I have a 7950x/3090 ti system with a Corsair shift 1200w power supply and it will just shut off by itself and reboot. it used to only shut off when I shut it off now the last stint only was 15 ish hours and it shut off / restarted itself been on for 8 ish hours now.
The only odd thing I have is a m.2 to u.2 for my Cd6 r 3.84 tb ssd
Side note Windows says can not connect to the update service and hasn’t been able to update in over 2 ish months.
DISM /Online /Cleanup-Image /CheckHealth
then DISM /Online /Cleanup-Image /ScanHealth
then DISM /Online /Cleanup-Image /RestoreHealth
then run sfc /scannow
once done check your memory for errors.
if its erroring its possible the o.s just crashed due to sustained bad writes (would explain your o.s. not connecting to update, windows wont update files that fail validation).
your saying odd thing? as in your not sure of?.. are we talking from a reputable source or ebay for that connector?.
i would pull it and see if the problem remains.
the 1200w psu should have way more grunt than you need.
sure both parts aint what you call eco friendly :D. but you shouldn’t be taking more than 650w-700w from the wall with that gpu and cpu running full belt.
so unless its actually failed you should be good on that.
It will shut off any time 30 min into a stream (vr) after editing a video watching youtube, closing steam vr
If I remember right all that’s set for Ocing, is pbo and expo not expo tweaked this was rock solid stable, pc was built/upgraded just after last Black Friday and went from 3090 to the ti month or so ago because a friend needed an upgrade 1080 ti to 3090 and the ti wasnt more than non so I got the new used 3090 ti because his cpu would even hold the 3090 back.,
so its when your unloading the cpu and gpu your getting issues.
do you have hardware monitor on your system?..
check the sensors and see if your within spec.
especially on the gpu pci-e power.
if your getting silly high power spikes from the gpu (seemingly 3090 ti’s can ask over 350w for an instant) which then causes the psu to fall over if it happens repeatedly in a short time.
if this is the case a bios flash on the gpu might solve it.
but the likely upshot if this is the issue, you will likely need to rma your psu and have it exchanged for a different 1.
i would try to exchange it for a higher quality lower power 1100 or 1000w unit. rather than taking a direct replacement as the replacement is likely to develop the same issues over time.
anyways heres steve to explain the situation with the 3090/4090 power draw.
By any chance did you just update your BIOS? I was having a similar issue where my system would shut off and hard lock after 30 seconds to 3 hours. Also running a 7950x and 3090.
Not the most current update just before x3D support I think, but it’s updated from the factory, issues started way after so long after Id say its, not the issue
Have you used bluescreenview or the event log to see if any entries point to anything? Could get you pointed in the right direction on what bit of software or hardware is causing it.
What motherboard? Amd has had some vendors over voltong soc which has been leading to cpu degredation and in worse cases exploding. Asus and MSI are the worse offenders. You might try updating the bios to agesa 1.0.0.7. Also disable expo if enabled. You could try 1 stick of ram at a time. To rule out memory problem.
hmm.
you used a fresh image?
if not download a fresh one. the error might be with your previous install media.
have you enabled your routers firewall to block ms ips?.
check your hosts file to make sure its not redirecting ms ip’s to 127.0.0.1 or 0.0.0.0
c:windows/system32/drivers/etc/hosts
your running on an admin account? or a user account with admin privs?.
if your on a normal user account the group policy for updates will likely be the issue.
but in this case its better to log in to an admin account, do your updates (if it lets you).
log out and log back into your user account (rather than increasing the privs of a base user)
lastly retry the fix in the link above to the 10forums.