Tech Support Guy banner
Status
Not open for further replies.

Computer crashes randomly

897 views 14 replies 3 participants last post by  crjdriver 
#1 ·
Hi all, I suspect this is a hardware issue however it might also very well be a software issue - or a combination thereof. I'm at a bit of a loss really, I hope this is an OK place to start this thread but to let me know if I should move it!

Quick background. I have a gaming rig that I built myself. Simple specs (gathered with the TSG utility) are below:
Tech Support Guy System Info Utility version 1.0.0.9
OS Version: Microsoft Windows 10 Pro, 64 bit, Build 19041, Installed 20200825122105.000000+060
Processor: Intel(R) Core(TM) i7-4790K CPU @ 4.00GHz, Intel64 Family 6 Model 60 Stepping 3, CPU Count: 8
Total Physical RAM: 16 GB
Graphics Card: NVIDIA GeForce GTX 1070
Hard Drives: C: 930 GB (855 GB Free); D: 3725 GB (2405 GB Free); G: 447 GB (228 GB Free);
Motherboard: ASRock Z97 Extreme6/3.1, s/n M80-54012000083
System: American Megatrends Inc., ver ALASKA - 1072009, s/n To Be Filled By O.E.M.
Antivirus: Windows Defender, Disabled


I have also provided more extensive specs as an attachment, an XML with way more detail than this but it's a good start. Oh and I am running Avast! Antivirus, not sure why it didn't pick that up. Both C and G run on SSDs, the C SSD is purely for Windows and other installs (I keep this easily format-able in case I need to reinstall Windows). D is a hard drive for storage which is backed up to the cloud, G is for game libraries (Steam and the likes).

It's been working well for a long while, I put her largely together in 2015 and since then have only really swapped out the water cooler when that went on the fritz.

However a little while back, maybe a few months, she'd randomly crash. The screen would just turn off, the fans would go to 100% and... that's it. I'd have to do a manual hard reboot. I did a little bit of research and found... not a lot. I decided to just reinstall Windows, it had been a little while so might as well. I did just that, however then my Steam started acting up - I have the Index and that would suddenly not work anymore. Fine, another Windows reinstall later she was up and running again. She worked well for several weeks, maybe even a month or 2, but then yesterday she suddenly crashed again. And then this morning, when I was trying to gather diagnostic data, she crashed three times in a row - but now she's again been doing fine for a few hours.

I have attached the Event Viewer logs of the last few days. I can't see a pattern there to be honest, but I must also admit that I'm not a Windows expert (I'm a software developer so have some knowledge but am definitely not an expert on this). Oh good to note; I'm unable to boot in safe mode as the screen won't turn on. I have a BenQ 1440p gaming monitor hooked up to the graphics card with a DisplayPort cable, but hooking it up with an HDMI to the motherboard directly didn't help either (it would just say 'no input').

Now I'm just not sure if it's hardware or not. I started suspecting my Graphics card but at the same time I can't be sure because I'm still able to use my Index without issue (although once I forgot to turn SteamVR on and that caused a crash with an Application error pointing at SteamVR in Event viewer). If it's the graphics card I don't mind replacing it, but I don't have another one lying around to swap in and see if that fixes the issue. I'd hate to replace it with the 3080 (which I'm kinda planning on) but then find out it's another expensive component causing the issues and then be stuck with my whole computer-upgrade-budget looking pretty in my case but the rig still crashing.

Anyone out there who wants to take a crack at figuring out what it might be?
 

Attachments

See less See more
#2 ·
You have not included the most important part; the exact pw supply you are running.
Next start with some basic troubleshooting. Go to the support page for your motherboard and download/install the A-tuning utility. Here is a link to the support page;
https://www.asrock.com/mb/Intel/Z97 Extreme63.1/#Specification
Once that is installed, post the following info; cpu temp, 12V, 5V, and 3.3V values. Do this twice. Once under a normal load and again while running a stress test. Use one of the following for the stress test; prime95, OCCT, or Realbench. Let the stress test run for at least two min to stabilize temps/voltages. Post screenshots of A-tuning rather than typing the numbers.
 
#3 ·
Hi crjdriver, thanks for the reply! I did my best to follow your instructions, I hope this is what you had in mind.

The power supply I have is a CoolerMaster V1000 (link).

The screenshot of the system under normal load (it's been running for a few hours but not doing anything strenuous):
Font Screenshot Software Rectangle Electronic device


Using a stress test by RealBench (been running for 3 minutes, haven't closed any other programs just started the stresstest on top of whatever it was doing during the above screenshot):
Font Screenshot Software Multimedia Electronic device


In case that's too small to read here's just the A-tuning screen:
Font Rectangle Screenshot Technology Circle


I hope that's the information you required, please let me know if there's anything else!
 
#8 ·
Ha thanks, I completely missed that - the big red line distracted me.

I currently have 2x8GB Corsair DDR3 (link). One is in the DDR3_A1 mobo slot, the other in DDR3_B1. A2 and B2 are empty.

This is the A1, the B1 has the exact same specs when I hover over it in A-tuning:
Font Magenta Brand Darkness Signage


Would it be worth testing the RAM individually? It comes in a 2x8GB pack, I thought you can't really change one of them out for another 8GB.

Is it possible there is another faulty component, so should I at least test my other hardware before I settle on buying new RAM? I'm mainly asking because the mobo isn't the youngest and doesn't support DDR4 as far as I know, so if I were to upgrade the RAM I'd likely also update the Mobo and CPU (cause new socket) but I don't want to drop almost a thousand quid on that and then find out the GPU is also faulty...
 
#9 ·
I see one thing that is a problem; vdimm or ram voltage. The specs on your ram show 1.55V to 1.65V and A-tuning shows your ram running @1.5V. Before replacing anything, here is what I would do;
1 System OFF and pw cord removed, open the case
2 Pull out both ram chips and blow out the slots with a can of air
3 Reinstall ram and be sure it is fully seated
4 Replace the pw cord and pw ON. Enter the bios. Set your ram voltage @1.6V [read your motherboard manual as to bios options/menus] Save settings and restart
5 Rerun the test and see if you still have errors. If you do, replace the ram
 
#10 ·
I still saw an error after about 1,5 hours of running the memtest after changing the voltage to 1.6V. I guess that means it's the RAM that's on the frits!

Thank you so much for your help, I never would've suspected the RAM otherwise. You've definitely made me want to run some diagnostics on the other hardware though, especially before I start throwing money at replacements like I said - do you have any tips on GPU diagnostics? Or perhaps other parts I should check as well (CPU, MOBO chipset or something?)

Thanks again, I really appreciate it!
 
#11 ·
CPUs seldom fail unless you overclock/overvolt them. It does happen however not often. Motherboards fail much more often than processors. A simple cpu test is to run prime95. Your cpu has 4 cores and 8 threads. If you see one or more threads not working, that would require more investigation.

Motherboards. Generally you give the board a VERY close inspection with a bright light. Look for any leaking or bulging capacitors. If you have checked all other hardware ie ram, hard disk, etc and still have problems, that would point at a motherboard problem.

Video card. Usually you just pull the video card out and use the onchip video that is part of most intel processors. If you still have a problem, you know it is not your video card. Problem goes away, it may be the video card.

The good news is that most ram carries a lifetime warranty. You would have to go to the mfg's site and request a RMA for your ram. It might take a while however you should get it replaced under warranty.
 
#13 ·
I'll definitely try those things crjdriver, thanks! I noticed that when I plugged my monitor into the mobo HDMI port it wouldn't come on, kept saying it had no signal - but I hadn't pulled out my graphics card so that could be a good next thing to try.

I'll start bugging Corsair then, I did purchase it when I was living in another country so let's see how that goes haha!
 
#15 ·
When you have a real video card installed, onchip video is disabled. You would have to remove your card in order to use the intel onchip video.
If you purchased the ram online, you can usually print your receipt after you login.
 
Status
Not open for further replies.
You have insufficient privileges to reply here.
Top