#Sudden black screen while playing a game
1 messages · Page 1 of 1 (latest)
Check your temperatures under load.
Should i do it in a game or in a benchmark application? In the game in playing right now temps seem fine, but when i opened up userbenchmark my CPU temp was up to 77 degrees celsius and im not sure if this will go any higher
I would say that you should do a good gaming session while having some temperature monitor on at the same time.
As long as the temperatures keep rising you kinda have to keep going until it reaches a baseline where it doesn't go up anymore.
You want to monitor both CPU and GPU temperatures during this.
Also I'm going to need your full system specs to confirm what the thermal design limits are.
And to also like... Figure out if your system combination is just bad.
like all parts?
All parts yeah
If you don't feel like doing a gaming session, you don't want to use a benchmark.
Instead you would want to use a stresstest.
Stresstest is quicker obviously, gives us faster results.
ryzen 5 5500
stock amd cooler
msi b550 gaming gen3
G skill ram 16gb 2x8 3200mhz (i dont know the exact model)
rx 6650 xt msi MECH
kingston NV2 1tb m.2
barracuda HDD 1tb
coolermaster MWE 650W White V2
how do i stresstest my system?
You would actually use a different stresstest app for different tests.
Furmark to get GPU temps. CPU temps with like Prime95.
I'll try prime95, because my GPU temps are always below 55 degrees
Prime 95 is super brutal though, so it will definitely hit the thermal limit with a stock cooler
That doesn't sound quite right, I would recommend a Furmark test too.
wdym?
would that be an accurate measure then?
GPU temps are typically at around 60-70°c degrees under load with a good cooler.
The fact that it's under 55 just doesn't sound right.
alright ill try the furmark test too then
Yup.
Do i change anything here?
Try just Smallest FFTs first
I don't want you just cooking the chip immediately.
do i stop it if it gets too high??
its 90 now
Stop the test.
You probably didn't stop the test
can i shut down pc??
Yeah the test isn't halted by just closing the window
Oh
I’ll let it cool down for a few minutes maybe
Like not powering it
Or is that worse
It should be fine to power it on.
It's not really worse, the load is off so the system has already started to cool down.
Even the stock cooler is efficient enough to dissipate heat relatively quickly when there's no load.
I did kinda expect this to happen though
yeah
i mean, it restarted itself pretty quick when playing apex
i think i was maybe 5 minutes in game
5 minutes at TjMax could cause it to power off.
Try and get some thermal readings in Apex then.
alright
I think that's a more realistic stress test in this scenario.
Even the lowest load Prime95 already went to TjMax.
sure
@crude violet nothing as of yet, cpu temps averaging around 68 degrees, GPU temps at 55 degrees
gpu temps look a little more expected now
I still have a hard time believing those temps
why is that
also, no display after i tried tabbing out of apex legends
As I explained before
I want to furmark this puppy
I want to see what the actual temperatures would be like under load
alright ill do the furmark test
well so only issue this time was no display after tabbing out to discord
but no random reboot or whatever
Hmm... Yeah it seems like we're dealing with a driver related issue.
I ran it for 2 minutes
And?
61 degrees max
2 minutes is not nearly enough btw for a proper test
oh
But I don't think it really matters, since this looks like a driver issue not thermal
Yeah well that's not nearly enough.
You would want to run a DDU to completely nuke the old drivers out of your system, a simple reinstall isn't enough.
Once you've done that, it's basically like doing a driver install on a brand new system.
Some other steps I usually recommend people take is updating their BIOS, installing the latest chipset drivers and doing a system integrity test.
how do u do a system integrity test
You open the command prompt as admin and type in sfc /scannow
BIOS should be latest, I'm pretty sure and chipset drivers were automatically installed when I built the system, which is around 2 weeks ago
No... That's not how it works.
What do you mean
When you first install the OS, it installs basically what ever the F drivers.
Same goes for stuff like chipset drivers.
No no, there was this MSI app that came up and it installed all drivers for me
Download and install the latest one.
Also we can check which BIOS revision you have.
how
If you open up Msinfo
It will tell you the BIOS revision you have installed onto the motherboard
For example in my system
Msinfo states it right below the CPU
do i install all?
Yes all
American Megatrends Inc. 3202
Which is also the latest BIOS available for my motherboard on Asus website
That's how you check what BIOS you have and if it's the latest
haha okay i dont have the latest bios, i thought my brother updated it, I guess not
Yup
I'll have to update it tomorrow because my brother has the usb and he's asleep
You will do that then
do you still think using DDU is necessary?
Honestly yeah
I like to play it safe with driver related issues
if the initial installation was f'd somehow, it wouldn't matter if you updated the driver or did a reinstall
the problem would only go away after running the DDU and doing a fresh install of the latest drivers
@crude violet alright, so I have to do DDU and bios update right
Yup
should my usb drive be empty when installing the new bios version?
Preferably yes and it should also be formatted in FAT32
alright ill be updating it now
Ok, that's done
Time for DDU
My memory went back to default speeds, do I change it back to 3200mhz
@crude violet also, I don't know if this helps you know what the exact problem is, but sometimes when I open up google chrome my pc freezes for a bit
Doesn't really help narrowing it down.
alright, well im about to uninstall drivers
@crude violet sorry for ping
remember to use DDU in safe mode
Yep I am , thanks though!

Okay, just got done installing the new drivers
I'll let you know if I have any problems still
@desert barn should I enable XMP again? It got disabled after installing the new bios update
yes
alright
I haven't played any games yet, but do you think it really couldve been my cpu overheating? After the 2nd random restart I put my pc down and reseated the ram, maybe the cooler went "back into place" and sticks better to my cpu and that caused it to not crash the 3rd time i tried apex
We've only isolated the drivers for now, the temperatures seemed to be okay for your workload.
I wouldn't start drawing any conclusions here.
Also if reseating RAM somehow fixed a temperature issue, that would mean your cooler is mounted completely wrong.
Like I said, don't draw any conclusions.
Because the theory of reseating RAM fixing temps is frankly kinda dumb ngl
no but because of gravity that my cooler sits better on my cpu
nothing to do with the ram i formulated it bad
Is your computer still laying on it's side?
nope
Then it's not the mounting.
alright
Ok no issues as of yet, just apex crashed, the game itself not pc
@crude violet i booted up a game, and screen turned black. Audio was kind of static
Alright, I think we've more or less ruled out drivers now
Basically we're at a point where I want to double check the CPU temperatures one more time
I opened up the Radeon software, and just clicked on something and pc rebooted
Interesting
Wait so drivers arent the issue you say?
We've already ruled them out haven't we?
I guess yeah
Hmm... I still kinda want to check the CPU temps one more time.
But it looks to me like we're either looking at a power related issue or there's an issue with the graphics card.
Power related or thermal issues are a lot more likely to cause these kinds of symptoms.
So I want to double check the CPU temps and then we start digging into the PC itself.
Okay
Do you want me to check the temps again with the same app?
Or something else?
Since Prime95 would just peg it at TjMax, use Cinebench instead.
alright im installing it
okay its installed
Run the multicore test, monitor your thermals at the same time.
went up to 89c
but that was a short period of time
So under load for an extended period of time it will go to TjMax
I guess so
How do you want to approach this?
Do you want to declare it a thermal issue and just get a new cooler or do you want to also explore the potential of a power failure?
I guess testing everything would be best
Well then, you would next have to gut the system.
I would try reseating the graphics card and the power connectors leading to it.
Okay doing it right now
How should I test it now that’s it’s reseated? Just a game?
Crashed just now
Welp.
Was a black screen and then reboot
At this point it's most likely the CPU overheat protection kicking in, or the PSU is failing to feed the graphics card.
Are there any PC hardware stores locally you could visit?
Nope
Basically what I would like you to do is buy a DeepCool AG400 cpu cooler or similar.
As well as a Seasonic B12 PSU.
Eliminate both potential issues at once, you can try with just the cooler first. If that works, hurray and you can return the PSU back to the store.
However if it doesn't work, install the PSU as well and try again.
If i order it today it'll come in tomorrow
If that works for you
Would the CPU overheat from just booting up a game tho?
I trust what you say
But I can't comprehend it
yes
This problem has occured with Apex legends specifically, correct?
Honestly. I don't know how heavy it is on the CPU to load this games, it's not likely that it's enough to trip the overheat protection on your CPU but it is possible non the less.
I would play it safe and get the AG400 since it's only like 30 bucks to make sure you won't have thermal issues period.
so the ag400 and the b12 seasonic psu right
That's correct
And like of course make sure it's possible to return them in case I was completely wrong with my diagnosis.
Or like at least the PSU. The cooler you might as well have even if it doesn't solve the issue.
alright
750W is more than enough right
Pff overkill as hell for your setup
You could honestly just buy a 550W unit and it would be more than enough.
okay
What do you think the problem would be if it still happens even with the new parts?
Problem with gpu?
We'll cross that bridge when we get there. I don't want to draw conclusions at this point.
alright
Would frequent power outages mess up my psu?
Generally speaking PSUs don't like to be cut from the wall while supplying power.
Unless you have surge protection, it's definitely possible.
I don’t think I have it
At my previous home we had frequent outages
Like a few times a week sometimes
And I used this psu in my previous built
You would always want to use a surge protector with your computer.
They're not that expensive and can really save your ass.
I've had two surge protectors that died protecting my computer.
Surge protector costs next to nothing when compared to how much my computer costs.
Would you say that most likely would’ve messed up my psu
Only thing power wise in this house was this random lack of power
But that was once
And only at my power outlet
Maybe that had something to do with it
Honestly I can't say for sure, hence I'm going to say what I've said before
I don't want to draw any conclusions.
Let's see if replacing the PSU even fixes the issue first, then you should consider buying a surge protector.
Okay
the seasonic b12 is reliable right
PSU tierlist gives it like a B ranking.
But I always recommend Seasonic B12s since my personal experience with them have been very good.
Especially if you're on a budget, it's very hard to beat Seasonic B12
Also I have to ask you to chill with the pings, especially on low priority questions.
550W is enough yeah
You can go higher if you're planning to upgrade your system in the near future
but the current setup can work out of a 550W unit
Yeah I'm probably not upgrading my system for a while
550W it is then
Alright, it should come in tomorrow
So, I should install the cooler first right and see if there's still any problems?
I think that's the way it should be done.
Would be pretty painful to go through the PSU replacement if it turns out you don't actually need it.
👍 I'll keep you updated, thank you for your time
Just curious about the thing you said, CPU overheating protection, does it shut down the system once it hits 90c, or when theres just a large temp spike?
It basically shuts down the system when there's literally nothing else it can do to get it below TjMax.
Say it's in a worst case scenario type of situation, the cooling fan is already spinning at max rpm. It has been downclocking like crazy but it just can't get to a controlled temperature...
It just cuts the power as a last restort to try and save the CPU.
However it is not a reliable method of making sure your CPU doesn't get damaged, the moment you hit TjMax is when you risk damage.
What could possible issues be when your CPU reaches TjMax?
Either the cooler is clogged in dust, the thermal paste is bad, the mounting is bad, the cooler itself is bad, the fan is broken...
It typically is the cooling that's failing in that situation.
interesting
Also, using my pc without stressing it too much should be fine right? I need it for school assignments
hopefully 🙏 but i have faith in your knowledge
Scrolling through Netflix Made my screen flicker and then it rebooted itself
Damn
The parts will arrive soon, right?
yes, in 5-6 hours
They will arrive once I'm back home so luckily I don't need my pc right now
Yeah, if the issue is getting more frequent it does honestly sound like we might be dealing with a PSU issue.
Light desktop loads shouldn't be enough to crash a system, even if it's the stock cooler with crusty thermal paste.
Alright, again, thank you for your time and patience🙏
Also how should I remove thermal paste, I don’t have isopropyl alcohol
You can use just coffee filters.
Or even papertowel as long as you double check you didn't leave lint behind.
Okay thanks
Is there a chance some other hardware got damaged because of the psu, or is it a small chance
It's always a possibility with PSUs
That's why you never want to cheap out on the unit.
Because it can and will destroy other components in the system.
It honestly really depends.
Each component has a separate warranty, if your system is a prebuilt you voided the warranty pretty much the moment you opened up the system.
My brother build it
We bought all parts individually, except psu and ram
Those came from my old built
Well in that case, each component has a separate warranty.
It should cover it, I’ll check later
Well it doesn't matter anyway.
What matters most is diagnosing and repairing the system from here.
Yes
Just the fact that it crashed during just browsing, I would skip straight to replacing both the PSU and the CPU cooler.
Once you've done that, the system should function fine.
Alright
@crude violet both cooler and psu are installed, still crashed
Well shit
yeah
Well that's great, so we've got basically nothing to work on
The only thing I could think of now is that it's probably the graphics card
I can ask my brother to put his graphics card in my computer, we might be able to rule that out
sounds good to me
it's an nvidia card, that shouldnt matter too much right?
You'll have to run DDU and install Nvidia drivers on your system.
okay
Other than that it will work obviously.
Put in a gtx 960, we had it lying around, booted up into windows fine, no crash. Booted up a game, no crash.
Keep using it for a little while longer, see if the system suddenly crashes.
Remember this?
At first it seemed okay, but it started crashing again.
Right
Use the system as it is for now. Let's say if it doesn't crash over the weekend, then put the AMD card back in and see if it starts crashing again.
Okay, will do
Crazy if it actually is the GPU, worked fine for like 2 weeks and now this haha
Well, we'll see what happens
The fact that the DDU helped a bit was a pretty strong indication of course.
Uhm, we're trying to boot into safe mode, there's no signal
With the GTX960?
yes
What in the...
Hold on a second here
This is the motherboard in your system right?
The graphics card is installed in the silver slot?
top one, yes
Hmm...
If you move the graphics card down to the second slot
Does it still behave weirdly?
Haven't tried that
Let's kill two flies with one strike
Take the AMD card, put it in the second slot and try using your system again as usual.
Okay will do
No crash as of yet, I booted up a game, I'll let it sit for a while
Like 10-15 minutes
And see if there's any issues
Nothing happened. I can try launching apex aswell, but last time I tried these things there was no crash either until the day after.
Yeah it's super weird because it's inconsistent like that.
That's what made me suspect either a PSU failure or overheat
What if I try the AMD gpu in the second slot for the weekend, and see if there's any crashes or abnormalities?
You can try that
Now we're at a point where we're kinda trying everything since we have nothing to work with
I don't think that will make a difference
Reseating the graphics card is the first thing we tried
thats true
Do you have any extra thermal paste on hand?
I do not
Because I'm thinking you might have to also try just reseating the CPU.
But since you don't have thermal paste, let's not do that then.
You could however reseat the RAM.
My brother said he reseated CPU when installing the cooler
And we've reseated ram too
and put in other sticks
Your bro did a gj doing all of that just in case
Yeah, he studied IT and all that in school I think
I think it's safe for us to focus on the graphics card for now.
and he built both my systems
So we're going to second slot with the AMD card
It's in, we installed the drivers again
Let's see how it goes
If it just immediately starts crashing, switch back to GTX960 and see if the problem disappears completely
Alright
And if it does, then you should claim the warranty on that graphics card.
Which I have to say... Haven't seen many RX cards fail like this.
right'
Oh yeah and obviously if you don't have any use for that new PSU, you might as well return it since it made no difference whatsoever.
Might keep it, just in case. My old PSU went through a lot of torture with the power outages haha
Better safe than sorry
Honestly that's a fair point, the B12 is significantly more reliable too
Alright, I'll be testing the pc now
No issues today. Also no flickering when I go fullscreen in a twitch stream or show on netflix (I assume the flickering would have to do with the gpu or whatever)
My brother changed my gpu back to first pcie slot, and where the system would reboot when i opened AMD software, nothing happens anymore.
I would've kept testing it a bit longer tbh.
Keep using the system as usual now and we'll see what happens.
also 50c gpu temp and 45c cpu temp seems good no or is it still unbelievable
I dont really have much to say at this point
If switching between two PCI-E lanes back to back somehow fixed the issue I'm just annoyed and confused
Either there was some dirt on the contacts that got rubbed off when going back and forth or there was something in the PCI-E slot...
If it's fixed now then that's great, I'm just annoyed I couldn't figure out what caused the issue in the first place.
It's fine man, you did your best and I appreciate that! Thank you for your time! If any issues occur again I'll send it in this thread, for now it seems to be working fine
You're human after all, you can't know everything haha
@crude violet display turned black 2 times today, first time I fixed it by pressing CTRL + shift + windows + b, 2nd time that didn't work had to reboot pc with power button
Random reboots started happening again
@crude violet sorry for ping i just dont know what to do
You're going to take a step back in diagnosis, last time it stopped rebooting after switching to the secondary PCI-E slot. So you should put it back there and keep using the system as usual and see if it starts rebooting with that slot in use.
alright
I feel like I have really bad performance in game for my system specs
I know the 2nd lane is slower, but even in the first slot my performance was bad
@crude violet I could try the same thing that made it crash last time? That might be faster to rule out things
Definitely if there's a consistent way to crash it, I was under the impression that it's random
And the way you did it has crashed the system previously every time?
yes
i installed new chipset drivers and then my pc crashed, this was 2 days ago, when i went back to old drivers it was stable ish, and now when the card is in the bottom slot i can play games
I think your mobo is broken.
can a broken motherboard cause bad performance in game?
If the PCI-E lane is f'd up then yeah
It could also be a CPU related issue but it's a bit less likely than a mobo failure.
I don't think it's an issue with the graphics card either, the problems would've persisted in the secondary slot.
bad performance is in 2nd slot too but that shouldnt matter too much right
well ill just rma the motherboard
and see if that fixes things
It's the second slot so... Worse bandwith, worse performance.
You can confirm the graphics card functionality for sure in another system, you could also try a different CPU.
But other than that I would say it's most likely the motherboard.
alright thank you
@crude violet they tested my motherboard and concluded it’s not broken, I’ll rma my gpu then
Weird but alright
Yeah idk I got a mail saying they tested the motherboard and it was in perfect shape
I’ll just rma the gpu and hopefully that’s the problem
Well the thing is that them "testing" the motherboard is usually just hooking it up and turning it on
Graphics Card might just be dealt with in a similar fashion
ah
Graphics card was the defect part, they sent me a 4060 as replacement, system is working fine now