• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

RX 9070 XT freezing/locking up only on desktop, anyone else?

I had the 5700 XT, the Gigabyte Aorus, it suffered from black screen issues and random software timeouts.
The weird part is it happened with my RTX 3070 too, it was the displayport cable I used at the time being too crappy for even 1080P 144hz.
Today I am on 1440P 180hz and have had the 6900 XT and 7900 XTX, rock solid stable.


Video on this.

 
Last edited:
RejZoR said:
I have this weird issue with it that I just can't seem to figure out and it's driving me crazy. I can game for hours from mild load of Overwatch 2 to heavy load of Oblivion Remastered or Doom The Dark Ages and it will never have any problems. But on desktop, what I usually do is have Youtube or other streaming video in PiP (Picture in Picture) and do other things like edit photos or browse web and suddenly, system would just lockup/freeze, sound would turn into static noise loop and this would go on for up to 30 seconds after which system unfreezes and everything is back in order like nothing happened. GPU crash detection is not triggered, there is nothing in Windows Event logs other than network reinitializing yellow exclamation trinagles after the lockup/freeze. I thought Firefox was causing this, but after switching to Vivaldi for several days and it just happened again. I just don't get it.

I've tested with everything stock, CPU, RAM, graphic card, still happening. Done memtest on RAM, all fine. I've removed my AE-9 dedicated soundcard and ran onboard audio for days and it still happened. I can't think of anything else being the issue other than graphic card which is the most "recent" addition and I've never had such issue with RTX 3080 prior (I've done clean system reinstall for Radeon). And the weird thing is, heavy load in games is not an issue where I could potentially blame PSU for it. But on desktop with minimal load. How does that make any sense?! I've also gone through several drivers, but I really can't go back to release drivers where I don't remember the issues, but they were bad for everything else, especially Radeon Chill and AFMF.

Any other idea because this is really stupid and it's annoying me so much.
Click to expand...
In my case old PC CPU 2500K and R9 380 Nitro, old legacy Adrenaline 22.6, but also with older windows driver from 2020. Locked sound and screen, green screen. 3-4 times in a month.
Hardware acceleration is enabled in the browser.
Only when watching youtube, not high loads.
I got notification in Adrenaline Watman something.

I clear GPU caches and and set HW info to log. All went away.
 
Turning off ULPS. I remember this thing causing problems on either HD6950 or HD7950 that I had years ago. Lets see...

There is one potential thing that I didn't even think about before, but a while ago I replaced my PSU PCIe power cables (after I already had RX 9070 XT). Thing is, the old ones I had, original ones that came with my Dark Power 11 Pro PSU were 2 cables per connector. My PSU has 4x PCIe power outputs and each can have output with 2 PCIe cables going directly from PSU connector (not daisy chaned at the end) so I basically had just 2 connected for 3 inputs on graphic card. I've replaced those with brand new ORIGINAL individually sleeved BeQuiet cables that have 1 PCIe cable per 1 power output. So now each cable that goes into graphic card comes from individual PCIe output on the PSU. But I find it hard to believe that brand new cables would be causing this. Not only connectors are new unlike my old ones that were replugged tons of times, each cable operating on individual rail should improve things and not make them worse by sharing them. But I honestly can't remember these issues when I had the old cables. Which would be super bizarre coincidence and timing, but I'm magnet for bizarre shit so it's a possibility. I'm so not looking forward to rewiring the whole thing with old fat stubborn cables...

UPDATE:
Holy s**t, disabling ULMP didn't go well at all. Marvel Rivals, even its launcher had just green rectangle instead of video with ULMP disabled. When I entered the game anyway, entire screen was green when splash videos were suppose to be and when it entered main menu, it was freezing like the issue I had on desktop, but each freeze took 1 second and worked for 2 seconds, freeze for 1 second and worked for 2 seconds. Enabled ULMP back and all that is normal again. WHAT THE ACTUAL F ?!
 
Last edited:
I've decided to go back to original PSU cables and it just locked up. So, the new individually sleeved cables from BeQuiet were not the issue.
 
RejZoR said:
I've decided to go back to original PSU cables and it just locked up. So, the new individually sleeved cables from BeQuiet were not the issue.
Click to expand...
Is the Be Quiet PSU ATX 3.0 or 3.1? As Wattage and reliability I see no issues.

Just a thought: it was mentioned somewhere in an article that older ATX 2.4 PSU can't normalize the current towards MBO and CPU after a GPU spike, producing a crash of application or GPU driver.
 
rusty caterpillar said:
Is the Be Quiet PSU ATX 3.0 or 3.1? As Wattage and reliability I see no issues.

Just a thought: it was mentioned somewhere in an article that older ATX 2.4 PSU can't normalize the current towards MBO and CPU after a GPU spike, producing a crash of application or GPU driver.
Click to expand...

Apparently it's ATX 2.4. Thing is, I didn't have any issues with it when I bought the RX 9070 XT, you'd think I'd have these issues from day 1 because I was using same programs that cause problems now even back then. I still have my RTX 3080 which I didn't sell yet so I might resort to that for testing if I run out of other ideas...

I'm now going to go totally out of the established presumption that Radeon is the problem here and try something with my RAM despite memtest86 giving it a clear result.

I've had 2x32GB RAM and added 2x32GB not long ago, I found identical RAM I already had so it's 4x identical sticks. Then I had weird total system crashes and found out 1 stick from the old original set was actually defective and memtest86 indicated that. G.Skill replaced the kit and sent me a brand new one fresh from the production line based on dates on it. However this kit is a bit more picky and wouldn't run at overclocked speeds or timings the old one did. Thinking back, I got the new kit around end of May 2025 and assuming I didn't work with tools that exhibit the problem, it might be acting up despite being ok with the memtest. Anyway, it's totally blind assumption I'm gonna try out. I've disabled DOCP and is now running at only 2166MHz instead of rated 3600MHz. To rule out any memory controller or RAM issues. Because when I was having those crashes because of RAM I couldn't find any logs ever or problematic drivers, it was just "Unexpected" error in the logs every time that was of no help to me. Lets see...

You got to be kidding me... Tested at rated DOCP settings. 2x sticks are basically new, bought in October 2024. I got new 2x from G.Skill directly in May 2025. Sigh. I've tested it and it was fine back then. Considering how it errored once and didn't in next 2 tests is fishy to me...
 

Attachments

  • 20250710_081915.jpg
    20250710_081915.jpg
    773.8 KB · Views: 40
Tested RAM sticks, shuffled them, tried the old and the new kit only, same issue.
 
Sounds more like a software issue, i would reinstall windows, not clean install, just a inplace upgrade from an iso or flash drive. Try with different drivers for graphics, audio and chipset.
 
Was there anything you did to the PC before this all started? Like an update, new software, new part, etc?
 
RejZoR said:
Apparently it's ATX 2.4. Thing is, I didn't have any issues with it when I bought the RX 9070 XT, you'd think I'd have these issues from day 1 because I was using same programs that cause problems now even back then. I still have my RTX 3080 which I didn't sell yet so I might resort to that for testing if I run out of other ideas...

I'm now going to go totally out of the established presumption that Radeon is the problem here and try something with my RAM despite memtest86 giving it a clear result.
Click to expand...
IMO is GPU, GPU driver or cable not the RAM, since you got it you probably installed newer drivers and that was changes in software. So revert driver to the first one you had initially.

If you did already this and you have same issues than send back the GPU.

Regarding the cable if 3080 works correctly with your cable doesn't mean 2.1 port on AMD will work, even that is back compatible, IMO.
 
At this point I'm suspecting NOTHING I've done is the culprit, I think Microsoft updated something in Windows 11 and fucked it all up (wouldn't be the first time). Because this is the only variable that's out of my control and it's just what it is.

The nature of the issue is so idiotic because it doesn't leave any trail of data behind. It's like person blacking out. The data is just missing for those few seconds system freezes and only thing that even notices is Battle.net that keeps reconnecting when it happens, the rest of system doesn't even react to it. If I have Event Viewer filled with errors, at least I can get info out of that. With these freezes, I have nothing.
 
RejZoR said:
At this point I'm suspecting NOTHING I've done is the culprit, I think Microsoft updated something in Windows 11 and fucked it all up (wouldn't be the first time). Because this is the only variable that's out of my control and it's just what it is.

The nature of the issue is so idiotic because it doesn't leave any trail of data behind. It's like person blacking out. The data is just missing for those few seconds system freezes and only thing that even notices is Battle.net that keeps reconnecting when it happens, the rest of system doesn't even react to it. If I have Event Viewer filled with errors, at least I can get info out of that. With these freezes, I have nothing.
Click to expand...
I know is a pain. Install a Win 10 or a Linux on a separate drive, see how it goes but pull out the win 11 one.
 
Freeze, hum, probably IRQ being used by 2 devices, otherwise you'd have something in the event viewer.
In some of the situations I know the problem could be in the sound driver, try reinstalling it.
Try checking it with LatencyMon, it will show you which device is faulty so you can reinstall it's drivers.
 
AVATARAT said:
Freeze, hum, probably IRQ being used by 2 devices, otherwise you'd have something in the event viewer.
In some of the situations I know the problem could be in the sound driver, try reinstalling it.
Try checking it with LatencyMon, it will show you which device is faulty so you can reinstall it's drivers.
Click to expand...
How?! I've not had IRQ issues for like over 20 years. How is this suddenly a thing again? Especially without me doing anything to the system.
 
RejZoR said:
How?! I've not had IRQ issues for like over 20 years. How is this suddenly a thing again? Especially without me doing anything to the system.
Click to expand...
Do you think your Windows doesn't use them or that they are gone?
And yes, during a crash/file restore for some reason the wrong IRQ may be associated, so when you reinstall the driver it will re-set the correct one. Similarly, the wrong service permission can be set during crash/repair, etc.

1752255664860.png
 
@AVATARAT
What I was saying is that I've not had issues with IRQ for like 20 years. Also for that to be consistently set wrong across TWO clean system wipes and countless driver reinstalls, I find that incredibly unlikely.

I'm currently running system with MPO disabled, but none of the IRQ's seem to be shared using this MSI mode utility. My Radeon has "Undefined" priority, yours seems to have "High". What's that for exactly?
 
Sorry, but I missed that you reinstalled.
OK :)
Yes, in some cases the MPO helps.

Have you tried LatencyMon?
ParkControl for stopping core parking can also help.

You can also try lowering the card frequency, set -500 with -20% power limit, if that helps then RAM or something bottlenecking.
 
AVATARAT said:
Sorry, but I missed that you reinstalled.
OK :)
Yes, in some cases the MPO helps.

Have you tried LatencyMon?
ParkControl for stopping core parking can also help.

You can also try lowering the card frequency, set -500 with -20% power limit, if that helps then RAM or something bottlenecking.
Click to expand...

From the looks of it, after testing bunch of stuff, it seems to be MPO thing. Since disabling it, I've not had any issues, but I need to be testing more time to be sure. Before, just using Firefox PiP with Youtube or any other video streaming and it would freeze sooner or later. This didn't happen entire afternoon yesterday. It happened at least once a day guaranteed before. If it won't happen entire today, it's this nonsense.

Stupid thing about it is that I didn't have this issue initially which means either Microsoft or AMD fucked it up sometime between launch of RX 9070 XT and this moment. And I'm suspecting it's Microsoft with one of their updates because AMD's drive from early May 2025 still has the issue and I know I didn't have issues back then. Because I was using Firefox in PiP for videos for last few years now and it would appear if the issue was there. I guess I'll have to bitch at AMD and they'll have to reach to Microsoft because reaching Microsoft is mission impossible...
 
Freezing is mostly hardware related. Did you try bios default and stress test?
 
texh said:
Freezing is mostly hardware related. Did you try bios default and stress test?
Click to expand...
Yes and yes. From the looks of it, it is in fact the MPO issue. After disabling MPO, not a single freeze entire evening yesterday and entire day today and I was doing the most problematic things again, image editing in Paint.NET and video streaming in Firefox using PiP. It would freeze up in this time for sure

What I don't understand is how I'm the only isolated case or people just don't have a clue how to troubleshoot this stupidity. Then again, how many people are out there who run Windows 11, with Radeon RX 9070 XT using Firefox and watching something in picture in picture function who sort of have idea and are trying to figure it out and then also report it to Microsoft/AMD ?
 
I just disabled MPO on my system as well - I had horrendous microstutter with the 9800x3d in games and I couldn't figure it out -- im on a 4090 btw. Since disabling stalker 2 no longer screenflashes and microstutters, and everything is running as it should.

Whats weird is it didn't do it on my intel board (Z890...) so i think it's some combination of MPO+ bios or mobo.
 
I've disabled MPO this morning, doing same problematic tasks as before and I've not had an issue so far. So, what the actual F ?! Has Microsoft updated something in between?!

EDIT:
Nope, system just froze up again...
 
Last edited:
RejZoR said:
I have this weird issue with it that I just can't seem to figure out and it's driving me crazy. I can game for hours from mild load of Overwatch 2 to heavy load of Oblivion Remastered or Doom The Dark Ages and it will never have any problems. But on desktop, what I usually do is have Youtube or other streaming video in PiP (Picture in Picture) and do other things like edit photos or browse web and suddenly, system would just lockup/freeze, sound would turn into static noise loop and this would go on for up to 30 seconds after which system unfreezes and everything is back in order like nothing happened. GPU crash detection is not triggered, there is nothing in Windows Event logs other than network reinitializing yellow exclamation trinagles after the lockup/freeze. I thought Firefox was causing this, but after switching to Vivaldi for several days and it just happened again. I just don't get it.

I've tested with everything stock, CPU, RAM, graphic card, still happening. Done memtest on RAM, all fine. I've removed my AE-9 dedicated soundcard and ran onboard audio for days and it still happened. I can't think of anything else being the issue other than graphic card which is the most "recent" addition and I've never had such issue with RTX 3080 prior (I've done clean system reinstall for Radeon). And the weird thing is, heavy load in games is not an issue where I could potentially blame PSU for it. But on desktop with minimal load. How does that make any sense?! I've also gone through several drivers, but I really can't go back to release drivers where I don't remember the issues, but they were bad for everything else, especially Radeon Chill and AFMF.

Any other idea because this is really stupid and it's annoying me so much.
Click to expand...
The symptoms you describe seems to be when the GPU is running 2D mode.

This really sounds like Cpu/Ram/Board issue, not concerning gaming or heavy tasks.

Like back when using shit memory (corsair lpx) with a 3600X Cpu on AM4. I believe a partial fix was to disable power down mode in the memory timings menu.
 
You must log in or register to reply here.
Back
Top