Sapphire RX 6800 XT Pulse Crashing with Ray Tracing enabled?

[Drumroll]

Received the same (original) card back, apparently no issues. Result, same issue.
Reinstalled Windows, updated drivers and everything. Nothing else but the benchmarks to test and crash.
First tried the Port Royal, black screen, back to desktop.
Then tried the DirectX Raytraicing feature test, crash to desktop, but now gives an AMD timeout error. Some progress.
Will try to use the switch to the other BIOS and see if anything changes.
If no success, may try another PSU.

Same issue with second BIOS.
Tried to run only one RAM at a time, as that would rule out RAM issue. RAM running at 2400MHz, so no overclock/XMP profile. Same issue.
Guess is time to try another PSU and see how it behaves.

 

[eeii]

Can you go into windows event viewer and tell us what the error message is.

 

[Drumroll]

Will do this afternoon. Thank you for the guidance so far.

 

[Drumroll]

Fault bucket , type 0
Event Name: LiveKernelEvent
Response: Not available
Cab Id: 0

Problem signature:
P1: 141
P2: ffff89083e87f010
P3: fffff8056d1316f0
P4: 0
P5: 3bd4
P6: 10_0_22000
P7: 0_0
P8: 256_1
P9:
P10:

Attached files:
\\?\C:\WINDOWS\LiveKernelReports\WATCHDOG\WATCHDOG-20220505-2137.dmp
\\?\C:\WINDOWS\SystemTemp\WER-443046-0.sysdata.xml
\\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER.218a1680-607d-4660-b138-96b123474d88.tmp.WERInternalMetadata.xml
\\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER.af8c6e8e-4dea-43b9-8c54-eac1b2e25d7b.tmp.xml
\\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER.3d0b21a3-629d-4956-8db0-4788ae58dbc7.tmp.csv
\\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER.07d65650-3cb5-495a-bd06-874823621d43.tmp.txt

These files may be available here:
\\?\C:\ProgramData\Microsoft\Windows\WER\ReportQueue\Kernel_141_81bcfd5626bb1d37edd2267a3a3d4121b6794_00000000_cab_c9fb6fac-4e16-40ad-b6dc-b69c33656e7f

Analysis symbol:
Rechecking for solution: 0
Report Id: c9fb6fac-4e16-40ad-b6dc-b69c33656e7f
Report Status: 4
Hashed bucket:
Cab Guid: 0

Tried the oldest driver available for this card + Windows 11, also January's and using last driver, getting exactly the same error.
Also, the error only happens with Raytracing.

 

[ASMB]

I do remember awhile back that some people were experiencing issues with their AMD card not adjusting GPU frequency correctly and that it was best corrected by adjusting the gpu frequency setting manually in the AMD software. I don't know if this is related to your crashing in ray traced games but it might help. If you select a game in the AMD software, and then select performance it will give you the option to manually tune your card. If you enable the gpu tuning it will allow you to adjust the gpu frequency, perhaps try reducing this to 85% maximum.

 

[Drumroll]

I do remember awhile back that some people were experiencing issues with their AMD card not adjusting GPU frequency correctly and that it was best corrected by adjusting the gpu frequency setting manually in the AMD software. I don't know if this is related to your crashing in ray traced games but it might help. If you select a game in the AMD software, and then select performance it will give you the option to manually tune your card. If you enable the gpu tuning it will allow you to adjust the gpu frequency, perhaps try reducing this to 85% maximum.

Appreciate the advice. Will try this now.
Above someone suggested capping the card via driver.
Answering

He could halve the clocks in the drivers, which will pin the voltage down too.

I suppose it is possible the ray tracing hardware is broken (it has separate cores doesn't it?), but that seems unlikely?

Are you running any other features in the drivers? Does it run different apis, like vulcan?

Tried one demo running Vulcan, crashed, Ray Tracing via DirectX too.
The symptom now is slightly different. The screen goes black, as if the benchmark will start, the GPU’s fans spin up fast, about a second or two, still on black screen, then back to desktop.
AMD reports as driver time out, and Windows Event Viewer still reporting the same error code.
Will try to cap the GPU via driver and see how it goes.
I’m in the verge of getting rid of this card, carry on with the onboard and use the PS5 for now and wait next gen. Don’t see reason to buy another GPU now as still over expected prices from launching time.

 

[Drumroll]

RESULT!
I was wrong expecting that simply capping FPS in a game would limit the GPU on its auto boost.
The GPU boosts happily way higher than the advertised max frequency under normal usage.
Even capping the FPS, the GPU would, even for a fraction of time, boost to some point causing the crash. And the first boost was enough to crash my GPU here.
The crash only happened with Ray Tracing. For any other scenario, is solid. On auto, the GPU would spike past 2500MHz, then stay just under 2400MHz. Allowing some room for max frequency I've seen it boost past 2700MHz, then stay just under 2600MHz.
Went down the route of capping via driver, only the max frequency, as suggested by @Tetras and no crashes.

Just found very odd that under any scenario, except Ray Tracing, the GPU would be fine at such frequencies, but would crash only under Ray Tracing.
Not sure if the issue is a combination of various factors here, but unless I manage to get some time off to investigate, I'm just glad the feature works, as long as max frequency is dialed back.

Just in case someone comes across this, my specs:
Intel 12700K
MSI Z690 Tomahawk DDR4
Team Group 8Pack 32GB (4x8) DDR4 3200MHz CL14
Sapphire RX6800 XT Pulse
Seasonic Focus PX850W Platinum
Samsung 980Pro 2TB.
Windows 11
AMD 22.4.2
Plus few bits.

Thank you very much everyone!