|
Problem description: Occasionally, while playing certain games, my computer will suddenly shut down and start up again. So far, I've had it happen with three games: Shadow Tactics: Blades of the Shogun, Mafia III, and just recently, Payday 2. Attempted fixes: Usually I can avoid the issue if I just turn down the system settings. However, I'm not certain which setting is causing the problem. With Shadow Tactics, I think it either had to do with the framerate, or with Vsync, and the crash would always happen whenever I quick-saved. Mafia III it happened not too long after I started playing it. I turned Vsync off, and turned the visuals down to medium, but I was able to keep the framerate at 60FPS without issue. Payday 2 is what convinces me that the issue is with the video card. The last time I played the game was back in December 2016, and I was able to play it at pretty high settings. I reinstalled it today, but it crashed twice not long after starting the game. I turned the visual setting down to medium and lowered the resolution, and i had no problems after that point. However, I never had it crash the computer like with the other games before now. Recent changes: The only thing I can guess could be the cause is an update to the video drivers. However, I'm not sure if that's the cause. I will say that I use NVidia GeForce Experience to update the drivers, but I don't know if it causes any issues. Part of me wonders if it has something to do with overheating, but, there's been no change to the area around where the computer ventilates. I'm also not sure how to gauge the temperature. I started having the problem with Shadow Tactics back in February 2017. I don't know if there were any drivers with problem released around that time. -- Operating system: Windows 7 Home Premium 64-bit System specs: CPU: Intel i7-4790K Video: EVGA NVIDIA GeForce GTX 970 HDD: Samsung SSD 850 EVO 500GB(primary drive w/ OS) & WD2003FZEX-00Z4S(used for file storage) Motherboard: ASUS Z97-A-USB31 Memory/RAM: HyperX (16GB) (Can't recall manufacturer at the moment) Power Supply: Thermaltake SP-850AHCCB (850W) Location: USA I have Googled and read the FAQ: I've read the FAQ, and I searched for the GTX 970. If it is because of the video card and the drivers, the solution might be to use Display Driver Uninstaller to clean out older driver version that are essentially left over. However, I wanted to run it by here first, since I've never given a message about the display driver failing, and I'm not sure if it's the problem. EDIT: I also found this video (https://www.youtube.com/watch?v=3ulQYC_HP9s), which makes me think the issue is with GeForce Experience. Max Wilco fucked around with this message at 01:10 on Apr 5, 2017 |
#
?
Apr 4, 2017 06:14
|
|
|
|
| # ? Apr 20, 2024 14:28 |
|
|
Seriously, its GeForce Experience. I was crashing in Ark and Fallout4 with nearly the same setup you had and uninstalling all drivers + GF experience and installing completely fresh drives and no GF experience solved it for me. Also if you are overclocking (cpu or gpu), return to default after the new drivers, run the same game(s) and watch your cpu/gpu core heat to make sure its all nominal before cranking the OC'd back up slightly. Some years ago i had an older machine that was barfing when at full load due to a fan being dead and thus no cooling... took alot of effort to discover that, let me tell you!
|
|
#
?
Apr 4, 2017 09:01
|
|
|
Slayerjerman posted:Seriously, its GeForce Experience. I was crashing in Ark and Fallout4 with nearly the same setup you had and uninstalling all drivers + GF experience and installing completely fresh drives and no GF experience solved it for me. Also if you are overclocking (cpu or gpu), return to default after the new drivers, run the same game(s) and watch your cpu/gpu core heat to make sure its all nominal before cranking the OC'd back up slightly. Thanks for replying. I'll try uninstalling GeForce Experience and I'll use Display Driver Uninstaller to clear things out and install the latest driver. I don't overclock anything, but I should probably check the fans and dust the computer out just to be safe. If I uninstall GeForce Experience, can I still use Shadowplay to record video?
|
|
#
?
Apr 4, 2017 11:17
|
|
|
I apologize for the double post, but I uninstalled GeForce Experience and use Display Driver Uninstaller to clean out the older drivers, and I installed the latest driver from NVIDIA. I started up Payday 2 and set things back to the higher settings and 1920x1080 resolution. However, not long after, the computer crashed and restarted. Should I try installing a older driver or is it possible that I did something wrong while using Display Driver Uninstaller? EDIT: I tried DDU again, and I made sure to clear out the GeForce Experience drivers, and then I cleaned out the drivers again and installed an older driver version from October 2016. I started up Payday 2 again and it crashed again  . I can't figure out what else could be causing this issue.
Max Wilco fucked around with this message at 20:40 on Apr 4, 2017 |
|
#
?
Apr 4, 2017 20:07
|
|
|
It's either overheating or your power supply. I'd suggest opening the case and thoroughly dusting it out, paying attention to all the vents, fans, and heatsinks. While doing that, note the brand and model of the power supply. If that fixes it or significantly reduces how often it happens, it was heat and you should check temperatures thoroughly to make sure it was completely fixed. If it made no difference, it's probably a failing power supply.
|
|
#
?
Apr 4, 2017 21:49
|
|
|
One other thing you can do is go into the NVidia control panel, and change the PhysX settings from 'auto-select' to the GTX 970.
|
|
#
?
Apr 5, 2017 00:21
|
|
|
GuavaMoment posted:One other thing you can do is go into the NVidia control panel, and change the PhysX settings from 'auto-select' to the GTX 970.
|
|
#
?
Apr 5, 2017 00:32
|
|
|
Alereon posted:It's either overheating or your power supply. I'd suggest opening the case and thoroughly dusting it out, paying attention to all the vents, fans, and heatsinks. While doing that, note the brand and model of the power supply. If that fixes it or significantly reduces how often it happens, it was heat and you should check temperatures thoroughly to make sure it was completely fixed. If it made no difference, it's probably a failing power supply. I opened up the computer and dusted out the inside. I didn't notice any problems with the vents or heat sinks. There was a fair deal of dust stuck on one of the heat sinks, and I had some trouble getting it all out, but it didn't seem like it would be enough to cause a major problem with overheating. The fans tend to get dust stuck on the blades, but it doesn't seem like it'd be enough to cause them to spin slowly, though it could be possible. Regardless, I feel like I was pretty thorough with the dusting. I'll test it again and make another post to see what results I get. I was also able to get the power supply model, as well as the RAM manufacturer and the video card manufactuer. I'll add them here, and add it to the OP as well. Video Card: EVGA Nvidia GTX 970 Power Supply: Thermaltake SP-850AHCCB (850W) RAM: HyperX How would I go about measuring the GPU/CPU temperature? Do I need specific equipment for it, or do I just need a certain kind of software? Also, if it is an issue with the power supply, and not with overheating, then is there a way I can gauge or test it? Is there also a way I can fix it, or do I just need to replace it entirely. GuavaMoment posted:One other thing you can do is go into the NVidia control panel, and change the PhysX settings from 'auto-select' to the GTX 970. Alereon posted:That wouldn't have anything to do with rebooting during games. PhysX was already set to 'Auto-Select'. Max Wilco fucked around with this message at 01:23 on Apr 5, 2017 |
|
#
?
Apr 5, 2017 01:08
|
|
|
CPUID Hardware Monitor is good for temperatures in general and will probably read your GPU correctly, the GPU-Z sensors tab may be able to additional temps like the VRM temperatures instead of just the GPU. If you don't find any high temperatures, are there any errors in the Windows System Event Log (Start, run, eventvwr.msc, Windows Logs, System) at the time of the crash? Overall, unless you see any high temperatures my money is on the power supply. That model had barely-acceptable DC quality under load when brand new, and a few years will not have been kind to it. Do you have another power supply you can test with?
|
|
#
?
Apr 5, 2017 01:48
|
|
|
Alereon posted:CPUID Hardware Monitor is good for temperatures in general and will probably read your GPU correctly, the GPU-Z sensors tab may be able to additional temps like the VRM temperatures instead of just the GPU. If you don't find any high temperatures, are there any errors in the Windows System Event Log (Start, run, eventvwr.msc, Windows Logs, System) at the time of the crash? I don't have another power supply, unfortunately, and dusting it out didn't seem to help; Payday 2 still crashed after playing it for 5-10 minutes or so. I downloaded the two applications you listed above. Is there something specific you want me to record and post to help in determining the problem? I can start Payday 2 and try to see readings I get, but I don't know if there's a way to record the reading at the point where it crashes. Is there also a power supply you'd recommend as a replacement? EDIT: I also pulled up the Windows Logs. Did you want me to look for something specific? I keep seeing an error along with the unexpected shutdowns with the message, "Crash Dump Initialization Failed." Max Wilco fucked around with this message at 02:29 on Apr 5, 2017 |
|
#
?
Apr 5, 2017 02:25
|
|
|
Just check the CPU core/package (package is the highest of the cores) temperatures, as well as the GPU temperatures. CPU and GPU should be under 80C. If they are above this it is still overheating and you'll need to fix that.
|
|
#
?
Apr 5, 2017 02:47
|
|
|
Alereon posted:Just check the CPU core/package (package is the highest of the cores) temperatures, as well as the GPU temperatures. CPU and GPU should be under 80C. If they are above this it is still overheating and you'll need to fix that. Here's what I recorded with HWMonitor while running Payday 2 right before it crashed again: 
|
|
#
?
Apr 5, 2017 03:24
|
|
|
That's downright cool. There's not really much of a way to be sure, but based on all these symptoms I think it's the power supply. If it was the motherboard or RAM you'd typically see bluescreens and application errors, and not just in games, and reducing the quality settings wouldn't have the same effect. Similarly, if the videocard was just dying, I think you'd see wider issues. The Parts Picking Megathread is a good choice for finding a new power supply, they probably have suggested models listed in the OP.
|
|
#
?
Apr 5, 2017 03:54
|
|
|
So I ended up taking my computer into the shop where I got it built. It turns out that the problem was with the video card. They ran a test on it, and the issue was that whenever the card was under heavy load, it would crash. They replaced it with a new card, and so far, I haven't had any problems. Regardless,thanks again to everyone who offered advice.
|
|
#
?
Apr 15, 2017 01:30
|
|
|
Shoot. So the computer's back to crashing again, despite the new video card (NVIDIA GTX 1070). It was working fine with Payday 2, but then I tried playing Fallout 4 and it started crashing again. Same thing happened with Mafia 3. I tried downloading the 3DMark demo from Steam to see if I could assess the issue, but only the full version has stress tests, and the benchmark demo ran without crashing. I tried disabling NVIDIA Shadowplay and the Steam Overlay, and it seemed like that worked; I was able to play both Fallout 4 and Mafia 3 at ultra/high settings (save for Vsync being on) for quite a while. I just tried playing Mafia 3 again today though, and it keeps crashing after 3-4 minutes in, even after turning the graphic settings down to medium. I called the shop where I got it built, and they recommended trying a different outlet. I unplugged everything from the power-strip, and tried it on a different power-strip on a different outlet, but no luck. Is it possible that there's still an issue with the power supply? I'm going to see if there's an issue with the outlets where I have the computer, but I still feel like that's the problem.
|
|
#
?
Apr 21, 2017 23:40
|
|
|
Max Wilco posted:Is it possible that there's still an issue with the power supply? I'm going to see if there's an issue with the outlets where I have the computer, but I still feel like that's the problem. Yes, it's much more likely to be a PSU issue than an issue with an outlet or a power strip.
|
|
#
?
Apr 28, 2017 05:30
|
|
|
|
| # ? Apr 20, 2024 14:28 |
|
|
Zogo posted:Yes, it's much more likely to be a PSU issue than an issue with an outlet or a power strip. Yeah, I took it in again, and the was power supply going dead. Putting the new video card is what finally did it in. Regardless, it's been replaced and I haven't been having any issues so far. Thanks again to everyone.
|
|
#
?
May 3, 2017 08:28
|
|