Question Every game crashes on self-built PC, help would be great :)

Mr.Frisfruit

Prominent
Jun 17, 2019
37
5
545
Hi guys,

So here's the deal... I've built a new pc with all new components except for the PSU and my SSD and HDD, which I took out of my old PC. Note: My old PC ran fine bit did show some issues such as occassional stutters and VERY brief pink screens/hues. My new PC runs great, except when gaming for about 20 minutes, after which pretty much every game crashes.

My current setup:
Asrock B450M Pro4 mobo
Ryzen 2600
GB Aorus RX 580
2x 8GB DDR4 RAM (Vengeance 2400Mhz)
Silverstone Strider Essential 600W (4 years old)

To further elaborate on the issue: games crash after 20 minutes and, just as with my old PC, I am having occassional stutters and brief pink screens. The stutters are barely noticable but you can mostly hear them, also when watching YT videos. I've come at a point where I am quite sure it is hardware related since I've tried basically everything. Wiped and reinstalled all mobo, gpu, cpu, and other peripheral drivers. Reinstalled windows. Rebuilt the entire PC to check for loose connections. Ran the PC with 1x 8 gb Ram (tested both, without success). Updated mobo bios and GPU bios. Tried running older drivers. Unplugged all non-essential peripherals. And some more minor changes. None of my components are overclocked and my temps are fine; CPU @ 60 under heavy load and GPU @ 75 under heavy load. My bios settings are also correct so that shouldn't be a problem.

My main problem is that I don't have someone with a similar system to borrow some components from, and I built the PC myself so I can't just send it back. Additionally, if I send something for repair, which appears to work fine, I still have to pay 'research-costs'. I am currently doubting whether the PSU could be the issue since it's the only components I've kept from my old pc (except for the hd's) and the stutters and pink screen issues are still present. I'm not sure if these issues can be caused by a faulty PSU since it kind of sounds like a bad GPU to me but I'm not an expert. If anyone has experience with similar cases I'd love to hear it before I go and spend even more of my hard-earned money..

Cheers.
 

Ralston18

Titan
Moderator
I would suspect the 4 year old PSU.

Very likely that PSU is not able to keep up with the power requirements of the new components.

Use Task Manager and Resource Monitor to observe system performance. First without gaming and then while gaming.

Event Viewer and Reliability History/Manager may also be capturing some relevant error codes or warnings. Especially just before or at the time of the crashes.
 

Mr.Frisfruit

Prominent
Jun 17, 2019
37
5
545
I would suspect the 4 year old PSU.

Very likely that PSU is not able to keep up with the power requirements of the new components.

Use Task Manager and Resource Monitor to observe system performance. First without gaming and then while gaming.

Event Viewer and Reliability History/Manager may also be capturing some relevant error codes or warnings. Especially just before or at the time of the crashes.

Thanks for the reply. What should I look for in the task manager? I did check voltages in HWmonitor which seem fine but I'm not sure if that says too much. My event viewer doesn't say much except for an occassional generic error I can't do anything with. I just checked the Reliability History and I do see that it says I had a hardware related problem yesterday:

Source
Windows

Summary
Hardwarefailure

Date
‎16/‎06/‎2019 00:06

Status
Report sent

Description
A problem with hardware has prevented Windows from functioning properly. Een probleem met de hardware heeft ervoor gezorgd dat Windows niet meer goed werkt.

Problem signature
Problem Event Name: LiveKernelEvent
Code: 141
Parameter 1: ffffda0b5a835050
Parameter 2: fffff80774390714
Parameter 3: 0
Parameter 4: 18c8
Versie besturingssysteem: 10_0_18362
Servicepack: 0_0
Product: 256_1
Versie van besturingssysteem: 10.0.18362.2.0.0.256.48
Landinstelling-id: 1043

Extra information about the problem
Bucket ID: LKD_0x141_Tdr:6_IMAGE_atikmpag.sys
Server information: 8c2c328b-f644-461e-a892-8b26071e1ecc
"



Not sure if this is useful in any way though, I'm kind of clueless at this point.
 

Ralston18

Titan
Moderator
Noted the following:

Summary
Hardwarefailure

In either Task Manager or Resource Monitor (one at a time) simply continue to observe system performance.

What is running, what resources are being used, and to what extent.

Again, first while not gaming - just explore the tool, learn how to navigate around, and what information is being presented.

Then slide the window to one side and leave open. Do others things except gaming and watching videos.

Next, game as usual and pay attention to the monitored window especially around the 20 minute mark.

The objective is to identify the circumstances that eventually lead up to the crash. Perhaps no crashes at all unless game "X" is being played and then only after the "20 minutes".

Need to narrow down the possible reasons by observation and elimination.

PSU is, in my mind anyway, the leading candidate with respect to the problems that are occuring.
 
You have 4 years on a PSU that Silverstone didn't have enough confidence in to offer anything more than a 1-year warranty on.

Just that much should tell you to replace the PSU--yesterday.

PSUs are SO important that I will not reuse one that has more than a year on it, and won't buy a PSU that isn't covered by a 10-year warranty from a reputable brand, and in the high end of their line-up. Yes, high-quality power supplies are expensive, but spending more money on a solid PSU is cheaper than potentially replacing an entire system when a low-quality PSU nukes itself.
 

Mr.Frisfruit

Prominent
Jun 17, 2019
37
5
545
Noted the following:

Summary
Hardwarefailure

In either Task Manager or Resource Monitor (one at a time) simply continue to observe system performance.

What is running, what resources are being used, and to what extent.

Again, first while not gaming - just explore the tool, learn how to navigate around, and what information is being presented.

Then slide the window to one side and leave open. Do others things except gaming and watching videos.

Next, game as usual and pay attention to the monitored window especially around the 20 minute mark.

The objective is to identify the circumstances that eventually lead up to the crash. Perhaps no crashes at all unless game "X" is being played and then only after the "20 minutes".

Need to narrow down the possible reasons by observation and elimination.

PSU is, in my mind anyway, the leading candidate with respect to the problems that are occuring.

I've monitored my gpu, cpu and ram activity & temps while gaming but couldn't see anything noteworthy when it crashes. I guess I'm just going to take the shot and buy a new PSU. Thanks for the advice.
 

Mr.Frisfruit

Prominent
Jun 17, 2019
37
5
545
You have 4 years on a PSU that Silverstone didn't have enough confidence in to offer anything more than a 1-year warranty on.
Well my car had 1 month of warranty and it's still going strong after 5 years lol. But yea I get your point. I'm probably ordering a new one today. The Coolermaster MWE Gold 650 is currently on sale at my retailer, would you say that's a good choice?
 
  • Like
Reactions: ElectrO_90
There is a very good Google Docs-based PSU list, done by one of the members, here (which I must not have bookmarked, and cannot seem to find, and I'm ready to head out the door to get my wife to work, so cannot search further). Once you find that list, you can compare your PSU selection with that list.

Personally, I won't go with anything that isn't a "Tier 1" PSU, but you be the judge of what you are willing to shoot for, in terms of the "quality : cost" ratio.

Perhaps another person can post a link to that spreadsheet, and we can both benefit from their information.
 

https://www.tomshardware.com/reviews/best-psus,4229.html

Don't think these are the links allangh was talking about, but it's a start.
 

Mr.Frisfruit

Prominent
Jun 17, 2019
37
5
545
I have installed a new PSU and sadly the problem still exists... I've also reinstalled windows just to be sure but no change.. Anyone has some other suggestions? Could this be a GPU malfunction?
 
I've reinstalled windows several times, so I doubt that is the problem. The only thing I could try is to install it via an USB instead of using W10's integrated reset function. But I think that's the same right?
Installing from a usb would be considered a clean install, starting completely fresh. I believe using the integrated reset function is different. But I may be mistaken.
 

ufslider

Distinguished
Dec 31, 2007
3
1
18,515
If this is happening while gaming, would lead to say it is the video card. I had issues with the AMD Wattman software. There are two issues to look into, you might want an older version of the software and then with the RX 580's, you want to undervolt the card for performance(custom settings-youtube this as well). I have the Sapphire RX580 and the normal clock is around 1340 and the boost is around 1430 but at 1200mv and it was crashing all the time. I don't have the steps I used on hand but you can do a search, the last 2 steps should be dropped by 100mv and the prior should be adjust accordingly as well. You can set the fans to custom speeds as well to help the cooling. The memory can be undervolted by 100mv as well. It takes some time to see how the settings alter the card but too many of these cards are not ready out of the box and cause people to think their builds are bad.
 

Mr.Frisfruit

Prominent
Jun 17, 2019
37
5
545
I think I finally managed to fix it... For some reason the Event Viewer now said something about audiodrivers or something alike. I've disabled Stereo Mix, which came with the motherboard drivers, and now everything seems to work fine.. Still wonder what the underlying issue is though, disabling stuff should never be a solution. I feel kind of stupid for buying a new PSU but atleast this one is futureproof for the next few years..
 

Mr.Frisfruit

Prominent
Jun 17, 2019
37
5
545
I cheered too soon... I'm still having the same issue. I reinstalled windows from a usb this time but still without any luck. I have no clue what to do next... I can't find anything useful that I haven't tried already.. I'm considering to send back my GPU but I'm afraid they won't be able to find something since it does work fine with stresstests. The issue is just so weird and I can hardly find anyone with a similar problem. After a fresh windows install games run fine for 1.5-2 hours but then crash, and after that keep crashing every 10-20 minutes. I'm clueless at this point.
 

Mr.Frisfruit

Prominent
Jun 17, 2019
37
5
545
If this is happening while gaming, would lead to say it is the video card. I had issues with the AMD Wattman software. There are two issues to look into, you might want an older version of the software and then with the RX 580's, you want to undervolt the card for performance(custom settings-youtube this as well). I have the Sapphire RX580 and the normal clock is around 1340 and the boost is around 1430 but at 1200mv and it was crashing all the time. I don't have the steps I used on hand but you can do a search, the last 2 steps should be dropped by 100mv and the prior should be adjust accordingly as well. You can set the fans to custom speeds as well to help the cooling. The memory can be undervolted by 100mv as well. It takes some time to see how the settings alter the card but too many of these cards are not ready out of the box and cause people to think their builds are bad.

I don't really feel comfortable changing these settings, and I shouldn't have to. It should work out of the box, as it does in most cases.
 

ufslider

Distinguished
Dec 31, 2007
3
1
18,515
I don't really feel comfortable changing these settings, and I shouldn't have to. It should work out of the box, as it does in most cases.

You are absolutely right....and I agree with you....But.....these cards do seem to have overheating issues. I understand not feeling comfortable and my advice to make it easier is that by changing these settings, you are not Overclocking these settings. You can overclock by changing the settings but the goal here is to test the card and verify if the card is the issue, you are actually lowering the settings and making sure the card is running stable. A safe test is to lower the mv across the board by 50 mv. You can also lower the frequency by 50Mhz as well. I think this works better but another option may be in the watman is to run under the silent or the preset lower settings as well. If you pc is only crashing during games and nothing else...I had the same issue. I returned a Asus Strix Vega 64...I had a great deal on an open box, I liked the looks but I didn't want to have to lower the settings to use it. This card has a reputation for being great after changing thermal paste and using a thicker thermal strip but it crashed too much for me to want to take the risk. If you search undervolting on the 580 cards, you will see how many people say the same thing about not wanting to change the settings but seeing that is the issue.
 
  • Like
Reactions: Mr.Frisfruit

Mr.Frisfruit

Prominent
Jun 17, 2019
37
5
545
You are absolutely right....and I agree with you....But.....these cards do seem to have overheating issues. I understand not feeling comfortable and my advice to make it easier is that by changing these settings, you are not Overclocking these settings. You can overclock by changing the settings but the goal here is to test the card and verify if the card is the issue, you are actually lowering the settings and making sure the card is running stable. A safe test is to lower the mv across the board by 50 mv. You can also lower the frequency by 50Mhz as well. I think this works better but another option may be in the watman is to run under the silent or the preset lower settings as well. If you pc is only crashing during games and nothing else...I had the same issue. I returned a Asus Strix Vega 64...I had a great deal on an open box, I liked the looks but I didn't want to have to lower the settings to use it. This card has a reputation for being great after changing thermal paste and using a thicker thermal strip but it crashed too much for me to want to take the risk. If you search undervolting on the 580 cards, you will see how many people say the same thing about not wanting to change the settings but seeing that is the issue.
Hm okay. I have a friend who has done such things before, I'll ask him to help me. Hopefuly it'll work. I'll post an update soon.

Thanks for the advice.
 

Mr.Frisfruit

Prominent
Jun 17, 2019
37
5
545
Hi guys it's been a while since I've given an update due to holidays but sadly the problem still persists... I have swapped the GPU with one of my friend's and games still keep crashing. So I've basically ruled out the GPU, RAM, PSU and any software/driver related issues. The only remaining components that could cause the problem are the CPU (which seems unlikely), my OS-storage drive or my mobo. Does anyone have any advice on what to test next? Could this be caused by a faulty mobo or ssd? I can't swap these components since none of my friends own compatible ones, so I probably have to guess which component is faulty, return it, and hope that it is indeed the culprit. So any advice on the 'most likely' broken component is appreciated, since it might save me quite some bucks.