Warframe restarts PC

[Dragoban]

I recently returned to Warframe and I'm crashing on average once per hour, with the game causing my PC to restart once every 2-3 hours. PC spec and prior troubleshooting below:

Operating System: Windows 11 Home 64-bit (10.0, Build 22621)
Language: English (Regional Setting: English)
System Manufacturer: To Be Filled By O.E.M.
System Model: To Be Filled By O.E.M.
BIOS: P3.80 (type: UEFI)
Processor: AMD Ryzen 9 5900X 12-Core Processor (24 CPUs), ~3.7GHz
Memory: 32768MB RAM
Available OS Memory: 32690MB RAM
Page File: 8138MB used, 26599MB available
Windows Dir: C:\Windows
DirectX Version: DirectX 12

GPU: XFX Speedster MERC 319 AMD Radeon™ RX 6800 XT BLACK
PSU: SF Series™ SF750 — 750 Watt 80 PLUS® Platinum SFX PSU
MEM: HyperX Predator DDR4 RGB 32GB Kit 3200MHz CL16 DIMM XMP RAM

1. Lowered graphics settings
2. Verified game files. Reinstalled game. Installed game on a different drive.
3. GPU Bios and drivers are updated
4. Motherboard BIOS remains on P3.80 for reasons I cannot recall while typing this. Researching later
5. Computer has been dusted a few days ago
6. Fresh Windows install, no viruses. Malwarebytes, Bitdefender, and Windows Defender concur.
7. All Windows updates ran as of 10AM this morning (10 hours ago).
8. CPU averages around 70-75C, peaks at 80C. CPU max allowed temp is 90C. GPU averages around 72C, unknown peak, max allowed is 110C.
9. Ran CPU and GPU at max load for 3-4 hours to verify PSU is not failing, no crashing or errors.
10. Drive tests from morning of 04/09/23 and 04/16/23 each show healthy.
11. CPU/GPU run at base clock and mem runs at XMP. Underclocking CPU, GPU, and running JEDEC on RAM did not help the crashing.
12. Changing RAM configuration, IE removing or moving around DIMMs, did not help
13. Stress tests have been run on CPU (P95, OCCT, HeavyLoad, 3DMark), GPU (OCCT, Furmark, Kombustor, HeavyLoad, 3DMark), and RAM (P95, OCCT, HeavyLoad, MemTest86, HCIMemTest, y-cruncher) with no errors or crashing.
Side note: Upwards of 2.5TB-3.0TB of games only contain one other game with crashes that are known and recognized by its Devs (Division 2). Said game does not restart the entire PC.

Unfortunately, event viewer is not giving me any information beside the generic critical and error of:

• The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.
• The previous system shutdown at 12:55:00 PM on ‎4/‎16/‎2023 was unexpected.

I don't experience any performance issues aside from crashing. I can keep a solid 120FPS on 5120x1440 or unlock and reach a few hundred FPS. I only drop to say, 60-70, in areas such as the loading area between Necralisk and Cambion Drift. I do not use VSync because it sets FPS to 160 on my 120Hz display.

Any input would be appreciated so that I could continue playing.

[Hobie-wan]

You could try running GPU-Z and watch if temps are creeping up or if there are any other anomalies. You could also turn on logging and see if anything happens before the restart. Though I thought I saw some other threads about AMD issues recently that might have been related to a recent GPU driver version.

[Dragoban]

On 2023-04-19 at 3:56 PM, Hobie-wan said:

You could try running GPU-Z and watch if temps are creeping up or if there are any other anomalies. You could also turn on logging and see if anything happens before the restart. Though I thought I saw some other threads about AMD issues recently that might have been related to a recent GPU driver version.

Temps were stated in #8. I was on the last reportedly stable driver prior to updating post-crashes to the most recent driver. Keep in mind that I've multiple terabytes of games, and this is the only one with unexplained crashes.

[Nekomian]

Could you generate logs from the launcher (or go to %LOCALAPPDATA%/Warframe in file explorer) and see if the EE log has any useful info about the crash?

You've already done basically all the troubleshooting I would've suggested, so it must be a very specific issue with your hardware and this game, or with this game and some conflicting software or drivers. It reasonably should not ever restart your entire computer unless there's some critical failure (related to kernel or unrecoverable driver, or hardware issues), and software generally crashes within' its own scope so this is rather rare.

You may also need to contact support for how specific this is - they may have you run specific diagnostics or tools written in-house to detect what specifically is happening and debug it, or generate more verbose logs if the normal ones aren't assisting in any way.

[MillbrookWest]

On 2023-04-17 at 12:51 PM, Dragoban said:

Any input would be appreciated so that I could continue playing.

In general, a program can't outright knock a computer out. It's usually a hardware issue that does that, and a program simply exposes it.

As a shot in the dark, check your event viewer for any WHEA-Logger events (Windows Logs > System > Sort by 'source')

[Dragoban]

On 2023-04-24 at 10:36 AM, MillbrookWest said:

In general, a program can't outright knock a computer out. It's usually a hardware issue that does that, and a program simply exposes it.

As a shot in the dark, check your event viewer for any WHEA-Logger events (Windows Logs > System > Sort by 'source')

Please read the entire original post before replying, thank you. 

Quote

8. CPU averages around 70-75C, peaks at 80C. CPU max allowed temp is 90C. GPU averages around 72C, unknown peak, max allowed is 110C.
9. Ran CPU and GPU at max load for 3-4 hours to verify PSU is not failing, no crashing or errors.
10. Drive tests from morning of 04/09/23 and 04/16/23 each show healthy.
11. CPU/GPU run at base clock and mem runs at XMP. Underclocking CPU, GPU, and running JEDEC on RAM did not help the crashing.
12. Changing RAM configuration, IE removing or moving around DIMMs, did not help
13. Stress tests have been run on CPU (P95, OCCT, HeavyLoad, 3DMark), GPU (OCCT, Furmark, Kombustor, HeavyLoad, 3DMark), and RAM (P95, OCCT, HeavyLoad, MemTest86, HCIMemTest, y-cruncher) with no errors or crashing.
Side note: Upwards of 2.5TB-3.0TB of games only contain one other game with crashes that are known and recognized by its Devs (Division 2). Said game does not restart the entire PC.

Unfortunately, event viewer is not giving me any information beside the generic critical and error of:

• The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.
• The previous system shutdown at 12:55:00 PM on ‎4/‎16/‎2023 was unexpected.

Additionally, there are certainly others reporting it such as this post that's been ongoing since last year: 

 

Edited April 26, 2023 by Dragoban
Adding context / more info

[Dragoban]

On 2023-04-21 at 2:48 AM, Naroxas44 said:

Could you generate logs from the launcher (or go to %LOCALAPPDATA%/Warframe in file explorer) and see if the EE log has any useful info about the crash?

You've already done basically all the troubleshooting I would've suggested, so it must be a very specific issue with your hardware and this game, or with this game and some conflicting software or drivers. It reasonably should not ever restart your entire computer unless there's some critical failure (related to kernel or unrecoverable driver, or hardware issues), and software generally crashes within' its own scope so this is rather rare.

You may also need to contact support for how specific this is - they may have you run specific diagnostics or tools written in-house to detect what specifically is happening and debug it, or generate more verbose logs if the normal ones aren't assisting in any way.

I did send copies of the launcher diag files from before and after a crash to support. Their only response since was to retry shutting down background processes to see if they are interfering with warframe. I'm not sure what to look for in the files to be of any use given they're upwards of 9000 lines. 

[MillbrookWest]

20 hours ago, Dragoban said:

Please read the entire original post before replying, thank you. 

No reference to what i asked in OP.

20 hours ago, Dragoban said:

Additionally, there are certainly others reporting it such as this post that's been ongoing since last year: 

Ironically, that's the same thread i replied in saying the exact same thing. OP from that thread was unwilling to test with an FPS cap.

Also ironically, is that i said the same thing there as i did here; 'what's the event log say?' Specifically the WHEA-Logger as AMD will throw an error here if it's CPU. If it's not CPU, you can look elsewhere.

But, as i say, a program can't crash a system (see user-mode). It merely exposes an issue. It's why Cinebench (or whatever) will crash your system with a bad overclock. The program does not care, it did the same thing is both cases. It's that hardware that took issue with the task asked, and the strain given.

Edited April 27, 2023 by MillbrookWest

[Skyline159]

Hi, i'm having the same problem.. while playing warframe my pc just randomly hard restart ... this just happens on warframe and not in any other games..

I did the sames checks and tests as the owner of this topic and everything is ok. I can run other games and stream with no problem, can run stress tests and benchmarks no problems, but when playing warframe just randomly hard restart out of nowhere..

Specs:

• Lian Li O11D ROG XL Edition Black
• ASUS ROG Maximus XII Hero
• Intel Core i9 10900k
• Arctic Liquid Freezer II 360
• G.SKILL Trident Z RGB 32GB (4x8GB) DDR4-3200MHz CL14
• Asus ROG Strix GeForce RTX 3080 Ti 12GB GDDR6X OC Edition
• Lian Li UNI AL120 RGB PWM Preto 120mm x7
• Seasonic Focus PX-850W 80+ Platinum
• Samsung 970 Evo Plus 500GB MLC V-NAND NVMe
• Samsung 1TB EVO 860 Series 2.5 SSD SATA III
• Kingston 2TB SSD M.2 2280 KC2500 3D TLC NVMe

[Dragoban]

On 2023-04-27 at 7:56 AM, MillbrookWest said:

No reference to what i asked in OP.

You asked about errors. I stated the only two errors that occurred, neither of which are what you wanted. 

On 2023-04-27 at 7:56 AM, MillbrookWest said:

Ironically, that's the same thread i replied in saying the exact same thing. OP from that thread was unwilling to test with an FPS cap.

I always run a 120fps cap. 

On 2023-04-27 at 7:56 AM, MillbrookWest said:

Also ironically, is that i said the same thing there as i did here; 'what's the event log say?' Specifically the WHEA-Logger as AMD will throw an error here if it's CPU. If it's not CPU, you can look elsewhere.

Already stated there were no such errors. 

On 2023-04-27 at 7:56 AM, MillbrookWest said:

But, as i say, a program can't crash a system (see user-mode). It merely exposes an issue. It's why Cinebench (or whatever) will crash your system with a bad overclock. The program does not care, it did the same thing is both cases. It's that hardware that took issue with the task asked, and the strain given.

It's the game. No other game, stress test, or benchmark causes it. Hence why I'm simply going to play other more stable games such as Destiny 2 or Monster Hunter. 

On 2023-04-28 at 12:29 PM, Skyline159 said:

Hi, i'm having the same problem.. while playing warframe my pc just randomly hard restart ... this just happens on warframe and not in any other games..

I did the sames checks and tests as the owner of this topic and everything is ok. I can run other games and stream with no problem, can run stress tests and benchmarks no problems, but when playing warframe just randomly hard restart out of nowhere..

Specs:

• Lian Li O11D ROG XL Edition Black
• ASUS ROG Maximus XII Hero
• Intel Core i9 10900k
• Arctic Liquid Freezer II 360
• G.SKILL Trident Z RGB 32GB (4x8GB) DDR4-3200MHz CL14
• Asus ROG Strix GeForce RTX 3080 Ti 12GB GDDR6X OC Edition
• Lian Li UNI AL120 RGB PWM Preto 120mm x7
• Seasonic Focus PX-850W 80+ Platinum
• Samsung 970 Evo Plus 500GB MLC V-NAND NVMe
• Samsung 1TB EVO 860 Series 2.5 SSD SATA III
• Kingston 2TB SSD M.2 2280 KC2500 3D TLC NVMe

I recommend finding other games to replace warframe, DE support has been of no assistance in my support ticket. All they've done is told me to repeat the steps I stated were already taken in my initial contact. 

[Nekomian]

46 minutes ago, Dragoban said:

It's the game. No other game, stress test, or benchmark causes it. Hence why I'm simply going to play other more stable games such as Destiny 2 or Monster Hunter. 

I think the point Millbrook is making here, which I made as well, is that the program itself cannot restart the system, as it's run in user mode and does not have permissions to do so in any capacity. It's more semantics, but something related to the kernel level (either a driver interfacing with it or the hardware itself) is having some panic or hard crash that is unrecoverable, and Warframe specifically exposes whatever underlying issue is present while other software may or may not. Software in and of itself runs in a certain scope, and generally crashes without taking the OS or computer down with it. This is why I suggested support assist with specific troubleshooting tools because it's a combination of things causing this to happen, it's not as simple as "the game makes it crash restart" or tons more users would be having this issue.

This is difficult to troubleshoot because it's happening seemingly randomly across variations of hardware (intel & amd processors, amd & nvidia GPUs of all different models), but also sometimes not for users with the same hardware (i.e. users with exact same hardware listed aren't experiencing these issues too). Generally that means it's specific to that system then if only some users are having issues, as in a hardware or driver issue that only affects that user, otherwise everyone with Ryzen 9 5900X & Radeon RX 6800 XT BLACK setups would be hard crashing, and we'd be seeing a lot more reports from the player base (though tbf I don't know many people using those exact specs, but other users' systems with RTX cards are more common).

Anecdotally, I'm seeing a lot of recent postings about your specific GPU and the exact same issues (PC "shuts off") users are having: https://community.amd.com/t5/drivers-software/amd-radeon-6800xt-xfx-speedster-keeps-crashing-and-disabling/td-p/592747 & https://community.amd.com/t5/drivers-software/6800xt-black-screen-amp-drivers-crashing/m-p/602386#M173929 & https://www.reddit.com/r/buildapc/comments/11ouxmh/amd_radeon_6800xt_xfx_speedster_keeps_crashing/ .

This makes me think it could also possibly be just some weird defect in it that only Warframe triggers, but honestly I'd need more testing / information to know for certain. Please try to be patient with support and ask them if they can escalate it, and provide as much info to them as possible. They're probably just as frustrated as they don't want users to have these issues in the first place, but given the extensive testing and knowledge an agent would need for this issue it probably doesn't fall under the scope of work that most of them are trained to deal with, and they have to go through procedural steps and process of elimination based on their material.

[MillbrookWest]

15 hours ago, Naroxas44 said:

but something related to the kernel level (either a driver interfacing with it or the hardware itself) is having some panic or hard crash that is unrecoverable, and Warframe specifically exposes whatever underlying issue is present while other software may or may not. Software in and of itself runs in a certain scope, and generally crashes without taking the OS or computer down with it.

Pretty much this.

16 hours ago, Dragoban said:

You asked about errors. I stated the only two errors that occurred, neither of which are what you wanted. 

The event viewer is filled with errors, even for a brand new install. They aren't always generated close together either. The error i was asking after gets logged a little while after booting - hence the clarity (as i have no way of knowing what you looked for; the two listed are almost always visible on the same section and for people, that's usually enough). 

But as i say, a system dropping is a sign of a hardware issue of some sort; it's unlikely its a terminal issue [the hardware itself] as a terminal issue would affect more processes. More likely the drivers - These can be for your keyboard, mouse, sound etc,.

If you have a spare GPU laying around, try swapping it in and seeing if the issue goes away.

EDIT:

If swapping the GPU still does nothing, then start with the other things you plug-in, failing that try (if you even remember which) removing programs that asked for admin privileges (either to run, or when installing [e.g. Afterburner]).

Edited May 2, 2023 by MillbrookWest

[DimkaTsv]

В 01.05.2023 в 20:48, Dragoban сказал:

It's the game. No other game, stress test, or benchmark causes it. Hence why I'm simply going to play other more stable games such as Destiny 2 or Monster Hunter.

Warframe is surprisingly optimized game that can load system harder than most games will do. It also can detect instabilities when others can't do. Granted it isn't perfect and can crash without hardware issues (which happens for me occasionally on SP Circuit for some reason. But it isn't driver crash, just game errors with DXGI_DEVICE_HUNG)

And knowing that such thing as transient load power spike exists i would advice you to look onto power related issues. 

Any game related issue would've either crashed game or driver.  But if there is nothing going on with your system game cannot simply cause shutdown event. 

 

Just to emphasize my point.  Google for Nvidia "new world incident". Load caused system to shut down but bad GPU (RTX 3080 or 3090) power management overlayed on top and fried quite a number of people's cards. GPU AIB's and Nvidia are even rolled out new VBIOS versions with stricter limits to fight that.

Trying framerate cap is good point to start testing. 

Edited May 2, 2023 by DimkaTsv

[Dragoban]

 

18 hours ago, DimkaTsv said:

Warframe is surprisingly optimized game that can load system harder than most games will do. It also can detect instabilities when others can't do. Granted it isn't perfect and can crash without hardware issues (which happens for me occasionally on SP Circuit for some reason. But it isn't driver crash, just game errors with DXGI_DEVICE_HUNG)

And knowing that such thing as transient load power spike exists i would advice you to look onto power related issues. 

Any game related issue would've either crashed game or driver.  But if there is nothing going on with your system game cannot simply cause shutdown event. 

 

Just to emphasize my point.  Google for Nvidia "new world incident". Load caused system to shut down but bad GPU (RTX 3080 or 3090) power management overlayed on top and fried quite a number of people's cards. GPU AIB's and Nvidia are even rolled out new VBIOS versions with stricter limits to fight that.

Trying framerate cap is good point to start testing. 

Running the sys at max possible power load for 4 hours did not cause issue. This game doesn't run the PC remotely close to max load. I will guarantee you dont have a recommended troubleshooting step that hasn't already been taken. 

 

On 2023-05-02 at 5:17 AM, MillbrookWest said:

Pretty much this.

The event viewer is filled with errors, even for a brand new install. They aren't always generated close together either. The error i was asking after gets logged a little while after booting - hence the clarity (as i have no way of knowing what you looked for; the two listed are almost always visible on the same section and for people, that's usually enough). 

But as i say, a system dropping is a sign of a hardware issue of some sort; it's unlikely its a terminal issue [the hardware itself] as a terminal issue would affect more processes. More likely the drivers - These can be for your keyboard, mouse, sound etc,.

If you have a spare GPU laying around, try swapping it in and seeing if the issue goes away.

EDIT:

If swapping the GPU still does nothing, then start with the other things you plug-in, failing that try (if you even remember which) removing programs that asked for admin privileges (either to run, or when installing [e.g. Afterburner]).

Issue also occurs on an entirely separate PC (3060ti with either 12600k or 13600k, idr which.) Also tested in this sys with an rx560. 

 

On 2023-05-01 at 3:08 PM, Naroxas44 said:

I think the point Millbrook is making here, which I made as well, is that the program itself cannot restart the system, as it's run in user mode and does not have permissions to do so in any capacity. It's more semantics, but something related to the kernel level (either a driver interfacing with it or the hardware itself) is having some panic or hard crash that is unrecoverable, and Warframe specifically exposes whatever underlying issue is present while other software may or may not. Software in and of itself runs in a certain scope, and generally crashes without taking the OS or computer down with it. This is why I suggested support assist with specific troubleshooting tools because it's a combination of things causing this to happen, it's not as simple as "the game makes it crash restart" or tons more users would be having this issue.

This is difficult to troubleshoot because it's happening seemingly randomly across variations of hardware (intel & amd processors, amd & nvidia GPUs of all different models), but also sometimes not for users with the same hardware (i.e. users with exact same hardware listed aren't experiencing these issues too). Generally that means it's specific to that system then if only some users are having issues, as in a hardware or driver issue that only affects that user, otherwise everyone with Ryzen 9 5900X & Radeon RX 6800 XT BLACK setups would be hard crashing, and we'd be seeing a lot more reports from the player base (though tbf I don't know many people using those exact specs, but other users' systems with RTX cards are more common).

Anecdotally, I'm seeing a lot of recent postings about your specific GPU and the exact same issues (PC "shuts off") users are having: https://community.amd.com/t5/drivers-software/amd-radeon-6800xt-xfx-speedster-keeps-crashing-and-disabling/td-p/592747 & https://community.amd.com/t5/drivers-software/6800xt-black-screen-amp-drivers-crashing/m-p/602386#M173929 & https://www.reddit.com/r/buildapc/comments/11ouxmh/amd_radeon_6800xt_xfx_speedster_keeps_crashing/ .

This makes me think it could also possibly be just some weird defect in it that only Warframe triggers, but honestly I'd need more testing / information to know for certain. Please try to be patient with support and ask them if they can escalate it, and provide as much info to them as possible. They're probably just as frustrated as they don't want users to have these issues in the first place, but given the extensive testing and knowledge an agent would need for this issue it probably doesn't fall under the scope of work that most of them are trained to deal with, and they have to go through procedural steps and process of elimination based on their material.

Also crashing with rx560 and another rig containing either 12600k or 13600k and 3060ti. 

 

As stated earlier, I simply won't play the game. Between PC crashing even before Duviri and the mistake called Duviri, I've more reason to leave than stay. 

[Nekomian]

39 minutes ago, Dragoban said:

As stated earlier, I simply won't play the game. Between PC crashing even before Duviri and the mistake called Duviri, I've more reason to leave than stay. 

That's your decision to make then, but if you decide to stick with it, both the community here and DE's own support will still try to assist in any way we can. It's unfortunate that this issue seems extremely difficult to track down, I know it's rather frustrating so I completely understand if you'd rather spend time playing something else than dealing with the many hours of troubleshooting (on top of what you've already done). I do hope it's resolved eventually.

[DimkaTsv]

В 03.05.2023 в 18:10, Dragoban сказал:

Running the sys at max possible power load for 4 hours did not cause issue.

Load is not an issue. Transients are.

And Nvidia RTX 3000 series are known for terrible power consumption spikes. Even if your GPU is rated for 230W, power spikes can reach 500-550W. JUST for GPU!. It happens for very short time, and such spike won't EVER be reported to system, especially as all Nvidia power management is average based. But PSU's are definitely feeling it, and if power spike overwhelms capacitance - system will shut down. 

Not sure about RX 500 series though. But RX 6000 is not that much better than RTX 3000 in that matter (even though yet still better)

And all "constant max power load stress test" like Furmark + Prime95 - yes, it gives a lot of power load... But it is consistently high power load. So your transients won't ever show up. 

Moreover some PSU models are much more susceptible to power transients than others.

 

For some reason even running days Warframe doesn't shut down system for me, meaning there is something for you on hardware level. Because GAME SIMPLY DON'T HAVE ACCESS TO ABILITY TO SHUT DOWN YOUR PC. At least normal games don't and Warframe definitely wasn't hacked by someone who injected trojan to shut systems down (and if it was like that, issue would've been much more widespread)

[Huggyos]

Hi,
- Do you have a HDMI or other cable connected to your VGA that doesnt the monitors you used for the game?

- Did you try to force PCIe generation to set to gen1 in BIOS?

- Tried to lower the RAM frequency in BIOS?

If stress testing are ok, I really dont think so that would be a PSU issue.

[HunterBLacK90]

i have the same problem and i have nvidia and intel if you fix it give me a call

[Tesla_Reloaded]

8 минут назад, HunterBLacK90 сказал:

i have the same problem and i have nvidia and intel if you fix it give me a call

Your PC is likely overheating or having power issues.

[Nekomian]

22 hours ago, HunterBLacK90 said:

i have the same problem and i have nvidia and intel if you fix it give me a call

I've stated it earlier but - it's a bit more complex of an issue than that. Crash logging & system info would be very helpful here, as force PC restarts are generally not caused by applications but at the kernel level or lower.

[PerseusNZ]

I had this issue as well. 

I have a Nvidia RTX 3080. After a driver update I did the 'restart later' thing then started Warframe. Warframe refused to load telling me that it no longer supports DX10 and to buy a new graphics card. After restarting my system to get the updated drivers going properly, Warframe launched but re-downloaded all of the game files and reinstalled itself. 

Every time I launched it after that my PC rebooted just before Warframe got to the login screen. No warning messages, no prompts, just a straight reboot.

Fixed it my adjusting the graphics settings. It was on DirectX 11 and High Performance. Reset to DirectX 12 (Beta) and Let Windows Decide and the game launched with no crash.

EDIT: Ohhh and putting it back to DirectX 11 / High Performance causes the crash to come back.

Edited February 28 by PerseusNZ

[Nekomian]

@PerseusNZ This sounds like some sort of hardware or OS issue if any application is making your system restart, let alone suddenly failing to read / verify files. More specifically, do you get any stop code when the system reboots, or what does event viewer have logged regarding it? If it was an application fault I would expect it to crash just the app - the system doing so sounds indicative of an OS, driver, or hardware issue (like a failing component or PSU suddenly cutting out when insufficient power or unregulated current is provided).