Game Crash When Launching from Super Carrier in F/A-18

[Relics]

This has happened in the past to me but was fixed before. This crash related bug has returned. 

 

Has happened twice in 10 Cat Shots. Hookup to CAT 2. Launch off the deck. As I do my clearing turn the game completely crashes a few seconds after leaving the deck. Happens so often I started to test it with the F-14A&B to see if it happens. The F-14 has never crashed. Each time I have done this I was on a Multiplayer Server.

 

Here is the following crash dump from both incidents today. 
Crash #1
dcs.log-20210130-054454.zip
Crash #2

dcs.log-20210130-010204.zip

Edited January 30, 2021 by Hushpuppy

[jpinard]

Is it with the F-18A (not the other variants)?  Because I just discovered when the F-18A tries to retract its landing gear it crashes the game 100% of the time.  The other F-18's are fine.

 

 

Edited February 1, 2021 by jpinard

[Flappie]

No, his crash seems directly linked to the C variant, indeed.
 

2021-01-30 01:02:03.005 WARNING FLIGHT: carrier technician [1: white2] created, id 2
2021-01-30 01:02:03.005 WARNING FLIGHT: carrier technician [1: green_hbb] created, id 3
2021-01-30 01:02:03.005 WARNING FLIGHT: carrier technician [1: green_nw] created, id 1
2021-01-30 01:02:03.818 WARNING FLIGHT: procCrewAssignAircraft client id 16778262, crew_id 1, hot_start 0
2021-01-30 01:02:03.818 WARNING FLIGHT: wCarrierCrew::set_LA() assigned aircraft **ae) [CAP] F/A-18C (CVN-7X) #005 on runway 1
2021-01-30 01:02:04.525 INFO    EDCORE: try to write dump information
2021-01-30 01:02:04.530 INFO    EDCORE: # -------------- 20210130-010204 --------------
2021-01-30 01:02:04.533 INFO    EDCORE: DCS/2.5.6.60720 (x86_64; Windows NT 10.0.19042)
2021-01-30 01:02:04.536 INFO    EDCORE: C:\WINDOWS\System32\DriverStore\FileRepository\nv_dispi.inf_amd64_e44a509137af5639\nvwgf2umx.dll
2021-01-30 01:02:04.539 INFO    EDCORE: # C0000005 ACCESS_VIOLATION at CAA87790 00:00000000
2021-01-30 01:02:04.549 INFO    EDCORE: SymInit: Symbol-SearchPath: '.;F:\Program Files\Eagle Dynamics\DCS World OpenBeta;F:\Program Files\Eagle Dynamics\DCS World OpenBeta\bin;C:\WINDOWS;C:\WINDOWS\system32;SRV*C:\websymbols*http://msdl.microsoft.com/download/symbols;', symOptions: 530, UserName: 'traye'
2021-01-30 01:02:04.551 INFO    EDCORE: OS-Version: 10.0.19042 () 0x100-0x1
2021-01-30 01:02:05.259 INFO    EDCORE: 0x0000000000437790 (nvwgf2umx): NVDEV_Thunk + 0xBC850
2021-01-30 01:02:05.259 INFO    EDCORE: 0x000000000106912F (nvwgf2umx): cask::ShaderList<cask::SoftmaxShader,cask::SoftmaxOperation>::availableShaders + 0x2E2CF
2021-01-30 01:02:05.259 INFO    EDCORE: 0x00000000010B3FCF (nvwgf2umx): cask::ShaderList<cask::SoftmaxShader,cask::SoftmaxOperation>::availableShaders + 0x7916F
2021-01-30 01:02:05.259 INFO    EDCORE: 0x00000000010B3F4A (nvwgf2umx): cask::ShaderList<cask::SoftmaxShader,cask::SoftmaxOperation>::availableShaders + 0x790EA
2021-01-30 01:02:05.260 INFO    EDCORE: 0x00000000010A916D (nvwgf2umx): cask::ShaderList<cask::SoftmaxShader,cask::SoftmaxOperation>::availableShaders + 0x6E30D
2021-01-30 01:02:05.260 INFO    EDCORE: 0x000000000067F25A (nvwgf2umx): cask::SafeEnum<cask::GemmChip_ENUMCLASS_SCOPEWRAPPER>::SafeEnum<cask::GemmChip_ENUMCLASS_SCOPEWRAPPER> + 0x165CA
2021-01-30 01:02:05.260 INFO    EDCORE: 0x000000000163552C (nvwgf2umx): OpenAdapter12 + 0x3A84AC
2021-01-30 01:02:05.261 INFO    EDCORE: 0x0000000000017034 (KERNEL32): BaseThreadInitThunk + 0x14
2021-01-30 01:02:05.261 INFO    EDCORE: 0x000000000004D0D1 (ntdll): RtlUserThreadStart + 0x21
2021-01-30 01:02:08.451 INFO    EDCORE: Minidump created.

 

 It looks like @-Quetzalcoatl-'s buddy's crash ("SoftmaxOperation"):

 

Do you happen to have an MSI Bazooka B450M motherboard?

Edited February 8, 2021 by Flappie

[Flappie]

Hi @Hushpuppy. Did you get rid of this bug or not? We have a new identical case here. I'm looking for solutions.

[Relics]

I seemed to have fixed the issue. I turned on Auto Page Filing and it managed to do ok. But the moment I turn it off I still get the issue. Again is only when you bring the gear up. NineLine told me about the Page File solution for a few issues and it works but you'll need to have space on your hard drive for it.

[Flappie]

Great news. You mean you ticked "Automatically manage paging file size for all drives"?

[Relics]

Yah, because any other way will cause a crash

[Flappie]

Interesting... it's the first time I hear about it. Usually, it's the other way around.

Thanks for your feedback.

[Relics]

Welp sad to say the page file fix only worked for a little while. The bug has now continued past the page file fix solution. I had been flying for 3 hours and did several launches prior to this and then crash on a fresh multiplayer restart on the syria map. Same conditions.

 

 dcs.log-20210314-064724.zip

[eLgargantuan]

This apparently happened to me as well this evening. 

dcs.log-20210314-041913.zip

[robgraham]

I haven't had this one happen, but this looks like 2 issues based on the logs I've just looked at 

 

Hushpuppy's appears to be a Nvidia driver crash with a shader command as can be seen here

2021-01-30 01:02:04.525 INFO    EDCORE: try to write dump information
2021-01-30 01:02:04.530 INFO    EDCORE: # -------------- 20210130-010204 --------------
2021-01-30 01:02:04.533 INFO    EDCORE: DCS/2.5.6.60720 (x86_64; Windows NT 10.0.19042)
2021-01-30 01:02:04.536 INFO    EDCORE: C:\WINDOWS\System32\DriverStore\FileRepository\nv_dispi.inf_amd64_e44a509137af5639\nvwgf2umx.dll
2021-01-30 01:02:04.539 INFO    EDCORE: # C0000005 ACCESS_VIOLATION at CAA87790 00:00000000
2021-01-30 01:02:04.549 INFO    EDCORE: SymInit: Symbol-SearchPath: '.;F:\Program Files\Eagle Dynamics\DCS World OpenBeta;F:\Program Files\Eagle Dynamics\DCS World OpenBeta\bin;C:\WINDOWS;C:\WINDOWS\system32;SRV*C:\websymbols*http://msdl.microsoft.com/download/symbols;', symOptions: 530, UserName: 'traye'
2021-01-30 01:02:04.551 INFO    EDCORE: OS-Version: 10.0.19042 () 0x100-0x1
2021-01-30 01:02:05.259 INFO    EDCORE: 0x0000000000437790 (nvwgf2umx): NVDEV_Thunk + 0xBC850
2021-01-30 01:02:05.259 INFO    EDCORE: 0x000000000106912F (nvwgf2umx): cask::ShaderList<cask::SoftmaxShader,cask::SoftmaxOperation>::availableShaders + 0x2E2CF
2021-01-30 01:02:05.259 INFO    EDCORE: 0x00000000010B3FCF (nvwgf2umx): cask::ShaderList<cask::SoftmaxShader,cask::SoftmaxOperation>::availableShaders + 0x7916F
2021-01-30 01:02:05.259 INFO    EDCORE: 0x00000000010B3F4A (nvwgf2umx): cask::ShaderList<cask::SoftmaxShader,cask::SoftmaxOperation>::availableShaders + 0x790EA
2021-01-30 01:02:05.260 INFO    EDCORE: 0x00000000010A916D (nvwgf2umx): cask::ShaderList<cask::SoftmaxShader,cask::SoftmaxOperation>::availableShaders + 0x6E30D
2021-01-30 01:02:05.260 INFO    EDCORE: 0x000000000067F25A (nvwgf2umx): cask::SafeEnum<cask::GemmChip_ENUMCLASS_SCOPEWRAPPER>::SafeEnum<cask::GemmChip_ENUMCLASS_SCOPEWRAPPER> + 0x165CA
2021-01-30 01:02:05.260 INFO    EDCORE: 0x000000000163552C (nvwgf2umx): OpenAdapter12 + 0x3A84AC
2021-01-30 01:02:05.261 INFO    EDCORE: 0x0000000000017034 (KERNEL32): BaseThreadInitThunk + 0x14
2021-01-30 01:02:05.261 INFO    EDCORE: 0x000000000004D0D1 (ntdll): RtlUserThreadStart + 0x21
2021-01-30 01:02:08.451 INFO    EDCORE: Minidump created.
2021-01-30 01:02:08.464 INFO    DCS: enumerating loaded modules

while the last one by eLgargantuan appears to be a DX 11 crash caused possibly when a shader was called. 

 

 

2021-03-14 04:19:12.688 INFO    EDCORE: # -------------- 20210314-041913 --------------
2021-03-14 04:19:12.689 INFO    EDCORE: DCS/2.5.6.61527 (x86_64; Windows NT 10.0.19041)
2021-03-14 04:19:12.690 INFO    EDCORE: C:\WINDOWS\SYSTEM32\d3d11.dll
2021-03-14 04:19:12.690 INFO    EDCORE: # C0000005 ACCESS_VIOLATION at 05F889C1 00:00000000
2021-03-14 04:19:12.702 INFO    EDCORE: SymInit: Symbol-SearchPath: '.;F:\Program Files\DCS World OpenBeta;F:\Program Files\DCS World OpenBeta\bin;C:\WINDOWS;C:\WINDOWS\system32;SRV*C:\websymbols*http://msdl.microsoft.com/download/symbols;', symOptions: 530, UserName: 'Tom'
2021-03-14 04:19:12.702 INFO    EDCORE: OS-Version: 10.0.19041 () 0x300-0x1
2021-03-14 04:19:14.709 INFO    EDCORE: 0x00000000001289C1 (d3d11): CreateDirect3D11SurfaceFromDXGISurface + 0xF431
2021-03-14 04:19:14.710 INFO    EDCORE: 0x00000000000176BD (dx11backend): RenderAPI::DX11Renderer::setShaderResources + 0xCD
2021-03-14 04:19:14.710 INFO    EDCORE: 0x0000000000062ADD (dx11backend): createRenderer + 0x4AB8D
2021-03-14 04:19:14.710 INFO    EDCORE: 0x000000000006237D (dx11backend): createRenderer + 0x4A42D
2021-03-14 04:19:14.710 INFO    EDCORE: 0x000000000005FB59 (dx11backend): createRenderer + 0x47C09
2021-03-14 04:19:14.711 INFO    EDCORE: 0x0000000000011E84 (dx11backend): RenderAPI::DX11Renderer::draw + 0x364
2021-03-14 04:19:14.711 INFO    EDCORE: 0x0000000000026F1F (NGModel): model::load_json_mat + 0x56FF
2021-03-14 04:19:14.711 INFO    EDCORE: 0x000000000002A09E (NGModel): model::DeckMaterial::draw + 0x314E
2021-03-14 04:19:14.711 INFO    EDCORE: 0x000000000002A397 (NGModel): model::ModelMaterial::draw + 0x2D7
2021-03-14 04:19:14.712 INFO    EDCORE: 0x0000000000043FDB (NGModel): model::NGModelLods::reload + 0x21B
2021-03-14 04:19:14.712 INFO    EDCORE: 0x00000000001214C7 (Visualizer): RenderParserImpl::DrawZWrited + 0xA7
2021-03-14 04:19:14.712 INFO    EDCORE: 0x000000000012616E (Visualizer): RenderParserImpl::isEmpty + 0x420E
2021-03-14 04:19:14.712 INFO    EDCORE: 0x0000000000125A6D (Visualizer): RenderParserImpl::isEmpty + 0x3B0D
2021-03-14 04:19:14.712 INFO    EDCORE: 0x00000000001697E2 (Visualizer): smSceneManager::regLua + 0x25C2
2021-03-14 04:19:14.712 INFO    EDCORE: 0x000000000015A7EE (Visualizer): smSceneManager::CreateSceneManager + 0x614E

 

 

Now to ask the normal set of questions have you all done the following:

 

1. Deleted Metashaders and FXO in your savedgames folder and allowed DCS to rebuild those files?
2. Done a repair of DCS to make certain nothing has become corrupted? (and if it does repair .. do step 1 again) 

3. Updated from the late Dec/Jan/early Feb Nvidia Drivers? as there was a fault in these drivers impacting multiple games?

4. Tried reducing Shadows or other graphic items to see if that has an impact on the crash itself. 

 

That lets you start to see if any of those items are the causes of the issues, the fact that page file seemed to fix it for a short period of time could indicate a memory issue especailly given it's at the SC which with shadows on high tends to be really GPU intensive.. but you've 2 different cards there a 6gb 1060 and an 8gb 1080.

 

But I'd try those things and see if any of them have an impact..