Desktop Keeps Rebooting During Gaming

[MatrimC7]

Hello everyone. I just built a PC, specs are:

MSI B450 Tomahawk Max, AMD Ryzen 5 3600X, ASUS RTX 2070 Super, geiL 16GB(2x8), Corsair CX750M.

My PC reboots during certain games, at certain points. Specifically:
Modern Warfare Multiplayer Gunpoint Mode reboots right after starting loading the level.
PES 2020 reboots various situations. Sometimes at the beginning of a match, sometimes when I take a freekick. Sometimes doesn't reboot and run smoothly.
Outer Worlds, when I try to use the workbench.
CS GO used to reboot, it ran smoothly last couple of tries. I didn't try it recently though.

I ran every benchmark, stress test(Heaven, furmark, kombustor, OCCT, Aida64), ran MemTest86, no issues whatsoever. I constantly check temp when trying to play the games that causes reboot, always around 60-65 at the point of reboot.

I installed all the latest drivers and I updated my BIOS. I haven't OC'd anything. I tried undervolting and underclocking my GPU. I tried enabling and disabling XMP. Nothing worked. Still get rebooted during PES 2020.

I can run other games perfectly. GTA V, Doom, TESO, Battlefield V, FIFA 20, Apex Legends, Divinity Original Sin 2, The Witcher 3 Wild Hunt, Dota, Stellaris. All running pefectly in highest settings, for hours.

I have no idea what is causing this and I am about to lose my mind.

 

[basco]

did ya look at temps during stress?
could ya borrow some hardware from a friend?

 

[Regeneration]

Reboots or BSODs? Check the event log (Control Panel -> Administrative Tools -> Event Viewer -> Check both critical and error).

If its BSODs, upload the DUMP files: C:\Windows\Minidump or C:\Windows\Memory.dmp.

Did you run stresst test for both GPU and CPU?

Check if the PSU is overheating by running FurMark and CPU stress test in the same time.

 

[MatrimC7]

did ya look at temps during stress?
could ya borrow some hardware from a friend?

Yes, temps are all normal during stress. During benchmarks and stress tests, GPU and CPU generally around 65-70. CPU sometimes go as much as 75 but never above.

Also I monitored temps during said games causing reboot, both temps never passed 70.

Unfortunately I don't have any friends that I can borrow hardware and try each part.

Reboots or BSODs? Check the event log (Control Panel -> Administrative Tools -> Event Viewer -> Check both critical and error).

If its BSODs, upload the DUMP files: C:\Windows\Minidump or C:\Windows\Memory.dmp.

Did you run stresst test for both GPU and CPU?

Check if the PSU is overheating by running FurMark and CPU stress test in the same time.

Completely reboots. System shuts down and restarts.

Each time after restart, I see kernel power error on event log with 0x8000400000000002 or 0x8000000000000000

No dump files generated after any of the reboots. And never seen any other indication on the eventviewer.

I actually ran Furmark and CPU stress test at the same time, CPU temps were around 70 maximum. Also ran OCCT PSU stress test for an hour, no faults. I have a screenshot of the result here:

 

[basco]

if ya tell us where ya from then maybe you will find a friend in techpowerup forums.
i am from austria and i would help ya if a ya not too far
but from your name i would go to matrimonio and think you are some spanish south american guy or girl- could be all wrong

 

[MatrimC7]

if ya tell us where ya from then maybe you will find a friend in techpowerup forums.
i am from austria and i would help ya if a ya not too far
but from your name i would go to matrimonio and think you are some spanish south american guy or girl- could be all wrong

Thank you but I'm affraid we are a little far away from each other . I'm from Turkey/Istanbul. And "Matrim" is from a character from The Wheel of Time series( Matrim Cauthon)

 

[FreedomEclipse]

The curse of the CX line of Corsair power supplies - Ive replaced a lot of them that would crash and bootloop machines till they were replaced. Ive even replaced a faulty CX with another factory sealed new CX that developed the same problem a few months later. Problem went away when i replaced them with an EVGA GQ.

The higher end corsair units tend to be pretty good though but i would avoid the CX line and anything below that range. Stick with the HX, TX range if you can.

The same company that builds a lot of high end corsair PSUs builds these CX PSUs. So you'd expect them to be good but they are not.

I think the problem we have here is the workmanship and build quality is 10/10 but the components inside are probably more like a 4/10 or 6/10 because they cheaped out on it.

::EDIT::

However if youre building a super cheap low-end machine that doesnt pull a lot of power then it doesnt matter if you run with the cheapest corsair money can buy so long as you meet the wattage requirements.

 

[Regeneration]

If you want a proper CPU stress test, try Linpack Xtreme.

Sounds like your PSU is overheating or failing. Touch it with your hand while gaming.

As @FreedomEclipse mentioned, the Corsair CX series aren't that great.

 

[MatrimC7]

Couple of other forums suggested it could be from PSU, I have one with a gray logo(I saw some posts about green ones being worse before.) I do hear some strange noises time to time(I am %99 sure it's coming from PSU), but I thought it's normal under stress.

I have one more question that just came to my mind. I used 1 PCIe cable to connect my RTX 2070 to PSU, like this:


Should I try to use 2 different cable sets?

 

[Bill_Bright]

I just built a PC

I would try another PSU. As FreedomEclipse notes, the CX series does not have the best reputation. But even the best brands can have a problem unit occasionally. So beg or borrow another from somewhere and try another PSU.

I don't see this being a stress issue since sometimes it works fine.

If not the PSU, I would pull the motherboard and check for extra standoffs or another foreign object under the board. Note that cases are designed to support 1000s of different motherboards. So, it is common for cases to have more motherboard mounting points than boards have mounting holes. A common mistake by the less experienced and distracted pros alike is to insert one or more extra standoff in the case under the motherboard. Any extra standoff creates the potential for an electrical “short” in one or more circuits. The result ranges from "nothing" happening (everything works perfectly) to "intermittent” problems to "nothing" (as in nothing works at all ). So, you might want to verify you only inserted a standoff where there is a corresponding motherboard mounting hole.

While software based memory tester are good at telling us if RAM is bad, they are not conclusive when it comes to telling us the RAM is good. That is, they can report no problems and the RAM still fails when in use. So try running with one stick of RAM at a time.

 

[basco]

there is a thread with these probs (0x8000400000000002 ) and it was the power supply \\ its post 85 in this thread

PC restarts when playing GTA 5 and DotA 2 after Windows 10 Anniversary Update

Lol u play Dota too Yeah. Add me up. Same as my user name here :)

www.techpowerup.com

 

[Regeneration]

Couple of other forums suggested it could be from PSU, I have one with a gray logo(I saw some posts about green ones being worse before.) I do hear some strange noises time to time(I am %99 sure it's coming from PSU), but I thought it's normal under stress.

I have one more question that just came to my mind. I used 1 PCIe cable to connect my RTX 2070 to PSU, like this:
View attachment 151980

Should I try to use 2 different cable sets?

Yes!! split the load between 2 cables.

When you say 'strange noises', you mean buzzing right?

 

[MatrimC7]

Yes buzzing. Exactly like this:

I will try splitting 2 PCIe cables and see if that will help.

BTW, completely unrelated but, is there anyway turning off all the RGB? I have no idea why I bought rams with RGB to be honest.

 

[Caring1]

Looking at the OCCT graphs above, I would say thermal throttling of the GPU, the power remains consistent during the GPU drop outs and temperature drops.

 

[Waldorf]

@Bill_Bright FreedomEclipse

its usually the M series that's bad, not the non modular ones.
but even if, i've never seen a unit that does everything fine, except for one game.

 

[Regeneration]

@Bill_Bright FreedomEclipse

its usually the M series that's bad, not the non modular ones.
but even if, i've never seen a unit that does everything fine, except for one game.

VS series are bad too. Overheating and failing out of the box.

 

[Bill_Bright]

its usually the M series that's bad, not the non modular ones.

Note sure I agree - but I will say all wired (not just Corsairs) PSUs have fewer parts. So, as a general statement, fewer problems with wired PSUs would make sense.

I do agree with Regeneration about VS series too - though "bad" may be a bit harsh. Not sure any are "bad". The problem is, it used to be if the PSU had the Corsair brand on it, you knew it was a quality PSU. Then they changed OEMs for their entry series supplies and failure rates went up. That tarnished their reputation, upsetting many users.

The facts are, there are many CS and VS series Corsair PSUs out there that are servicing their user's needs just fine. Perhaps saying the CS and VS series are "not as good" as top tier Corsair models would be more appropriate?

Even if the OP had a top tier EVGA or Seasonic PSU in there, until Man can create perfection 100% of the time, there will always be a unit that fails to meet specs. So swapping in another, known good PSU would still be an appropriate first troubleshooting step, regardless the PSU in there.

 

[MatrimC7]

Well, I completely assembled everything except CPU. I noticed I missed 1 standoff screw and screwed that one properly this time. I used 2 different PCIe cables to connect GPU to PSU. PES 2020 caused reboot again. This time when I try to reboot my PC, the VGA LED on MOBO didn't go off and after another restart it booted up. Now I'm doing a fresh OS install to see if it would help or not.

 

[SomeOne99h]

Well, I completely assembled everything except CPU. I noticed I missed 1 standoff screw and screwed that one properly this time. I used 2 different PCIe cables to connect GPU to PSU. PES 2020 caused reboot again. This time when I try to reboot my PC, the VGA LED on MOBO didn't go off and after another restart it booted up. Now I'm doing a fresh OS install to see if it would help or not.

Guys correct me if that is wrong, prolong use of a faulty power supply can damage the computer's components?

 

[MatrimC7]

Nope.

After clean OS install and drivers, still reboots during Modern Warfare multiplayer.

I have no idea what is the problem and at this point, I am just feeling so broken.

 

[FreedomEclipse]

Guys correct me if that is wrong, prolong use of a faulty power supply can damage the computer's components?

If a powersupply cant keep voltages regulated - Things would start smoking up or catching fire.

Damage tends to be instant. A faulty power supply can fry everything hooked up to it and even if your parts survive, there is no telling what sort of damage that has been done to them. Even replacing the faulty powersupply would not guarantee continued functionality of parts that have been fried by a bad power supply.

 

[MatrimC7]

If a powersupply cant keep voltages regulated - Things would start smoking up or catching fire.

Damage tends to be instant. A faulty power supply can fry everything hooked up to it and even if your parts survive, there is no telling what sort of damage that has been done to them. Even replacing the faulty powersupply would not guarantee continued functionality of parts that have been fried by a bad power supply.

Is there anyway to tell that it is certainly PSU's fault though?

Edit: I was running linpack Xtreme to stress test and realized that there is a strange clicking sound coming from the case in every 10 seconds or so. I am not sure but it sounds like coming from the CPU area.

It's coming from the HDD.

 

[Bill_Bright]

Guys correct me if that is wrong, prolong use of a faulty power supply can damage the computer's components?

Depends on what the fault is - but yes prolonged (or even short term) exposure to incorrect voltages can, in some cases, result in premature failure of connected components.

I noticed I missed 1 standoff screw and screwed that one properly this time.

A missing standoff is not really a problem - other than, depending on its location, it might allow the motherboard to flex too much when inserting RAM or a card, or the heatsink fan assembly. It is extra standoffs that might short voltages together or to ground that can result in intermittent disruptions or even permanent damage.

Is there anyway to tell that it is certainly PSU's fault though?

Yes - as mentioned above by swapping in another known good PSU. Beyond that, there is no way normal users can conclusively test a PSU as (1) that must be done under a variety of realistic loads and (2) PSUs need to be tested for proper regulation and ripple suppression too - not just that voltages are present and within the allowed ±5% tolerances. In other words, to properly and conclusively test a PSU, it must be done by a qualified person under a variety of loads using an oscilloscope or dedicated power supply analyzer. Since most users don't have that sort of test equipment, the know-how to use it or training to interpret the findings, swapping in a known good PSU to see if the problem goes away is really the only way to do it.

 

[MatrimC7]

So, then I have 2 options. 1. Contacting the retailer to change my unit. 2. Buying a new PSU.

I already contacted with Corsair, they replied that I should contact with the retailer since it is recently purchased unit. I contacted with retailer I bough the parts from, they replied after 2 days by saying just "send the unit to us via cargo, we will check" in Friday afternoon. With the weekend curfew going on here in Turkey, it basically means that my package will wait in shipment company for another 5 days. Geez. And beyond that, I'm pretty sure they will just say "ah, your unit is fine, no problem whatsoever" after 2 weeks at best. The customer service and customer rights are not really great over here.

If I am to buy a new PSU, would you guys have any suggestions?

Addition: I just ran Linpack Xtreme and furmark at the same time for 20 minutes or so. No crash, no reboot. Temps are all around 65 maximum.

Ah, also, when I install the motherboard again, I realized there is 1 strange standoff in the case(corsair 270R). It is sort of reversed standoff that a small screw just stay inside of motherboard. And it seems like there is no way to unscrew it, I tried.

 

[Bill_Bright]

You don't have any trusted friends or relatives with a computer you can temporarily swap PSUs with?

Since that PSU is under warranty, I would send it back. If you can get your money back, I would go for an EVGA SuperNova or Seasonic, if available in your area. If not, then an upper tier Corsair.