BSoD from just gaming to browsing, BCCode: 116

Page 2 of 3 FirstFirst 123 LastLast

  1. Posts : 10
    Windows 7 x64
    Thread Starter
       #11

    Probably irrelevant now, but I've only run skype with audio. Just ran furmark. Computer died when the hotter card reached 86-87 degrees. This worries me, since it has definitely been stable under slightly higher temperatures in the past, and the temperature hadn't seemed even close to peaking in the furmark. So one or maybe even both of the video card are essentially half-dead in some way :/

    Not too happy with that. They're both a little over a year old. Will do individual furmarks with 1 card at a time, which might rule out a power issue. Sorry for wasting your time with a vid card(s?) problem.

    Edit1; Ran FurMark on one video card (other video card is literally out of the comp). Let it run for 10 minutes, temperature maxed out at 94 degrees. I think(/hope) that this was the hotter card. Running the next one shortly.

    Last edited by johnratchet3; 09 Mar 2012 at 19:50. Reason: Results of test edited in to avoid dbl post
      My Computer


  2. Posts : 11,269
    Windows 7 Home Premium 64 Bit
       #12

    Not wasting my time. We're here to help. Let us know how things go with one card at a time. You may also consider running without the SLI bridge. Sometimes the bridge can fail, too.
      My Computer


  3. Posts : 10
    Windows 7 x64
    Thread Starter
       #13

    Thanks, that's really nice of you. GPU 1 results are above, and I just ran GPU 2 through 20 minutes of FurMark under the same conditions.



    There's a few interesting things I noticed. Firstly, neither vid card crashed. GPU 1 only went through 10 minutes of testing, but hey, in SLI the comp didn't even reach the 2 minute mark. This points to a power supply issue, perhaps one that's evolved from the PSU becoming aged and falling to a point where it can't run both GPUs under high load. This doesn't FEEL right to me, but it makes some sense.

    Something I noticed; GPU 1 has a slightly higher voltage of something; VDDC 1.012V against GPU 2's VDDC 1.000V. GPU 2 ran slightly cooler, would that explain a slightly different voltage? I ran them both in their native positions in the case, with the case open. This meant slightly different positioning, and both were on their usual supplementary power cables.

    I'm not sure how to proceed from here. Can't find the 2nd SLI bridge, dimly recall wondering where it was when unpacking one of the vid cards. Any ideas on what to investigate next?

    Edit; found the PSU model; Seasonic SS-850HT. It's 850W, and from memory, the amps are more than sufficient.
    Last edited by johnratchet3; 09 Mar 2012 at 21:57. Reason: Added info, avoiding dbl post
      My Computer


  4. Posts : 11,269
    Windows 7 Home Premium 64 Bit
       #14

    850 W is sufficient assuming it is running as it should. However, if the PSU is failing, it can start to exhibit odd behavior. The best way to find out is to swap out a PSU that is rated similarly and see if you still have the same problems. If you have any friends with a good PSU ~850 W or better, that would be the best test.

    A higher voltage will result in greater temperatures; the voltage increase is actually why overclocking requires better cooling. Not sure if voltages that are that small in difference should exhibit that behavior, but it may be an explanation. Depends on how much of a temperature difference we are speaking of. The 3 C in the pictures is pretty small and probably is attributed at least in part to the voltage difference. One card may also have more dust in it (you cannot always see the dust since it can get in the crevices) and the manufacturing cannot be exactly identical, so there are other contributing factors to temperature, as well.

    Can you run the two cards together without the bridge, or does it still crash?
      My Computer


  5. Posts : 10
    Windows 7 x64
    Thread Starter
       #15

    Times have changed it seems; can't activate SLI without the bridge. For the sake of it, ran another FurMark on GPU1 with GPU2 in place and idling in the background. TLDR; Disturbing temperature, surprising stability.



    10 minute FurMark, most of it was spent at 97 degrees. Typing this out in the same session; comp didn't crash. Problem is, can't do an SLI FurMark without an alternate SLI bridge. I found a rigid Gigabyte 3-way SLI bridge that doesn't fit, because it has 2 adaptors for each card. Looks like this;

    http://www.overclockzone.com/zolkorn...p/IMG_0238.JPG

    Until I find some way to have an alternate PSU test the comp, what do you think the odds are that cutting the bridge up the middle would result in a working alternate? xD I doubt I'd ever use 3-way SLI after experiencing the trouble behind 2-way, so I'm quite tempted try it out and see if it's just 2x tri bridge stuck together.
      My Computer


  6. Posts : 11,269
    Windows 7 Home Premium 64 Bit
       #16

    Not sure if that will work. I am concerned with how hot your cards are getting, though. They max at 104 C, so you are getting very close to that mark. Do they have any dust in them, or is it normal for them to run that hot?
      My Computer


  7. Posts : 25,847
    Windows 10 Pro. 64/ version 1709 Windows 7 Pro/64
       #17

    In my opinion you must cool both video card down. That is the first step.
      My Computer


  8. Posts : 10
    Windows 7 x64
    Thread Starter
       #18

    To be fair, FurMark is not exactly a normal experience for any GPU. To also be fair, when I look up reviews on the exact same vid card, and "load" temperatures don't even exceed 80 I find myself wondering if I even have the same thing (although load temps were recorded with a different test).

    EVGA GTX 460 1GB Superclocked EE (External Exhaust) Review - Page 17

    I haven't opened up either vid card (no idea how), but they've both been removed from the case a few times and dusted as best I can manage. Can of compressed air and a vacuum cleaner to do so.
      My Computer


  9. Posts : 11,269
    Windows 7 Home Premium 64 Bit
       #19

    It is not a good idea to use a vacuum cleaner on electronics. It can damage components. A can of compressed air or a low pressure compressor are the best tools for removing dust(by low pressure, I mean something that is designed for electronics and will not cause fans to spin faster than intended).

    The temperature of the graphics card will vary from motherboard to motherboard, as well. It may be yours is not providing the proper cooling controls. I could also be a power supply issue as you suspect, and that is carrying over to the cooling controls not working as they are intended to work.
      My Computer


  10. Posts : 10
    Windows 7 x64
    Thread Starter
       #20

    Sorry for the long delay in report, just haven't had any developments since last post. Gonna find time to take the comp down to the shop probably, since none of my friends have a power supply that's SLI safe, and let them know what's probably gone wrong. All things considered, between a simple cross-over cable, in perfect visual condition, and 2 video cards that held out individually at insane temperatures under FurMark, I'm pretty sure now that this is a power supply issue.

    Immense thanks to everyone who helped out this technical novice find out the cause of a very generic BSoD error. Especially to writhziden, who put up with my durpy self the whole way, you have my deepest thanks :) Been running on non-SLI until I can get the PSU examined, and likely replaced. Will report with confirmation once everything is running smoothly. My greatest thanks again :) everyone was really helpful and understanding.
      My Computer


 
Page 2 of 3 FirstFirst 123 LastLast

  Related Discussions
Our Sites
Site Links
About Us
Windows 7 Forums is an independent web site and has not been authorized, sponsored, or otherwise approved by Microsoft Corporation. "Windows 7" and related materials are trademarks of Microsoft Corp.

© Designer Media Ltd
All times are GMT -5. The time now is 05:04.
Find Us