• Hi there and welcome to PC Help Forum (PCHF), a more effective way to get the Tech Support you need!
    We have Experts in all areas of Tech, including Malware Removal, Crash Fixing and BSOD's , Microsoft Windows, Computer DIY and PC Hardware, Networking, Gaming, Tablets and iPads, General and Specific Software Support and so much more.

    Why not Click Here To Sign Up and start enjoying great FREE Tech Support.

    This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Solved Gaming PC crashes when taxed in any Video Game

Status
Not open for further replies.

ghostrider1982

PCHF Member
Nov 3, 2022
15
1
42
All,

I have been having a problem with my PC for several weeks now that it will work perfectly fine until I try to play video games with it. I can watch movies/videos, play emulator games, stream videos, surf the web etc. and rarely does it every have a single issue. However, the moment I fire up a game in some short period of time the system will crash. Normally to black screen (various errors in event viewer related to nvlddmkm) where it will either recover with the game crashed out or the system will reboot. This can happen 1-2minutes in or 1hr in it's very sporadic but once it happens if the system recovers and I try to go back in it generally crashes again very quickly. I tried some fix that was on YouTube about BlackScreen nvlddmkm where it had me ensure the video driver is the most current and made sure power settings are maxed but it had little effect. I've also monitored temperatures but nothing gets too high (CPU taxed 40-50C) (GPU taxed 70-80C).

I don't think the issue is RAM as I've ran MemTest multiple times on my RAM with no faults and also i've tested my PC with 8gb/16/32gb and get the same crashes. It actually seems like it crashes faster when I have all 4 sticks/32gb of RAM in then when I run 16gb. So this leads me more towards a MB or Power issue but I can't rule out the graphics card either. I have a 750W power supply so power should be adequate if it's quality power. Another weird thing that I don't know if it's related at all or not is I have RGB ram and very quickly it goes out of color sync which makes me again wonder if there is an MB/Power issue or if it's not related at all. When I start the system it's synced but in no time it goes out of sync in a rainbow configuration.

Any additional troubleshooting steps, guidance or feedback would be greatly appreciated. I've attached the speccy report results link below. Thank you.

http://speccy.piriform.com/results/XTJUpPPxjZOT0AZWidPze6R
 
Looked at Speccy and saw a couple of things but need to get back to you about them when I get back home.

Your PSU while of good quality is too weak to support your GPU, ASUS state a minimum of an 850W PSU for your GPU.
 
A good quality 850W PSU is not what I would call cheap tbh, ASUS makes various versions of the RTX 3080, a couple require two 8 pin supplemental power dongles from the PSU but most of them require three, example card here requires two, all versions require 850W minimum.

Best value I could find CORSAIR RM850 CP-9020235-NA 850 W ATX 80 PLUS GOLD Certified Full Modular Power Supply amazon also have it at the same price.

Speccy is not giving us your RAM specs, can you post the brand and model name or number for us.

Power Profile
Active power scheme: Norton Utilities Premium - Ultra Performance

Never heard of such garbage, Norton should stick to what they know and leave everything else well alone.

Change the Windows Power Plan to Balanced, Ultra and High Performance are a form of overclocking that is known to cause stability and overheating issues, the setting should only be used for gaming type notebooks that have a discrete GPU that needs the extra power.
 
Norton and the likes are commonly referred to as snake oil that no good ever comes from, why have you even got it on your machine.

Did you never notice that the specs for your card say 850W.

That is the wrong RAM for your CPU, Intel state here up to 3200MHz and if you have XMP enabled the RAM will get auto OCd past what the CPU can handle and the PC fall over.

Go into the BIOS, disable XMP and then manually set the RAM to run at 3200MHz and the voltage to 1.35V, save the new settings and exit the BIOS.

The EVGA was a great PSU and being that it is only eight years old it still has two years of its ten year warranty period left, if you never had problems with it previously it should be safe to swap in, spec wise it is all there but only you know of its condition so you must make the final decision.
 
So as you might be able to tell this wasn't a custom PC build. I bought it via CyberPower. My mistake to think they wouldn't give you specs in a model that don't go together such as the RAM with the CPU and the GPU with the power supply.

I've made the changed in the BIOS for the RAM and am going to swap out the PSU now and see how it goes. Will let you know.
 
No joy. Replaced the power supply. Have some other issues to sort out when it comes to the CPU cooler and the fans not working on the radiator (don't think they have) but when it comes to crashing it still happens maybe even more so now. I think it's the video card. I try to run a stress test on the video card and literally within 10-30seconds it crashes out every time either it recovers or goes straight into a PC reboot. I tried moving the video card to a different slot but it didn't change anything. Think i'll have to send it in as its under warranty.
 
An easy test to help, pull the GPU and run off the video output on the CPU (The video output on the motherboard), go back and try the same test.

If that solves your issue, it might be worth doing a display driver reinstall before shipping it out for warranty repair.

It's possible the fan curve in the BIOS is set to maintain low-speed or no-speed fans until the computer heats up a little, I would make sure that you have them plugged into the proper headers first though, overheating can get expensive very quickly.


Be mindful of what you can and cannot tamper with in accordance with your warranty/warranties.
 
I did what you suggested and while it was blazing at 6FPS at 4K Res and 16FPS at 1080 the system never crashed whereas with the video card it gets to full power and then crashes in 10-30seconds.

What would you suggest for reinstalling drivers. I've had various iterations of the Nvidia drivers over the past few months and nothing really changed the issue i just didn't have time to deal with it until now and I'd really like to use my Gaming RIG for more than just watching shows and surfing the web which is all I can really do with it right now.

As for the CPU Cooler, it's a Thermaltae Floe Riing RGB 240 seen here: https://www.amazon.com/Thermaltake-...ocphy=9008568&hvtargid=pla-451648940878&psc=1

It has 2 fans on the radiator and the block on the CPU. The block on the CPU has a separate connection to the Motherboard and I can tell it's working when i touch the connectors to and from the radiator but the Fans only have a connection that goes to a TT controller box. It allows me to go into the Thermtake software and change the RGB colors just fine the but the fans never spin up at all. I'm not sure what to look for in the BIOS as far as Fan Curve, all the ASUS bios has is something about QFAN controls or whatever and i tried to set it all to just run 100pct to see if i could get the fans to spin but they just won't. I mean the processor never seems to get hot but I can't believe that never spinning is a common state for those fans on a radiator designed to dissipate heat. I have 3 other case fans but they are not adjustable as they are all controlled together. What's weird is there is a reading for CPU fan and I can't figure out what it's tracking when it talks CPU fan as there is no other fan and it varies in speed. In the Aura Sync I can vary the speed of said fan but I do not know where this mystery fan resides because as far as I know there is no other fan, unless it's somehow referring to the block on the CPU.
 
I put the Video Card back in and uninstalled all Nvidia software and then went to Device Manager and uninstalled the Graphics Card but it finds the driver again and installs it moment later. I then ran the same stress test and was getting 325FPS at 1080 but then after about 8 seconds artifacts pop up all over the screen and the system goes black and immediately restarts. It's at 100pct Power when this happens. Again we though power issue earlier but i've now swapped to a 1000W PSU and it's making on difference.
 
This suggests to me there's either a GPU issue or a power issue.

Since you seem to have a competent power supply by the sounds of it (a model # would confirm this) my next guess is something with the GPU as you have stated.

If you have another GPU (preferably a similar one, such as a 20 series or another 30 series) you can swap that in and double over your work.


When you removed your display drivers, did you do it correctly?



Run DDU and download a fresh installer executable from here.

An important note: You MUST download the driver executable and disconnect your computer from the internet before running DDU, otherwise Windows will automatically download drivers (which you don't want).

Once you have ran DDU in safe mode, boot up into a normal Windows session (ensuring that you aren't connected to the internet for even a moment) and run the NVIDIA installer.

Restart your computer from the power menu and see if this helps.
 
  • Like
Reactions: Pyro
Tried the driver thing no difference, if anything the problem is getting worse as my screen has flickered on me a few times and even did lock up once not even trying to play a game. But any stress test and like clockwork within 10-30seconds it crashes out. Guess we've really ruled everything out and the culprit has to be the video card. Glad it's on a 3yr warranty but too bad I have to pay for shipping there and back lol.
 
Both ways? Usually shipping is paid for by the sender (you pay shipping there and the repair depot pays shipping back), who makes the card?

If it works fine otherwise the CPU should hold over some light gaming (low settings 1080p/lower, nothing AAA) for the week or two it takes to get your GPU back.

Unless someone else has any ideas it sounds like you're on the right track. :)
 
Have some other issues to sort out when it comes to the CPU cooler and the fans not working on the radiator

This should have been mentioned in your OP.

when it comes to crashing

Software such as Windows can crash and when it does crash you get a BSOD and when enabled a crash dmp is generated, programs or games when they crash can on occasion close to the desktop but the computer will still be 100% functional.

Hardware failure such as a weak power supply and/or overheating are not software related and when a computer for example suddenly turns off, freezes or the screen goes black etc the behaviour should be described as the "computer shut down unexpectedly" or froze etc and not as having crashed as the latter implies a software issue as opposed to an obvious hardware issue when described properly.

Having the correct info means that helpers will not be looking for a software issue when the problem is clearly hardware related

Being that a weak PSU was previously being used CyberPower may have trouble with any RMA, the bench gear that Nvidea techs have at their disposal can detect when an item has as they may well put it has been " inappropriately used " means that any warranty is null and void, they did this when cards cost a couple of hundred of your currency so I don`t expect them to be any different when cards are now costing just shy of and sometimes more than 1K.

Any warranty terms that you may have still had were voided the minute that you started tinkering with the internal components and sorry to say that the minute that you had concerns over the radiator cooling fans you should have stopped using the PC and sought technical assistance and if still under warranty with CyberPower first.
 
Status
Not open for further replies.