RTX 2080 Spontaneous Reboot Crashes From Certain Games...

S

Sithtiger

Weaksauce
Joined
Jun 23, 2016
Messages
96
Hi, so I'm having spontaneous reboots with certain games. So far it's happened with COD:AW and Battletech: Urban Warfare or Portal 2. Interestingly for Portal 2, it only crashed when playing Co-op. I believe I fixed that game. It seems Portal 2's in-game voice chat was causing it. I disabled it and used Discord and it seems fine now! I have the newest version of Windows 10 (was using the older version before the version that broke VPN's, same problem) and I have reason to believe it's my RTX 2080, which is my 2nd. The first one I sent back in February of this year. I don't believe it's my PSU because I have an older 1070 that I used when I was waiting for my new 2080 to get sent to me. The 1070 I have didn't have any problems running the same games my previous 2080 had problems with.

It's not heat because I've played other games like Battlefield V or Anthem, etc and it worked fine for hours. I realize Anthem is a newer game (and dead now) but you'd think, it would crash on it. In addition to all of that, I had my old system running my 1070 with an MSI Z97 Gaming 5 mobo, 32GB of DDR3 RAM, same SSD's HDD's and Blu-ray and same PSU and OS. No problem with the 1070 in either system. I think it's a driver problem but I don't recall what driver I used a while ago to fix some of the issues.

Anyway, does anyone have ANY idea as to why this is happening? In almost every instance, when it would shut down, it would only be after a couple of minutes and actually after I updated to the newest Windows update I tested COD: Advanced Warfare, which can actually be VERY demanding if you have everything turned up like Supersampling, which I thought was causing the problem, but wasn't because I monitored temps and it was running at 1870Mhz and 72 C. The hottest I've seen it for around an hour and it didn't crash once. I also ran MSI Kombuster graphics test....never crashed. Every time it would crash, I never got a BSOD, it would always just reboot. There was nothing in the Event logs of an error because of the GPU, but the only problem it logged was this:
Event 10, EnhancedStorage-EhStorEcgDrv

A TCG Command has returned an error.
Desc: AuthenticateSession
Param1: 0x1
Param2: 0x60000001C
Param3: 0x900000006
Param4: 0x0
Status: 0x1

The ONLY relevance I found with his was a Microsoft answer that it was this guy's NIC. In this guy's case, it was never resolved. I don't believe their advice will help me because if that was the problem then why doesn't my 1070 have any problems with ANY games? Currently, I'm using the new 431.36 Nvidia driver and before that, the 430.86 driver. Same problems. I don't get it!?!

Thanks in advance!
 
So I updated my NIC, there was an update available,but Battletech: Urban Warfare still reboots. I'll have to check COD: Advanced Warfare again though. Again, I don't think it's the PSU because if it were that, it would crash and reboot anytime it was stressed, but it doesn't. I wonder if the new RTX 2000 series Super cards do this as well?
 
Have you tried a few games locally and not online to see if they crash?
 
I am going to bet power supply. I had this exact issue on one of my cards there is a way to test this for sure. install an overclocking app like afterburner or precision x and turn the power section down to 80 % or 60% and so on until it no longer reboots if it stops rebooting then power is for sure the issue if this does not help then you can look elsewhere. I had this issue running an rtx 2080 on an 400whatt PSU had to turn power down to stop rebooting so I could test a few things before putting it in my main rig.

Thent
 
Thently said:
I am going to bet power supply. I had this exact issue on one of my cards there is a way to test this for sure. install an overclocking app like afterburner or precision x and turn the power section down to 80 % or 60% and so on until it no longer reboots if it stops rebooting then power is for sure the issue if this does not help then you can look elsewhere. I had this issue running an rtx 2080 on an 400whatt PSU had to turn power down to stop rebooting so I could test a few things before putting it in my main rig.

Thent
Click to expand...

Hi, yes I have Afterburner on. I fear that you're right. I've had my Corsair HX 850 since 2012. I had no problems with my GTX 1070 with Afterburner. I'll try that tonight. It was happening with COD:AW and I turned off Shader Preload, and turned Super Sampling down from 16x to 4x. I had a similar problem with COD 4: Modern Warfare Remastered...well Shader Preload I turned down. I know I have enough power. I notice with HW Monitor with Afterburner on when testing Furmark my 2080 was running up to 1875 MHz, which is REALLY high, but the temps aren't necessarily up there. I know it's not a thermal issue with the CPU or GPU. When using Furmark, I've only tested for 10 minutes but the temp got up to 85 C but it never crashed. On COD:AW though (before I lowered those settings) it would crash within a few seconds once the level loaded.

I know my PSU is powerful enough to power the GPU, so I'm wondering if the auto O/Cer in Afterburner is pushing the card too hard? I mean 1875 MHz is really high. I think the voltages look Ok. That is, it's getting 12 volts, but I honestly don't know what to look for other than the 12v rail and noticeable fluctuations in the power, which I don't see. I think it reboots without Afterburner on as well, but I'll test it with and without and I'll set the power level to 80% or lower. I could also lock the speed to a certain frequency because I remember testing the card and it couldn't hold 1875 MHz.

I'll report back when I know something new.
 
Thread then should be should called:
"OC'ed via third party app RTX 2080 Spontaneous Reboot Crashes From Certain Games..." ;)
 
It is anecdotal, but I had two Corsair HX850 go bad on me (the original and the RMA replacement). It took a few years in between failures each time.

Conversely, I have an AX850 (bought to replace the HX850 while waiting on the RMA) and it has been flawless (now in a family member’s computer).
 
I had a titan X (Maxwell) go bad on me this way. It would reboot randomly when playing certain games. I had replaced the PSU because that initially looked like the culprit but that didn't solve the issue. Turned out the card was going bad and it eventually got to the point where the card will either run for hours on end without problems or immediately cause a reboot that triggers the anti surge message at boot. This may not be the case with yours but it's something to check.
 
Sithtiger said:
spontaneous reboots
Click to expand...
Sithtiger said:
it would shut down
Click to expand...
So which is it? Rebooting or shutting down? Or both?

Sithtiger said:
I know my PSU is powerful enough to power the GPU, so I'm wondering if the auto O/Cer in Afterburner is pushing the card too hard?
Click to expand...
If you are experiencing shutdowns where the computer actually powers off (not BSOD, not reboot, but actually turns off and stays off very suddenly), then it will more than likely be a PSU issue. Likely an over current protection (OCP) or under voltage trip inside the PSU. The PSU is trying to protect itself and your computer from detected low voltage drops/sags or excessive current draw. Even though the AVERAGE power draw is less than what the PSU is rated for, the INSTANTANEOUS power draw in certain loads (ie. certain games, doing things in certain games, etc.) can exceed the PSU's built-in OCP thresholds and the computer will spontaneously and immediately power off with no warning whatsoever. The only practical solution is to buy a higher wattage PSU. The OCP threshold will be higher (most likely) in the higher-wattage PSU, thus preventing the shutdowns. Sometimes switching from one manufacturer to another in the same wattage range can solve it as well - each brand/design of PSU has different OCP designs and thresholds. But the more reliable choice is to bump up to the next wattage level of PSU.

This kind of thing is going to become more and more common as the GPUs and CPU get more powerful. The new AMD CPUs can alter the current draw in as little as 1ms intervals. Modern GPUs are similar in design. With more and more parallelization of higher numbers of cores/threads and Cuda cores/Shading units/TMUs/ROPs/SM Count/Tensor cores/RT cores, it is possible for games to suddenly "turn them all on at once" with a correspondingly high call for increased voltage and current. When it was just Quad core CPUs and a max of 400 GPU cores it was no problem. Now it's different.
 
Azrak said:
So which is it? Rebooting or shutting down? Or both?


If you are experiencing shutdowns where the computer actually powers off (not BSOD, not reboot, but actually turns off and stays off very suddenly), then it will more than likely be a PSU issue. Likely an over current protection (OCP) or under voltage trip inside the PSU. The PSU is trying to protect itself and your computer from detected low voltage drops/sags or excessive current draw. Even though the AVERAGE power draw is less than what the PSU is rated for, the INSTANTANEOUS power draw in certain loads (ie. certain games, doing things in certain games, etc.) can exceed the PSU's built-in OCP thresholds and the computer will spontaneously and immediately power off with no warning whatsoever. The only practical solution is to buy a higher wattage PSU. The OCP threshold will be higher (most likely) in the higher-wattage PSU, thus preventing the shutdowns. Sometimes switching from one manufacturer to another in the same wattage range can solve it as well - each brand/design of PSU has different OCP designs and thresholds. But the more reliable choice is to bump up to the next wattage level of PSU.

This kind of thing is going to become more and more common as the GPUs and CPU get more powerful. The new AMD CPUs can alter the current draw in as little as 1ms intervals. Modern GPUs are similar in design. With more and more parallelization of higher numbers of cores/threads and Cuda cores/Shading units/TMUs/ROPs/SM Count/Tensor cores/RT cores, it is possible for games to suddenly "turn them all on at once" with a correspondingly high call for increased voltage and current. When it was just Quad core CPUs and a max of 400 GPU cores it was no problem. Now it's different.
Click to expand...

Ok, so while I can't be 100% sure until I put a new PSU you, I feel pretty certain that's what it is now! Before 123Lanoix replied I was reading other posts from other people. I posted with the same problem earlier this year with the same card. I was brand new. I got it replaced with the same model and it seemed fine for a while, but certain games would cause it to do this, so this has been going on for a while. That's why it's been VERY difficult to pinpoint. I even ran MSI Afterburner OC Scanner just to push the hell out of it and it never crashed, not once and I'd actually never done that with this particular GPU before. It was pushed harder and was clocked higher than ever and it only failed when it was pushed to 2025MHz Boost clock! Even then the temp only got to 74C! When it would crash before in certain games, it sometimes crashed at 69C and the Boost clock was only 1730MHz or so. I've played for hours with the temps as high 77 to 80C and it's been fine. It's not a driver, I've tested virtually all of them.

I read what you posted here and it was the same question that you asked: "Rebooting or Shutting down"? It was very similar and it hit me. If it's shutting down and you're not O/Cing and it isn't a driver problem and I'm pretty sure it's not a GPU problem or RAM or CPU, the most likely culprit is the PSU. I've never seen a PSU that would work with some games, while others it has a hard crash, but it does reboot, but the key is it shuts down without warning and no error messages. The only errors are in Event Viewer and it just says the PC was shut down incorrectly...nothing else.

This has been a long time coming though, because I had this with my 1070 too, but it rarely happened....like maybe 3x a year at most, but looking at all of it, it makes sense now. Also, it happened when I was O/Cing my new i5 9600k, but I had the same response. It just shut down and restarted without warning. Before you say it's heat, it only got to 85C at 4.8GHz and it also shut down and rebooted at 4.7GHz too with of a temp of something similar..don't recall exactly. I have a Hyper 212 too. It's not the best but you can see those temps shouldn't have caused that. Anyway, I should have reworded the title of the thread differently, but I wasn't really thinking it through and initially when I read that question probably a day before 123Lanoix and you and other ppl responded, I might have thought it was just restarting until I read that question and I triggered the problem to make sure it was shutting down and it was and then it all clicked. As I said, I've never seen this type of behavior before.

Sorry, this is a long response but I have an interesting ending (although I suppose it's not actually over until I update you, as I will so bear with me). So I was looking at the new PSU's to buy. I found an 850 watt Seasonic Gold at Newegg on sale for $120!!! 5 out of 5 stars/eggs. I looked back to see how much mine cost and I noticed it had a 7 year warranty on it. I looked at the date and I had 2 days left before it ran out!!! So I spoke with Corsair and they didn't have an HX850 Silver (yeah Newegg sold me the wrong one...should have been a Gold) but they asked if an HX850 Platinum would do and I emphatically said yes since the Platium has a higher and cleaner power output at 50%. I think it's supposed to be rated at 95% or so, but even 90% is fine. I just want it to work. The only downer is I have to send my PSU to them first and then they'll send the replacement out.

So, assuming this works, not only will I FINALLY be able to play all my games with no problems, I can quite possibly O/C my CPU to 5GHz or more!!! So, thanks for replying Azrak as you're further reinforcing my confidence that this is indeed the problem. I have another PSU I'm going to use until my replacement comes. It's my sons and it's pretty old. It's a Corsair HX620 watt PSU. It has an 8-pin connector, but 2080 needs an 8+6 pin connector. Couldn't I plug a connector from my HX850 into the HX620 PSU into the 2080 and have it work? If not, I can always put my 1070 back in for a bit. I live near Riverside and Corsair's RMA dept. is in Fresno. It shouldn't take that long for it to get there, yes? I just want to make sure I'm not going to damage anything by plugging a connector from the 850 to the 620. They are both Corsair and both HX models, so I'd think it would be fine, but if there's a problem I'd like to know. Thanks!

Factum said:
Thread then should be should called:
"OC'ed via third party app RTX 2080 Spontaneous Reboot Crashes From Certain Games..." ;)
Click to expand...

Well, I was about to do that, but I tested without Afterburner and it crashed all the same, so Afterburner isn't the cause. It appears to be my PSU. I'll report back when I get the replacement or perhaps with my son's if it's stable itself. It is old.

123Lanoix said:
I would check either Power supply or your RAM run memtest86
Click to expand...
I ran memtest86 when I first encountered this problem at the beginning of the year and the RAM was Ok. Thanks for the suggestion though. Had I not tried that, it might have been the culprit.

jolli said:
It is anecdotal, but I had two Corsair HX850 go bad on me (the original and the RMA replacement). It took a few years in between failures each time.

Conversely, I have an AX850 (bought to replace the HX850 while waiting on the RMA) and it has been flawless (now in a family member’s computer).
Click to expand...

Yeah, the next time I buy a new PSU, I will buy a Seasonic. I'll probably buy the one I mentioned above, the Seasonic Focus Plus 850 watt Gold Series. It was on sale for $120, but it's $140 right now, but they're always on sale. I'd be willing to bet that it will be $100 around Black Friday or so.

Pulsar said:
I had a titan X (Maxwell) go bad on me this way. It would reboot randomly when playing certain games. I had replaced the PSU because that initially looked like the culprit but that didn't solve the issue. Turned out the card was going bad and it eventually got to the point where the card will either run for hours on end without problems or immediately cause a reboot that triggers the anti surge message at boot. This may not be the case with yours but it's something to check.
Click to expand...

I'm hoping that doesn't happen to me. I have to say I'm concerned about that. I'm sure they will send me a refurbished PSU and it won't work! I'll be buying a Seasonic PSU the next time. I'm not pleased that Corsair doesn't offer an option of getting your CC # and sending your the replacement PSU first and then if they don't receive the RMA'd PSU, then charge the customer's card! Oh no, they require you to pay for the PSU replacement if you want it first and then when they get your PSU, they'll refund you. If I had the money to do that, I'd just buy that Seasonic. They wanted to charge me $185 for that replacement PSU, so I'm not sure if it's a refurbished one or not because that's what the new HX850 Platinum costs.

Azrak said:
So which is it? Rebooting or shutting down? Or both?


If you are experiencing shutdowns where the computer actually powers off (not BSOD, not reboot, but actually turns off and stays off very suddenly), then it will more than likely be a PSU issue. Likely an over current protection (OCP) or under voltage trip inside the PSU. The PSU is trying to protect itself and your computer from detected low voltage drops/sags or excessive current draw. Even though the AVERAGE power draw is less than what the PSU is rated for, the INSTANTANEOUS power draw in certain loads (ie. certain games, doing things in certain games, etc.) can exceed the PSU's built-in OCP thresholds and the computer will spontaneously and immediately power off with no warning whatsoever. The only practical solution is to buy a higher wattage PSU. The OCP threshold will be higher (most likely) in the higher-wattage PSU, thus preventing the shutdowns. Sometimes switching from one manufacturer to another in the same wattage range can solve it as well - each brand/design of PSU has different OCP designs and thresholds. But the more reliable choice is to bump up to the next wattage level of PSU.

This kind of thing is going to become more and more common as the GPUs and CPU get more powerful. The new AMD CPUs can alter the current draw in as little as 1ms intervals. Modern GPUs are similar in design. With more and more parallelization of higher numbers of cores/threads and Cuda cores/Shading units/TMUs/ROPs/SM Count/Tensor cores/RT cores, it is possible for games to suddenly "turn them all on at once" with a correspondingly high call for increased voltage and current. When it was just Quad core CPUs and a max of 400 GPU cores it was no problem. Now it's different.
Click to expand...

As I mentioned earlier, it completely shuts down and then restarts. The way a CPU should react when it gets too hot if you're testing it and I'm talking about in Windows would be the CPU will lock up, right? If it's O/Ced really high, then yeah, the PC might shut down and then reboot or just shutdown but if you're careful like I am it just locks up before it reboots. So, with this build, I was still using my 1070 because I had to save a couple months so I could buy the 2080 in January in this year. With my 1070 with this system, I had some occasional problems with games maybe 2 or 3 times a year and yeah, it just shut down completely and rebooted with no warning message. The thing is, I ran the game that did this and it was fine after that and I changed nothing. So the same thing happened with my previous system, but it only happened once or twice in a year. My point is my PSU starting failing about 3 years ago. I was not aware PSU's could just slowly die in that manner. It's frustrating and infuriating when you test it and the voltages and everything is fine. My friend has a multimeter for PSU's (I need one badly) and naturally it was just fine. I'll include a pic of my CPU, PSU and GPU temps. I have a log file too, but I won't post that. I watched it and it showed nothing because it never crashed. I'm referring to the Afterburner OC Scanner test. I also looked at a log file from HWiNFO when I had a crash and everything looked normal.

I'd like to thank everyone that read, replied and offered suggestions. It helps. I now have more insight into PSU problems and how they can slowly go out and otherwise work fine until pushed using certain games! I'll get back to you guys to let you know if the PSU is in fact the problem. I feel confident it is but I won't say until I'm sure.
 
Sithtiger said:
Ok, so while I can't be 100% sure until I put a new PSU you, I feel pretty certain that's what it is now! Before 123Lanoix replied I was reading other posts from other people. I posted with the same problem earlier this year with the same card. I was brand new. I got it replaced with the same model and it seemed fine for a while, but certain games would cause it to do this, so this has been going on for a while. That's why it's been VERY difficult to pinpoint. I even ran MSI Afterburner OC Scanner just to push the hell out of it and it never crashed, not once and I'd actually never done that with this particular GPU before. It was pushed harder and was clocked higher than ever and it only failed when it was pushed to 2025MHz Boost clock! Even then the temp only got to 74C! When it would crash before in certain games, it sometimes crashed at 69C and the Boost clock was only 1730MHz or so. I've played for hours with the temps as high 77 to 80C and it's been fine. It's not a driver, I've tested virtually all of them.

I read what you posted here and it was the same question that you asked: "Rebooting or Shutting down"? It was very similar and it hit me. If it's shutting down and you're not O/Cing and it isn't a driver problem and I'm pretty sure it's not a GPU problem or RAM or CPU, the most likely culprit is the PSU. I've never seen a PSU that would work with some games, while others it has a hard crash, but it does reboot, but the key is it shuts down without warning and no error messages. The only errors are in Event Viewer and it just says the PC was shut down incorrectly...nothing else.

This has been a long time coming though, because I had this with my 1070 too, but it rarely happened....like maybe 3x a year at most, but looking at all of it, it makes sense now. Also, it happened when I was O/Cing my new i5 9600k, but I had the same response. It just shut down and restarted without warning. Before you say it's heat, it only got to 85C at 4.8GHz and it also shut down and rebooted at 4.7GHz too with of a temp of something similar..don't recall exactly. I have a Hyper 212 too. It's not the best but you can see those temps shouldn't have caused that. Anyway, I should have reworded the title of the thread differently, but I wasn't really thinking it through and initially when I read that question probably a day before 123Lanoix and you and other ppl responded, I might have thought it was just restarting until I read that question and I triggered the problem to make sure it was shutting down and it was and then it all clicked. As I said, I've never seen this type of behavior before.

Sorry, this is a long response but I have an interesting ending (although I suppose it's not actually over until I update you, as I will so bear with me). So I was looking at the new PSU's to buy. I found an 850 watt Seasonic Gold at Newegg on sale for $120!!! 5 out of 5 stars/eggs. I looked back to see how much mine cost and I noticed it had a 7 year warranty on it. I looked at the date and I had 2 days left before it ran out!!! So I spoke with Corsair and they didn't have an HX850 Silver (yeah Newegg sold me the wrong one...should have been a Gold) but they asked if an HX850 Platinum would do and I emphatically said yes since the Platium has a higher and cleaner power output at 50%. I think it's supposed to be rated at 95% or so, but even 90% is fine. I just want it to work. The only downer is I have to send my PSU to them first and then they'll send the replacement out.

So, assuming this works, not only will I FINALLY be able to play all my games with no problems, I can quite possibly O/C my CPU to 5GHz or more!!! So, thanks for replying Azrak as you're further reinforcing my confidence that this is indeed the problem. I have another PSU I'm going to use until my replacement comes. It's my sons and it's pretty old. It's a Corsair HX620 watt PSU. It has an 8-pin connector, but 2080 needs an 8+6 pin connector. Couldn't I plug a connector from my HX850 into the HX620 PSU into the 2080 and have it work? If not, I can always put my 1070 back in for a bit. I live near Riverside and Corsair's RMA dept. is in Fresno. It shouldn't take that long for it to get there, yes? I just want to make sure I'm not going to damage anything by plugging a connector from the 850 to the 620. They are both Corsair and both HX models, so I'd think it would be fine, but if there's a problem I'd like to know. Thanks!



Well, I was about to do that, but I tested without Afterburner and it crashed all the same, so Afterburner isn't the cause. It appears to be my PSU. I'll report back when I get the replacement or perhaps with my son's if it's stable itself. It is old.

I ran memtest86 when I first encountered this problem at the beginning of the year and the RAM was Ok. Thanks for the suggestion though. Had I not tried that, it might have been the culprit.



Yeah, the next time I buy a new PSU, I will buy a Seasonic. I'll probably buy the one I mentioned above, the Seasonic Focus Plus 850 watt Gold Series. It was on sale for $120, but it's $140 right now, but they're always on sale. I'd be willing to bet that it will be $100 around Black Friday or so.



I'm hoping that doesn't happen to me. I have to say I'm concerned about that. I'm sure they will send me a refurbished PSU and it won't work! I'll be buying a Seasonic PSU the next time. I'm not pleased that Corsair doesn't offer an option of getting your CC # and sending your the replacement PSU first and then if they don't receive the RMA'd PSU, then charge the customer's card! Oh no, they require you to pay for the PSU replacement if you want it first and then when they get your PSU, they'll refund you. If I had the money to do that, I'd just buy that Seasonic. They wanted to charge me $185 for that replacement PSU, so I'm not sure if it's a refurbished one or not because that's what the new HX850 Platinum costs.



As I mentioned earlier, it completely shuts down and then restarts. The way a CPU should react when it gets too hot if you're testing it and I'm talking about in Windows would be the CPU will lock up, right? If it's O/Ced really high, then yeah, the PC might shut down and then reboot or just shutdown but if you're careful like I am it just locks up before it reboots. So, with this build, I was still using my 1070 because I had to save a couple months so I could buy the 2080 in January in this year. With my 1070 with this system, I had some occasional problems with games maybe 2 or 3 times a year and yeah, it just shut down completely and rebooted with no warning message. The thing is, I ran the game that did this and it was fine after that and I changed nothing. So the same thing happened with my previous system, but it only happened once or twice in a year. My point is my PSU starting failing about 3 years ago. I was not aware PSU's could just slowly die in that manner. It's frustrating and infuriating when you test it and the voltages and everything is fine. My friend has a multimeter for PSU's (I need one badly) and naturally it was just fine. I'll include a pic of my CPU, PSU and GPU temps. I have a log file too, but I won't post that. I watched it and it showed nothing because it never crashed. I'm referring to the Afterburner OC Scanner test. I also looked at a log file from HWiNFO when I had a crash and everything looked normal.

I'd like to thank everyone that read, replied and offered suggestions. It helps. I now have more insight into PSU problems and how they can slowly go out and otherwise work fine until pushed using certain games! I'll get back to you guys to let you know if the PSU is in fact the problem. I feel confident it is but I won't say until I'm sure.
Click to expand...

I didn’t read all that, but personally I like single 12V rail PSUs. I have 1600W EVGA P2’s running my rigs. I also size my PSUs so I don’t go over 75% capacity. Most people say 80%.

I’d make sure whatever you buy has a single rail. I know the multirail caused some issues for people with RTX cards, but I had single rail for ~6 years now so I was good.
 
Ok, so I'm temporarily using my son's PSU until the replacement PSU comes in. He's only been using his laptop anyone for around 6 months. Anyway, I wanna tell everyone thanks so much and yeah, it was the PSU, because I'm using this Corsair HX620 and I immediately tried a benchmark that caused my PC to instantly shut down and restart before and did the same with the 620 and it was fine. I also played COD:AW with the lowered settings as I had before and it was Ok sometimes with the 850, but would eventually crash and then when it would boot up again I'd try it and it would crash not a minute into the game. Anyway it ran fan with the lowered settings and I restored everything to what I was running before the crash and then even turned Supersampling up to 16x and it ran for a long time and never crashed!!! I eventually put it down to 8x and that's fine but it still too much for a 2080. If I disable verticle sync, I can run 8x but 4x is optimal.

I just want to say thanks to everyone again for helping and this is case close. This is officially solved! It was a bad PSU. It's just weird the way it was going bad. I ran certain games (that would usually crash it) and even MSI Afterburner when it got hotter than ever before, except maybe a Furmark test, but it clocked higher than before, which was 2025 MHz for the Core for Boost mode. The memory maxed at 8000 MHz. So Afterburner decided on a clock speed of +125 for the core and 7000 MHz for the memory. I don't know if that translates to actually +125 or not. I haven't tried it yet. I want to wait until I get my PSU in.

What's funny is I've had PSU problems for 2 years but I didn't realize it was the PSU because I was using my 1070 and my PC only crashed 2 or 3 times last year but now I realize it was the PSU because it shut down and restarted. If it was the GPU getting too hot, it would have just frozen and then Windows would recover or reboot without losing power (as I've seen it do this many times before working on computers) or either spontaneously shut down and reboot. I also can look back and see that this was causing CPU issues when I built this PC in December of 2018 and the CPU would spontaneously shut down and restart and the temp for the CPU wasn't hot enough to cause it to reboot, but now I realize that it was the PSU all along.

I'd just like to know why it ran fine on certain games but on others like COD 4: Modern Warfare Remastered, COD:AW and Darksiders 3 (to name a few) would all cause it to crash. Even testing it with a multimeter showed it as running fine. Ya know I used HWiNFO64 and all the voltages were solid....that is until they weren't, but when that happened it didn't record that in the logs and Windows doesn't record that. Well, at least the problem is fixed now and I know the cause.
 
You must log in or register to reply here.
Back
Top