Jump to content

F1 2019 - DX12 Stability Issues

Recommended Posts

Seems like I'm having similar issues to others here.

I'm not sure when the game became unstable as I hadn't played for a while but at the moment I can sometimes play for an hour or more with no problems then other times like tonight I play for less than half an hour before I get the ego dump crash. 

I've already tried disabling the shader cache which I thought had worked as I had no crashes in a 1.5 hour session yesterday but alas, it was not to be...😢

I've attached the relevant files here and well, I'm really hoping someone can help! 

Thanks 

info.txt

info.txt

DxDiag.txt

Edited by TheRichie2k19

Share this post


Link to post
Share on other sites
On 11/16/2019 at 12:47 AM, TheRichie2k19 said:

Seems like I'm having similar issues to others here.

I'm not sure when the game became unstable as I hadn't played for a while but at the moment I can sometimes play for an hour or more with no problems then other times like tonight I play for less than half an hour before I get the ego dump crash. 

I've already tried disabling the shader cache which I thought had worked as I had no crashes in a 1.5 hour session yesterday but alas, it was not to be...😢

I've attached the relevant files here and well, I'm really hoping someone can help! 

Thanks 

info.txt

info.txt

DxDiag.txt

Post the make and model of your motherboard.

Update Nvidia drivers by DDU and then install latest.

Try troubleshooting without the Webcam and the XBox controller attached.

Share this post


Link to post
Share on other sites

Evenin' Steve,

Thanks for the response.

My Motherboard is ASUS Strix B350-F Gaming. 

The only drivers I've updated since recently resetting windows are my AMD chipset drivers and the latest NVIDIA driver (using DDU for clean install as you suggest).

The game works fine (for 3 days of a couple of hour sessions so far) in DX11.

Strangely enough a Ubisoft game 'The division 2' destroyed my system stability due to something to do with DX12 too hence my recent windows 10 reset. Other DX12 games 'Control' and 'Shadow of the Tomb Raider' seem to be fine.

Can I ask why you think the webcam/controller may be an issue?

Share this post


Link to post
Share on other sites
9 hours ago, TheRichie2k19 said:

Evenin' Steve,

Thanks for the response.

My Motherboard is ASUS Strix B350-F Gaming. 

The only drivers I've updated since recently resetting windows are my AMD chipset drivers and the latest NVIDIA driver (using DDU for clean install as you suggest).

The game works fine (for 3 days of a couple of hour sessions so far) in DX11.

Strangely enough a Ubisoft game 'The division 2' destroyed my system stability due to something to do with DX12 too hence my recent windows 10 reset. Other DX12 games 'Control' and 'Shadow of the Tomb Raider' seem to be fine.

Can I ask why you think the webcam/controller may be an issue?

Controllers interfere with other controllers, the webcam is an audio source. Traditionally, these are things that tend to upset the game in some way or another.

Latest UEFI - these contain AGESA on AMD platform (basically underpins Zen performance / compatibility on AM4) - I see you do in fact have it already. You also have latest audio drivers.

I have no reason to believe you don't also have the latest chipset drivers if you assert that as everything else you say checks out.

I would go down the game uninstall / reinstall route.

If you're using any tweak / OC utility, again please try removal for troubleshooting and incidentally if you are using any RGB control / lighting software, please remove it temporarily also.

The memory QVL is here. Can't hurt to be on it. Try Memtest on RAM sticks individually.

That's about all I have. If the game continues to crash, then mail your files directly to CM. The more reports they get of genuine crashes the better.

Edited by steviejay69

Share this post


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

Controllers interfere with other controllers, the webcam is an audio source. Traditionally, these are things that tend to upset the game in some way or another.

Latest UEFI - these contain AGESA on AMD platform (basically underpins Zen performance / compatibility on AM4) - I see you do in fact have it already. You also have latest audio drivers.

I have no reason to believe you don't also have the latest chipset drivers if you assert that as everything else you say checks out.

I would go down the game uninstall / reinstall route.

If you're using any tweak / OC utility, again please try removal for troubleshooting and incidentally if you are using any RGB control / lighting software, please remove it temporarily also.

The memory QVL is here. Can't hurt to be on it. Try Memtest on RAM sticks individually.

That's about all I have. If the game continues to crash, then mail your files directly to CM. The more reports they get of genuine crashes the better.

Thanks Steve,

Thanks for the advice. To be honest I think I've already done much of it however. Since the reset I made sure drivers were indeed up to date and I haven't reinstalled MSI afterburner (which I only ever used for monitoring anyway) or Asus AURA or any other non essential apps etc. 

Interesting point re the controller/webcam though I haven't had any issue with either peripheral whilst playing this/any other game before.

With regards to RAM..As I said earlier I was having severe problems with 'The Division 2' in DX12 too (I haven't reinstalled that yet to try Dx11 as it completely BORKED my system). It was because of the problems with that game that I did the following the steps prior to my system reset;

*DDU nvidia driver and reinstalled latest version

*DDU nvidia driver again and rolled back to a previously known stable version

Reseated RAM & GPU

Swapped out PCIE PSU Cable

*Completed 4 passes @ around 3 hours of Memtest86 and 2 runs of windows 10 memory diagnostic with no errors (I didn't test 1 stick at a time but I assume an error would still be flagged, I just wouldn't know which stick/slot was the problem). My current RAM modules are not on the QVL buuuut RAM support was greatly improved between Ryzen 1000 and 3000 series CPU's and their BIOS revisions and the QVL hasn't been updated in a long time. I have bought another RAM kit that is on the QVL but I'm reluctant to swap them out as I will lose a little performance and as only these two games in DX12 seem to have problems. 

Would I be right in thinking that if a game is unstable in DX12 but fine in DX11 and all other games seem to be ok that my RAM is most likely ok?

AS you've suggested I've emailed all the  crash dumps I have along with DXDIAG info to CM and they've said they're looking into it whilst being happy that I can play the game without problems in DX11 😏

Apologies for the long post, just wanted to be as detailed as possible

Share this post


Link to post
Share on other sites
18 hours ago, TheRichie2k19 said:

Thanks Steve,

Thanks for the advice. To be honest I think I've already done much of it however. Since the reset I made sure drivers were indeed up to date and I haven't reinstalled MSI afterburner (which I only ever used for monitoring anyway) or Asus AURA or any other non essential apps etc. 

Interesting point re the controller/webcam though I haven't had any issue with either peripheral whilst playing this/any other game before.

With regards to RAM..As I said earlier I was having severe problems with 'The Division 2' in DX12 too (I haven't reinstalled that yet to try Dx11 as it completely BORKED my system). It was because of the problems with that game that I did the following the steps prior to my system reset;

*DDU nvidia driver and reinstalled latest version

*DDU nvidia driver again and rolled back to a previously known stable version

Reseated RAM & GPU

Swapped out PCIE PSU Cable

*Completed 4 passes @ around 3 hours of Memtest86 and 2 runs of windows 10 memory diagnostic with no errors (I didn't test 1 stick at a time but I assume an error would still be flagged, I just wouldn't know which stick/slot was the problem). My current RAM modules are not on the QVL buuuut RAM support was greatly improved between Ryzen 1000 and 3000 series CPU's and their BIOS revisions and the QVL hasn't been updated in a long time. I have bought another RAM kit that is on the QVL but I'm reluctant to swap them out as I will lose a little performance and as only these two games in DX12 seem to have problems. 

Would I be right in thinking that if a game is unstable in DX12 but fine in DX11 and all other games seem to be ok that my RAM is most likely ok?

AS you've suggested I've emailed all the  crash dumps I have along with DXDIAG info to CM and they've said they're looking into it whilst being happy that I can play the game without problems in DX11 😏

Apologies for the long post, just wanted to be as detailed as possible

Cheers mate, it's good to get such detailed feedback.

It's hard to predict how RAM errors will manifest themselves due to the way memory is loaded these days, depends which addresses (if any) are faulty.  Also it depends on whether it is a timing issue. Much RAM is sold as (e.g.) 3000MHz when in fact it is 2400 or 2666 RAM with XMP @3000. XMP is an Intel spec not AMD, that kind of RAM should be run at it's default speed not the XMP speed. 

As you say, the improvements to the platform have resolved many memory issues, but it's entirely possible that people have not updated their system through ignorance or "it works" (anecdotally with other titles that just haven't crashed 'yet').

Stuff on the QVL is stated to work, so you can take that as gospel (unless that memory tests faulty). 

Memtest can't always simulate every condition as it works on addressing how the chipset determines the memory is located. Using sticks singlely can speed up the elimination process. If a stick tests good (I would go for 4 passes), not only have you verified the stick you have verified the slot. You can use the verified stick to test the other slots and the verified slot to test the other stick. Then finally to be 100%, you can test the sticks together. Of course, if you have matched memory this is probably the law of diminishing returns. People will still use mismatched memory and as long as the machine POSTs then they're happy....even and especially matched memory with differing capacities can cause issues if the chipset is not configured to reject SS/DS pairings. It's usually buried in the manual they did not read.

So yeah, your hardware combo is not that uncommon - CM are aware of issues with Nvidia that have been addressed previously. I don't have AMD/Nvidia, I am Intel/AMD so it's hard for me to get upclose and personal with the actual hardware to find or recreate / disprove any issues. 

But generally, if the hardware is DX12 there is no reason that with 100% hardware and 100% configuration the game should work either way....unless CM just haven't found the DX12 bug....yet.

Share this post


Link to post
Share on other sites
27 minutes ago, BarryBL said:

Hi @Racinghero90,

I've merged your created thread as we already have a DX12 related topic active.

HDR output is not related to DX12 stability. Many of us do not have HDR capability.

Share this post


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

HDR output is not related to DX12 stability. Many of us do not have HDR capability.

Haha sorry all. (good spot @steviejay69)

Moved back out of thread.

  • Like 1

Share this post


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

Cheers mate, it's good to get such detailed feedback.

It's hard to predict how RAM errors will manifest themselves due to the way memory is loaded these days, depends which addresses (if any) are faulty.  Also it depends on whether it is a timing issue. Much RAM is sold as (e.g.) 3000MHz when in fact it is 2400 or 2666 RAM with XMP @3000. XMP is an Intel spec not AMD, that kind of RAM should be run at it's default speed not the XMP speed. 

As you say, the improvements to the platform have resolved many memory issues, but it's entirely possible that people have not updated their system through ignorance or "it works" (anecdotally with other titles that just haven't crashed 'yet').

Stuff on the QVL is stated to work, so you can take that as gospel (unless that memory tests faulty). 

Memtest can't always simulate every condition as it works on addressing how the chipset determines the memory is located. Using sticks singlely can speed up the elimination process. If a stick tests good (I would go for 4 passes), not only have you verified the stick you have verified the slot. You can use the verified stick to test the other slots and the verified slot to test the other stick. Then finally to be 100%, you can test the sticks together. Of course, if you have matched memory this is probably the law of diminishing returns. People will still use mismatched memory and as long as the machine POSTs then they're happy....even and especially matched memory with differing capacities can cause issues if the chipset is not configured to reject SS/DS pairings. It's usually buried in the manual they did not read.

So yeah, your hardware combo is not that uncommon - CM are aware of issues with Nvidia that have been addressed previously. I don't have AMD/Nvidia, I am Intel/AMD so it's hard for me to get upclose and personal with the actual hardware to find or recreate / disprove any issues. 

But generally, if the hardware is DX12 there is no reason that with 100% hardware and 100% configuration the game should work either way....unless CM just haven't found the DX12 bug....yet.

Much appreciated Steve,

I think since F1 2019 wasn't crashing for me in DX12 until recently, and as long as I can play it in DX11 I'll wait and see what happens. I cant really judge recent events on the instability I had with 'The division 2' as one look at Ubisoft forums show how much of a mess that game is! Bit by bit I'm reinstalling my game library and will put in as much testing time as my kids will allow, if something does go wrong again I'll switch out my Corsair LPX kit with the G-Skills I have here. I'd be surprised if this doesn't end up being some kind of driver/dx/windows issue or maybe even AMD AGESA related (similar to the recent destiny 2 problems)?

Whenever I read that people still chuck in mismatched ram modules I'm reminded of trouble I had with an earlier build when I found corsair would sell multiple ram kits with the same part number but different revisions that would cause all kinds of problems if thrown together. I learned that lesson HARD.

One more question if I may? If I do swap out my Corsair for Gskill ram will I have to reset my CMOS via the jumper/battery or is 'load optimised defaults' from within the BIOS sufficient?

 

Share this post


Link to post
Share on other sites
2 hours ago, TheRichie2k19 said:

Much appreciated Steve,

I think since F1 2019 wasn't crashing for me in DX12 until recently, and as long as I can play it in DX11 I'll wait and see what happens. I cant really judge recent events on the instability I had with 'The division 2' as one look at Ubisoft forums show how much of a mess that game is! Bit by bit I'm reinstalling my game library and will put in as much testing time as my kids will allow, if something does go wrong again I'll switch out my Corsair LPX kit with the G-Skills I have here. I'd be surprised if this doesn't end up being some kind of driver/dx/windows issue or maybe even AMD AGESA related (similar to the recent destiny 2 problems)?

Whenever I read that people still chuck in mismatched ram modules I'm reminded of trouble I had with an earlier build when I found corsair would sell multiple ram kits with the same part number but different revisions that would cause all kinds of problems if thrown together. I learned that lesson HARD.

One more question if I may? If I do swap out my Corsair for Gskill ram will I have to reset my CMOS via the jumper/battery or is 'load optimised defaults' from within the BIOS sufficient?

Never use 'load optimised defaults'. At best, you'll get 'compatibility mode'. At worst, the PC maybe won't boot!

Defaults will be when the power is removed and the CMOS battery jumper is shorted. The PC will start so you can configure it properly.

I would just set the memory to Auto/SPD. RAM will have a datasheet or you can read the SPD information from utilities like HWiNFO. There you can read the timing information and if SPD checks out as claimed then no further config is necessary. 

I hardly ever use the default settings in UEFI, there are things to turn on, off and configure optimally or preferentially.

Share this post


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

Never use 'load optimised defaults'. At best, you'll get 'compatibility mode'. At worst, the PC maybe won't boot!

Defaults will be when the power is removed and the CMOS battery jumper is shorted. The PC will start so you can configure it properly.

I would just set the memory to Auto/SPD. RAM will have a datasheet or you can read the SPD information from utilities like HWiNFO. There you can read the timing information and if SPD checks out as claimed then no further config is necessary. 

I hardly ever use the default settings in UEFI, there are things to turn on, off and configure optimally or preferentially.

Thanks Steve, lots of useful info there.

I'm running my RAM at it's DOCP (AMD's term for XMP) settings @ 3200Mhz, so yeah in CPUz/HWINFO and BIOS all timings check out. Great to be able to bounce info off someone so knowledgeable & helpful.

Hopefully, I wont need this spare ram kit anyway but time will tell.....I'll probably have a whole new exciting problem with F1 2020 :classic_biggrin: 

EDIT:

Ah....nevermind, after all these days the game just crashed in DX11 mode after about 1.5hrs of gameplay. Same behaviour as the crashes in DX12.

now I'm really stumped....and stressed :classic_blink:

Edited by TheRichie2k19
Game crashed in DX11

Share this post


Link to post
Share on other sites
21 hours ago, TheRichie2k19 said:

Thanks Steve, lots of useful info there.

I'm running my RAM at it's DOCP (AMD's term for XMP) settings @ 3200Mhz, so yeah in CPUz/HWINFO and BIOS all timings check out. Great to be able to bounce info off someone so knowledgeable & helpful.

Hopefully, I wont need this spare ram kit anyway but time will tell.....I'll probably have a whole new exciting problem with F1 2020 :classic_biggrin: 

EDIT:

Ah....nevermind, after all these days the game just crashed in DX11 mode after about 1.5hrs of gameplay. Same behaviour as the crashes in DX12.

now I'm really stumped....and stressed :classic_blink:

If you're feeling adventurous and wanna buy some MX4 you can always go on YT for a teardown of your GPU heatsinks and renew the TIM. Did mine and dropped temps 8-9C all round. OEMs use pads and they're rubbish. HWiNFO has a sensor log feature you can leave open while you play. Also, did you notice FPS drops any time before you crashed? Could be an indicator of throttling which is the usual precursor to overheating. But if you crash out and can play other games straightaway, it's not overheating.

Was watching an interesting video on Gamers Nexus about stuttering in RDR2, seems the more performance you have the game is worse. Something similar may be afflicting the EGO engine as most of the people with stutter or crashing are on higher end cards (usually nVidia) and higher end CPUs (usually Intel).

 

Share this post


Link to post
Share on other sites
On 11/22/2019 at 7:22 PM, steviejay69 said:

If you're feeling adventurous and wanna buy some MX4 you can always go on YT for a teardown of your GPU heatsinks and renew the TIM. Did mine and dropped temps 8-9C all round. OEMs use pads and they're rubbish. HWiNFO has a sensor log feature you can leave open while you play. Also, did you notice FPS drops any time before you crashed? Could be an indicator of throttling which is the usual precursor to overheating. But if you crash out and can play other games straightaway, it's not overheating.

Was watching an interesting video on Gamers Nexus about stuttering in RDR2, seems the more performance you have the game is worse. Something similar may be afflicting the EGO engine as most of the people with stutter or crashing are on higher end cards (usually nVidia) and higher end CPUs (usually Intel).

 

Evenin' Steve,

Sorry for the late reply, I had to take some time away from the PC or I think I might have taken a bite out of it by now.

It is funny and very true how you mention higher end setups having issues....my 1050ti and 1070 never have any problems.

My 2070 Super is an NVIDIA founders edition card which haven't had the best reputation so I've monitored it with HWinfo and afterburner for a while, though I removed afterburner as I've heard it's overlay can cause issues itself. I would crack it open and take a look at the pads etc but I've only had the card since Sept so if it is having a crisis I'll leave it to NVIDIA to sort it out. 

My system locked up briefly the other day and I found the attached errors/warnings in the event viewer (no thanks to NVIDIA). I've that I DDU'd the NVIDIA driver and reinstalled the latest one and my system seems fine. NVIDIA have gone from saying there doesn't seem to be anything wring with the card but RMA it anyway to those errors refer to the card so RMA it. I've asked them on multiple occasions if there's a benchmark I can use to prove that the card is OK or not and they ignore the question EVERY time.

I'm playing other really taxing games like 'control' and 'shadow of the tomb raider' with no issues and after already shelling out over £80 for another ram kit I don't want to let NVIDIA potentially replace this card with a refurb unless I KNOW it's borked. 

I realise I'm going off topic here and I apologise but is there any benchmarks/stress tests that would confirm that card condition? It's already passed multiple runs of timespy stress test temps are all within norms and I'm going to reinstall valley, heaven and superposition tonight.

With regards to your other question re stuttering before crashing...nope. No stuttering or artifacts so far.

Thanks Steve

 

 

 

Display Warnings both at 0749 this morning 22.1119.txtnvlddmkm error 22.11.19 0746.txtnvlddmkm error 22.11.19 0254.txt

 

Share this post


Link to post
Share on other sites
41 minutes ago, TheRichie2k19 said:

Evenin' Steve,

Sorry for the late reply, I had to take some time away from the PC or I think I might have taken a bite out of it by now.

It is funny and very true how you mention higher end setups having issues....my 1050ti and 1070 never have any problems.

My 2070 Super is an NVIDIA founders edition card which haven't had the best reputation so I've monitored it with HWinfo and afterburner for a while, though I removed afterburner as I've heard it's overlay can cause issues itself. I would crack it open and take a look at the pads etc but I've only had the card since Sept so if it is having a crisis I'll leave it to NVIDIA to sort it out. 

My system locked up briefly the other day and I found the attached errors/warnings in the event viewer (no thanks to NVIDIA). I've that I DDU'd the NVIDIA driver and reinstalled the latest one and my system seems fine. NVIDIA have gone from saying there doesn't seem to be anything wring with the card but RMA it anyway to those errors refer to the card so RMA it. I've asked them on multiple occasions if there's a benchmark I can use to prove that the card is OK or not and they ignore the question EVERY time.

I'm playing other really taxing games like 'control' and 'shadow of the tomb raider' with no issues and after already shelling out over £80 for another ram kit I don't want to let NVIDIA potentially replace this card with a refurb unless I KNOW it's borked. 

I realise I'm going off topic here and I apologise but is there any benchmarks/stress tests that would confirm that card condition? It's already passed multiple runs of timespy stress test temps are all within norms and I'm going to reinstall valley, heaven and superposition tonight.

With regards to your other question re stuttering before crashing...nope. No stuttering or artifacts so far.

Thanks Steve

 

 

 

Display Warnings both at 0749 this morning 22.1119.txtnvlddmkm error 22.11.19 0746.txtnvlddmkm error 22.11.19 0254.txt

 

I think if you can run TimeSpy with good temps there's nothing to worry about except maybe how long it is until the next CM patch and hope it sorts your issue 🙂

 

Try using the Frame Limiter and peg it somewhere around where the 1070 tops out (<=120FPS?)

Edited by steviejay69

Share this post


Link to post
Share on other sites
On 11/24/2019 at 9:55 PM, steviejay69 said:

I think if you can run TimeSpy with good temps there's nothing to worry about except maybe how long it is until the next CM patch and hope it sorts your issue 🙂

 

Try using the Frame Limiter and peg it somewhere around where the 1070 tops out (<=120FPS?)

Y'know I was just thinking that even though I use vsync @ 74hz maybe I should try capping the framerate too. I'll give it a shot this evening.

Cheers mate 👍

  • Agree 1

Share this post


Link to post
Share on other sites
3 hours ago, TheRichie2k19 said:

Y'know I was just thinking that even though I use vsync @ 74hz maybe I should try capping the framerate too. I'll give it a shot this evening.

Cheers mate 👍

If you use VSYNC@74Hz then that must be your display's max frequency, VSYNC will limit frames. You'll need a device you can run adaptive sync on that's good to 120Hz or better.

I would try turning the settings up so 75Hz is barely achievable, but I would have thought with that combo you'll get that easily. Not sure what to suggest really.

I have a 144Hz Freesync and even I can hit 144FPS with RX580 on a lot of tracks with High settings.

  • Agree 1

Share this post


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

If you use VSYNC@74Hz then that must be your display's max frequency, VSYNC will limit frames. You'll need a device you can run adaptive sync on that's good to 120Hz or better.

I would try turning the settings up so 75Hz is barely achievable, but I would have thought with that combo you'll get that easily. Not sure what to suggest really.

I have a 144Hz Freesync and even I can hit 144FPS with RX580 on a lot of tracks with High settings.

yeah i'm running a 2070 super on a 1080p/75hz display so I'm already running everything maxed out :classic_biggrin: without vsync I hit average 130-140 fps with everything cranked up. Who knows, maybe locking the framerate too will have some effect?  Remember the 2017/2018 games having those weird stutter issues when using vsync?

Incidentally I really should get a 1440p or 144hz monitor at some point 😫

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

×