Newly built rig, BSoD's aplenty..

Maccers · 09 Feb 2011

Hello all

For the last two weeks (every since I assembled my new rig) I've been getting a large number of BSoD's. Sometimes I can go a day or two without seeing one, then I can get 3 in the space of 25 minutes! Normally I'm very savvy on technical and software issues, but this has me very confused.

The specs for the rig in question are as follows:

AMD Phenom II X6 1090T Black Processor
Asus M4N75TD Motherboard, most up-to-date BIOS
G.Skill Ripjaws, 2x4Gb DDR3 1333 kit, Matching pair, sequential serial numbers. 7.5Gb is available, as the onboard graphics is using 512Mb for Hybrid-SLI
Gigabyte GTX460 1Gb GDDR5 discrete graphics card
OCZ ModXtreme-Pro 600w PSU, although this was only installed today
1TB RAID primary disk ,comprised of a Western Digital Caviar Black 500 and Samsung Spinpoint 500
Windows 7 Home Premium x64. Most recent updates installed.

I've had a number of theories as to why this is happening and right now I'm just losing my mind. The most recent of which was noticing that the 12v rail was dropping sometimes below 9.0v and that might be having some effect, after blowing £70 on an 80plus approved PSU and getting stable voltages on the 12v rail I'm still getting BSoD's.

I'm also able to rule out the RAM, did 6 passes with Memtest86 and another windows based testing program last week. Testing with each stick individually and both together passed flawlessly.
Checked the BIOS and played around with any overclocking settings that were present. both auto OC and disabling all OC causing crashes.

The first time I installed Win 7 it threw up a BSoD and I thought that the install might of been corrupted. I've since reinstalled and it's still temperamental. One current theory which I'm willing to entertain might be that the DVD drive I am using is on the way out, which would explain why I had the first BSoD on install, and why I had to install Supreme Commander and a few other games up to 6 times before it would complete successfully. Not because of BSoD's, but because files on the discs were either corrupt or mistakenly detected as malicious software.

Aside from replacing the DVD drive and reinstalling off that I'm completely stuck on solutions. As I will need to buy a new drive I'm not going to spend any more money until I'm absolutely sure it's the cause and that a replacement will quell the BSoD's. Doing backups also takes a huge amount of time for me and I don't want that to be for nothing.

I have a number of .dmp files, I've been able to view them in WhoCrashed and understand what has crashed in each instance, but I'm not able to draw up a pattern and come to a definitive conclusion. As said before, the going theory is that the install disc can only be partially read because of a defective DVD Drive, and some files are going into the install corrupt. but it's only a theroy :)

If someone here who understand .dmp files better than I do can look and confirm, or provide a solid reason otherwise I would greatly appreciate it.

Thanks

skunksmash · 09 Feb 2011

hmm.......

its pretty clear you've done most run of the mill troubleshoots...... i suppose booting from USB would rule out the DVD drive as a problem but if you've got a retail disc that wont be much good for you.

in this situation (unless anyone can sift something from your dump files), i'd start stripping the machine back to basics.... mobo, 1 stick of RAM, primary HDD, onboard GFX, optical drive..

see if it will fire up ok without the Hybrid SLI..???

Max Peck · 09 Feb 2011

Well, it does sound like you're having a hardware problem all right. Over the years I've noticed that system problems that "don't make sense" generally come from 2 sources:

1) Virus or Malware
2) Hardware issues - flaky/cracked motherboard, bad I/O connector or port.

It is possible that your processor is having heat problems. Check to make sure the heat-sink/fan is properly attached and functioning.

I noticed that you're using an AMD processor. I dunno about anyone else but I have found over-the-years that systems based on that brand of processor tend to be a little bit more finnicky. I never could put my finger on it but over a 25+ year period working with PC's I never had stability or reliability problems with Intel processors. EVERY time I would get an AMD I would have weird problems - with exactly the same software. The first time or two it happened I figured it was a fluke but then after 10 or 15 years I saw a trend. Too weird.

I'm not trying to say that the AMD processors are crap or anything, but after being burned by them a half-a-dozen consecutive times I stopped buying those. They work most of the time (or at-least they seem to) but tend to be a bit flaky, particularly when they get hot.

If you can afford to do so - try replacing the processor with an Intel (if the board supports it) or replace the motherboard entirely with one that has an Intel. I'm no vendor fanboy or anything but that's what my experience is with stuff like this. We tend to look at the mainboard last when we have problems like this. Good luck.

FWIW ...

-Max

yowanvista · 09 Feb 2011

The Saitek Raid driver is outdated, update it
http://www.saitek.com/uk/down/drivers.php

Code:

SaiH0464.sys        fffff880`042c9000    fffff880`042f3400    0x0002a400    0x47ed28be    28/03/2008 21:19:58

Code:

dump_nvraid.sys        fffff880`019cb000    fffff880`019f4000    0x00029000    0x4a13a5ac    20/05/2009 10:39:40

skunksmash · 09 Feb 2011

Max Peck said:

Well, it does sound like you're having a hardware problem all right. Over the years I've noticed that system problems that "don't make sense" generally come from 2 sources:

1) Virus or Malware
2) Hardware issues - flaky/cracked motherboard, bad I/O connector or port.

It is possible that your processor is having heat problems. Check to make sure the heat-sink/fan is properly attached and functioning.

I noticed that you're using an AMD processor. I dunno about anyone else but I have found over-the-years that systems based on that brand of processor tend to be a little bit more finnicky. I never could put my finger on it but over a 25+ year period working with PC's I never had stability or reliability problems with Intel processors. EVERY time I would get an AMD I would have weird problems - with exactly the same software. The first time or two it happened I figured it was a fluke but then after 10 or 15 years I saw a trend. Too weird.

I'm not trying to say that the AMD processors are crap or anything, but after being burned by them a half-a-dozen consecutive times I stopped buying those. They work most of the time (or at-least they seem to) but tend to be a bit flaky, particularly when they get hot.

If you can afford to do so - try replacing the processor with an Intel (if the board supports it) or replace the motherboard entirely with one that has an Intel. I'm no vendor fanboy or anything but that's what my experience is with stuff like this. We tend to look at the mainboard last when we have problems like this. Good luck.

FWIW ...

-Max

lol...

lets see if we can get his current gear working first shall we......

AMD make good gear nowadays.... (given, its not as rugged as intel) but there should be no reason this system wont work when brand new!!

Maccers · 09 Feb 2011

Thanks for the replies.

It's always nice to put up a big, descriptive post up and not immediately be fobbed with tl:dr and 'Cool story, bro':)

Yowanvista. Thanks for checking the crash logs. I've now updated both the Saitek driver from their site and the Nvidia Raid Driver from the mobo disc.

Skunksmash. I can run the system with a single stick of memory, but that's the limit of stripping it back. The onboard graphics is strictly for Hybrid SLI only, the motherboard doesn't have a VGA, DVI, S-Video or HDMI port on it. Booting without the graphics card results in the usual POST bleeps telling me there's no display adaptor. I was running the system without the Hybrid SLI for a few days after I first built the computer and was receiving the crashes so I'm not sure that's the culprit. In any event, if I'm still receiving BSoD's I'll drop down to 4Gb and disable the Hybrid SLI to free up the extra 512Mb (if I'm gonna halve my memory I'll need all the RAM I can get lol).

Max Peck. I'm using a custom cooler, and keep a watchful eye on both the temperatures and fans with Speedfan. The processor has never gotten above 35C, even during a 4 hour match on SupCom with 5 opponents and a 1,000 unit limit on each team. Admittedly this will now be a problem if the processor is faulty and need to return it to AMD on warranty but I'll cross that bridge when (and if) it comes to that. As for AMD themselves, I've been using them since the days of the K6 and no processor (excluding this one for the moment) I've bought from them has ever put a foot wrong.

And I've just remembered, there's also the issue of backwards compatibility. My last system was an AM2+ and that, along with AM2 and AM3 are all compatible. If I get stuck further on I can pull out the 1090T and replace it with my old processor and see how stable it is.

Anyways, I'll keep going with the driver updates and see how things go, If I'm posting here with more crash logs you know I've not had any joy, but I'll still keep an eye on the thread if any more ideas pop up.

Thanks again

yowanvista · 09 Feb 2011

and the Nvidia Raid Driver from the mobo disc.

Mobo disks usually contain old drivers, I'll suggest to download a newer version from Nvidia

Max Peck · 10 Feb 2011

skunksmash said:

lol...

lets see if we can get his current gear working first shall we......

AMD make good gear nowadays.... (given, its not as rugged as intel) but there should be no reason this system wont work when brand new!!

Well ... of course. I just had to provide a different angle there!

-Max

Max Peck · 10 Feb 2011

Maccers said:

Max Peck. I'm using a custom cooler, and keep a watchful eye on both the temperatures and fans with Speedfan. The processor has never gotten above 35C, even during a 4 hour match on SupCom with 5 opponents and a 1,000 unit limit on each team. Admittedly this will now be a problem if the processor is faulty and need to return it to AMD on warranty but I'll cross that bridge when (and if) it comes to that. As for AMD themselves, I've been using them since the days of the K6 and no processor (excluding this one for the moment) I've bought from them has ever put a foot wrong.

Good. It's probably not the processor, just wanted to add a little perspective, that's all. I've struggled for weeks with systems, replacing everything in the box only to find out that's what it was all along. Good luck with it.

-Max