Sign in with
Sign up | Sign in
Your question
Solved

Computer is giving random BSOD's, on same build I've been using for a year.

Last response: in Systems
Share
October 14, 2013 2:50:09 AM

Neb here.
Alright, so this is one that has stumped me for a while now.

REFERENCES:

Here are my specs:
CM 690 II Advanced case with FULL fan system, in good directional array and airflow.
MSI MS-7522 mobo
Socket 1366 Intel i7 920 CPU
Radeon HD 5770 (I have a Gigabyte SOC, and I have tried single card and 2 way crossfire [with a sapphire 5770], no difference at all)
Corsair 3x2GB High Performance RAM
500g WD standard HDD
Apevia 680w Standard Non-Modular PSU
(Addons below)
Corsair H100i Liquid CPU cooler
Full set of Cooler Master Blue series varied-size fans.
And one PCI USB expansion bay. Because I have too many peripherals.

And what I have tried:
Yes, turning it off and turning it back on.
Disabling all overclocks, resetting bios and setting voltages and clocks to recommended settings.
Checking and monitoring temps, turning fans up, maximizing airflow… it isn’t (all) heat issues and I’ll get to that in a sec..
Switching to Single monitor.
Upgrading GPU drivers for both cards
Changing out peripherals and updating drivers for such.
Taking out expansion bay and booting without it.
Tweaking liquid settings and checking compatibility issues – found nothing.
Tethering my WiFi to my laptop via a bridge on Ethernet just in case it was my WiFi USB adapter.
Several long and strenuous stress tests.
Enabling HPET
Banging my head on my desk.
Taking a snack break after 20 hours.
Quitting out of frustration.

Ok, now that I’ve gotten that out of the way, I’d like to explain the full extent of my problem. I’ve been getting blue screen errors for a few days now. The off part is that I’ve had the same exact setup for about a year now, with no changes ever since I started college (due to lack of liquid funds) and this problem came out of nowhere.
As you can see, I’ve tried many things. And at the point of “quitting out of frustration”, I decided to get help, so I took notes. I’ve been getting several messages of BSOD’s, like:

IRQL_NOT_LESS_OR_EQUAL
STATUS_IMAGE_CHECKSUM_MISMATCH
And several kernel errors that I will update as soon as I can find them in logs

Now the things that I’ve managed to get down are that in about half of my BSOD errors, I get graphical tearing, or no visual at all. Sometimes it will just be a blue screen with splotches of white in random places. I decided it may be the GPU, so I looked more into it. It turns out that the GPU WAS causing the graphical tearing.
My 5770 (I have been running a single for a while) was in the middle of its first stress test, and at around 58, it started flashing out and getting ready to quit. At this point I hurried to the “stop” button and made it in time. I lowered the clock speeds, turned up the fan and lo and behold, I got no BSOD for about three hours, and then it happened again.
Right now as I am typing, I have gone through this data sheet with no errors, BSOD’s, or hiccups. But any time that I start up a game, I will last anywhere from 3 minutes to 3 hours and it will inevitably crash.
Now, one’s first thought into this issue is that it would be a heat issue. But, surprisingly, just to test this theory, I cooled my GPU down to about 54c at full load and my CPU to 50c (because my cooling is awesome when it needs to be) and played a game. About half an hour in, it crashed, and I got a BSOD.
This is the part where I noticed something very peculiar. My errors have all been happening at stages of input. I hit a key, I click, this always happens directly before my computer crashes. I’ve NEVER had a crash with this issue while I’ve been standing still in nuketown or reading something, It’s tied to my input in some way.
Alright. That was a page and a half on 11-font in word, not to include those lists I’ve given. Please reply back with any input you have, kudos to whoever can help, and have a good one.
October 14, 2013 2:58:26 AM

go into bios and find HPET mode enable it.
as its an older 1366 socket it will be disabled by default. the upshot of this is if you use win 7 or vista you will get random bsods.
i had the same issue when i built my system for about 3 months till i found out what hpet was for...
hpet is often refered to as precision events/ event timer or high precision events timer.
m
0
l
October 14, 2013 3:11:53 AM

HEXiT said:
go into bios and find HPET mode enable it.
as its an older 1366 socket it will be disabled by default. the upshot of this is if you use win 7 or vista you will get random bsods.
i had the same issue when i built my system for about 3 months till i found out what hpet was for...
hpet is often refered to as precision events/ event timer or high precision events timer.


Alright. Trying it now. Will get back with results.
m
0
l
Related resources

Best solution

October 14, 2013 4:09:32 AM

even if it doesnt fully fix the issue you should keep this enabled it will make your system more stable. im just suggesting it because it fixed a similar problem for me.
Share
October 14, 2013 6:25:08 PM

Alright, i enabled said setting and my computer ran smoothly for about 10 hours, and then quit on me again. Stress tests returned no abnormal results. Just ran Windows Memory Diagnostic and nothing came up. This is getting intense.
m
0
l
October 14, 2013 8:23:23 PM

open the events viewer (ctrl pannel/admin tools/events) and look at the critical list. look for anything just before you get an error 41 (pc powered off for no reason)
m
3
l
October 15, 2013 2:12:05 AM

IT seems like different kinds of things happen each time I get a bsod. One of them was a HAL 15 error having to do with a "periodic tick", before another failure there are 8 threading events about my CPU, there's an error saying that bootmgr waited 0ms for user input...

How do I export this as a file so that it may be looked at by others?
m
0
l
October 15, 2013 2:16:33 AM

Actually, scratch that. I just noticed something that I think may be very relevant. This periodic tick is showing up relatively chose to every critical error. I did some research and it is linked to PCI based devices, and all of my input devices are based out of this newer PCI-based usb card. We might be onto something....
m
0
l
October 15, 2013 4:34:51 AM

cool. looks like your on your way...
hpet i think will solve that as its a hardware clock based thing... when you enabled it, did you have to set it to 32 or 64. because sometimes when its set to 64(which it should be on 64bit o.s) it can cause issues with some motherboard/cpu configs. if thats the case it should work fine on 32 as windows just needs to know its there and turned on, it doesnt really care what bitrate its set at.
it could be the reason the pc was stable for 10 hours then went off.
worth a try if you notice the time between bsods increases 1s you set hpet.
m
0
l
October 15, 2013 5:16:48 AM

I had my HPET set to optimal config (automatic detection) But I ended up fixing the machine by just pulling the power and pulling out the PCI card!

It always seems like the simplest fixes for the most complicated problems. Oh well.

Thank you very much for you help hexit, have a really good one.
m
0
l
!