[SOLVED] Unable to find culprit of RANDOM black screen

[srkcalb]

Hello. It is the first time I post here and I've read many posts before in this forum about similar issues regarding black screens. I got a 1080 ti not long ago and it's been an amazing card so far but recently I've been getting RANDOM black screens. I used to think it was my HDD due to it making an odd sound(one that resembles forcing shutdown a computer/laptop) and the game freezing/black screening. It DOES NOT happen with any game, just with a few ones:

-Star Wars Jedi: Fallen Order (100% random)
-Black Desert Online (has only happened in Dark "bosses" Rifts)
-MechWarrior 5: Mercenaries (after finishing loading a custom match)

I've seen through Windows Event Log that my Nvidia driver stops working and "recovers" but that doesn't happen. After the black screen occurs, my PC is still on. The sound sometimes stops, sometimes it just keeps going and it never recovers of the black screen. It even reboots itself sometimes without me doing anything. I tried re-starting the video driver with WIN + CTRL + SHIFT + B, but it didn't help, didn't do anything. No artifacts anywhere.

I've played Apex(everything maxed out) many times but it has never happened before, just the game getting stuck in character selection which I thought it was the game's fault(I still don't know)

I've managed to reproduce this black screen in BDO three times which helped me to find the one that I thought was the culprit, the HDD(it wasn't the HDD that had Windows installed, it had just games inside, including BDO). I changed it and I made a clean Windows install just in case. I tried to reproduce the black screen once again with the new HDD and clean Windows install fighting one of the Dark Rift bosses and the black screen didn't happen so I thought it was fixed. I kept playing as I do and then I reached another Dark Rift Boss, after he got to half HP, the black screen returned. It has never crashed outside of those games I mentioned before and I've never done OC to it(I know it comes OC'd by default but besides that, I've done nothing).

TEMPS ARE OK, GPU reaches about 72c max temp, and CPU is about 66c to 70c while gaming. While being idle GPU is at 30c and CPU 40c to 50c. I tried running Heaven benchmark for 15 minutes(max temp was 71c) and nothing happened. Although there's a Temp reading in HWMonitor called TMPIN1(It's called Aux in HW iNFO) which reads 52c at idle and 90c or more while gaming. I've read online that this could be the VRM of the processor or the Northbridge. My knowledge is limited here.
RAM IS OK, I've run MemTest for 15 minutes
DRIVERS ARE OK, I've reinstalled them twice using DDU in my past Windows installation and it kept happening
GPU ~, I don't know if this is a sign that my GPU is dying. Could be. But I want to be sure that is not anything else.
MOBO ~, I'm not sure if my MOBO is bad, I doubt it.
PSU ~, My PSU is quite old, I've been using it for many years so it could be dying. I'm gonna check the voltages tomorrow with a Multimeter.

I must mention that my hardware components are quite old and this could be the reason for the black screens, though it didn't happen before. I currently live in a country with many political and economic issues so there's not much I can do. At least I'd like to find the culprit of this black screen. Any guidance would be really helpful. I will try to add a memory dump file later when another black screen occurs.

THANKS!

MOBO: MSI H61M-P20
CPU: i5-2400
GPU: EVGA GTX 1080 Ti SC2 Gaming iCX
RAM: Ramaxel 1333MHz 4gb x2 Dual Channel
OS: Windows 10 Pro 1909
PSU: CoolMax CUL-750B
Using a DVI Cable on a 1600x900 monitor and an HDMI cable for a second TV monitor at 1920x1080

PD: I know my GPU is heavily bottlenecked by my CPU but I can't change it yet due to economic reasons but still, this black screen shouldn't be happening.

 

[Nemesia]

I would point your problem to that PSU you said is quite old. This is quite a low tier PSU.

 

[Frag Maniac]

The specific part of a VRM to worry about is the MOSFET, which typically have a max safe temp limit of 150c. Even if the thermal resistance rating is poor, you're still OK up to at least 130c. So I doubt it's your VRMs causing the problem.

That PSU is really garbage though. Hardware Secrets, whom are fairly trustworthy on hardware reviews, after thoroughly testing it said "We can’t understand how a manufacturer label a 450 W power supply as a 750 W and get away with it. "

Source: https://www.hardwaresecrets.com/coolmax-cul-750b-750-w-power-supply-review/10/

Mind you a 1080 Ti requires 600w total system power minimum to function properly, so this is a real concern. It's enough of one in fact to immediately replace it before it does damage to your GPU or other parts, especially since it's 6 years old. The only safe alternative would be playing only very low resource use games until you can afford to replace it.

This is a classic case of using a PSU with a claimed power rating above what you need, vs just sufficing with a decent PSU that has just what you need. In this case however you'd even get MORE power with a decent 600w PSU.

A PSU is not something to add in as last item and go cheap on. Like I said, a poorly designed one can destroy other parts. So at the very least, STOP playing those 3 games, or any others with similar hardware requirements, until you can get a decent PSU.

 

[Dcopymope]

That PSU is really garbage though. Hardware Secrets, whom are fairly trustworthy on hardware reviews, after thoroughly testing it said "We can’t understand how a manufacturer label a 450 W power supply as a 750 W and get away with it. "

Source: https://www.hardwaresecrets.com/coolmax-cul-750b-750-w-power-supply-review/10/

Yeah, well they should know perfectly well how they get away with it, its because there are no actual laws explicitly stating they can't. I mean, holy dog piss in heaven, in certain other industries this would be a law suit waiting to happen, WTF. I would never trust another power supply from Coolmax ever again in life, or any other electronic device for that matter. Where are the standards at? If the I.T industry really wants to make the 80+ rating for instance truly mean something, then it needs to be a requirement, not voluntary, not for critical components like the damn power supply. I can't imagine the aviation industry being this unregulated, lives would certainly be at stake, for crying out loud.

 

[Frag Maniac]

Yeah, well they should know perfectly well how they get away with it, its because there are no actual laws explicitly stating they can't. I mean, holy dog piss in heaven, in certain other industries this would be a law suit waiting to happen, WTF. I would never trust another power supply from Coolmax ever again in life, or any other electronic device for that matter. Where are the standards at? If the I.T industry really wants to make the 80+ rating for instance truly mean something, then it needs to be a requirement, not voluntary, not for critical components like the damn power supply. I can't imagine the aviation industry being this unregulated, lives would certainly be at stake, for crying out loud.

Didn't want to open a whole can of worms on the utter lack of responsible manufacturing laws, but I totally get your point. I just hope OP takes it seriously and researches the next PSU purchase carefully.

 

[Dcopymope]

Didn't want to open a whole can of worms on the utter lack of responsible manufacturing laws, but I totally get your point. I just hope OP takes it seriously and researches the next PSU purchase carefully.

Yeah, the can of worms needs to be opened, wide open, to warn the customer about the lack of oversight. If I had influence, there would be a ton I'd change about the standards in the I.T industry, I would just use the laws governing aviation as inspiration. From my personal experience as a flight dispatcher, the FAA and equivalent international agencies of other countries does not screw around when it comes to oversight, not very nice people.

 

[Frag Maniac]

Well it's quite frankly understandable that the aviation industry has such strict regulations, lives depend on it. That said, we hear on the news all too often that certain airplane manufacturers and airlines are not heeding what the inspectors say, as evidenced recently with Boeing and one of their top employees submitting his resignation after the heat came down.

If that kind of thing can happen in the airline industries, you can bet it's going to happen in an industry where it's customers are treated like ne'er do wells. I mean no one's going to cry foul if Johnny's gaming PC goes tits up, except for a few of his gaming friends.

Again though, as I implied above, this is all very off topic. A topic worth discussing, but perhaps in a thread devoted to it.

 

[srkcalb]

Thanks for the replies and I understand your concern. When I got this PSU I barely knew anything about this stuff, I was a teenager so I didn't pay attention to it. As long as I could play the games I wanted, I was satisfied.

Update about my issue: I tested the PSU today and it seems to be okay. I also tested another 1080ti and the issue still persists so it's not a bad GPU, which makes me feel good. You guys say that it is a terrible PSU so I'll bring tomorrow another PSU that I have just to be sure. And I understand the risks, thanks. I will not play those games again unless I need to test if the issue has been fixed.

There's another thing that I think is the culprit here, the TMPIN1. I'm not entirely sure what it is but at being idle its temperature is about 50c and when I enter in any of those games, the temperature raises until 116c to 120c. Right at that point, the black screen occurs. I think the backplate of my 1080ti is heating up components of my MOBO, my case is really small so it may not have enough refrigeration to cool down. What are your thoughts about this?

Checking onto system error logs and stuff, I managed to find that sometimes the black screen comes along with a BSOD, which explains why sometimes my computer reboots itself, and sometimes it doesn't. Basically, the BSOD doesn't occur all the time.

 

[Frag Maniac]

There's another thing that I think is the culprit here, the TMPIN1. I'm not entirely sure what it is but at being idle its temperature is about 50c and when I enter in any of those games, the temperature raises until 116c to 120c. Right at that point, the black screen occurs. I think the backplate of my 1080ti is heating up components of my MOBO, my case is really small so it may not have enough refrigeration to cool down. What are your thoughts about this?

Like I said, normally a mosfet of a VRM can handle at least 130c, but this is also a 6 yr old MB, and not made for enthusiast gaming. It could be it's just getting too weary for even 120c heat. You also have to factor in that software temp readings have a margin of error.

If you can't afford a MB replacement, especially since doing so properly would involve a new CPU and DDR4 RAM, a possible temporary workaround would be to get a micro sized heatsink kit that has heatsinks that fit your mosfets. They can lower temps 5c to 10c or more.

Just make sure you measure your mosfets first. They should be the parts I highlighted in red below.

Here's just one of many examples of a mosfet heatsink kit. Measure first! If in doubt they're the right parts, you should be able to tell by how hot the above parts feel to touch after gaming for a while. Make sure you clean them with rubbing alcohol before applying the HSs. The higher the percentage of alcohol, the better, but do it when they're cool to the touch. I would also highly recommend replacing the PSU first.

https://www.amazon.com/Easycargo-He...X421W42EXZR&psc=1&refRID=YVDDJ7P14X421W42EXZR

 

[srkcalb]

Like I said, normally a mosfet of a VRM can handle at least 130c, but this is also a 6 yr old MB, and not made for enthusiast gaming. It could be it's just getting too weary for even 120c heat. You also have to factor in that software temp readings have a margin of error.

If you can't afford a MB replacement, especially since doing so properly would involve a new CPU and DDR4 RAM, a possible temporary workaround would be to get a micro sized heatsink kit that has heatsinks that fit your mosfets. They can lower temps 5c to 10c or more.

Just make sure you measure your mosfets first. They should be the parts I highlighted in red below.

Here's just one of many examples of a mosfet heatsink kit. Measure first! If in doubt they're the right parts, you should be able to tell by how hot the above parts feel to touch after gaming for a while. Make sure you clean them with rubbing alcohol before applying the HSs. The higher the percentage of alcohol, the better, but do it when they're cool to the touch. I would also highly recommend replacing the PSU first.

https://www.amazon.com/Easycargo-He...X421W42EXZR&psc=1&refRID=YVDDJ7P14X421W42EXZR

Thanks for the quick answer. I understand what you said about the VRM and I think you're right. They should be able to maintain high temperatures but I'll keep that in mind. I can't get the heatsinks sadly, Amazon can't reach my country, I'd need a courier but anyways. I'll use another PSU tomorrow and I'll post again.

Now, here's the thing. I suspect the GPU is heating up what I think is the Northbridge(I assume by what I've seen on the internet that is this thing right here

https://imgur.com/cxBnqkK

View: https://imgur.com/cxBnqkK
. Considering the size of the 1080 Ti and that the heatsink of the Northbridge has a slight bump, the card could be transferring heat to it. I have 0 knowledge about this but yesterday I noticed that the Northbridge is right behind my GPU backplate. If my backplate were touching the Northbridge and transferring heat to it until it reaches 120c, what would happen? As far as I know, the Northbridge communicates with graphics or PCI-E. What do you think? Would it cause these black screens? Is that even the Northbridge to begin with?

I found a MOBO with the same socket number(1155) that I could use to determine if mine has some kind of issue. It is a Gigabyte P67X-UD3-B3.

 

[Frag Maniac]

Considering the size of the 1080 Ti and that the heatsink of the Northbridge has a slight bump, the card could be transferring heat to it. I have 0 knowledge about this but yesterday I noticed that the Northbridge is right behind my GPU backplate. If my backplate were touching the Northbridge and transferring heat to it until it reaches 120c, what would happen? As far as I know, the Northbridge communicates with graphics or PCI-E.

That seems a bit of a stretch, although like I already said, the easy way to determine where that 120c is coming from is wait until it gets good and hot and touch those various parts. You should be able to feel which are the hottest. Again though, I would do this after replacing the PSU.

The backplate of a GPU serves as both a heatsink and a stiffener, and the edge of it is far from being the hottest part of it. If anything were it actually touching the Northbridge heatsink, it would just act as a larger sink to help shed heat off the NB. Pretty sure you're worrying too much there.

 

[srkcalb]

That seems a bit of a stretch, although like I already said, the easy way to determine where that 120c is coming from is wait until it gets good and hot and touch those various parts. You should be able to feel which are the hottest. Again though, I would do this after replacing the PSU.

The backplate of a GPU serves as both a heatsink and a stiffener, and the edge of it is far from being the hottest part of it. If anything were it actually touching the Northbridge heatsink, it would just act as a larger sink to help shed heat off the NB. Pretty sure you're worrying too much there.

Sorry for taking so long to answer. I decided to follow your advice after switching my motherboard since it did nothing. I turned off my computer and I dismounted the PSU.

Now I’ve come back to ask for advice. There’s two guys selling two decent PSU here where I live.(Both are new and sealed) Here’s thing, I don’t know which one should I get. Since you guys have more experience with gaming PSUs in general, which one would you recommend me buying between those two? I feel a little bit more inclined to EVGA but I don’t know

-Thermaltake Toughpower Grand RGB 850w 80+ Gold
-EVGA Supernova 750w G3

thanks

 

[Frag Maniac]

Sorry for taking so long to answer. I decided to follow your advice after switching my motherboard since it did nothing. I turned off my computer and I dismounted the PSU.

Now I’ve come back to ask for advice. There’s two guys selling two decent PSU here where I live.(Both are new and sealed) Here’s thing, I don’t know which one should I get. Since you guys have more experience with gaming PSUs in general, which one would you recommend me buying between those two? I feel a little bit more inclined to EVGA but I don’t know

-Thermaltake Toughpower Grand RGB 850w 80+ Gold
-EVGA Supernova 750w G3

thanks

I'd definitely go with the EVGA Supernova. It's rated a 9.8/10 on jonnyGURU.

https://www.jonnyguru.com/blog/2016/12/26/evga-supernova-750-g3-power-supply/6/

 

[srkcalb]

I'd definitely go with the EVGA Supernova. It's rated a 9.8/10 on jonnyGURU.

https://www.jonnyguru.com/blog/2016/12/26/evga-supernova-750-g3-power-supply/6/

Thank you. I’ll go get it and check if the issue still persists.

 

[zx128k]

Hello. It is the first time I post here and I've read many posts before in this forum about similar issues regarding black screens. I got a 1080 ti not long ago and it's been an amazing card so far but recently I've been getting RANDOM black screens. I used to think it was my HDD due to it making an odd sound(one that resembles forcing shutdown a computer/laptop) and the game freezing/black screening. It DOES NOT happen with any game, just with a few ones:

-Star Wars Jedi: Fallen Order (100% random)
-Black Desert Online (has only happened in Dark "bosses" Rifts)
-MechWarrior 5: Mercenaries (after finishing loading a custom match)

I've seen through Windows Event Log that my Nvidia driver stops working and "recovers" but that doesn't happen. After the black screen occurs, my PC is still on. The sound sometimes stops, sometimes it just keeps going and it never recovers of the black screen. It even reboots itself sometimes without me doing anything. I tried re-starting the video driver with WIN + CTRL + SHIFT + B, but it didn't help, didn't do anything. No artifacts anywhere.

I've played Apex(everything maxed out) many times but it has never happened before, just the game getting stuck in character selection which I thought it was the game's fault(I still don't know)

I've managed to reproduce this black screen in BDO three times which helped me to find the one that I thought was the culprit, the HDD(it wasn't the HDD that had Windows installed, it had just games inside, including BDO). I changed it and I made a clean Windows install just in case. I tried to reproduce the black screen once again with the new HDD and clean Windows install fighting one of the Dark Rift bosses and the black screen didn't happen so I thought it was fixed. I kept playing as I do and then I reached another Dark Rift Boss, after he got to half HP, the black screen returned. It has never crashed outside of those games I mentioned before and I've never done OC to it(I know it comes OC'd by default but besides that, I've done nothing).

TEMPS ARE OK, GPU reaches about 72c max temp, and CPU is about 66c to 70c while gaming. While being idle GPU is at 30c and CPU 40c to 50c. I tried running Heaven benchmark for 15 minutes(max temp was 71c) and nothing happened. Although there's a Temp reading in HWMonitor called TMPIN1(It's called Aux in HW iNFO) which reads 52c at idle and 90c or more while gaming. I've read online that this could be the VRM of the processor or the Northbridge. My knowledge is limited here.
RAM IS OK, I've run MemTest for 15 minutes
DRIVERS ARE OK, I've reinstalled them twice using DDU in my past Windows installation and it kept happening
GPU ~, I don't know if this is a sign that my GPU is dying. Could be. But I want to be sure that is not anything else.
MOBO ~, I'm not sure if my MOBO is bad, I doubt it.
PSU ~, My PSU is quite old, I've been using it for many years so it could be dying. I'm gonna check the voltages tomorrow with a Multimeter.

I must mention that my hardware components are quite old and this could be the reason for the black screens, though it didn't happen before. I currently live in a country with many political and economic issues so there's not much I can do. At least I'd like to find the culprit of this black screen. Any guidance would be really helpful. I will try to add a memory dump file later when another black screen occurs.

THANKS!

MOBO: MSI H61M-P20
CPU: i5-2400
GPU: EVGA GTX 1080 Ti SC2 Gaming iCX
RAM: Ramaxel 1333MHz 4gb x2 Dual Channel
OS: Windows 10 Pro 1909
PSU: CoolMax CUL-750B
Using a DVI Cable on a 1600x900 monitor and an HDMI cable for a second TV monitor at 1920x1080

PD: I know my GPU is heavily bottlenecked by my CPU but I can't change it yet due to economic reasons but still, this black screen shouldn't be happening.

If vt-d is enable in BIOS then disable it and hopefully the black screens will go away. vt-d yes i5 2400 Basically if vt-d is enabled, the drivers will turn the monitor off randomly.

 

[srkcalb]

If vt-d is enable in BIOS then disable it and hopefully the black screens will go away. vt-d yes i5 2400 Basically if vt-d is enabled, the drivers will turn the monitor off randomly.

As far as I recall I did turn it off but thanks. I’ll double check that after I buy the new PSU.

 

[srkcalb]

Okay, I mounted the PSU today and it hasn't happened so far so thanks for the help guys. I'd say the culprit was always the PSU.

The black screen has been successfully resolved.

 

[DSzymborski]

You were fortunate that it doesn't appear to be a more expensive lesson! Crossing my fingers that you've got this resolved now.

 

[Frag Maniac]

Thank you. I’ll go get it and check if the issue still persists.

Not sure what country you're from, but Newegg just sent me an email showing the 650 and 750 supernovas for $67 and $87. Not bad for a gold PSU with 10 yr warranty.

It's definitely worth it to sign up for their email subscriber deals.