Page 1 of 2
1
2
LastLast
  1. #1
    Fluffy Kitten Colmadero's Avatar
    10+ Year Old Account
    Join Date
    Jul 2011
    Location
    Miami, FL
    Posts
    1,759

    Computer Rebooting - Possible PSU Problem

    So currently my computer is rebooting whenever I place the GPU under heavy stress.

    If I open up WoW and just chill, it's fine. But if I join a dungeon with heavy AoE, effects, etc. The computer eventually reboots.

    I've been monitoring the GPU temperature and I don't think it's that as it hovers around 70~ degrees and stays there. CPU temps are even lower at around 30-40 degrees.

    HOWEVER, I downloaded the MSI Kombustor program and as soon as I hit stress test with the test being (GL) msi-01, the program loads up and then the computer IMMEDIATELY reboots. Doesn't even run for a bit, just straight up reboots.

    I've read in some forums now that if the PSU is not providing the energy need to the GPU, the GPU will send a reboot signal to the computer.

    NVIDIA Geforce GTX 1060 3GB
    Don't remember what the PSU is but it's a Thermaltake and the SN is TR600CUSMS004118 so I think it's a 600W
    Intel i5-4670k, non-overclocked.

    Thoughts?

  2. #2
    I have a 650 watt PSU with a 2070 super and it works perfectly fine so it's not due to lack of watts that's for sure, get HWInfo and check to see the PSU is running the 3v, 5v and 12v at the right volts. Is your GPU overclocked? or anything for that matter because that's the first thing I'd do is set everything to stock and try again.

    Or set up MSI Afterburner and Rivatuner to see exactly how much power the GPU is getting to rule out the fact it's not getting enough.
    Last edited by Naoto; 2020-11-06 at 06:10 PM.

  3. #3
    Fluffy Kitten Colmadero's Avatar
    10+ Year Old Account
    Join Date
    Jul 2011
    Location
    Miami, FL
    Posts
    1,759
    Quote Originally Posted by Naoto View Post
    I have a 650 watt PSU with a 2070 super and it works perfectly fine so it's not due to lack of watts that's for sure, get HWInfo and check to see the PSU is running the 3v, 5v and 12v at the right volts. Is your GPU overclocked? or anything for that matter because that's the first thing I'd do is set everything to stock and try again.
    I do not overclock anything as it's something I'm not comfortable with. Everything is running at stock.

    Let me check HWInfo and get back to you.

  4. #4
    Please wait Temp name's Avatar
    10+ Year Old Account
    Join Date
    Mar 2012
    Location
    Under construction
    Posts
    14,631
    Sounds like you're tripping some OCP or something. Do you have access to a secondary PSU to test?

    If not, then try downloading MSI afterburner or EVGA precision X1 and see about lowering your voltage slider a bit, to something like 90% and then running a stress test. If that's stable then you're triggering OCP and need a new PSU. If it still happens then.. Might still be OCP, or a number of other things

  5. #5
    Quote Originally Posted by Colmadero View Post
    I do not overclock anything as it's something I'm not comfortable with. Everything is running at stock.

    Let me check HWInfo and get back to you.
    Fair enough, no idea how old the PSU is but if the 3v, 5v and 12v are not what they should be or the GPU isn't getting enough power you're better off buying a new PSU as they are pretty damn cheap, I assume you're not getting any errors in event viewer?

  6. #6
    Fluffy Kitten Colmadero's Avatar
    10+ Year Old Account
    Join Date
    Jul 2011
    Location
    Miami, FL
    Posts
    1,759
    Quote Originally Posted by Naoto View Post
    Fair enough, no idea how old the PSU is but if the 3v, 5v and 12v are not what they should be or the GPU isn't getting enough power you're better off buying a new PSU as they are pretty damn cheap, I assume you're not getting any errors in event viewer?
    let me know if this is useful to you: https://imgur.com/a/B0lK8bW

  7. #7
    How old is your psu? An old psu will lose power as time goes on, even if it is a 600w it could be old and faulty and not actually providing that amount of power.

    I would assume this is a psu problem, as if your card is going bad normally something more than just a cold reboot happens (blue screen/flashing/distortion etc). It seems like it isn't a temp problem if those are you actual temps. Only other thing it could possibly be IMO is bad memory, but I am just going to go out on a limb here and say it's the psu.

  8. #8
    Fluffy Kitten Colmadero's Avatar
    10+ Year Old Account
    Join Date
    Jul 2011
    Location
    Miami, FL
    Posts
    1,759
    Quote Originally Posted by Temp name View Post
    Sounds like you're tripping some OCP or something. Do you have access to a secondary PSU to test?

    If not, then try downloading MSI afterburner or EVGA precision X1 and see about lowering your voltage slider a bit, to something like 90% and then running a stress test. If that's stable then you're triggering OCP and need a new PSU. If it still happens then.. Might still be OCP, or a number of other things
    What's interesting is that, I downloaded OCCT, ran the CPU AND GPU stress tests and it didn't crash the computer.

    - - - Updated - - -

    Quote Originally Posted by Synical123 View Post
    How old is your psu? An old psu will lose power as time goes on, even if it is a 600w it could be old and faulty and not actually providing that amount of power.

    I would assume this is a psu problem, as if your card is going bad normally something more than just a cold reboot happens (blue screen/flashing/distortion etc). It seems like it isn't a temp problem if those are you actual temps. Only other thing it could possibly be IMO is bad memory, but I am just going to go out on a limb here and say it's the psu.
    I've gotten a fair amount of PSUs throughout time. I'd say this PSU is... 4-5 years old.

  9. #9
    Please wait Temp name's Avatar
    10+ Year Old Account
    Join Date
    Mar 2012
    Location
    Under construction
    Posts
    14,631
    Quote Originally Posted by Colmadero View Post
    What's interesting is that, I downloaded OCCT, ran the CPU AND GPU stress tests and it didn't crash the computer.
    .. Huh. Interesting. Try Furmark. It's made to basically be a GPU power virus to stress your GPU to its breaking point.
    Also try Prime95, which will stress the CPU similarly.

    Do one at a time, if they're both stable then run them at the same time. If it's power delivery at fault, that will find it

  10. #10
    Fluffy Kitten Colmadero's Avatar
    10+ Year Old Account
    Join Date
    Jul 2011
    Location
    Miami, FL
    Posts
    1,759
    Quote Originally Posted by Naoto View Post
    Fair enough, no idea how old the PSU is but if the 3v, 5v and 12v are not what they should be or the GPU isn't getting enough power you're better off buying a new PSU as they are pretty damn cheap, I assume you're not getting any errors in event viewer?
    Event viewer doesn't say anything - just registers the system turn off time and turn on and classified it as a dirty shutdown. I've had chkdsk come up once due to this.

  11. #11
    Was that screenshot from max load? think it's under your motherboard where you see the PSU volts so just look for PSU 3v, 5v and 12v.

    Pretty sure the only time I had an issue like this it was because of my PSU being shit, as I said they are pretty cheap and you wouldn't need more than 650. Go with something like a Corsair or any other reputable brand and make sure it's gold.
    Last edited by Naoto; 2020-11-06 at 06:28 PM.

  12. #12
    Fluffy Kitten Colmadero's Avatar
    10+ Year Old Account
    Join Date
    Jul 2011
    Location
    Miami, FL
    Posts
    1,759
    Quote Originally Posted by Naoto View Post
    Was that screenshot from max load? think it's under your motherboard where you see the PSU volts so just look for PSU 3v, 5v and 12v.
    That is a screenshot of me running a not so intensive video game, 4 screens with multiple windows open.

    Mobo pic: https://imgur.com/a/ZgD7uIP

  13. #13
    Quote Originally Posted by Colmadero View Post
    That is a screenshot of me running a not so intensive video game, 4 screens with multiple windows open.

    Mobo pic: https://imgur.com/a/ZgD7uIP
    I'm no expert but it all looks fine to me, what's the max GPU power you get under full load? problem with PC's is it can either be the PSU or the GPU and it's almost impossible to find out which it is.

  14. #14
    Quote Originally Posted by Colmadero View Post
    That is a screenshot of me running a not so intensive video game, 4 screens with multiple windows open.

    Mobo pic: https://imgur.com/a/ZgD7uIP
    You are overthinking things when its the PSU.

    Ask a friend to bring his over, test it out the same way you trigger your restart, easy peesy, if it happens again, you already said your thermals are okay, so its only the motherboard left which is a very rare case but it can happen, but since you seem to have logical values at your screenshot voltage, its highly unlikely.

    I will assume the 1.7V at VCCIN and AUXTIN3 at 123C are the usually buggy reporting programs.
    Last edited by potis; 2020-11-06 at 06:38 PM.

  15. #15
    Fluffy Kitten Colmadero's Avatar
    10+ Year Old Account
    Join Date
    Jul 2011
    Location
    Miami, FL
    Posts
    1,759
    Quote Originally Posted by potis View Post
    You are overthinking things when its the PSU.

    Ask a friend to bring his over, test it out the same way you trigger your restart, easy peesy, if it happens again, you already said your thermals are okay, so its only the motherboard left which is a very rare case but it can happen, but since you seem to have logical values at your screenshot voltage, its highly unlikely.
    Sadly, I don't have a friend who can just bring a PSU like that.

    I can try purchasing one at Best Buy and just return it if we determine it's not the PSU.

    Should I purchase one with the same or more Watts?

  16. #16
    Quote Originally Posted by Colmadero View Post
    Sadly, I don't have a friend who can just bring a PSU like that.

    I can try purchasing one at Best Buy and just return it if we determine it's not the PSU.

    Should I purchase one with the same or more Watts?
    Unless you plan on going to 12 cores with a 3090 GPU I wouldn't go above 650, I have a 2070 super and i7 8700k overclocked and get nowhere near to maxing my 650 PSU out.

  17. #17
    Quote Originally Posted by Colmadero View Post
    Sadly, I don't have a friend who can just bring a PSU like that.

    I can try purchasing one at Best Buy and just return it if we determine it's not the PSU.

    Should I purchase one with the same or more Watts?
    If you have that capability, why not, maybe ask them to lent you their testing one or w/e :P I dont know how best buy works.

    Wattage wise make sure you get at least 550W and at least 80+Bronze if so or simply go for a similar 650W (even better 80+gold) for when you upgrade that PC, post the models you can find maybe when you are there? Or links?

  18. #18
    Check your GPU fan settings.
    according to that image,your fans are set to 20% for 60c range .
    I would suggest upping that to 70% , then try to run wow again.

  19. #19
    You should keep HWInfo open and load up a game benchmark to see how much GPU power output you get at full load, your card has a maximum of 120 watts and your 1st screenshot was 28 watts at 56% load.

  20. #20
    Please wait Temp name's Avatar
    10+ Year Old Account
    Join Date
    Mar 2012
    Location
    Under construction
    Posts
    14,631
    Quote Originally Posted by Colmadero View Post
    Sadly, I don't have a friend who can just bring a PSU like that.

    I can try purchasing one at Best Buy and just return it if we determine it's not the PSU.

    Should I purchase one with the same or more Watts?
    Get a 600-650W unit. Get something decent too, in case you have to keep it.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •