Hardware & Technical MSI Afterburner causing PC to reboot?

Yesterday I installed a 2080 TI card, a 700w PSU and an SSD. Everything was totally fine all of yesterday, but I noticed my fan noise is very high on the graphics card when inside stations, so I decided to download MSI Afterburners to monitor the card temp. The very next time I started up Elite, my PC rebooted after playing for less than a minute. I tried it again and it did the exact same thing. I uninstalled afterburner and now it’s been running fine for over an hour. Could afterburner really have caused this or is something probably wrong with the card?
 
When was the last time you cleaned your PC?
This sounds like a heat problem (fans ramping up, random reboots,...)

When you uninstalled the afterburner, it just means your GPU went back to its stock clocks so it doesn't overheat as much right now, but it only delayed the inevitable, I'd say and it will start overheat again once more dust builds up.
 
When was the last time you cleaned your PC?
This sounds like a heat problem (fans ramping up, random reboots,...)

When you uninstalled the afterburner, it just means your GPU went back to its stock clocks so it doesn't overheat as much right now, but it only delayed the inevitable, I'd say and it will start overheat again once more dust builds up.

I mildly cleaned it yesterday when it was open, however I didn’t have any compressed air. I ordered a Can today so I will open it back up and try to give it a good cleaning. According to afterburner my heat never got over 82c. The limit was preset at 84c.
 
I mildly cleaned it yesterday when it was open, however I didn’t have any compressed air. I ordered a Can today so I will open it back up and try to give it a good cleaning. According to afterburner my heat never got over 82c. The limit was preset at 84c.
That is still pretty hot, though. Be sure to give your card's fins a good blowing, once you get the can. CPU cooler, too.
Also check the front fans. If they have a dust filter, it could have gotten clogged up and you system may be starved of air altogether.
 
That is still pretty hot, though. Be sure to give your card's fins a good blowing, once you get the can. CPU cooler, too.
Also check the front fans. If they have a dust filter, it could have gotten clogged up and you system may be starved of air altogether.

the graphics card is brand new, it was literally unboxed yesterday. I did notice the processor fan had a little bit of dust on it, however. I’m going to give everything a good cleaning though. Is there a way to monitor gpu temp from nividia control panel?
 
the graphics card is brand new, it was literally unboxed yesterday. I did notice the processor fan had a little bit of dust on it, however. I’m going to give everything a good cleaning though. Is there a way to monitor gpu temp from nividia control panel?
If you want to monitor your PC, I very much recommend HWinfo64. It's a great tool that can help you monitor EVERYTHING. Temps, clocks, voltages,...

If the GPU is new, then the main suspect for resetting your PC is the CPU itself, then.
If it's more than 2 years that you had it in your system, it would also be a good idea (if cleaning the cooler won't help) to take off the cooler and replace the thermal paste, which could have dried out, especially if you've been running it hot for some time already.
 
82C is a relatively comfortable temperature for GPU, but you lose some clocks. Check power delivery in your overclocking software, it's often what causes instability. Try to give it more power. Try to reproduce the crash and check event viewer. I think it's a pretty common NVidia driver crash.
Also, according to the other thread you run Elite at 4k with supersampling thrown in. That ought to put any GPU on it's knees.
 
If you want to monitor your PC, I very much recommend HWinfo64. It's a great tool that can help you monitor EVERYTHING. Temps, clocks, voltages,...

If the GPU is new, then the main suspect for resetting your PC is the CPU itself, then.
If it's more than 2 years that you had it in your system, it would also be a good idea (if cleaning the cooler won't help) to take off the cooler and replace the thermal paste, which could have dried out, especially if you've been running it hot for some time already.

Thank you! I’ll definitely follow your advice, I really appreciate it. A little bit of an update also, windows 10 said my driver was up-to-date, but I went ahead and downloaded the latest driver from Nvidia’s website anyways.

After installing the driver the card is much quieter now, the fan noise has been cut in half. Frame rates are slightly lower, maybe five or 10fps less, but I can live with that.
 
82C is a relatively comfortable temperature for GPU, but you lose some clocks. Check power delivery in your overclocking software, it's often what causes instability. Try to give it more power. Try to reproduce the crash and check event viewer. I think it's a pretty common NVidia driver crash.
Also, according to the other thread you run Elite at 4k with supersampling thrown in. That ought to put any GPU on it's knees.

yes, I also took your advice and lowered the SS. Speaking of that, is anti-aliasing and supersampling related? If so, which AA setting is preferred? FXAA, SMAA or MLAA2-4?
 
yes, I also took your advice and lowered the SS. Speaking of that, is anti-aliasing and supersampling related? If so, which AA setting is preferred? FXAA, SMAA or MLAA2-4?
Antialiasing helps with smoothing the jagged edges of the picture. They all work a bit differently, with FXAA being the basic (least performance hit) and MLAA being the most precise (with bigger performance hit). Neither of these should give your 2080ti any trouble, though.
Supersampling renders the image in higher resolution and then downsizes it to fit your monitor. The resulting image is therefore sharper. So from a certain point of view it "kind of" does something similar to antialiasing (i.e. it results in better image without jagged edges and edge flickering) but it's a completely different process. It's more useful (and noticeable) on monitors with lower resolution (like 1080p). The higher the base resolution, the less of an effect it has on visual quality. If you have a 4K monitor, you don't really need it. It's very useful in VR, though.
 
If you weren't using MSI AB to load any profiles that changed the card's clocks or voltages, the most likely cause for the instability would be it's polling of temp sensors and such. You may want to mess with the "compatibility properties" settings, or shut down other applications that could be trying to monitor the same things.

yes, I also took your advice and lowered the SS. Speaking of that, is anti-aliasing and supersampling related? If so, which AA setting is preferred? FXAA, SMAA or MLAA2-4?

ED, like most applications that use differed shading, does not support conventional MSAA (multisampling anti-aliasing), so that leaves supersampling (rendering at a higher internal resolution then scalling to the output resolution, which is the best, but by far slowest way to reduce jaggies), or various post-processing AA filters (FXAA, MLAA, SMAA, etc).

Depending on what your display's native resolution is, and your desired frame rate target, you may or may not want to run supersampling, but a 2080 Ti can handle up to about 5k without dipping below 60 fps, so if you are on a lower resolution display, you can probably get away with considerable levels of supersampling (which in ED is a multiplier of the linear resolution values...eg if you have a 1080p display 2x supersampling will be 3840*2160 internal resolution, or 4k).

Unless you don't like the effect, I almost always recommend SMAA for the actual 'anti-aliasing' setting in-game. It generally looks the best and doesn't perform any worse than the other filters. FXAA tends to blur stuff too much, while the MLAA settings tend to be worse at mitigating jaggies while still being slower than SMAA.
 
Back
Top Bottom