Question Constant CTD on multiple games

[Scarecrow]

Hi,
I've been having CTD problems with three different games : Warhammer 40.000: Darktide, Cyberpunk 2077 and Sea of Thieves. Always looks the same : game suddenly freezes, then closes after ten seconds or so, sending me back to desktop with a box for sending a bug report. In Darktide, it'll happen after about 10 minutes during my first attempt of the day, and then repeatedly every 3 to 5 minutes after that. In Cyberpunk, it's frequent but more random, except in some specific sections of the game, where I crash consistently around the same time. In SoT, it happens after a couple minutes in game and I get an error message telling me to update my video drivers.

Specs:
CPU: AMD Ryzen 5 2600X
GPU: Nvidia RTX 2070
SSD: Crucial P1 3D NVMe PCIe M.2 500GB
HDD: WD Blue 500GB (2010)
RAM: G.SKILL F4 DDR4 3200 C16 2x8GB
MBD: Gigabyte X470 AORUS ULTRA GAMING
PSU: Corsair RM850 Gold

Windows 10 Pro 64bit
OS build 19045.2364

Userbenchmark link: https://www.userbenchmark.com/UserRun/57999503

More detailed explanation:
A month ago, I had a 600W PSU and a stock cooler on my CPU. I was running Cyberpunk on low with no ray-tracing at an average of 30 fps, but crashes were pretty rare (maybe one crash per 2-3 hour session on average). It was kind of working so I didn't look deeper into it.
Then I started playing Darktide and the game was crashing constantly, as described in first paragraph. I tried the following things: updated graphics drivers, verified integrity of game files, put all settings on the lowest value available, try to set the "worker thread count" option to "1" (dev suggestion). None of them worked, so I quit waiting for an update, cause the game is notoriously unstable for a lot of people.
After that, I went back to Cyberpunk and found that it started to crash repeatedly too. Went online for help, I was told that my PSU was too weak, that my CPU might be running hot and that I might have a memory problem. My CPU was indeed runnning hot (over 90C), so I replaced my stock cooler with a be quiet! Pure Rock 2, replaced my PSU with a new 850W one and did some RAM tests with no errors.
After the upgrade, I now run both Cyberpunk and Darktide on higher quality with better framerate, but they still both crash constantly. My CPU isn't hot anymore (never goes above 75C). I also enabled XMP for my RAM in the BIOS. No luck.
I tried excluding the games from my antivirus scan list: no luck.
I tried running Cyberpunk directly from the .exe as administrator: I got past the scanning sequence I was stuck on, but crashed 10 seconds later.
Sea of Thieves tells me to update my video drivers after the crashes, which I've already done multiple times over the last few weeks.
All games are installed on SSD.
I don't really know what to do, a lot of this stuff is already beyond my limited understanding of computers.
Will happily provide more information. Any help is greatly appreciated.

 

[Kaamos_Llama]

Hi and welcome, that sounds frustrating. Have you tried upgrading to the latest BIOS on your motherboard?

Assuming your motherboard is a rev1 (it will say on the board itself somewhere) Download from here and follow the instructions in your motherboard manual to update it. See if that helps.

X470 AORUS ULTRA GAMING (rev. 1.0) Support | Motherboard - GIGABYTE Global

Lasting Quality from GIGABYTE.GIGABYTE Ultra Durable™ motherboards bring together a unique blend of features and technologies that offer users the absolute ...

www.gigabyte.com

Also worth doing a full reinstall of GPU drivers just to be sure, using the steps in this article.

If that doesnt work, can you list the exact part number for the RAM? Older Ryzen CPU's can be very fussy with RAM compatibility.

Something like this: F4-3200C16D-16GVKB

If its not on the compatibility list at the Gigabyte website I would suspect RAM, you could try running the RAM at stock settings, disabling XMP in the BIOS and see if that stops the crashes.

 

[Scarecrow]

Thanks a lot for taking the time. I'm afraid I'm gonna be a bit of a bother.
I tried to update to the latest BIOS version, but I get an "invalid bios image" error. I'm apparently on version F3 and from what I've gathered, I need to first update to F31, then to F40, which requires EC FW Update Tool (B19.0517.1 or later version), and only then will I be able to update to the latest version available (F63c).
Two issues: I've never updated the BIOS and it says on the gigabyte website that it has risks, so I'm a little stressed out doing 3 times in a row, and I can't seem to find EC FW Update Tool on the gigabyte website, only here https://driverscollection.com/_5354...-Tool-v.B19.0606.1-for-Windows-10-64-bit-free and I don't know if it's a trustworthy source.

Edit: I checked my RAM (F4-3200C16-8GFX) and it's not on the compatibility list. I thought I checked that when I bought the components, Apparently I wasn't thorough enough.
I don't think it's the XMP though, because I had the crashes both with and without (I only turned it on after I started having the issue). So, new RAM it is...?

 

[Kaamos_Llama]

I dont know what that is, you shouldnt update your BIOS from inside Windows as theres a higher risk of a crash, but its perfectly safe to do in the BIOS which you would access by pressing delete at the prompt when the machine is booting up after being powered down.

As long as you dont have a power cut or pull the power cable out while its updating you should be fine, I did it myself today on my Gigabyte board using Qflash, which your manual says that your board has.

I'll try to keep it clear and simple

1. Download the BIOS versions you need from the motherboard support page.
2. Extract the files by right clicking and extract all.
3. Take a USB Memory stick and format it to FAT 32 by right clicking in File Explorer and selecting 'format' and in then 'FAT 32' in the drop down menu.
4. Cut and paste the BIOS file to the stick and nothing else.
5. Place the USB stick in a USB port on the back.
6. Restart the machine and enter BIOS by pressing delete at the prompt.
7. Enter the Qflash utility and follow the instructions there, the BIOS you downloaded should be listed as available.

8. Wait for the machine to reboot, it may power up and down a couple of times, and if you need to do the same again just follow the same steps with the second BIOS.

 

[Scarecrow]

Yeah, that's what I did, but when I select the X470AOUG.63c, nothing happens, and when I try to click on the right arrow to go to the next step, I get the "invalid bios image" error.

 

[Scarecrow]

I finally managed to update to the latest BIOS version by updating to an intermediate one first. Sadly, I just tested it in Cyberpunk 2077 and I crashed as usual. I'm going to try the full resintall of GPU drivers.

 

[Scarecrow]

Clean install performed, still crashing

 

[Kaamos_Llama]

Whats bothering me is you say that it was happening rarely and then started to happen a lot more for no apparent reason.

You didnt install anything program wise in that time that's running in your system tray constantly, monitoring software from Razer or Corsair for example? Updates to those kinds of things can be dodgy, try uninstalling anything like that.

Also what antivirus are you using? Have you tried turning it off entirely? Its totally fine to just use Windows Defender as long as youre a bit careful, so uninstalling whatevr 3rd party one you are using for a short while to test wouldnt be dangerous or anything.

Before replacing the RAM, I would also be looking at backing everything important up and doing a fresh Windows install to check its definitely not anything in software, unless I happened to have hardware on hand to test with. Obviously you would want to install minimal programs other than Steam and games to check.

You could also try switching the RAM to the empty slots in the motherboard and see if that changes anything, Obviously power it down and turn off at wall first. Try just one stick in the first slot as well and see if the crashes stop at least.

 

[Scarecrow]

Yeah it makes no sense to me either. The crashes in Cyberpunk got out of control when I updated my GPU drivers the first time to try and fix the crashes in Darktide (early decembre). But since then, there's been a couple new versions and the crash have persisted in both games. The only thing I installed recently was a backup software (EaseUS Todo) to make a full backup of my drives before tinkering with more advanced stuff, but that was already after the crashes had started.
No antivirus apart from Windows Defender.
I just tried moving the RAM around, 2 sticks, 1 stick. Weirdly, I was able to play a little longer with only 1 stick, but still crashed after a couple minutes.
I'm kind of dreading the clean Windows install, but it seems I'm running out of options.

 

[Brian Boru]

Non-expert poking nose in:

SSD: Crucial P1 3D NVMe PCIe M.2 500GB
HDD: WD Blue 500GB (2010)

Which drive is Windows on? The games?

Have you been using the HDD since 2010?

How much free space is on each drive?
Is the HDD defragged?

3 diff games is possibly an indication they're having difficulty loading extra stuff required as you progress in a game.

 

[Kaamos_Llama]

Yeah it makes no sense to me either. The crashes in Cyberpunk got out of control when I updated my GPU drivers the first time to try and fix the crashes in Darktide (early decembre). But since then, there's been a couple new versions and the crash have persisted in both games. The only thing I installed recently was a backup software (EaseUS Todo) to make a full backup of my drives before tinkering with more advanced stuff, but that was already after the crashes had started.
No antivirus apart from Windows Defender.
I just tried moving the RAM around, 2 sticks, 1 stick. Weirdly, I was able to play a little longer with only 1 stick, but still crashed after a couple minutes.
I'm kind of dreading the clean Windows install, but it seems I'm running out of options.

Just checking in to make sure you followed the steps in the article I linked above when clean reinstalling GPU drivers. It can help.

Could also try updating the AMD chipset drivers, and all of the network drivers etc from the motherboard support site to make sure none of that is corrupted.

 

[Scarecrow]

Non-expert poking nose in:


Which drive is Windows on? The games?

Have you been using the HDD since 2010?

How much free space is on each drive?
Is the HDD defragged?

3 diff games is possibly an indication they're having difficulty loading extra stuff required as you progress in a game.

Thanks for lending a hand.

Windows and the games are on the SSD.

I've been using the HDD for a while (2012 I think), but I only use it to store music, movies, pictures, etc. It used to be the main drive on my previous computer.

Before installing Sea of Thieves (the latest problem game), the SSD was at around 100 GB of free space. Now it's at 19. The HDD has 143.

Haven't manually defragged the HDD in a while, I don't really know how it could affect things on the SSD but I'll do it anyway just in case.

 

[Scarecrow]

Just checking in to make sure you followed the steps in the article I linked above when clean reinstalling GPU drivers. It can help.

Could also try updating the AMD chipset drivers, and all of the network drivers etc from the motherboard support site to make sure none of that is corrupted.

Yeah, I followed the guide step by step, ran DDU in safe mode and everything.

I updated the chipset, still crashing. I can't find any network driver on the gigabyte page. I tried updating the Realtek audio driver, the installer uninstalled my current driver, rebooted the PC, and now it keeps telling me it needs to uninstall and reboot before installing the new one. Did it three times, still not reinstalling... PC doesn't recognise my sound system anymore.

Edit: after canceling the installer and a couple reboots, it reverted back to the old audio driver and I have sound again. I think I'm gonna leave that one alone.

 

[Scarecrow]

I am new here. I've been trying to post in the forums but I am being blocked...any help?

On the main forum page, there's a "Post thread..." button on the right, just above the "Welcome to PC Gamer" banner with the search bar. Click it, then choose the relevant category.

 

[Brian Boru]

Before installing Sea of Thieves (the latest problem game), the SSD was at around 100 GB of free space. Now it's at 19

No worries re the HDD, if neither Windows nor games are on it. Can't hurt to defrag tho. Make sure you have its irreplaceable contents backed up somewhere else—it's a question of when it'll fail, not if.

19 is too little free space. The topic is too complicated for me, but I use 20% free space as my rule of thumb. See here and there and this place. This is more important on your Windows drive, as Windows needs extra room for various maintenance and update operations. So I advise you to get back towards 100 free on the SSD—don't stress if you can only get 50-60, you probably don't need 20% on larger drives… eg 100GB is likely more than enough on a 1TB drive. It's a grey area with a lot of variables

But this apparently isn't related to your problem, since it was happening when you still had 100 free.

PS

On the main forum page

Thanks for helping, I moved that post to an appropriate location. We get hundreds a day wanting to post links

 

[ipman]

I don't know enough to advise on this, but I occasionally get CTD's, freeze then CTD or even just freeze and can't even get back to desktop in one particular game(the latter was an audio driver problem).

(Loading the game from a desktop shortcut has helped for me).

I'm sure this is common knowledge but a good place to start it type RELIABILITY into SEARCH>View Reliability History(and Problem History). Click on the Critical Event by date, then View Technical Details. It might give you an insight.

(So for me it says, 'game stopped interacting with Windows and was closed').

Maybe worth a go! And I empathise.

 

[Kaamos_Llama]

I didnt even know they had replaced Event viewer in Win 11! Think its still Event viewer in Windows 10 though afaik. Good idea though @ipman , sometimes it points to something.

@Scarecrow
I had one other thought @Scarecrow try unplugging all USB devices and see if that changes anything. If it helps plug them back in one by one until you find the culprit.

 

[Scarecrow]

Reporting back after a clean Windows install: still crashing in Cyberpunk. Haven't tested the other games yet, my internet is not great and they'll take some time to download.

I don't know enough to advise on this, but I occasionally get CTD's, freeze then CTD or even just freeze and can't even get back to desktop in one particular game(the latter was an audio driver problem).

(Loading the game from a desktop shortcut has helped for me).

I'm sure this is common knowledge but a good place to start it type RELIABILITY into SEARCH>View Reliability History(and Problem History). Click on the Critical Event by date, then View Technical Details. It might give you an insight.

(So for me it says, 'game stopped interacting with Windows and was closed').

Maybe worth a go! And I empathise.

Thanks for joining in.
I did get an error in the reliability history:

Description
A problem with your hardware caused Windows to stop working correctly.

Problem signature
Problem Event Name: LiveKernelEvent
Code: 141
Parameter 1: ffff8007b57eb460
Parameter 2: fffff8015f33af20
Parameter 3: 0
Parameter 4: 625c
OS version: 10_0_19045
Service Pack: 0_0
Product: 256_1
OS Version: 10.0.19045.2.0.0.256.48
Locale ID: 2057

I'm searching for that error on Google and I find mostly fixes I've already tried: update GPU drivers, update BIOS, update Windows, check for corrupted files... Some articles also mention overclocking. Now I've never overclocked anything myself, but is it possible my GPU was "factory overclocked"? If that's the case, maybe I could try to underclock it back to its nominal value, although I have no idea what that would be and how to do it.

I didnt even know they had replaced Event viewer in Win 11! Think its still Event viewer in Windows 10 though afaik. Good idea though @ipman , sometimes it points to something.

@Scarecrow
I had one other thought @Scarecrow try unplugging all USB devices and see if that changes anything. If it helps plug them back in one by one until you find the culprit.

Doesn't hurt to try.

 

[Kaamos_Llama]

Unstable overclocking can also causes instability in RAM and CPU. The same kind of instability a slightly sketchy lower RAM timing setting might cause..

If your GPU was causing this I'm pretty sure the errors would show up a GPU driver crashes. Display driver nvlddmkm stopped responding etc. If there's none of those I dont think its the GPU.

I'd suggest ordering a RAM kit from maybe Amazon thats definitely on the motherboard QVL compatibility list. If it solves the problem great, if not you can return the kit and I'm out of ideas, sorry to say.

 

[ipman]

Yes at least having that Code 141 gives you focus. Just looking at a few of the suggested fixes.
Have you tried Windows Memory Diagnostic, run as administrator, that seems to run a diagnostic on RAM and tell you if modules need replacing.

It also suggests Command Prompt, run as admin>sfc/scannow>

DISM /online / Cleanup-Image /ScanHealth

DISM /online /Cleanup-Image /RestoreHealth

It seems to suggest you can remove overclocking software(Is that right?)> Settings>Apps>uninstall

Disconnect all external devices, and when reconnecting each + clean reinstall drivers.

Disable (non Microsoft) Startup Programs, Restart.



Good Luck. I thought my PC had died recently. Somehow a connector on the MB had worked it's way loose(I wish it was always that simple).

But thank the PC gods because generally the answer is out there somewhere and it's satisfying to fix it.

 

[Kaamos_Llama]

One last thought that should have been one of the first @Scarecrow. You did clear CMOS at some point?

 

[Scarecrow]

Yes at least having that Code 141 gives you focus. Just looking at a few of the suggested fixes.
Have you tried Windows Memory Diagnostic, run as administrator, that seems to run a diagnostic on RAM and tell you if modules need replacing.

It also suggests Command Prompt, run as admin>sfc/scannow>

DISM /online / Cleanup-Image /ScanHealth

DISM /online /Cleanup-Image /RestoreHealth

It seems to suggest you can remove overclocking software(Is that right?)> Settings>Apps>uninstall

Disconnect all external devices, and when reconnecting each + clean reinstall drivers.

Disable (non Microsoft) Startup Programs, Restart.



Good Luck. I thought my PC had died recently. Somehow a connector on the MB had worked it's way loose(I wish it was always that simple).

But thank the PC gods because generally the answer is out there somewhere and it's satisfying to fix it.

Tried all except for the clean peripheral driver install, cause I can't find specific drivers for my keyboard and mouse (Logitech G513 and G402) on the Logitech website and I just reinstalled Windows, so the default drivers should be as fresh as can be, so I'm not sure what I'm supposed to do.

One last thought that should have been one of the first @Scarecrow. You did clear CMOS at some point?

I had not, but I just did and it's still crashing.

I also checked my clock speeds with CPU-Z and GPU-Z and neither seems to be overclocked.

Also did a chkdsk which came out clean.

I'm gonna try and get my hands on a friend's GPU just to completely eliminate the GPU angle and order some QVL approved RAM, even though I can't understand how I could have ran so many different memory stress tests without a single error if that was the culprit.

I'm also gonna make a post on the Windows forums about the KernelLiveEvent 141 error, maybe someone there will know more details about it.

I'm getting really defeated :/ really appreciate your help though

 

[Kaamos_Llama]

Tried all except for the clean peripheral driver install, cause I can't find specific drivers for my keyboard and mouse (Logitech G513 and G402) on the Logitech website and I just reinstalled Windows, so the default drivers should be as fresh as can be, so I'm not sure what I'm supposed to do.



I had not, but I just did and it's still crashing.

I also checked my clock speeds with CPU-Z and GPU-Z and neither seems to be overclocked.

Also did a chkdsk which came out clean.

I'm gonna try and get my hands on a friend's GPU just to completely eliminate the GPU angle and order some QVL approved RAM, even though I can't understand how I could have ran so many different memory stress tests without a single error if that was the culprit.

I'm also gonna make a post on the Windows forums about the KernelLiveEvent 141 error, maybe someone there will know more details about it.

I'm getting really defeated :/ really appreciate your help though

As I understand from lurking in RAM tweaking and overclocking threads over the years while dabbling, Memtest is good for finding if a RAM stick is faulty, but not very good at finding intermittent instability in timings.

Good luck!

 

[Scarecrow]

Hey, dropping by with an update for posterity's sake: turns out my GPU was the problem. The exact reason is still unclear, but according to the guy who helped me confirm it on Microsoft forums, there's nothing to be done.
With current GPU prices, I guess I'll have to make due with 2D games and League of Legends for a little while.

At least I learned a few things in the process, right..? Anyway, thank you all for your time, really appreciate it.

 

[Kaamos_Llama]

Guess my hunch was wrong, sorry if I led you down the wrong path. Hope that it gets it fixed for you!