PC Tech help needed :(

Discussion in 'Computer Hardware, Devices and Accessories' started by pwsincd, Dec 27, 2015.

  1. pwsincd
    OP

    pwsincd Garage Flower

    Member
    3,302
    1,699
    Dec 4, 2011
    Manchester UK
    Bought my lad a low to mid range PC and have nothing but trouble with it prior to Christmas (where i thought the issue had been resolved) and now after xmas a whole new set of issues (possibly).

    thought it was a bargain for £260 advertised in aria PC as a refurb with 60day warranty (aprox 40 days left)

    Spec :
    AMD FX 8320 CPU with a stock fan
    Gigabyte AMD AM3+ 970A 4X DDR 3 USB 3.0 ATX Motherboard
    8 gig RAM (unknown brand)
    MSI r7 260x 2gig GPU (originally : GTX 670)
    60 gig SSD
    1TB HDD
    DVD writer
    500watt switching generic (by the looks) PSU

    So the problem(s) :

    initially installed win10 and struggled like fek to even install drivers for the GTX card , tried all sorts and then tried windows 7 , with the same issues. Took PC back and they swapped card to the r7 260x and installed drivers for me. I was suitably impressed , came home quick check of it and all seemed fine . Wrapped it up for xmas.

    Xmas day :

    Opens Pc sets it up : set up is win10 with latest AMD driver set installed. The PC runs fine until i try a game , it crashes after 5 or 10 mins the fan is running at full tilt . Different games provide different crashes , but all have crashed the PC so far .. CS:GO after five mins closes the PC . COD WAW freezes PC after 10 to 15 mins , minecraft runs ok but seems to take an age to render a complex world prior to being able to play ... install HD mods /optifine etc , it crashes pc after 5 mins.

    installed windows 7 : all the same faults are still present. Updating windows 7 did once also shut down the PC . not to sure if this was coincidence.

    have tried so far to disconnect the HDD and run games from SSD only , same problem occurs.
    before i go any further and as i cannot at this time of year simple return it yet i wondered if anyone could give me a heads up to what may be causing this and possible solutions. Hope you can help... as you can imagine my lad is a little pissed off :(
     
  2. mashers

    mashers Stubborn ape

    Member
    3,837
    5,153
    Jun 10, 2015
    Kongo Jungle
    Which fan? CPU? GPU? PSU? Try running with the case open and a desk fan blowing into the PC. Not a solution obviously, but will confirm my suspicion that it is overheating.
     
  3. pwsincd
    OP

    pwsincd Garage Flower

    Member
    3,302
    1,699
    Dec 4, 2011
    Manchester UK
    yeah i will do this , and report back my findings.
     
  4. Duo8

    Duo8 I don't like video games

    Member
    3,438
    1,138
    Jul 16, 2013
    What RAM? Run memtest or something.
     
  5. FireEmblemGuy

    FireEmblemGuy Celebrating a decade of shitposting

    Member
    2,400
    421
    Jul 6, 2007
    United States
    Michigan, USA
    Yeah, I'd say leave memtest running for at least twelve hours and see what comes up. Usually in a case like this it's either the RAM or the PSU, but RAM usually gives a bluescreen and most of the time (in my personal experience, granted) PSU failures have a bit more flair to them.

    I suppose there might also be something up with the memory on the GPU, but I don't think you can really take it out and test without it - as far as I'm aware neither that processor nor the motherboard have integrated graphics processing.
     
  6. Duo8

    Duo8 I don't like video games

    Member
    3,438
    1,138
    Jul 16, 2013
    Checking GPU memory is possible. http://mikelab.kiev.ua/index_en.php?page=PROGRAMS/vmt_en
     
  7. pwsincd
    OP

    pwsincd Garage Flower

    Member
    3,302
    1,699
    Dec 4, 2011
    Manchester UK
    memtest is running whilst were out for the night so will see what happens , does it create a log ?
     
  8. ody81

    ody81 GBAtemp Fan

    Member
    464
    229
    Aug 21, 2012
    Grab GPU-Z (google it) run it and check the log file box. That'll let you save the finding to a log while the game crashes.
    Then run a game until it crashes. This'll at least rule out GPU overheating.

    Next step is taking a look at your system events log just in case something obvious has borked up.
    If two separate cards fail something's wrong, hopefully a software issue.
     
  9. pwsincd
    OP

    pwsincd Garage Flower

    Member
    3,302
    1,699
    Dec 4, 2011
    Manchester UK
    Ok memtest ran and all reports are fine , running again just to be sure , ill look into the GPU-Z next thanks..
     
  10. pwsincd
    OP

    pwsincd Garage Flower

    Member
    3,302
    1,699
    Dec 4, 2011
    Manchester UK
    ok ran GPU-Z after 15 mins it crashed whilst running a cranked up minecraft optifine and HD shaders etc . the GPU-z's log file didnt really show any significant spikes in any of the parameters , all seemed ok . Currently running CPU-z using the stress test feature . After the PC crash while using GPU-Z the pc took an age before it would boot. Its starting to look like overheating , im thinking CPU/MOBO but im not sure how to verify this. any ideas ?
     
  11. ody81

    ody81 GBAtemp Fan

    Member
    464
    229
    Aug 21, 2012
    Well GPU-Z shows you the heat of the GPU. CPU-Z will log heat also, so same again, run CPU-Z, save to a log and run your test (minecraft in this case). That should clear up some things.

    After that, we look at what Windows is doing while this is all going on. And maybe a quick check of the bios, make sure any settings that affect fan speed are set for maximum speed. It's called cool 'n quiet on Intel boards, can't remember what it's called on AMD boards but it should stick out like a sore thumb under 'PC Health' settings or some re-wording of that.
     
  12. mashers

    mashers Stubborn ape

    Member
    3,837
    5,153
    Jun 10, 2015
    Kongo Jungle
    Try what I suggested - run it with the case off an a desk fan blowing into the case to disperse the hot air. If that improves the situation then it's overheating. Also, which fan is running? That will indicate what is overheating.
     
  13. pwsincd
    OP

    pwsincd Garage Flower

    Member
    3,302
    1,699
    Dec 4, 2011
    Manchester UK
    ok running cpu-z i couldnt seem to log cpu temp , so got another app(open hardware monitor) . not sure on its legitimacy but seems a similar app logs all to a csv file ... the results of which dont seem extreme at all all temps are sub 50c . i have been running the pc with the side off and it has slightly increased the time it takes before a crash , did have a fan (well hairdryer on cold setting) blowing cold air into the case , this hasnt made much difference , maybe 5 or 10 mins additional use. its the CPU fan that is going mental but temps dont seem extreme.

    @ody81 : what software checks are you reffering too ?

    edit : cant get into bios , think its a monitor issue . will set up a different monitor ...
     
    Last edited by pwsincd, Dec 28, 2015
  14. mashers

    mashers Stubborn ape

    Member
    3,837
    5,153
    Jun 10, 2015
    Kongo Jungle
    How do you know the CPU temperature isn't extreme if you can't log it? If the CPU fan is constantly running at high speed, this suggests that the CPU is overheating.
     
  15. pwsincd
    OP

    pwsincd Garage Flower

    Member
    3,302
    1,699
    Dec 4, 2011
    Manchester UK
    used a different app to log (i did say this :) ) . @ody81 : all fan speeds are set to max and the cool n quiet setting is enabled . Ive set biod to warn on over temp .. not sure how it does this visually ? .. any other pointers....
     
  16. ody81

    ody81 GBAtemp Fan

    Member
    464
    229
    Aug 21, 2012
    Yeah, disable cool n quiet. It's an option for people who have jet turbines as cooling systems, it'll pick it's times to slow the fan down and speed it up. Better to just shut it off, fans aren't really noisy, and they last a long time so go for it, disable it.

    For software checks, start with the basic event logs. Just hit Start and type 'view event logs'.
    On the left panel of that window you'll see the main ones, Application and System under the Windows Logs portion.
    These are what you'll want to focus on, note what time it is on your pc, start a game, when it crashes pop into event logs and check the time against the system events.
    Take note of the events.
    Just check to see what is actually occurring at the time of the crash, these logs can be surprisingly helpful sometimes. I emphasize sometimes though :)

    Another thing to check, in your bios it should tell you how many rpms the cpu fan is running at, you say it's going crazy, just look and see to make sure it's not too crazy.

    Edit: And I've seen even new drives have trouble, grab this:

    http://crystalmark.info/download/index-e.html

    It's a tool (not the absolute best but it's all I can link to here), to check the condition and health of your hard disk. Temperatures included, I've seen many a drive running at 95-105 degree celsius and causing all kinds of trouble (crashes, boot problems) without doing anything suspicious enough to draw attention to the problem (like major corruption etc.)

    Better safe than sorry.
     
    Last edited by ody81, Dec 28, 2015
  17. TotalInsanity4

    TotalInsanity4 GBAtemp Supreme Overlord

    Member
    6,912
    6,881
    Dec 1, 2014
    United States
    Under a rock
    Could you give us pictures of both the outside and the inside of the case? I'd like to get a good idea of what kind of airflow you have, and see if it needs to be improved on if it's contributing to overheating
     
  18. pwsincd
    OP

    pwsincd Garage Flower

    Member
    3,302
    1,699
    Dec 4, 2011
    Manchester UK
    conclusion : gpu was fukked..... replaced at the store .. all is well. ...
     
    TotalInsanity4 likes this.