Results 1 to 3 of 3

Thread: OCCT GPU test

  1. #1
    Joined
    Jan 2006
    Posts
    204

    OCCT GPU test

    So I have two Radeon HD3850 512MB cards, both overclocked to 730/850. One of them is made by ASUS with the Glaciator cooler, the other made by Sapphire with the stock cooler. While gaming, both cards will generally be around 75c.

    I just tried the OCCT GPU test for the very first time and I was shocked. Both cards started climbing.. 70c.. 80c.. 90c.. for the first time EVER, I heard the Sapphire card's fan spin up. Eventually, the ASUS card reached 100c and it was still climbing (its fan always runs at the same speed) while the Sapphire card was a little cooler at 98c, somewhat stabilized. At this point I turned it off because I didn't want to risk damaging my cards. I'm not sure if I should lower my overclocks, technically the cards can overheat but in reality this has never happened. In fact, before today I wasn't even aware the Sapphire card had fan control..

    I never knew this test would stress my GPUs that much more than any game I have ever tried before. A lot of reviews I read use games to determine the load temperatures of graphics cards. Does this mean a lot of the high end cards with load temps of ~90c would just overheat during the OCCT GPU test? In my opinion this test should be used by reviewers. I don't even want to know what will happen if you submit a GTX480 to OCCT, it would probably BSOD after 5-10 minutes.

    Has anyone else experienced similar results with this test, or is it just me? 25c higher than any 3d game and still climbing is quite a lot, I almost couldn't believe my eyes.

  2. #2
    Joined
    Oct 2009
    Location
    West Lafayette, IN
    Age
    30
    Posts
    79

    Re: OCCT GPU test

    All these benchmarks like FurMark and OCCT really just aren't realistic anyway - there is absolutely no real-world application that will generate the same sort of load that they will artificially generate. This is quite simply why most reviews don't bother using them for their load temp charts.

    Modern cards such as the GTX400 series and the Radeon HD 5000 series have protection mechanisms against this kind of thing.
    For Cypress, AMD has implemented a hardware solution to the VRM problem, by dedicating a very small portion of Cypress’s die to a monitoring chip. In this case the job of the monitor is to continually monitor the VRMs for dangerous conditions. Should the VRMs end up in a critical state, the monitor will immediately throttle back the card by one PowerPlay level. The card will continue operating at this level until the VRMs are back to safe levels, at which point the monitor will allow the card to go back to the requested performance level. In the case of a stressful program, this can continue to go back and forth as the VRMs permit.
    [Source Anandtech]
    Not sure how the GTX400s handle the protection....but I would assume NVIDIA has taken it into account....I hope...


    But I heard that people running these stress apps on the RV770 in particular was very troublesome...
    The overcurrent situation they created was too much for the VRMs on many cards, and as a failsafe these cards would shut down to protect the VRMs. At a user level shutting down like this isn’t a very helpful failsafe mode. At a hardware level shutting down like this isn’t enough to protect the VRMs in all situations. Ultimately these programs were capable of permanently damaging RV770 cards...
    Last edited by Yangorang; 04-14-2010 at 11:21 AM.

  3. #3
    Joined
    Nov 2002
    Location
    In bed with one of my avatar AMD girls :D
    Age
    39
    Posts
    8,876

    Re: OCCT GPU test

    Quote Originally Posted by Yangorang View Post
    All these benchmarks like FurMark and OCCT really just aren't realistic anyway - there is absolutely no real-world application that will generate the same sort of load that they will artificially generate. This is quite simply why most reviews don't bother using them for their load temp charts.
    OCCT gpu test is a lot better then furmark. I have seen it detect errors where furmark wouldnt.
    As far as them being realistic, I disagree. They show you what your cooling can do by acheiving max temp. And as for stability it seems to be 99.9% accurate. I dont think I can recall an OC passing OCCT that wouldnt pass with hours of actual gaming.

    This is why Im a OCCT fanboy

    Both its CPU and GPU tests I see superior to all others, catching errors prime wont

Posting Permissions

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