on again.
D.

You write probably not heat related. Do you or don't you
know it is staying cool enough, based on temp readings?
Speedfan is one app that will monitor temps.

idle motherboard temp @ 23C, idle CPU @ ~20-26C, idle video card @
~22-26C and I seem to top out during full load at = motherboard @ 25C,
CPU @ ~24-30C, video card @ ~28-33C

Games I can start up ANYTIME, and the problem occurs [actually every
game I've tried so far];
Call of Duty 4 - direct X 10 and DX9
Crysis [64bit and 32bit] - direct X 10 and DX9
Gears of War - direct X 10 and DX9
Bioshock - direct X 10 and DX9

Forget about DX info, leave out things nonapplicable. Make
sure windows is not set to reboot on error (Automatically
reboot on system failure).

only mentioned it to state I've tried the games in both vista x64 and
xp 32bit

Focus on WinXP SP2 only for now. By doing so you reduce the
variables and use the OS with the most issues resolved
already through length of time it's been out and patched.

I like this idea too - I tried both OSs hoping more debugging info to
read...

Leave only 1 memory module installed until the problem is
resolved. The key is reduction in variables, there could
actually be more than one thing wrong with a new build and
it is best to resolve your most significant problem with
other things as basic and minimal as possible. After you
fix the problem, THEN add back the other 3 modules and
retest with Memtest86+ overnight to reconfirm stability.

excellent advice!

Do not put parts on
anti-static sheets and foam, that will cause problems.
Anti-static sheets and foam are often deliberately designed
to conduct electricity, which is how they are static
dissipative. Often it won't cause a problem but it can
introduce further problems in some situations.

I appreciate the concern, I moved to cardboard instead, with
components exposed to air for me to troubleshoot/rule out heating
issues -- basement room temp = ~63F.

Forget about software/drivers/etc. It is not software
providing you have set windows to not shut down on error,
and that you then don't get a crash dump, bluescreen, etc.

I agree.

Forget about reinstalling OS. I'm not trying to be critical
of the process, just concise in what can be rulled out.

I appreciate the help! very much so

Leave system at stock, and confirm there is no bios setting
for temperature shutdown that is set too low, and that the
temp reading seems correct (some bios may not register temp
right for some CPUs). If the temp seems off by a LOT, seek
a bios update. Actually see what bios updates are available
either way and check their notes to see if anything seems
applicable to your parts.

BIOS is current, ASUS shows most recent = 02/22/08, thats what I got.
everything is at stock right now.

Check your bios memory and MCH voltage. Try bumping memory
to 2.05V at most (this with only one module installed
still), that is all it should need at stock speed.

Yep, been over and over voltage settings... days worth of tweaking! I
just don't think it's BIOS voltage settings...

I'd almost forgotten, examine all cards, cables, etc, the
whole interior of the system and reseat any cards you doubt,
including video card. Also pull AC power, remove battery
for 10 minutes and use clear CMOS jumper. Only set minimal
required things in bios from then on instead of a final
tweaked config.

done, done and done again =)
been over this _several_ times

Leave it at stock speeds till problem is resolved.
check.

but you don't tell us exactly how long, EXACTLY what moment
it happens and exactly what you see on screen then.

ANY game I try to load, boots and as soon as the start screen comes
on, or sometimes it will let me play the game for 2mins at MOST, the
computer shuts down, no warning - power off, fans off, screen black -
no moving parts... no blue screen, just black.

Don't use any kind of ntune or other software that interacts
with anything. If still in doubt about temps, leave the
case side panel off and point a fan at the video card
temporarily.

GPU fan is controled via ntune, by cranking the fan from 35%/default
to 100%/custom the GPU cools down 10-15C everytime, it is a precaution
I take to make sure it's not the GPU overheating or something, but I
have tried this with default/stock setting and with it bumped to 100%.

If still on the anti-static material, get it off of that.

ya, moved the motherboard to cardboard, PSU is still in the case,
screwed in...

We still don't know at what exact moment it's crashing or
whatever it's doing. We have a lot of info about what you
tried but not what the very specific exact thing you observe
is, only a summary of it not the detail of what eyes see,
ears hear, etc. Remember we are not there to know what you
see.

I hope my answers above makes it clearer now...

Forget about Vista X64 for now. It's a wildcard. Stick
with the simplest config, a clean XP 32bit installation with
only drivers, NO other hardware monitoring or logging or
ntune tweaking anything at all. Only in a bare minimal
config can you be sure nothing else is causing problems. I
realize the purpose for the logging but I mean now, since
nothing significant is being logged, tuned or adjusted, it
has shown to be of no use and could only have negative
consequences if it remains.

agreed - thats where it stands currently... still with the problem =(

Based on what's been provided and playing odds, it's most
likely the PSU is problematic. If you've a multimeter you
should hook ...

no multimeter unfortunately... I am taking a gamble and am gonna
return the PSU for the OCZ StealthXStream OCZ600SXS 600W Power
Supply , with fingers crossed this and the RAM+video card RMA will
resolve the issue. I am stumped... if someone like you can help me
troubleshoot it without sending it back, I'd be tickled pink!

thanks for the assistance, hope I answered your questions clearly!

(e-mail address removed)

 

[agentdcooper]

OPPS, I cut the last part of yr message off, google groups had a 'read
more' option I din't see at first...

Your temps are not correct. Motherboard is not 23C, maybe a
sensor meant to measure chassis temp is but not northbridge
which is the relevant motherboard temp. CPU is not 20-26C
under load which is the relevant time (like when running
Orthos for 30 minutes). 8800GTS video card is never going
to be 22-26C, not ever (Unless you installed these parts in
your freezer) so I think you need a different temp
monitoring program.

Likewise with the "top out" temps they are impossibly too
low with one exception, if your room temp were 62F as
mentioned previously, it is possible your CPU idle temp
could be 26C. It is not possible anything else would be so
low under load unless using very good water cooling.

I think you are right, I think it's more closer to 70F idle, 85+ under
load... but I will check that again with another program...

Do not overclock anymore until your problem is resolved!
done.

You must first determine if the parts are stable at stock
speeds and remain stable. After that has happened,
including games, then determine what o'c you can achieve and
fiddle with settings.

Please no more logs of an overclocked system. Overclocking
comes _after_ everything else is 100% working and is always
the first thing to undue when there is a problem like this.
That especially applies to the video card if it is also
overclocked.

On the issue of clock changing, if all else fails you might
consider underclocking. Do that by decreasing FSB to lowest
setting but manually setting the same memory timings
(5,5,5,18,2T) in the bios. This will reduce CPU heat, power
consumption, memory bus (keep memory ratio the same),
hopefully isolating the video card as the only part still
running at stock speed and output. If you do this just be
sure you have manually fixed the memory ratio and timings so
the bios isn't changing them to more aggressive ones due to
the lower FSB setting.

under clocking still causes the problem!

You don't mention what version of speedfan but if not the
latest (1.34 beta?) try that.

yes, 1.34beta

Compare it's readings and the
bios health/hardware monitor page to what Everest is
reporting, you need to determine what the actual values are
for voltages and temps since Everest seems wrong.

If your board has EIST and C1E bios settings, disable these
temporarily.

disabled in BIOS currently, done.

Given the short amount of time you have left to
troubleshoot, I would go to a local store (one with a good
return policy) and pick up a different PSU to test with. If
that is impossible, at least strip the system down to bare
minimum parts (1 memory module no other cards, no optical
drive, underclocked FSB, etc) and try same games that crash.

bare-bones - just 2 stick of 1gig RAM, CPU, motherboard, video, PSU,
keyboard, mouse, hard drive, monitor

Are there any remotely modern (3D) games you can load that
don't crash?

no, not so far...

I don't understand with only one day left to RMA, why you
are still overclocking. I realize you may want to
overclock, maybe it was the whole point of the parts but
even so, to have successful overclock the parts have to
prove themselves 100% stable before overclocked.

I dropped the OC entirely, running completely stock... still having
the trouble.

Thanks again for your time!

(e-mail address removed)

 

[baronepicusfuror]

Thanks for the reply! Any help is good help, the more brains the
better =)

The way usenet works, and free time working around sleep and
work and family, etc, you could easily need more than one
day even if we posted something that solves a problem.

Get the RMA NOW. You don't have to actually RMA it now, but
because you are running out of time you should go ahead and
get it then if you get the problem fixed, just let the RMA
expire instead of sending them the parts (contact them to
cancel it later if necessary).

excellent suggestion, this is exactly what I did today! thank-you
newegg! RMA'n memory [1 was DOA, but have to replace @ same time], the
video card and PSU. I am going with replacements for RAM and video
card, but requesting refund on the PSU -- I am going to replace the
hec Zypher 650W PSU for this instead =http://www.newegg.com/Product/Product.aspx?Item=N82E16817341010
-- it's an OCZ StealthXStream OCZ600SXS 600W Power Supply, reviews are
good, sounds like well-built PSU, I was very hesitant in the first
place to buy the hec PSU, but we shall see if it was the source of the
trouble...

Run Memtest86+ for a few hours. Run Orthos for at least 60
minutes. If either of these show any error you need correct
that subsystem fault. More likely, when the game starts and
the video card enters 3D mode it consumes more power and
this is enough to overwhelm your PSU. I'd have thought an
HEC 650W would be enough but it may be a budgetized design
or have some flaw or failure. Check voltages with a
multimeter if possible.

Everest stability tests, ATI Tool stability tests, and ntune stability
tests have been run, nothing seems to knock the computer off-line
except starting up ANY game, sometimes I can play a game for 2mins,
then iit powers the system off. when I say "powers the system off", I
mean it just shuts down with no warning, to an OFF state - I have to
push the power button, the screen just goes black, all fans stop - no
blue screen or warning.

Also do the standard windows things like checking Event
Viewer.

Event viewer just says the game took over desktop manager, then ends
there -- not much use, I turned on ALL reporting/verbosity(sp?), with
1second updates

What exactly does "shuts down" mean? Does it:

A) Reset and hang
B) Reset and boot
C) No reset, just hang/crash, then is there a bluescreen?
D) Turn off, then the switch on case front can turn it on.
E) Turn off, then PSU AC power switch or cord must be
pulled before it'll turn on again.
D.

You write probably not heat related. Do you or don't you
know it is staying cool enough, based on temp readings?
Speedfan is one app that will monitor temps.

idle motherboard temp @ 23C, idle CPU @ ~20-26C, idle video card @
~22-26C and I seem to top out during full load at = motherboard @ 25C,
CPU @ ~24-30C, video card @ ~28-33C

Games I can start up ANYTIME, and the problem occurs [actually every
game I've tried so far];
Call of Duty 4 - direct X 10 and DX9
Crysis [64bit and 32bit] - direct X 10 and DX9
Gears of War - direct X 10 and DX9
Bioshock - direct X 10 and DX9

Forget about DX info, leave out things nonapplicable. Make
sure windows is not set to reboot on error (Automatically
reboot on system failure).

only mentioned it to state I've tried the games in both vista x64 and
xp 32bit

Focus on WinXP SP2 only for now. By doing so you reduce the
variables and use the OS with the most issues resolved
already through length of time it's been out and patched.

I like this idea too - I tried both OSs hoping more debugging info to
read...

Leave only 1 memory module installed until the problem is
resolved. The key is reduction in variables, there could
actually be more than one thing wrong with a new build and
it is best to resolve your most significant problem with
other things as basic and minimal as possible. After you
fix the problem, THEN add back the other 3 modules and
retest with Memtest86+ overnight to reconfirm stability.

excellent advice!

Do not put parts on
anti-static sheets and foam, that will cause problems.
Anti-static sheets and foam are often deliberately designed
to conduct electricity, which is how they are static
dissipative. Often it won't cause a problem but it can
introduce further problems in some situations.

I appreciate the concern, I moved to cardboard instead, with
components exposed to air for me to troubleshoot/rule out heating
issues -- basement room temp = ~63F.

Forget about software/drivers/etc. It is not software
providing you have set windows to not shut down on error,
and that you then don't get a crash dump, bluescreen, etc.

I agree.

Forget about reinstalling OS. I'm not trying to be critical
of the process, just concise in what can be rulled out.

I appreciate the help! very much so

Leave system at stock, and confirm there is no bios setting
for temperature shutdown that is set too low, and that the
temp reading seems correct (some bios may not register temp
right for some CPUs). If the temp seems off by a LOT, seek
a bios update. Actually see what bios updates are available
either way and check their notes to see if anything seems
applicable to your parts.

BIOS is current, ASUS shows most recent = 02/22/08, thats what I got.
everything is at stock right now.

Check your bios memory and MCH voltage. Try bumping memory
to 2.05V at most (this with only one module installed
still), that is all it should need at stock speed.

Yep, been over and over voltage settings... days worth of tweaking! I
just don't think it's BIOS voltage settings...

I'd almost forgotten, examine all cards, cables, etc, the
whole interior of the system and reseat any cards you doubt,
including video card. Also pull AC power, remove battery
for 10 minutes and use clear CMOS jumper. Only set minimal
required things in bios from then on instead of a final
tweaked config.

done, done and done again =)
been over this _several_ times

Leave it at stock speeds till problem is resolved.
check.

but you don't tell us exactly how long, EXACTLY what moment
it happens and exactly what you see on screen then.

ANY game I try to load, boots and as soon as the start screen comes
on, or sometimes it will let me play the game for 2mins at MOST, the
computer shuts down, no warning - power off, fans off, screen black -
no moving parts... no blue screen, just black.

Don't use any kind of ntune or other software that interacts
with anything. If still in doubt about temps, leave the
case side panel off and point a fan at the video card
temporarily.

GPU fan is controled via ntune, by cranking the fan from 35%/default
to 100%/custom the GPU cools down 10-15C everytime, it is a precaution
I take to make sure it's not the GPU overheating or something, but I
have tried this with default/stock setting and with it bumped to 100%.

If still on the anti-static material, get it off of that.

ya, moved the motherboard to cardboard, PSU is still in the case,
screwed in...

We still don't know at what exact moment it's crashing or
whatever it's doing. We have a lot of info about what you
tried but not what the very specific exact thing you observe
is, only a summary of it not the detail of what eyes see,
ears hear, etc. Remember we are not there to know what you
see.

I hope my answers above makes it clearer now...

Forget about Vista X64 for now. It's a wildcard. Stick
with the simplest config, a clean XP 32bit installation with
only drivers, NO other hardware monitoring or logging or
ntune tweaking anything at all. Only in a bare minimal
config can you be sure nothing else is causing problems. I
realize the purpose for the logging but I mean now, since
nothing significant is being logged, tuned or adjusted, it
has shown to be of no use and could only have negative
consequences if it remains.

agreed - thats where it stands currently... still with the problem =(

Based on what's been provided and playing odds, it's most
likely the PSU is problematic. If you've a multimeter you
should hook ...

no multimeter unfortunately... I am taking a gamble and am gonna
return the PSU for the OCZ StealthXStream OCZ600SXS 600W Power
Supply , with fingers crossed this and the RAM+video card RMA will
resolve the issue. I am stumped... if someone like you can help me
troubleshoot it without sending it back, I'd be tickled pink!

thanks for the assistance, hope I answered your questions clearly!

(e-mail address removed)

Uninstall .NET Framework 1.1, 2.0, 3.0 and any hotfixes pertaining.
Then reinstall .NET Framework 1.1 Redistributable
(http://www.microsoft.com/downloads/details.aspx?FamilyId=262D25E3-
F589-4842-8157-034D1E7CF3A3&displaylang=en)
Lame? Yes. Broken? Yes. Works? Yes.
Peace.