Friends build problems.

Associate
Joined
21 Jun 2009
Posts
1,668
Location
Chester/Loughborough
I offered to build a system for a friend around christmas, telling him just give me a budget ill get back to him with a spec and if hes happy we'd take it from there.
At easter he asked me to come around and piece together a set of components he'd bought - sounded like trouble already.
Went around and found him with this :
Intel Q6600 (stock cooling)
ASUS P5Q Deluxe
4x1GB DDR2 GEiL (black dragon i think) 1066MHz
ASUS nVidia GTX285
Western Digital 500GB Caviar Green
SONY DVD R+/W 22x
Coolermaster 700W modular thing - looks like a Realpower M700
Coolermaster Storm Sniper case (this thing is butt ugly and loud as hell but airflow should be adequate).

I told him hes wasted money all over the place (spent £1300 altogether) but got on with building it and it all went well. Installed the OS (XP 32bit) with no issues and all the drivers - seemed to be working just fine.
He went ahead and put Crysis Warhead on there and just as it got to the end of the little intro movie it blue screened. Id heard about these GPUs from ASUS overheating a lot and it had a shrill screech coming from it so I installed RivaTuner and turned the fan upto 100%.
He played + completed Crysis (did say it blue screened 1 more time during i think).

Couple of weeks later he got in touch saying it had been blue screening a lot and is acting very sluggishly. I was in Uni at the time so couldnt do much about it but told him to download Intel Burn test, OCCT and Realtemp/core temp/speed fan and do some stress testing looking at the temps - they were skyrocketing on the CPU after very little time and would get through a 15minute stress test only 50% of the time.
I said it sounded like an overheating problem and some aftermarket cooling should sort it, then suggested the Thermalright Ultra or Megahalems.

When I got back from Uni he asked me to install his Zalman 9700 - I did and tested it and temps seemed to be good 50C after 20mins IBT stress test.
Whilst I had the PC I took a look at the software he had on it and there was piles of crap and clutter.
Gave the PC back to him and within that week he was complaining of blue screens and sluggishness (temps still solid). I suggested he got rid of any of the software he didnt need. He said he deleted half the stuff and there was still problems so gave it back to me to test.

There was one game in particular that would throw out a blue screen for him, i think it was Never Winter Nights or something, so I set his PC up back here and started doing some tests - system did not seem sluggish or problematic in any way and I couldnt get it to crash when running any program - game or stress tests.
So I started thinking it was likely a situational factor, since Id never had it crash on me. My room is always on the cool side - windows flung wide open and normally a bit of a breeze, whereas his is quite cramped and only has a small window. His computer is kept under a desk in the corner with his bed only 1-2 feet away from the front of it. However he was always reporting similar system temps at his house to what I was getting . . .
Other differences include his Samsung TV/monitor hybrid thing, cyborg mouse (least ergonomic mouse I have ever used) and some ricer keyboard thing. All of which required some software interface whereas the things I am using are simple plug and play.

After a bunch of tests and no problems I began to think the problems were related to his peripherals. Opened up the system (probably should have done this in the first place but figured I might aswell try and replicate the problem first) and noticed he had installed a Wireless PCI card which I asked him about and he said when installing he didnt unplug the PSU, just switched it off at the back and it gave him some issues when first going in - didnt seem to fit properly. When he got it in the software gave him more problems and then it wouldnt connect to anything for a while, and then it started working suddenly. He installed it a couple of weeks before I came back from Uni - about half way between me leaving and getting back so he had been saying there were issues before that.
Hes also removed one of the DIMMs of RAM so theres only 3 in there now - im not sure why.

Hes on holiday atm and so Ive got the system for a week more, I turned it on this morning and started doing some more tests and it finally BSOD's for me -
CPU:OCCT about 10 minutes in - temps all looked fine around 50.
MEMORY_MANAGEMENT
STOP: 0x0000001A (0x00B97201, 0x00000000, 0xC0883000)

CPU:LINPACK again about 10 minutes in with all temps looking fine.
IRQL_NOT_LESS_OR_EQUAL
STOP: 0x0000000A (0xC25926F8, 0x00000002, 0x00000000, 0x8051192F)

Now im running a GPU test, turned the fan back down to stock and its going fine (I do have the side of the case off though and surprisingly its quiter that way). Its 28 minutes in with no problems but it is hitting 81C - not too bad for a GPU right ?

Now im thinking its either CPU (but ive heard most people say they either work or they dont, theres no middle ground), motherboard (plan to try updating the BIOS although current one list full support for CPU etc) or the PSU.
I do have a spare GPU (9800GTX+) and power supply (OCZ 780W), and do have a Q6600 and some more GEiL ram in my bros system which I would prefer not to disassemble but can if needs be.

Going to do a memtest when once ive created the CD.

Any input would be appreciated.

EDIT: thats longer than I thought it would be as I was typing it lol.
 
The lack of responses goes to show that giving full detail on the problem straight off the bat isnt always the best idea.

I looked into the BSOD errors that have been occuring and got a few ideas from that.
Tried stripping the RAM down to just 2 DIMMs instead of the 3 and no change.

After a little thought I figured if it is the PSU its likely to be the connectors to the mobo, so took the 24pin out and checked it over - all looked fine. took the 8pin CPU connector and the top left pin clip was warpped pretty badly, got some tweezers in there and reshapped the clip hopping this will solve the problems. 8pin slotted back on much easier than it before.
 
Have you tried removing all but one stick of ram to see if that fixes the problem, and if not try the other 2 sticks individually? Check that the RAM has the correct voltage set too.

After googling the first STOP error you received, I found a windows update which is meant to fix the problem (although might not work in your case):

http://support.microsoft.com/kb/929338
 
Last edited:
Yeah ive tested with minimal components outside the case - although in this case its only really removing 2 sticks of RAM and the DVD drive. And it made no difference.
XP is fully updated including the patch linked.

With the PSU connector re-shaped its going further through the stress test than it would this morning. Just hope its not being tempermental.
 
Reshapping the 8pin connector helped, added 5 minutes onto the stress test before it blue screened. But it still blue screened.

Swapped the RAM out with some other stuff I had and tested it and it worked with A1,A2 and B1 populated, and A1, A2, B1 and B2 populated. lead me to believe there was an issue with the RAM he was running.
Put that in my system and it worked fine - completed all the tests i through at it.
So then I thought it must be an issue with the RAM/mobo interface. Opened up the BIOS and the ASUS mobo had decided that this 1066MHz RAM was running at 1200MHz with 1.9V (had checked in OS and it was reported as running at 1066MHz). Set these to manual at what the RAM was rated an voila, no issues. Its completed two 1 hour OCCT tests, and is now 2:15 through a LINPACK using 90% of available RAM.

Seems to be fixed.
 
Back
Top Bottom