Question Help games are crashing and my hair is being pulled out

[Joe Gallo]

Hello all, I tend to slowly cannabalize my pc parts into new pcs.

https://pcpartpicker.com/list/JNRNh7

This is my current full build

What has been slowly getting worse and worse is aside from the normal random games crashing. Certain games will frequently crash. They either lead to one of the following symptoms or multiple of them.

Sometimes I will be forced to sign out of windows to close the game, my screen will freeze and sounds will loop, my pc will only respond to fully turning the pc off by power button, i will get a random single pastel colored screen with a thin pixelated line (almost like a line of dead pixels on the monitor), or i will get clock client watchdog with a system restart.

I’ve tried changing ram slots, running xmp 1, 2 or no profile, gear ratio 1:1 and 1:2, overclocking my cpu, not overclocking my cpu.

I’ve ran memtest86, prime95, and furmark and had no overheating, errors or crashes. (No overheating during gaming either)

I’ve always had the most updated bios, windows, all drivers and programs

The games that are most severely affected are; Fallout 76 Hell Let Loose Vrchat

Other games that show similar signs but are more playable are; Total war Pharaoh, Total War Napoleon, NBA 2k25

One other thing of note is that when I use Msi afterburner to underclock vram, clock speed and power usage. Vrchat and hell let loose become playable, Fallout 76 will make it past the loading screen and let me play for a few minutes. However all 3 will exhibit these symptoms immediately when that afterburner profile is not enabled.

This wasn’t always the case i used to never crash on these games. Slowly but surely they all 3 slowly have started to get worse and worse. I’m pulling my hair out trying to understand this as I play other games like RDR2, fallout 3/NV/4 FH5 and a number of other games without issue.

PLEASE HELP!

 

[Lutfij]

How old is the PSU in your build?

I’ve always had the most updated bios
What BIOS version are you on for your motherboard?

windows,
You're on version 24H2?

Use DDU to remove all GPU drivers(intel, AMD and Nvidia) in Safe Mode, then manually install the latest GPU driver sourced from Nvidia's support site in an elevated command, i.e, Right click installer>Run as Administrator.

 

[Joe Gallo]

How old is the PSU in your build?

I’ve always had the most updated bios
What BIOS version are you on for your motherboard?

windows,
You're on version 24H2?

Use DDU to remove all GPU drivers(intel, AMD and Nvidia) in Safe Mode, then manually install the latest GPU driver sourced from Nvidia's support site in an elevated command, i.e, Right click installer>Run as Administrator.

My PSU is from 2016-2018 somewhere in there it’s been a while I’m not 100% sure.

Bios I’m on version 1202, the most recent from Asus Rog Ai suite 3

Windows I am on 24H2

And driver wise I’ve only ever used the stock windows drivers and whenever I update drivers on nvidia app, I do a clean install, installation

 

[Lutfij]

If you're working with this motherboard;

ROG STRIX Z590-A GAMING WIFI

ROG STRIX Z590-A GAMING WIFI motherboard features boosted power delivery, an optimized cooling design, PCIe® 4.0, WiFi 6, AI controls and cyberpunk-inspired aesthetics.

rog.asus.com

then;

ROG STRIX Z590-A GAMING WIFI

ROG STRIX Z590-A GAMING WIFI motherboard features boosted power delivery, an optimized cooling design, PCIe® 4.0, WiFi 6, AI controls and cyberpunk-inspired aesthetics.

rog.asus.com

you have a number of BIOS updates to go through. Make sure you're on the latest chipset driver version, then flash your BIOS to version 2302.

And driver wise I’ve only ever used the stock windows drivers and whenever I update drivers on nvidia app, I do a clean install, installation
Both of those methods are only going to land you with more hair loss. Try DDU and see if things change, after you've updated the BIOS.

 

[Joe Gallo]

If you're working with this motherboard;

ROG STRIX Z590-A GAMING WIFI

ROG STRIX Z590-A GAMING WIFI motherboard features boosted power delivery, an optimized cooling design, PCIe® 4.0, WiFi 6, AI controls and cyberpunk-inspired aesthetics.

rog.asus.com

then;

ROG STRIX Z590-A GAMING WIFI

ROG STRIX Z590-A GAMING WIFI motherboard features boosted power delivery, an optimized cooling design, PCIe® 4.0, WiFi 6, AI controls and cyberpunk-inspired aesthetics.

rog.asus.com

you have a number of BIOS updates to go through. Make sure you're on the latest chipset driver version, then flash your BIOS to version 2302.

And driver wise I’ve only ever used the stock windows drivers and whenever I update drivers on nvidia app, I do a clean install, installation
Both of those methods are only going to land you with more hair loss. Try DDU and see if things change, after you've updated the BIOS.

I’ll get that done and update you when it’s finished

 

[SkyNetRising]

i will get a random single pastel colored screen with a thin pixelated line (almost like a line of dead pixels on the monitor)

One other thing of note is that when I use Msi afterburner to underclock vram, clock speed and power usage.
Vrchat and hell let loose become playable, Fallout 76 will make it past the loading screen and let me play for a few minutes.
However all 3 will exhibit these symptoms immediately when that afterburner profile is not enabled.

That sounds like a graphics card problem.
What are gpu temperatures during gaming load?

Try with a different graphics card.

 

[Joe Gallo]

That sounds like a graphics card problem.
What are gpu temperatures during gaming load?

Try with a different graphics card.

Never higher than 84, typically 65-78. And I don’t have another card or friends that oc game unfortunately

 

[SkyNetRising]

And I don’t have another card or friends that oc game unfortunately

You'll have to spend some money then.
Your card is over 10 years old. It's basically begging for an upgrade.

 

[Joe Gallo]

You'll have to spend some money then.
Your card is over 10 years old. It's basically begging for an upgrade.

Yeah I know. This will be an unpopular opinion, but up until recently I’ve been running everything at 1080p ultra and in the games I play, when I had a 4K tv. 4K ultra with no issues. RDR2 was the only thing that struggled a bit but with a few turned down settings still played fine. I’ve been waiting for an affordable 4K ultra 140hz raytracing card for years now. Short of that it makes no sense at all to upgrade (I’m sorry for the people that spent money and will disagree or people who play 1440p 140hz). So I’m hoping it’s my PSU because my cables will need an update when I do update my gpu. Spending $600 usd in 2016/17 on a top end card running everything ultra still. it’s a tough sale to upgrade to today’s gpu market.

 

[SkyNetRising]

it’s a tough sale to upgrade to today’s gpu market.

Modern cards will run circles around your old 980 ti.

 

[Joe Gallo]

Modern cards will run circles around your old 980 ti.

I don’t disagree, but I’m not sitting here [vulgarity redacted] to benchmarks, I’m gaming. In the games I play I’m running things ultra with good frames. So there’s nothing tangible a new card can give me besides 1440p, 4K and raytracing. I’d rather just skip 1440p go straight to 4K as I think I’ve shown I like to buy things are very future proof. But this is all besides the point of troubleshooting my system.

 

[MEMOFLEX]

I don’t disagree, but I’m not sitting here fapping to benchmarks, I’m gaming. In the games I play I’m running things ultra with good frames. So there’s nothing tangible a new card can give me besides 1440p, 4K and raytracing. I’d rather just skip 1440p go straight to 4K as I think I’ve shown I like to buy things are very future proof. But this is all besides the point of troubleshooting my system.

The issues you are having does point towards it being a GPU issue but as mentioned above there are a few things that need addressing such as the BIOS and vendor chipset and general motherboard drivers. Let's see what the outcome of those other recommended updates are i guess.

4k 144hz monitors are a wonderful thing to have but in modern AAA titles with Ray Tracing turned on then even the best current cards can struggle to achieve that without the help of upscaling technologies.

Out of interest what would be your contingency plan be should the gpu be the problem and requires replacing?

 

[Lutfij]

I don’t disagree, but I’m not sitting here [vulgarity redacted] to benchmarks, I’m gaming. In the games I play I’m running things ultra with good frames. So there’s nothing tangible a new card can give me besides 1440p, 4K and raytracing. I’d rather just skip 1440p go straight to 4K as I think I’ve shown I like to buy things are very future proof. But this is all besides the point of troubleshooting my system.

If after all things are said and done, the culprit ends up being your GPU...then that needs replacing. Either through a donor GPU from a gracious relative, friend or neighbor or courtesy of a new GPU purchase. To add, a concurrent GPU would go hand in hand with your platform although the PSU in your build was released in mid-2015, making it a decade old unit. A new GPU would warrant a new PSU purchase as well.

 

[Joe Gallo]

If you're working with this motherboard;

ROG STRIX Z590-A GAMING WIFI

ROG STRIX Z590-A GAMING WIFI motherboard features boosted power delivery, an optimized cooling design, PCIe® 4.0, WiFi 6, AI controls and cyberpunk-inspired aesthetics.

rog.asus.com

then;

ROG STRIX Z590-A GAMING WIFI

ROG STRIX Z590-A GAMING WIFI motherboard features boosted power delivery, an optimized cooling design, PCIe® 4.0, WiFi 6, AI controls and cyberpunk-inspired aesthetics.

rog.asus.com

you have a number of BIOS updates to go through. Make sure you're on the latest chipset driver version, then flash your BIOS to version 2302.

And driver wise I’ve only ever used the stock windows drivers and whenever I update drivers on nvidia app, I do a clean install, installation
Both of those methods are only going to land you with more hair loss. Try DDU and see if things change, after you've updated the BIOS.

So I’ve done all the steps and the issue is still there. I should add I also tried each ram individually in different slots as it just occurred to me that I haven’t done that yet.

The issues you are having does point towards it being a GPU issue but as mentioned above there are a few things that need addressing such as the BIOS and vendor chipset and general motherboard drivers. Let's see what the outcome of those other recommended updates are i guess.

4k 144hz monitors are a wonderful thing to have but in modern AAA titles with Ray Tracing turned on then even the best current cards can struggle to achieve that without the help of upscaling technologies.

Out of interest what would be your contingency plan be should the gpu be the problem and requires replacing?

Yeah unfortunately those steps didn’t help.

Right now, if it is the gpu thats the issue. My fingers are crossed that C series intel arc gpus are a big update with them wanting to wow us. Given the news lately I think that’s unlikely.
There’s also rumors of a 50 series super update that has additional vram at a lower prices which is interesting. I’d probably hold off until both of those drop as I can bare this issue for a few more months.

If I had to purchase something today I’d probably get a ARC b580 or amd 9060xt 16gb if it’s on sale. then I’d upgrade to something higher end the year after next gen console release.

A new console release year gpu basically means you don’t need to update for 5-6 years in my book. Considering they’re dropping in the next couple of years. That what I was waiting to upgrade, if my parameters weren’t met sooner

If after all things are said and done, the culprit ends up being your GPU...then that needs replacing. Either through a donor GPU from a gracious relative, friend or neighbor or courtesy of a new GPU purchase. To add, a concurrent GPU would go hand in hand with your platform although the PSU in your build was released in mid-2015, making it a decade old unit. A new GPU would warrant a new PSU purchase as well.

Yeah that’s backup plan, and what prompted this forum post, is that I probably will be purchasing a new PSU today if I can’t solve this in here. I was trying to isolate the problem further and set up that gpu upgrade

 

[Lutfij]

Yeah that’s backup plan, and what prompted this forum post, is that I probably will be purchasing a new PSU today if I can’t solve this in here. I was trying to isolate the problem further and set up that gpu upgrade

I usually ask users to source(borrow, not buy) parts to help with the troubleshooting process. If we were all buying parts to troubleshoot a system, we'd all end up with two of each part, albeit one would be brand new and not concurrent('ish).

You could drop your GPU onto another known working, capable system which has more wattage than your current system. That would rule out your platform being the center of the issue, provided the GPU doesn't behave the same way on donor system.

If your ram was the issue, then memtest would've spat out an error within 10 passes of the test running. If you didn't get anything for 10 passes, it's not your ram.

 

[Joe Gallo]

I usually ask users to source(borrow, not buy) parts to help with the troubleshooting process. If we were all buying parts to troubleshoot a system, we'd all end up with two of each part, albeit one would be brand new and not concurrent('ish).

You could drop your GPU onto another known working, capable system which has more wattage than your current system. That would rule out your platform being the center of the issue, provided the GPU doesn't behave the same way on donor system.

If your ram was the issue, then memtest would've spat out an error within 10 passes of the test running. If you didn't get anything for 10 passes, it's not your ram.

Good to know, I had been worried given it was Corsair ram from when they were having all of those issues with their ram.

I was posting to make sure that it is the GPU and or PSU as those are my next upgrades anyways.
I really don’t mind upgrading my PSU as today is my bday and it’ll be like a birthday gift to myself. I was going to do a PSU upgrade as I do plan to upgrade my gpu soon anyways.

So just for my sanity is it kind of ruled out at that it’s my MoBo, cpu or ram? Which was the scary thing for me.

As for higher wattage, there’s two questions I have there.
I currently have a 850w Platinum would you suggest getting an even higher PSU? Or are you testing to see if the PSU is failing in that scenario?

Also I’m currently using a vga power supply cable that has a splitter of 6/6+2 could that be the culprit? I’ve run this like that this whole time so I never thought it’d be the issue.

 

[Lutfij]

Welp, Happy Birthday then! 🥳

So just for my sanity is it kind of ruled out at that it’s my MoBo, cpu or ram?
Not really. Troubleshooting isn't about pointing a finger at something and hoping it's the culprit, you swap parts and narrow down the list of possible culprits until you're left with the logical conclusion. At this moment of time, you're just working with your bubble of own parts(a closed loop if you will).

 

[Joe Gallo]

Welp, Happy Birthday then! 🥳

So just for my sanity is it kind of ruled out at that it’s my MoBo, cpu or ram?
Not really. Troubleshooting isn't about pointing a finger at something and hoping it's the culprit, you swap parts and narrow down the list of possible culprits until you're left with the logical conclusion. At this moment of time, you're just working with your own parts.

Thanks man,

and okay well I’ll swap out that power supply today and then later in the year probably get a graphics card, if still no luck

 

[gamerk316]

Yeah, anything involving dead/stuck pixels usually means the GPU is on the way out. That's one reason I still prefer to get CPUs with an integrated GPU built in, since it's great for confirming GPU failures.

 

[thestryker]

It's possible that the video card may need pads replaced as those tend to connect the VRM to the heatsink and aren't temperature monitored. If you're comfortable taking apart the card and sourcing new pads/repasting the GPU I think this would be a good step to take before buying a new video card.

 

[Joe Gallo]

Yeah, anything involving dead/stuck pixels usually means the GPU is on the way out. That's one reason I still prefer to get CPUs with an integrated GPU built in, since it's great for confirming GPU failures.

Yeah I skipped out on it this upgrade because I had a gpu upgrade in mind when I built it. But I unfortunately didn’t foresee this happening

It's possible that the video card may need pads replaced as those tend to connect the VRM to the heatsink and aren't temperature monitored. If you're comfortable taking apart the card and sourcing new pads/repasting the GPU I think this would be a good step to take before buying a new video card.

I did a cursory try and it wasn’t something I think I would be able to do. That is the answer I was expecting when I posted or that my psu was failing. I’ve been debating bringing it to a shop to repair it. But I think I’m going to see what GPU launches look like and the next few months and go from there. It’s not too bad on most games except fallout 76. And even underclocking it still gives me great results

 

[Joe Gallo]

It is not the power supply, all wired up and running. Still getting the crashes. Now powered by ATX 3.1 lol

Welp, Happy Birthday then! 🥳

So just for my sanity is it kind of ruled out at that it’s my MoBo, cpu or ram?
Not really. Troubleshooting isn't about pointing a finger at something and hoping it's the culprit, you swap parts and narrow down the list of possible culprits until you're left with the logical conclusion. At this moment of time, you're just working with your bubble of own parts(a closed loop if you will).