1. #1

    MOBO or GPU dying?

    Just started happening this last two weeks.

    About an hour into PC usage sometimes more, my screen suddenly freezes up, mouse pointer wont move. Screen then goes black and i get a message that there is no signal. Even though the PC is obviously still running. (its not a monitor fault as i have a laptop running through it too which works fine).

    CTRL/ALT/DEL does nothing, system cant be reset any other way than a hard reboot.

    Utilities are telling me that my MOBO temp is 114degrees (it isn't, id be able to smell it and the case is cool to the touch).

    To be fair on the system its now approaching 4 years old as it was built in spring 2013.


    Asus z87k mobo
    i5-4670k cpu
    Asus 760gtx


    Maybe its just getting a bit long in the tooth, but id like to have an idea of what is dying first.

    Cheers

    R

  2. #2
    Deleted
    Quote Originally Posted by redderz1 View Post
    Just started happening this last two weeks.

    About an hour into PC usage sometimes more, my screen suddenly freezes up, mouse pointer wont move. Screen then goes black and i get a message that there is no signal. Even though the PC is obviously still running. (its not a monitor fault as i have a laptop running through it too which works fine).

    CTRL/ALT/DEL does nothing, system cant be reset any other way than a hard reboot.

    Utilities are telling me that my MOBO temp is 114degrees (it isn't, id be able to smell it and the case is cool to the touch).

    To be fair on the system its now approaching 4 years old as it was built in spring 2013.


    Asus z87k mobo
    i5-4670k cpu
    Asus 760gtx


    Maybe its just getting a bit long in the tooth, but id like to have an idea of what is dying first.

    Cheers

    R
    You think that's old? My 3770k still runs like a train and has been running at 4.8ghz since I first installed it.

    Anyway, concerning your probem, it can be anything. Have you checked event viewer? Does it happen at random or only under load? What are your idle and load temps? What 'utilities' are telling you your "mobo temp" is 114(c°)? What does the BIOS say? Have you tried a fresh Windows install?

    etc.

  3. #3
    Not sure what im looking for in event viewer but there have been critical events from "kernel-power". These seem to tie up with my shut downs in terms of timing.

    I re-installed windows 8.1 a few days ago. Should have mentioned that.

    Ive used three different utils for MOBO temp. The one im currently looking at is "Speccy" and its reading 113 degrees.

    My idle temps are 35-37 degrees for graphics and cpu. Mobo is permanently at 113-114 degrees.

    The shut down happens randomly, under no load, often just while on net pages (it would most likely shut down whilst im typing this reply!)

    As far as BIOS im not sure what I would be looking for there.

    cheers

  4. #4
    The Lightbringer Evildeffy's Avatar
    15+ Year Old Account
    Join Date
    Jan 2009
    Location
    Nieuwegein, Netherlands
    Posts
    3,772
    Quote Originally Posted by redderz1 View Post
    Not sure what im looking for in event viewer but there have been critical events from "kernel-power". These seem to tie up with my shut downs in terms of timing.

    I re-installed windows 8.1 a few days ago. Should have mentioned that.

    Ive used three different utils for MOBO temp. The one im currently looking at is "Speccy" and its reading 113 degrees.

    My idle temps are 35-37 degrees for graphics and cpu. Mobo is permanently at 113-114 degrees.

    The shut down happens randomly, under no load, often just while on net pages (it would most likely shut down whilst im typing this reply!)

    As far as BIOS im not sure what I would be looking for there.

    cheers
    The description of your issues could indicate any number of things and needs to be conclusively tested.
    The question is are you capable or desiring to do a diagnostic of things? If not .. bring it to a computer store to do it and explain in detail what the problem is.
    If you are capable/wanting to then run a battery of tests starting with:

    Prime95 first (with temperature monitoring) to see if it remains stable or crashes immediately for at least an hour. (stop if temps go berserk of course (100°C+)).
    Furmark after (with temperature monitoring) to see if it remains stable or crashes immediately for at least an hour. (stop if temps go berserk of course (100°C+)).

    Then both Prime95 and Furmark together to see if it remains stable or crashes immediately for at least an hour. (stop if temps go berserk of course (100°C+)).
    This is to stress the PSU a bit here, problem is that a PSU is rather difficult to test and it powering on and working right now doesn't mean it's not broken.
    Nor is doing a paperclip test enough to determine anything of the sort, I say this to be clear and I see a lot of people making the wrong assumptions.

    After that you run at least 3 full passes with MemTest86+ from Hiren's Boot CD or something (this takes long).

    This takes care of some of the basic tests for you work with and should get you started nicely.

    As far as your mobo temps go... well when it's saying that temperature go NEAR (NOT TOUCH) the motherboard heatsinks.. if it really is at that high a temp you'll notice as it'll radiate enough heat for you to know it'll burn your fingers.
    Also check in the BIOS if it shows these temperatures there, if not it's likely a bogus reading.

  5. #5
    Deleted
    The Z87-K has a passive cooler for the chipset - maybe it's not properly attached anymore or the airflow has been blocked.


  6. #6
    if the system is still running as in turned on and the display says no signal you can verify by having some music running if its the gpu.
    If the musik is continung to play its likely that its just your gpu
    if the whole pc freezes as in musik stops or loops it can basically be anything.

  7. #7
    when was the last time you cleaned out the dust?

    sounds like the GPU is overheating to me
    <insert witty signature here>

  8. #8
    Quote Originally Posted by smokii View Post
    when was the last time you cleaned out the dust?

    sounds like the GPU is overheating to me
    I dust out every couple of weeks. Its spotless in there!

    - - - Updated - - -

    Quote Originally Posted by Evildeffy View Post
    The description of your issues could indicate any number of things and needs to be conclusively tested.
    The question is are you capable or desiring to do a diagnostic of things? If not .. bring it to a computer store to do it and explain in detail what the problem is.
    If you are capable/wanting to then run a battery of tests starting with:

    Prime95 first (with temperature monitoring) to see if it remains stable or crashes immediately for at least an hour. (stop if temps go berserk of course (100°C+)).
    Furmark after (with temperature monitoring) to see if it remains stable or crashes immediately for at least an hour. (stop if temps go berserk of course (100°C+)).

    Then both Prime95 and Furmark together to see if it remains stable or crashes immediately for at least an hour. (stop if temps go berserk of course (100°C+)).
    This is to stress the PSU a bit here, problem is that a PSU is rather difficult to test and it powering on and working right now doesn't mean it's not broken.
    Nor is doing a paperclip test enough to determine anything of the sort, I say this to be clear and I see a lot of people making the wrong assumptions.

    After that you run at least 3 full passes with MemTest86+ from Hiren's Boot CD or something (this takes long).

    This takes care of some of the basic tests for you work with and should get you started nicely.

    As far as your mobo temps go... well when it's saying that temperature go NEAR (NOT TOUCH) the motherboard heatsinks.. if it really is at that high a temp you'll notice as it'll radiate enough heat for you to know it'll burn your fingers.
    Also check in the BIOS if it shows these temperatures there, if not it's likely a bogus reading.
    Thanks. Ill give this a go. I like to sort things out myself.

    I'm convinced the mobo temp is bogus. Surely id be able to smell it and feel it through the case. The case is cool.

  9. #9
    Are capacitors good still? Is the video card having latest drivers? Is it going bad? What does the windows event log say? Are there any memory dumps? Have you reseated all the RAM, Harddrive cables, cards, etc? Have you tested the hard drive to see its state? Have you tested the power supply?

  10. #10
    Quote Originally Posted by redderz1 View Post
    I'm convinced the mobo temp is bogus. Surely id be able to smell it and feel it through the case. The case is cool.
    I know you can smell when a CPU burns out, but most of the time things will throttle or shut off before they get to a temp that you would smell. Obviously something is shutting off, so something is causing that, likely heat.

    Also, no, you wouldn't necessarily feel the heat on the case either. Unless the part that was overheating had direct contact with the case. If it's the chipset on the motherboard though, that has no contact to the case. The heat would have to be absorbed by the board itself, then transfer to the mounting screws, then transfer to the case. Not likely to happen as that heat will be dissipated to the air long beofre that happens.

    There is a possibility that the temperature is reading wrong, but that's a pretty rare case. You likely do have a heat issue and are just trying to convince yourself that you do not.

  11. #11
    Deleted
    Quote Originally Posted by redderz1 View Post
    Surely id be able to smell it and feel it through the case. The case is cool.
    While the temperature reading is most likely bogus, you won't necessarily be able to feel it by simply touching the case if a single component (chipset) runs hot depending on where the sensor is actually located.

  12. #12
    The Lightbringer Evildeffy's Avatar
    15+ Year Old Account
    Join Date
    Jan 2009
    Location
    Nieuwegein, Netherlands
    Posts
    3,772
    Quote Originally Posted by redderz1 View Post
    Thanks. Ill give this a go. I like to sort things out myself.

    I'm convinced the mobo temp is bogus. Surely id be able to smell it and feel it through the case. The case is cool.
    That's not true, depending upon size and location.

    Since chipsets (as an example) are located either bottom right (south bridge, most common) or below the CPU (north bridge, getting more exceedingly rare) they often have way bigger heatsinks than their actual size.
    This also means that if you have any airflow going through the case the heat gets dragged out.
    But... because it's such a tiny volume vs. large amount of air going out because of the physical size you might not feel it coming out of the fans' airstream, let alone the actual case as you'd only feel it if the chipset were somehow connected to the case as a heatsink since there's no giant heatsink nor fan design blowing against it.

    This is why I suggested checking the way I did and not via the case, I recommend you do this A.S.A.P. just to be certain it's not going ballistic.

    Take a 5W light bulb (not LED) f.ex. ... if you hold that in your hand when it's on for a while it'll cause burns... but lift up your fingers 0,1mm away from the glass and the radiation of heat is negligible .. but move back 0,1mm and BURN BABY BURN.
    Same principle applies for your motherboard's chipset (if that's indeed it).

    NOTE: Still check the BIOS for temps of course when this heat is supposedly that high in Windows.

    Also .. VRMs regularly go over 120°C on various components, especially graphics cards, you don't smell those either do you.

    The temperatures we state aren't "hard limits" ... they deteriorate very quickly and can crap out quickly at those temperatures but often they can take a lot of heat before you start smelling anything, I'm referring here to 150°C and higher.

  13. #13
    OK, thanks so much for the advice so far. I've been doing some testing....

    Prime95 first (with temperature monitoring) to see if it remains stable or crashes immediately for at least an hour. (stop if temps go berserk of course (100°C+)).
    Furmark after (with temperature monitoring) to see if it remains stable or crashes immediately for at least an hour. (stop if temps go berserk of course (100°C+)).
    The prime95 test got my CPU up to 61°C

    Fumark took my GPU up to 81°C

    MoBo Temp throughout both of these tests showed 113-115°C.

    Then both Prime95 and Furmark together to see if it remains stable or crashes immediately for at least an hour. (stop if temps go berserk of course (100°C+)).
    Running both tests together took my CUP up to 63°C and GPU up to 82°C

    Again MoBo temp read as 113-115°C

    After that you run at least 3 full passes with MemTest86+ from Hiren's Boot CD or something (this takes long).
    Ran this test. Came back with Zero errors.

    Also check in the BIOS if it shows these temperatures there, if not it's likely a bogus reading.
    The only temp shown in my BIOS is the CPU temp. As its at idle when BIOS is it showed 34-35°C.

    As far as your mobo temps go... well when it's saying that temperature go NEAR (NOT TOUCH) the motherboard heatsinks.. if it really is at that high a temp you'll notice as it'll radiate enough heat for you to know it'll burn your fingers.
    This is why I suggested checking the way I did and not via the case, I recommend you do this A.S.A.P. just to be certain it's not going ballistic.
    I could detect almost zero heat radiating when hovering my fingertip 2-3mm from the heatsink. I eventually touched the heatsink and I'd estimate it to be no more than 40-45°C. Warm but definately not hot.

    The temps shown on Speccy and Open hardware monitor were maxing at 115°C.

    I've tried booting the system form cold (switched off for 3 hours). Running Speccy straight away and it immediately shows 113-115°C.


    Just to make sure theres nothing else I might be missing.


    Full specs are:

    i5-4670k
    EVO212 cooler
    Asus z87-k Mobo.
    Corsair VengeanceLP 8gb (2x4gb) DDR3 1600mhz Ram
    Corsair CX600M PSU
    Asus GTX760 DirectCU11 CDDR5 2gb.

    I've made sure its clean, ive reaseated memory and various connectors around the MoBo

    Interestingly, the shutdowns I've been used to did not occur for the entire duration of any of the Stress tests......

    Thanks again

  14. #14
    The Lightbringer Evildeffy's Avatar
    15+ Year Old Account
    Join Date
    Jan 2009
    Location
    Nieuwegein, Netherlands
    Posts
    3,772
    The temperature reading then is basically from a fixed "bogus" sensor as all sensors are displayed in the BIOS as well.

    Regardless with load it seems to hold stable which honestly gives me 3 options from my own anecdotal experience:

    1: Windows install may be borked, you could attempt a clean install from scratch.
    2: The PSU might be having issues.
    3: The GPU might be having issues.

    Considering what you report ... all 3 are an option at this point.
    The only way to test the latter 2 is to replace them with a borrowed unit.

    The first ... well that's easy, a pain in the ass but easy.

    Personally I'm betting on option #3 at this point but it's only a guess at this point and as said... the only way to test this is to borrow a GFX card and replace it for a while.

Posting Permissions

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