Rx6950 problem or is it ?.

[AsRock]

Well it's been a on going issue, how ever since getting a higher wattage PSU seems like it's just one game that crashes, but not your normal CTD.

Game in question is Red Dead Online, not the typical 2 crashes what most get which are GFX err and the for on going #ffffffffffff error. This crash is green screen reboots under Vulcan and black screen reboots under DX12.

It's 99% chance it will happen with in 2 hours of playing the game, it's the only game that does it i have tried cinebench (both), furmark, OCCT even a combo of them with 0 problems.

W1zards mem test and the dos version too for 60 minutes, no issue.

removed overclocks used bios defaults slower ram settings too. all so tried each stick with the game different slots.

Games tried

All the AC2's (3 in all) AC4 ACO
FarCry 3, 4, 5
Death Stranding
Balders gate 3
Horizon Zero Dawn
Dying Light 1
Beyond two Souls


It don't matter what resolution RDO is or quality it will do it with in the 2 hours, tried the pipeline console command along with some .exe commands for it still happen.

Maybe some one has a idea as i am about to send the card in which i am not still 100% to blame, maybe some one has a AM5 system with the same card and the game could try if there is such a person.

Although i have yet to try it in the other system but it be a real pain, still hoping there is some thing else i could try.

 

[Toothless]

I'll be getting one here in the next week or so with my current config in the specs list. I can try to help out with testing on some of those games.

Which driver?

 

[Vya Domus]

Might be the game I remember RDR2 was crashing constantly exactly as you describe it for me back when I had a 1080, I forgot what I did to fix it sadly.

 

[AsRock]

I'll be getting one here in the next week or so with my current config in the specs list. I can try to help out with testing on some of those games.

Which driver?

My thinking it's more to do with AM5 and the RX6950, or even the newer AMD drivers which i have tried oldest to newest of a few days ago, but will be testing over the weekend.

Be interested still if you get any issue's with it though.

Might be the game I remember RDR2 was crashing constantly exactly as you describe it for me back when I had a 1080, I forgot what I did to fix it sadly.

GFX_err and #FFFFFFF the later being the one that's been around since launch. This problem is not those, it's hard reboots.

Yeah nothing else seems to do it even when bogging the system down with a bunch of apps to take the memory and vram.

 

[Space Lynx]

Games tried

All the AC2's (3 in all) AC4 ACO
FarCry 3, 4, 5
Death Stranding
Balders gate 3
Horizon Zero Dawn
Dying Light 1
Beyond two Souls

these games all work fine though?

 

[AsRock]

these games all work fine though?

Yep, might of been the odd crash or issue due to age of some. And i mean played them though 100% completed once again.

 

[Hardcore Games]

These days anything less than 1000W for a PSU just seems inadequate
reset windows can get ahead of system problems too

 

[Vya Domus]

These days anything less than 1000W for a PSU just seems inadequate
reset windows can get ahead of system problems too

I have a 750W PSU and with an overclocked 7900XT (500W max GPU power according to hwinfo) and Ryzen 7900 with PBO enabled (150W) I have no problems. 1000W is categorically not necessary even for the most power hungry systems out there.

 

[QuietBob]

Well it's been a on going issue, how ever since getting a higher wattage PSU seems like it's just one game that crashes,

Are you using the 750 W Prime with your Merc 319 6950XT? I think it may be cutting it too close, especially that you're seeing less frequent reboots now with a more powerful PSU. The Merc hasn't been tested on TPU, but one 6950XT AIB card spiked to 573 W. Your 7700X can draw 140 W with PBO, and we're already at 713 W. Try running the game with default GPU clocks and a 60 fps cap to see if the issue persists.

I have a 750W PSU and with an overclocked 7900XT (500W max GPU power according to hwinfo) and Ryzen 7900 with PBO enabled (150W) I have no problems

That GPU max power figure does not seem to be reported correctly. In games your reference 7900XT draws about 320 W stock, with spikes to 412 W. So in a worst case scenario your system probably consumes less than 600 W. You could easily verify that number with a watt meter.

That said, you're right in saying that 1000 W isn't a hard requirement for a modern system. My MBA 7900XTX rig is powered by a 650 W PSU and the highest total consumption I've recorded was around the 500 W mark.

 

[JrRacinFan]

@AsRock IM thinking driver immaturit. Have you tried DDU and clean install latest adrenalin? if not go for that and check for results there. Other than this, I have no other ideas.

 

[AsRock]

Are you using the 750 W Prime with your Merc 319 6950XT? I think it may be cutting it too close, especially that you're seeing less frequent reboots now with a more powerful PSU. The Merc hasn't been tested on TPU, but one 6950XT AIB card spiked to 573 W. Your 7700X can draw 140 W with PBO, and we're already at 713 W. Try running the game with default GPU clocks and a 60 fps cap to see if the issue persists.


That GPU max power figure does not seem to be reported correctly. In games your reference 7900XT draws about 320 W stock, with spikes to 412 W. So in a worst case scenario your system probably consumes less than 600 W. You could easily verify that number with a watt meter.

That said, you're right in saying that 1000 W isn't a hard requirement for a modern system. My MBA 7900XTX rig is powered by a 650 W PSU and the highest total consumption I've recorded was around the 500 W mark.

Same issue with my 750w and a new Corsair Shift 1000w. the game could be drawing as little is 200w and still do it.

@AsRock IM thinking driver immaturit. Have you tried DDU and clean install latest adrenalin? if not go for that and check for results there. Other than this, I have no other ideas.

Tried old and new drivers using DDU, even ones recommended by XFX. If it was as driver thing would it not be more likely effect other things too ?.

ASUS got a new mobo bios out in the last day or so was thinking about getting that but i bet it's just for the X3D chips that are on the way.

 

[AsRock]

Really started to think it's not the GPU

 

[tabascosauz]

Really started to think it's not the GPU

Did you get some VESA certified DP cables yet? When you DDU, are you going to safemode or just staying in Windows?

Had some annoying crashing every day/every other day in specifically 1 game (Sniper Elite 5) on the 7900XT, fixed by getting a new HX1000 to replace SF750, and using only 1.4 VESA certified DP cables. Unfortunately I didn't have time to test at the time which factor (PSU or cables) was specifically responsible for fixing the crashing. I didn't change drivers at the time, so I can say with certainty the crashing was solved. Wouldn't rule anything out though just because it's only happening in 1 game - none of the listed titles constitute that much GPU load at 4K if limited to 60Hz (43UN700).

How sure are you about stability on the 7700X's part? Touch Curve Optimizer at all? Event Viewer looks clean? Run some ycruncher?

How about mem stability on the 6000CL32 kit? Run some TM5?

 

[AusWolf]

If it's only one game where the errors happen, and the system is stable during literally all stability tests, then maybe it's the game, or the game together with your specific hardware combination?

 

[Shou Miko]

@AsRock are you using any extension or custom cables for your graphics card?

I got some nice pci-e 8pin and cpu extensions from Bitfenix I used up until I got my RTX 3090 last year and my Gigabyte Z590 Vision G I started to encounter shut downs and so wasn't any issue with Asus and MSI mobos even ref RX 6800 XT but after I dropped them I had no issues at all on my SeaSonic Prime 1.2KW.

I am currently using the AMD's Adrenalin 23.2.2 and newest bios on my Asus Prime X670-P WiFi and I got no stable issues so far. I am running my 7700 non X stock and undervolting my RX 6800 XT.

 

[AsRock]

Did you get some VESA certified DP cables yet? When you DDU, are you going to safemode or just staying in Windows?

Had some annoying crashing every day/every other day in specifically 1 game (Sniper Elite 5) on the 7900XT, fixed by getting a new HX1000 to replace SF750, and using only 1.4 VESA certified DP cables. Unfortunately I didn't have time to test at the time which factor (PSU or cables) was specifically responsible for fixing the crashing. I didn't change drivers at the time, so I can say with certainty the crashing was solved. Wouldn't rule anything out though just because it's only happening in 1 game - none of the listed titles constitute that much GPU load at 4K if limited to 60Hz (43UN700).

How sure are you about stability on the 7700X's part? Touch Curve Optimizer at all? Event Viewer looks clean? Run some ycruncher?

How about mem stability on the 6000CL32 kit? Run some TM5?

Using the Corsair Shift (1000w) right now, as i use a Receiver i have to use either 2 or 3 cables. i have tried it with a DP cable which would be directly from PC to monitor but to get sound all so used the HDMI cables.

But maybe the Corsair power supply is faulty too, as it's showing the same issue that my Seasonic 750w Plat did.

Resolution don't matter it just crashes sooner or later it will.

Ran the 7700X with defaults and lower power like setting the PPT to say a 100w makes no difference. Never tried ycruncher.

TM5 ?, i have ran heavy loads on ram though OCCT, W1zards memtest and the dos memtest64 without problems and the windows version ( HA). All so ran the memory at default as well (4800) no difference.

currently testing setting the PCI-e and other parts to Gen 3, was testing last night but one of the common errors of RDO popped up ( #FFFFFFFFFF) and thought uck it, it's bed time haha.

If it's only one game where the errors happen, and the system is stable during literally all stability tests, then maybe it's the game, or the game together with your specific hardware combination?

Thinking that too.

@AsRock are you using any extension or custom cables for your graphics card?

I got some nice pci-e 8pin and cpu extensions from Bitfenix I used up until I got my RTX 3090 last year and my Gigabyte Z590 Vision G I started to encounter shut downs and so wasn't any issue with Asus and MSI mobos even ref RX 6800 XT but after I dropped them I had no issues at all on my SeaSonic Prime 1.2KW.

I am currently using the AMD's Adrenalin 23.2.2 and newest bios on my Asus Prime X670-P WiFi and I got no stable issues so far. I am running my 7700 non X stock and undervolting my RX 6800 XT.

No, i don't agree doing it as it just makes another possible fail point.

That new bios i will give it a try but i have a feeling it's changes are for the newer chips only, ASUS are not all that clear on it.

 

[A Computer Guy]

Did you check windows event viewer for any issues leading up the game crash?

 

[AsRock]

No just the eventid 41, pretty much no data .

- <Event xmlns="

http://schemas.microsoft.com/win/2004/08/events/event

">
- <System>
<Provider Name="Microsoft-Windows-Kernel-Power" Guid="{331c3b3a-2005-44c2-ac5e-77220c37d6b4}" />
<EventID>41</EventID>
<Version>8</Version>
<Level>1</Level>
<Task>63</Task>
<Opcode>0</Opcode>
<Keywords>0x8000400000000002</Keywords>
<TimeCreated SystemTime="2023-02-28T06:34:25.1960701Z" />
<EventRecordID>29415</EventRecordID>
<Correlation />
<Execution ProcessID="4" ThreadID="8" />
<Channel>System</Channel>
<Computer>DESKTOP-</Computer>
<Security UserID="" />
</System>
- <EventData>
<Data Name="BugcheckCode">0</Data>
<Data Name="BugcheckParameter1">0x0</Data>
<Data Name="BugcheckParameter2">0x0</Data>
<Data Name="BugcheckParameter3">0x0</Data>
<Data Name="BugcheckParameter4">0x0</Data>
<Data Name="SleepInProgress">0</Data>
<Data Name="PowerButtonTimestamp">0</Data>
<Data Name="BootAppStatus">0</Data>
<Data Name="Checkpoint">0</Data>
<Data Name="ConnectedStandbyInProgress">false</Data>
<Data Name="SystemSleepTransitionsToOn">0</Data>
<Data Name="CsEntryScenarioInstanceId">0</Data>
<Data Name="BugcheckInfoFromEFI">false</Data>
<Data Name="CheckpointStatus">0</Data>
<Data Name="CsEntryScenarioInstanceIdV2">0</Data>
<Data Name="LongPowerButtonPressDetected">false</Data>

 

[Deleted member 185158]

Try a different set of memory. Same issue, RMA the cpu.

Event 41 is sudden loss of power, the system shuts off.

Probable 90%+, RMA the cpu and replace the memory.

This is the 3000 series rig correct? That's mainly having this problem? 2 rigs and I'm not sure which you're using...

 

[Vya Domus]

I really don't think it's worth trying to RMA anything for something that happens in just one game. I always had issues like this, one time Fortnite kept crashing and locking up my system for months, could never figure out why, then after an OS reinstall it no longer did.

 

[Shou Miko]

I really don't think it's worth trying to RMA anything for something that happens in just one game. I always had issues like this, one time Fortnite kept crashing and locking up my system for months, could never figure out why, then after an OS reinstall it no longer did.

Had the same issue with GTA V no matter if it was the Epic or Steam version and only that one game.

So I stopped using it and haven't tried it since.

 

[AsRock]

Try a different set of memory. Same issue, RMA the cpu.

Event 41 is sudden loss of power, the system shuts off.

Probable 90%+, RMA the cpu and replace the memory.

This is the 3000 series rig correct? That's mainly having this problem? 2 rigs and I'm not sure which you're using...

System never shuts off.

AM5

I really don't think it's worth trying to RMA anything for something that happens in just one game. I always had issues like this, one time Fortnite kept crashing and locking up my system for months, could never figure out why, then after an OS reinstall it no longer did.

Yeah that's a bit extreme even if it is the game i play the most.

Tried reinstalling, second time was without any windows updates going back a year or so.

 

[Chrispy_]

If you've had crashes due to insfuccient PSU Wattage in the past, there's a non-zero chance some of your libraries or system files are corrupted.

Have you run these two commands (or reinstalled Windows) since having an unstable system yet?

sfc /scannow​

dism /online /cleanup-image /restorehealth​

 

[AsRock]

If you've had crashes due to insfuccient PSU Wattage in the past, there's a non-zero chance some of your libraries or system files are corrupted.

Have you run these two commands (or reinstalled Windows) since having an unstable system yet?

sfc /scannow​

dism /online /cleanup-image /restorehealth​

Tried it and even reinstalled the OS twice.

 

[Vya Domus]

Try a heavier game than RDR2, maybe something with ray tracing, so you can rule out definitively that it's the game.