I can't stay in windows long enough crashes to no signal within mins keeping eye on overlay but nothing stands out.
Think it’s an RMA case unfortunately. What PSU are you using?
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
The Radeon RX9070XT / RX9070 Owners Thread
- Thread starter CAT-THE-FIFTH
- Start date
Think it’s an RMA case unfortunately. What PSU are you using?
EVGA SuperNOVA 750 G2, ive managed to stay in windows long enough to do occt, nothing stands out, no vram errors, nothing really
It’s rather odd, which is why I mentioned the PSU. You could possibly try doing the power test in OCCT and see if it trips the PSU. It could be faulty.
Faults that appear at different times usually point to a power/cable issue rather than a dead/dying card, but it’s never certain. But it’s rare that a dead/dying card would not post for a while and then suddenly handle VRAM tests.
Aside from the PSU possibly being faulty, I’d check the cables (especially if you’re using extensions). I had a build once that kept black screen restarting at random times. Turned out to be a poor connection to the cable extension. I stopped using extensions for that reason.
Remove all under volts, overclocks and reset bios.I can't stay in windows long enough crashes to no signal within mins keeping eye on overlay but nothing stands out.
Do a DDU run and reinstall drivers.
Then test without changing anything
ive checked a load of stuff, im wondering if something in background is causing this, so going through and uninstalling everything. strangely windows been an arse lately saying it need to restart to install an update, but fails to update, i have rufus on standby as may as well try a full wipe if nothing works and go from there, in 30yrs plus ive never had a faulty card, and boy ive had some cards
but will try that psu yesy to see if throws anything up, but its stable atm, which suggests something other than gpu tbhive checked a load of stuff, im wondering if something in background is causing this, so going through and uninstalling everything. strangely windows been an arse lately saying it need to restart to install an update, but fails to update, i have rufus on standby as may as well try a full wipe if nothing works and go from there, in 30yrs plus ive never had a faulty card, and boy ive had some cards
It’s not impossible for a GPU to intermittently work and not work, but very rare. And in my experience, a faulty GPU will absolutely throw errors in VRAM tests.
I personally think it’s power/cable related, but could be totally wrong. I’m doubtful it’s due to software as you say it’s a clean windows install. Was it a total clean OS install or just a refresh over an old install?
Can't even get back on the GPU into windows atm had to go to igp. So some weird stuff going on, it was fine during occt left for 20mins with no errors. So some weird voodoo shenanigans going on here, I do love a good mystery
Whack it in a 1x slot or whatever you have on your board, see if you get the same result.
Can't even get back on the GPU into windows atm had to go to igp. So some weird stuff going on, it was fine during occt left for 20mins with no errors. So some weird voodoo shenanigans going on here, I do love a good mystery
Reinstall Windows (totally clean install)
DDU the GPU/igpu driver when automatically installed by windows, and select ‘Disable drivers from windows update’ under advanced options. This was fixed in the latest AMD drivers, but it still happens from time to time.
If it’s still not fixed, it’s either almost certainly the PSU, or a very very weird faulty card. I would then -
Check cabling to GPU
Power test on OCCT to stress the PSU.
windows sure does love to install the old drivers on top though
I redid the paste on the GPU
GPU: 69c
Hotspot: 82c
VRAM: 72c
Much improved. Yet a few key points to consider.
1. The GPU clocks are higher 15c than the 9070 XT Pulse
2. Hotspot is about 15c cooler
3. VRAM about 15c cooler
Fan speed is about 46%. My 9070 XT barely exceeds 30%, so runs quieter. In fact it runs so quiet I can use a more aggressive fan curve and drop hotspot and VRAM temps about 2-3c.
RMA, this thing is busted IMO....
It's looking that wa. Psu ain't that old couple of yrs I've tried every cable replaced some works fine on igp. Yet crashes to no signal if if no load on psu. Only thing it could be is the GPU imo. I test psu tomorrow but I'm sure it's fine
Really can't think what else it could be. So one GPU is 30yrs, wasn't a bad run
Im thinking it's an HDMI port issue as dp works fine on monitor as does igp on with HDMI. It's only the GPU that has a paddy via hdmi
Really can't think what else it could be. So one GPU is 30yrs, wasn't a bad run
Im thinking it's an HDMI port issue as dp works fine on monitor as does igp on with HDMI. It's only the GPU that has a paddy via hdmi
Last edited:
Have you tried 25.3.1 drivers?
I've seen people reporting their GPUs lose signal with 25.3.2 and having to roll back. It's to do with VRR and certain applications such as Edge, Chrome etc.
Edit. Just seen it crashes when you install windows so obviously not latest driver related.
Last edited:
It's funny you should say tha.as never use edge, but yesterday started to use it. Since all hell been kicking off. But the 25.3.1 was a nightmare playing acc. Hence why switched to 25.3.2
what setting did you run?