Jump to content
Bug Reporting - PLEASE FOLLOW RULES AND COMPLETE REPORTS Read more... ×

Crash to Desktop during solo races

Recommended Posts

Hi team and others,

 

Since patch 1.12 I'm seeing an awful lot of Crash to Desktop errors in F1 2020. I can run the game fine in multiplayer, practice and qualifying, but when I run a single player race, I'll get a CTD anywhere between round 2 - 19. I'm currently on the Hungaroring and have been unable to complete a full race despite attempting the race five times in total. I've run in 50% mode as well - same issue. 

I'm running a My Team Career. Not sure if that makes any difference. 

I've tried DX12, DX11 and to turn down some settings to ease any load issue I might be seeing, but nothing helps. My PC is not overclocked, not tweaked in any particular way and it is stable in every other application and game I'm running. 

 

hardware_settings_info.xml hardware_settings_config.xml

I've added the recent EGO dump, a DXDiag and my current hardware settings config. 

Please let me know what I can do to speed this along from my end. 

DxDiag.txt

egodumper.log


EDIT: Added Q/A:

1.       A detailed description of the issue. Please include any error codes here. - Crash to Desktop errors happen mid-race. Not in qualifying or practice, though. I can run the game fine in multiplayer, practice and qualifying, but when I run a single player race, I'll get a CTD anywhere between round 2 - 19. I'm currently on the Hungaroring and have been unable to complete a full race despite attempting the race five times in total. I've run in 50% mode as well - same issue. 

2.       Platform - PC
3.       What version of the game you are using (Shown on the start screen in the bottom left corner of the screen) 1.12.
4.       Game-mode? Single player, My Team mode.
5.       What are your replication numbers? Minimum test attempts are 4. Please also add EXACT replication steps for us to try too. 5. Not sure you guys can replicate. 
6.       What troubleshooting have you attempted? Please always try to attempt to fix the issue - Run DX 12, DX 11, Lower all graphics settings, run safemode. 
7.       What peripherals are you are using (gamepad, wheel make & model etc) - Logitech G29 Racing Wheel, Tobii Eyetracking 4C
8.       Any screenshots or video of the issue? If the video is long, please include a timestamp for when the issue appears. Cannot screenshot while CTD. Happens too fast, and Ego Dumper does not always trigger. 

Edited by JWDalsager

Share this post


Link to post
Share on other sites

I suggest you use a recommended driver rather than the latest optional.

20.9.1 (or the one I use, 20.5.1 this has proved to be stable on my old RX580 and my RX5700XT)

AMD chipset drivers are important as is the updated AGESA but I need the board make and model (some kind of ASUS B450/B550/X570)

The EGO dumper log file is less helpful than the actual crash dump.

Noticed and old driver for a TP-Link WiFi adapter in the dxdiag, try looking for an update or playing wired.

 

Share this post


Link to post
Share on other sites

you have to create a swap file and set it to at least 8 GB. The games from Codemasters want to move to a swap file. NO MATTER whether you have 37 GB of RAM.

Edited by MicMiller

Share this post


Link to post
Share on other sites
7 hours ago, MicMiller said:

you have to create a swap file and set it to at least 8 GB. The games from Codemasters want to move to a swap file. NO MATTER whether you have 37 GB of RAM.

If you read the dxdiag you will see that the swap file is sized at around 27GB, so that's not the issue. But I do question the use of such high amounts of RAM in general. No idea why you would need 32GB of RAM to game. Most games run in 8-16GB just fine and don't even use that much. In fact the whole question of swapfiles is just to leave Windows be.

Share this post


Link to post
Share on other sites

@steviejay69: Will look into the optional driver, of course. Reverting should be a simple fix. 

Chipset drivers are downloaded directly from AMD and updated as needed, so I don't think there's any issue there. The TP-Link WiFi is probably a good call. I believe it's running on Windows standard drivers right now. 

Will look into all of it to be absolutely sure.

 

I will ignore your questioning of my 32GB memory 😉 This is my personal/project computer which is used for gaming occasionally but for production efforts and video editing quite regularly. That memory comes in handy for many other things than gaming. I've never seen an excessive amount of memory being an issue.  

Share this post


Link to post
Share on other sites
34 minutes ago, JWDalsager said:

@steviejay69: Will look into the optional driver, of course. Reverting should be a simple fix. 

Chipset drivers are downloaded directly from AMD and updated as needed, so I don't think there's any issue there. The TP-Link WiFi is probably a good call. I believe it's running on Windows standard drivers right now. 

Will look into all of it to be absolutely sure.

 

I will ignore your questioning of my 32GB memory 😉 This is my personal/project computer which is used for gaming occasionally but for production efforts and video editing quite regularly. That memory comes in handy for many other things than gaming. I've never seen an excessive amount of memory being an issue.  

I was responding more to the person who suggested the page file. It is set correctly (automatic I believe). Certainly no-one should adjust it without good reason.

I'm not being critical of your choice of memory, merely stating that it is not required for games. But yes, with the default settings memory management is creating a very large swapfile, due to the way memory handling is with dealt with on Windows. I can certainly assert that the majority of telemetry received at Redmond will back up the fact that the majority of W10 devices will probably contain 4-8GB of RAM. 

Power users and gamers are simply a user subset. 🙂

I was around during Windows 98, yes the 512MB limit was an issue. With 32-bit OS (particularly XP) over 3GB was an issue. I have been around longer than most!

Edited by steviejay69

Share this post


Link to post
Share on other sites

Can you also add your in-game graphical settings to the report @JWDalsager. Maybe turning something like Async Compute off may help.

 

Share this post


Link to post
Share on other sites

@steviejay69: Oh, yes. I remember. Running more memory than 4GB on my 32-bit Windows never caused issues, but I was stupid enough at the time to do it because "future proofing". 

 

@BarryBL: I honestly thought this was in the config files attached? 

If not, here you go. As expected, everything is maxed out, and Async Compute is on. 

image.thumb.png.1b29f14c7cacabd0dd98c8ff32e0d38d.pngI'll try switching it off as well.

 

Ideally, I'll run each of these changes separately, then test game performance. Otherwise we don't really have an RCA. 

Edited by JWDalsager
  • Thanks 1

Share this post


Link to post
Share on other sites

Yeah turning some of these down might be the first options here. Start with Asynchronous Compute and then SSRT Shadows will help.

 

Share this post


Link to post
Share on other sites

@BarryBL: The odd thing is that I've already tried running most of these on medium'ish and disabled. SSRT Shadows caused issues in F1 2019 for me, but as soon as I turned it off here I got crashes in lap 1 instead of somewhere around lap 5-15.

 

Didn't touch Async Compute, though. Will try as last change apart from the driver etc. setup. 

Share this post


Link to post
Share on other sites

@BarryBL:

Alright. I've updated all drivers as recommended and set my AMD drivers to 20.9.1., which is the last official driver set.

I've also disabled Async Compute. None of this makes a difference, I'm afraid. I still get random crashes, even in practice laps where I wasn't seeing an issue earlier. 

 

This is the latest EGO Dump:

[28-10-2020 18:10:41] - ##################################################
[28-10-2020 18:10:41] - EGO dumper
[28-10-2020 18:10:41] - processId=18972, threadId=8324, exceptionPointers=000000000B16EE90, buildId=692646
[28-10-2020 18:10:41] - ##################################################
[28-10-2020 18:10:41] - Suspending threads of process 18972
[28-10-2020 18:10:41] - tempPath=C:\Users\jwdal\AppData\Local\Temp\
[28-10-2020 18:10:41] - dumpFilename = C:\Users\jwdal\AppData\Local\Temp\692646-20201028-181041.dmp
[28-10-2020 18:10:41] - offsetsFilename = C:\Users\jwdal\AppData\Local\Temp\692646-20201028-181041.off
 

I sent the report and went to the dumpfile location, but was unable to find said dump. All I can find is this one from two days ago. 

692646-20201026-222803.dmp

Share this post


Link to post
Share on other sites

OK, I tried out one more thing. Have yet to confirm, but I was suddenly able to complete a few practice laps without incident.

 

Do any of you guys have experience with Radeon Anti-Lag? It was enabled in my driver by default, so I tried to untick it. 

Share this post


Link to post
Share on other sites

So... Crashes came back. Tried disabling a variety of settings (Shadows, Async, etc.) and checking cooling etc. on my build. No change and no immediate issue. 

I managed to run a full race before this happened, and now I'm unable to run few laps in Monza.


EDIT: Removed a chunk of text here, as recent dump is in my newer post below. 
 

Edited by JWDalsager

Share this post


Link to post
Share on other sites

Alright. Most recent crash logs and files attached. I've tried lowering all settings one at a time, running in windowed mode, running in safe mode and switching between DirectX 11 and 12. I've even tried switching my G29 to a different USB port and removing my Eyetracker from the mix. 

Without any impact. Still getting these crashes roughly mid-race. I'm at a loss. 

 

692646-20201101-090212.dmp

692646-20201101-090212.off

dred.txt

hardware_settings_config.xml

hardware_settings_info.xml

info.txt

Share this post


Link to post
Share on other sites

As far as I can see you haven't responded to my request for information on the motherboard, nor updated those UEFI (AGESA) and drivers. Or maybe this is just a Vega GPU issue.

I have absolutely no technical issues with this game on RX580 or RX5700XT (despite me being apprehensive after the RX580 ran so well). Just gameplay bugs Codemasters!!!

Other things to consider are power draw (PSU) and RAM timings (again to do with the UEFI). But hey, I've only been building and maintaining PCs for about 30 years, so what do I know, right?

Edited by steviejay69

Share this post


Link to post
Share on other sites

@steviejay69

That is true, I haven't. It's an ASUS x470 Prime Pro running the most recent BIOS update as per 28/10/2020. AGESA is included. Chipset drivers updated. 

Total system power draw is 582 peak wattage on a 750W Corsair RMX power supply. Not saying I haven't considered it, but it being the culprit in this game and nothing else seems unlikely to me. 

If it's a VEGA issue I'm having trouble explaining why I'm seeing this in 1.12 but not in any previous version of the game.

I've run memory tests via both windows memory diagnostics and memtest and I've received no errors, nor do I have crashes in any other games and applications, so I'm sorry if I didn't bother replying to that bit of advice. To me it seemed irrelevant. 

What bugs me is that I'm running out of options. I've tried running low, medium and high settings. I've tried older graphics drivers. I've tried running my G29 and eyetracker in other USB ports. I've tried running without the eyetracker enabled. I've tried monitoring temperatures to check for spikes or clock throttling. I'm not seeing any sequences that I can replicate.

 

Edited by JWDalsager
  • Thanks 1

Share this post


Link to post
Share on other sites
2 minutes ago, JWDalsager said:

@steviejay69

That is true, I haven't. It's an ASUS x470 Prime Pro running the most recent BIOS update as per 28/10/2020. AGESA is included. Chipset drivers updated. 

Total system power draw is 582 peak wattage on a 750W Corsair RMX power supply. Not saying I haven't considered it, but it being the culprit in this game and nothing else seems unlikely to me. 

If it's a VEGA issue I'm having trouble explaining why I'm seeing this in 1.12 but not in any previous version of the game.

I've run memory tests via both windows memory diagnostics and memtest and I've received no errors, nor do I have crashes in any other games and applications, so I'm sorry if I didn't bother replying to that bit of advice. To me it seemed irrelevant. 

What bugs me is that I'm running out of options. I've tried running low, medium and high settings. I've tried older graphics drivers. I've tried running my G29 and eyetracker in other USB ports. I've tried running without the eyetracked enabled. I've tried monitoring temperatures to check for spikes or clock throttling. I'm not seeing any sequences that I can replicate.

 

OK, if you have done it then the confirmation helps. Squarely with CM to sort it then possibly, because I know the Vega is architecturally different to the older GCN and not a Navi card.

Remove Armoury Crate. DDU the GPU driver. Go with the recommended 20.9.1 (I think 20.4.2 was the last recommended before that if a rollback is necessary). Avoid using the ASMedia USB ports (uninstall any downloaded driver and let WU/Inbox driver install). Get the AMD chipset drivers from AMD, not ASUS (IDK where that build is from). 

The only other option is to nuke Windows 10 and go again but I doubt you'll have the 1909 media (I too am staying on 1909 'til the game's lifecycle is done).

Don't think there's much else your end, you are using GHub and as long as you're avoiding the ASMedia USB ports there should be no latency issues.

I would suggest letting the driver install vanilla and going with the highest settings in-game and the desktop refresh set as high as possible with the game set at the same refresh. Consider using FPS counter and capping if the framerate is soaring. Deleting the hardware_settings_config.xml should let everything redetect after the DDU and reinstall.

Other than that CM might want to suggest which features to start disabling on the GPU from there.

 

  • Agree 1

Share this post


Link to post
Share on other sites
2 hours ago, steviejay69 said:

OK, if you have done it then the confirmation helps. Squarely with CM to sort it then possibly, because I know the Vega is architecturally different to the older GCN and not a Navi card.

Remove Armoury Crate. DDU the GPU driver. Go with the recommended 20.9.1 (I think 20.4.2 was the last recommended before that if a rollback is necessary). Avoid using the ASMedia USB ports (uninstall any downloaded driver and let WU/Inbox driver install). Get the AMD chipset drivers from AMD, not ASUS (IDK where that build is from). 

The only other option is to nuke Windows 10 and go again but I doubt you'll have the 1909 media (I too am staying on 1909 'til the game's lifecycle is done).

Don't think there's much else your end, you are using GHub and as long as you're avoiding the ASMedia USB ports there should be no latency issues.

I would suggest letting the driver install vanilla and going with the highest settings in-game and the desktop refresh set as high as possible with the game set at the same refresh. Consider using FPS counter and capping if the framerate is soaring. Deleting the hardware_settings_config.xml should let everything redetect after the DDU and reinstall.

Other than that CM might want to suggest which features to start disabling on the GPU from there.

 

Thanks for the followup.

1. Have removed Armoury Crate. Didn't use it anyway. 

2. 20.9.1. is my current driver, so I'll keep that 🙂

3. I'll try removing the ASMedia drivers. 

4. Chipset was installed from AMD 🙂

Not seeing latency on my USB per sé, I think, but since I don't know if they're causing crashes specifically, it's hard to pinpoint, of course. 

Good suggestions. Will try them out. 

// 

  • Thanks 1

Share this post


Link to post
Share on other sites

Update:

I've removed Armoury Crate and ASMedia USB drivers. Ran the game with my Tobii 4C eyetracker disabled simply because I forgot about it. Accidentally removed my My Team campaign, so I had to start over...

And I've been able to run the entire Melbourne and Bahrain races with only one incident during Melbourne qualifying. So it's still unstable, but apparently much less so. Vietnam ran fine too.

Edited by JWDalsager
  • Like 1

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×