PC problem. Page 5

  • Page

    of 6 First / Last

  • Deleted user 30 November 2012 22:07:55
    @Iain815

    I've experienced the exact same symptoms you've described (with faulty GPUs). The ATI mobility Radeon 7500 in my old Thinkpad R40 (gave out on its 7th year), and a ATI FireGL-X2-256 workstation GPU(£700), after about 12months that went faulty part way through the winter, return to normal through the spring, summer, autumn and then repeated the problem the following Winter.

    Try the card with the old Doom3 demo, with settings at maximum, as the intense stencil shader, hard shadows will start to cause graphical corruption if it is a faulty GPU, little red or white dots, were the result of Doom3 demo on my X2 .

    It is the same RRoD/YLoD fault, so you can try and fix it with a heat gun. But in all likelihood it is a bin job.
  • Iain815 15 Jan 2013 21:02:42 436 posts
    Seen 2 hours ago
    Registered 7 years ago
    Iain815 wrote:
    God I'm losing my mind over this.

    MBoard: Z77-D3H
    GPU: MSI R7850 Power Edition
    CPU: i5-3570k
    PSU: Coolermaster GX 650w

    Basically a few weeks ago my computer would start crashing randomly with a blank blue screen forcing me to power off at the wall. This would happen randomly, from gaming to just browsing and could occur 10 minutes after turning on, or hours. I spent many days going through trail and error trying to figure which components were causing this. I've completely ruled out:

    The RAM (No errors found, changed modules)
    HDD (No errors found)
    Overheating
    PSU
    Windows 7/8 (reinstalled, reformatted)

    Then I moved onto the graphics card. Took it out and for a whole day had no problems. "Well that must be it!". But to make sure sure it truly was the card, I popped it into the other (x4) PCI-E slot and it worked fine for a day on it. :/

    Faulty (x16) PCI-E slot? But now I'm currently using my old GPU now in it (GTS 450) and so far have had no problems.

    What the hell is it, dodgy motherboard or GPU? I'm leaning towards GPU, because maybe only running at x4 means it cant get enough going to make it crash, but I don't know if the logic in that is correct.
    Just in case anyone is having this problem, Googled it and landed here, I think I found a solution. I sent it to SCAN for RMA, they tested it and found no faults, asked them to test again and they did, still no faults.

    Anyway, after tearing my hair out for another day or two, I installed Afterburner and deceased the core clock and memory clock (down to 930 from 950 and 1150 from 1200 respectively) and I haven't any problems in the past 24 hours. I'm going to slowly creep it up everyday until is bites back.

    So basically I'm stuck with a card which not only I cant overclock (main selling point) but performs slightly below what it should. And two RMA's that came up clean, so no way to prove it. Not much I can do about it seems.

    Lowering the clock rate might have been an obvious thing to do for some, but given the crashes were random, I didn't assume it was the reason. Who knows, maybe it's just teasing me and will fuck up again in an hour.
  • dsmx 16 Jan 2013 00:58:50 7,686 posts
    Seen 2 hours ago
    Registered 9 years ago
    The only other thing I can think of there is possibly the power supply isn't giving you enough juice, underclocking the graphics card would reduce the amount of power needed to run the system.

    "If we hit that bullseye the rest of the dominoes will fall like a a house of cards, checkmate." Zapp Brannigan

  • Dirtbox 16 Jan 2013 01:03:41 78,771 posts
    Seen 2 hours ago
    Registered 12 years ago
    Depends, if it's a soft overclock, that wouldn't be the case.

    Always best to set everything to defaults if you're running into problems though.

    Also check all the connections on the motherboard, make sure the board doesn't have any popped capacitors, kill as many background operations as possible and have a good long root through your error logs.

    I'd say the video card is overheating though.

    Edited by Dirtbox at 01:08:27 16-01-2013

    +1 / Like / Tweet this post

  • Iain815 16 Jan 2013 01:31:54 436 posts
    Seen 2 hours ago
    Registered 7 years ago
    Nah, I doubt it's the PSU. Before it started kicking off it ran perfectly fine for weeks on a 450w. I have a 650w now which should be plenty. Motherboard looks good, never really have any background processes running and crashes never turn up details on error logs.

    The annoyance is that I wasn't overclocking it in the first place. Well, technically the card itself comes from the factory overlocked, but it designed to be able to work at that level and beyond.

    I've had to lower it below the defaults. By lowering the clock I'm basically downgrading. It's not much, but still, it's annoying that I won't be able to overclock it in the future.

    Edit: Actually, it seems I can adjust the voltage controls with Afterburner as well. Would upping that help? Would fall under the power theory.

    Edit:
    Just gave it a small increment. We'll see how she goes.

    Edited by Iain815 at 02:23:19 16-01-2013
  • RobTheBuilder 16 Jan 2013 02:34:50 6,521 posts
    Seen 1 year ago
    Registered 10 years ago
    @Iain815 if you are being forced to run it below the settings it was advertised at surely you should be eligible for a refund?

    Edited by RobTheBuilder at 02:35:55 16-01-2013
  • mal 16 Jan 2013 02:46:52 22,787 posts
    Seen 22 minutes ago
    Registered 13 years ago
    Sounds like it might be PCI bus instability to me, if shifting things about on that affects it. I had a motherboard with that before, and apparently the fix was to bridge a resistor that had been added late in development to meet emissions standards, but it seems it did something to the bus dynamic that at certain frequencies it'd occasionally introduce a bad bit into the mix and the system would hang.

    Cubby didn't know how to turn off sigs!

  • Iain815 16 Jan 2013 03:11:12 436 posts
    Seen 2 hours ago
    Registered 7 years ago
    Yeah, cranking up the voltage didn't help, it crashed. Booted up again and nudged the voltage even more and crashed mere minutes later. Back to below defaults now.

    RobTheBuilder wrote:
    @Iain815 if you are being forced to run it below the settings it was advertised at surely you should be eligible for a refund?
    Problem is, I sent to back to get checked and it came up non-faulty twice. It could be bad luck both times, but now it's most likely thereís something else wrong, like mal suggested the motherboard slot.

    This guy seems to be having a similar problem and fixed it:
    http://forums.overclockers.co.uk/showthread.php?t=18438818

    Iíll do what he suggests at the end of the thread and will report back.

    Update: Been a few hours, no problems so far. Slightly overclocking a little just to be certain. Ran Kombuster for an hour with a movie playing in the background.


    Update 2: Seeing this thread being bumped reminded me to leave a final update. It seems the problem was with my Memory Clock on the GPU. I have to underclock it to 1150MHz (default 1200) and it no longer crashes. I can overclock the Core Clock all the way to the top with no problems. Regardless if the problem lies with the GPU or MBoard connections, it's working fine now so I'm happy to live with it.

    Edited by Iain815 at 19:12:34 19-02-2013
  • ecu 19 Feb 2013 18:56:06 77,228 posts
    Seen 9 minutes ago
    Registered 10 years ago
    So I think I just cooked my CPU. After some gaming, I restarted my computer and it wouldn't boot back up. I checked the mobo and the red CPU light was lit up, so I opened the case up and the heatsink was caked in dust. I took some compressed air to it and cleaned it out as best I could, and it's up and running again now.

    I haven't done any permanent damage have I? And can anyone recommend a CPU temperature monitor program?
  • Rusty_M 19 Feb 2013 19:13:04 4,864 posts
    Seen 1 hour ago
    Registered 8 years ago
    Hwmonitor, coretemp and speedfan all work quite well.

    Coretemp will display on an LCD display on your keyboard if you ave one of those.

    The world is going mad. Me? I'm doing fine.
    http://www.twitch.tv/rusty_the_robot
    http://twitter.com/Rusty_The_Robot

  • superdelphinus 19 Feb 2013 19:42:05 8,123 posts
    Seen 18 hours ago
    Registered 9 years ago
    @Iain815 my 7850 is a bit like that too, though I don't have to underclock it and I've not had a crashing problem as standard. The core clock will go miles over but if I push the memory clock more than about 10 it becomes extremely unstable. That his on a sapphire OC edition
  • ecu 20 Feb 2013 01:19:15 77,228 posts
    Seen 9 minutes ago
    Registered 10 years ago
    Rusty_M wrote:
    Hwmonitor, coretemp and speedfan all work quite well.

    Coretemp will display on an LCD display on your keyboard if you ave one of those.
    Cheers. Using Core Temp, seems to be decent. Having the window next to my desk open drops my temperatures by 15C. Think I'll just throw on an extra layer when I play Crysis. :D
  • Aretak 23 Feb 2013 11:31:51 10,374 posts
    Seen 45 minutes ago
    Registered 12 years ago
    So despite being perfectly fine when I went to bed last night, when I switched my PC on this morning I was greeted with the fans running at full blast, a loud beep and no output on the monitor. After about an hour of fiddling I managed to make it boot without one of the RAM sticks in, so I assumed that was the problem. But having now put the 'faulty' RAM back in (even in the same slot it was before), it all seems to be working just fine. Rebooted a handful of times without issue.

    Just wondering if it's something I should be worried about long-term? An early sign of the RAM stick failing or something? I don't really get why simply taking it out and putting it back in again fixed anything.
  • Dirtbox 23 Feb 2013 13:47:03 78,771 posts
    Seen 2 hours ago
    Registered 12 years ago
    Run a memory test.

    The fan thing suggests it could be the motherboard though.

    +1 / Like / Tweet this post

  • Aretak 24 Feb 2013 11:43:09 10,374 posts
    Seen 45 minutes ago
    Registered 12 years ago
    Yup, seems you're right and it's fucked. Won't boot again this morning despite being fine 7 hours ago. No amount of wiggling or replacing bits looks like it'll help this time either. What a pain in the arse. =/
  • THFourteen 24 Feb 2013 12:29:29 34,206 posts
    Seen 18 minutes ago
    Registered 9 years ago
    Short term solution, don't switch it off.
  • Aretak 24 Feb 2013 13:06:44 10,374 posts
    Seen 45 minutes ago
    Registered 12 years ago
    THFourteen wrote:
    Short term solution, don't switch it off.
    Well yeah, I shouldn't have done last night, but too late now as it won't boot at all today. Thinking of just buying some old P4 system for 40 quid on eBay to browse the net on. Can't afford a proper replacement right now and using this PS3 browser must be what they use to torture you in Hell.
  • urban 24 Feb 2013 13:26:12 11,031 posts
    Seen 14 minutes ago
    Registered 11 years ago
    sounds like a motherboard issue to me
  • Aretak 26 Feb 2013 22:19:41 10,374 posts
    Seen 45 minutes ago
    Registered 12 years ago
    Huh. It's alive again. Despite spending hours trying unsuccessfully to get it to boot yesterday, I randomly pushed the button a couple of hours ago almost by mistake (had a senior moment and was wanting to check something online, forgetting the PC was dead), and after a brief period of the fans whirring up and down erratically, it booted. It's been running absolutely fine for the past couple of hours now.

    I guess that does confirm that it's whatever part of the motherboard that controls booting the system which is having problems, and the rest is fine. Now that I'm actually up and running there seem to be no issues whatsoever. I guess it'll be okay for now, as long as I don't have another brain fart and shut it down. Or get a power cut. Or burn alive when it catches fire during the night.
  • monkehhh 19 May 2013 14:40:29 3,476 posts
    Seen 1 hour ago
    Registered 6 years ago
    Similar issues for me.. once I get the PC on, it's fine if I don't turn it off. Reboots are fine, but a full power off will inevitably lead to it not switching on again. Drives and fans spin, but the graphics card isn't doing its 'loud fans for a few seconds' thing and no video signal gets to the screen. It doesn't attempt to post or anything and there's no beeps (even if I take the memory out, etc). It will probably work tomorrow if I unplug everything then try again tomorrow, but then I can't turn it off again. Pain I the arse to diagnose without spare bits.. and expensive. Knackered motherboard?
  • Dirtbox 19 May 2013 14:58:35 78,771 posts
    Seen 2 hours ago
    Registered 12 years ago
    Popped capacitors do that, they're essentially kickstarts for the transistors that control the voltages for the chips, one fails and you're done.

    +1 / Like / Tweet this post

  • monkehhh 19 May 2013 15:11:14 3,476 posts
    Seen 1 hour ago
    Registered 6 years ago
    Cheers. It'll happily still not complain when I remove the memory and the graphics card, so it would seem it is buggered. Stupid obsolete X58 / LGA1366 socket thingy!
  • monkehhh 19 May 2013 17:43:50 3,476 posts
    Seen 1 hour ago
    Registered 6 years ago
    Working again.. after a complete disconnection from power and a rest for a few hours.
  • monkehhh 25 May 2013 10:52:21 3,476 posts
    Seen 1 hour ago
    Registered 6 years ago
    Bought a new board on Ebay.. thanks to the seller for testing it after I'd paid (turns out it was faulty). So now I don't have a board on the way and I'm waiting for my money back from PayPal. Balls.
  • kentmonkey 27 May 2013 22:44:58 20,863 posts
    Seen 5 hours ago
    Registered 10 years ago
    Like a child I've just punched my laptop in anger and now the hdd won't read. It makes a sound like a really really quiet car revving.

    Have I lost everything? It started making that noise as soon as I twatted it and wouldn't respond. Turned it off and now won't boot. It's my only computer as well.
  • X201 27 May 2013 22:54:54 15,684 posts
    Seen 2 hours ago
    Registered 9 years ago
    You've probably caused the drive head to scrape the disc surface. Not good.

    Edit:Typo scrap scrape, either word would have done

    Edited by X201 at 22:58:01 27-05-2013
  • kentmonkey 27 May 2013 22:58:51 20,863 posts
    Seen 5 hours ago
    Registered 10 years ago
    Fuck it. Photos, work..... All lost. Haven't backed up for weeks. :-(
  • X201 27 May 2013 22:59:46 15,684 posts
    Seen 2 hours ago
    Registered 9 years ago
    Does it attempt to boot?
    Anything on screen?
  • kentmonkey 27 May 2013 23:01:43 20,863 posts
    Seen 5 hours ago
    Registered 10 years ago
    Laptop tries and then says there isn't a bootable disc connected.
  • X201 27 May 2013 23:04:15 15,684 posts
    Seen 2 hours ago
    Registered 9 years ago
    Hard drive it is then.

    :-(


    In future, drop it instead of punching it. Modern drives can detect when they're falling and park the disc in a protective mode.
  • Page

    of 6 First / Last

Log in or register to reply