Announcement

Collapse
No announcement yet.

Help! Ripjaws X F3-12800CL9D, constant lockups and restarts while gaming

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Help! Ripjaws X F3-12800CL9D, constant lockups and restarts while gaming

    Help!

    I am experiencing non-stop (as many as 4-5 per hour) lockups/crashes when gaming. Sometimes the lockups are sporadic (can go a day or two lucky,) usually get a few per day, and sometimes am mercilessly crashing all day. The system is generally stable outside of gaming (only frozen a few times when not gaming in the last year) and is able to POST reliably as well. The lockups are never BSODs, just the system freezes and exorcist noises come out of my speakers. Sometimes the system will then reboot after 15 seconds or so, but most of the time I have to hard reset the computer. Games affected have been... all of them: Left4Dead2, Skyrim, Assassin's Creed: Revelations, Starcraft 2, Battlefield 3, Star Wars: The Old Republic, Mass Effect 2 & 3, etc etc.

    My specs are as follows:
    RAM: 16GB (4x4GB) Ripjaws X F3-12800CL9D-8GBXL (Running with XMP @ 1600MHz, timing: 9-9-9-24-2T)
    CPU: i5 2500K @ 4.2 GHz (have experimented with the stock 3.3GHz up to 4.2 and everything in between)
    Motherboard: Gigabyte P67A-UD3P-B3 rev 1.1 running latest F9 BIOS (have experimented with previous builds since F4 to no avail)
    GPU: Gigabyte 560GTX Ti OC (have swapped in a Radeon HD3850 and Geforce 8800GT as well)
    SSD: Intel 510 120GB (latest firmware)
    PSU: Antec Truepower New 650W
    OS: Windows 7 Ultimate SP1 64bit
    GPU Driver: 301.42
    DirectX 11.0

    I am 99.999% certain it is the memory that is posing the issue as I have swapped or RMA'd every component listed above except the RAM (I had erroneously assumed until now that the RAM was probably fine because it had passed a couple overnight runs of memtest86.) I have also tried various permutations of fresh OS installations and various driver versions with no success. CPU and GPU both pass prime95/Furmark respectively with zero issues, and temperature monitoring/logs over the last year show that temperatures are stable and not a cause of the crashes I'm experiencing. Windows Event Viewer/Log also shows nothing particularly crazy going on.

    As far as what I've tried with the memory, as per other support posts on the forum:
    -XMP/no XMP (no difference)
    -Disable XMP and manually set timings to 9-9-9-24-2T (despite the status screen showing that the motherboard is reading the XMP correctly)
    -Enable XMP and manually set timings/voltages
    -Set DRAM voltage to 1.55-1.56V
    -Set VTT to 1.13V

    None of the above has worked. As of today I have tried dropping from the rated 1600MHz to 1333MHz (kept 9-9-9-24-2T timings and increased DRAM/VTT voltage) to see if I can get any stability at all but I won't know if it has helped until I've managed to go a day or two crash free at least. Even then, the memory would be failing to meet its specified performance of 1600MHz.

    Please help! I've been pulling my hair out for a year and have gone without my computer for weeks at a time while waiting to RMA various parts to no avail. With my thesis due very soon I can't afford to go without the computer again!

  • #2
    mVidia has a new beta out driver out 304.79 I've just been looking at, might give it a try

    http://downloads.guru3d.com/download.php?det=2947


    Pls offer comments on support I provide, HERE, in order to help me do a better job here:

    Tman

    Comment


    • #3
      I will of course install the latest driver, but it makes no difference -- the same issues occurred when using completely different drivers/gpu (Radeon 3850HD, ATI Catalyst.)

      Comment


      • #4
        Yep, still crashing

        Comment


        • #5
          May well be the sticks, easy way to check would be if you have any or can borrow a couple similar sticks and try them


          Pls offer comments on support I provide, HERE, in order to help me do a better job here:

          Tman

          Comment


          • #6
            Update:
            While the beta driver didn't help, it looks like things are stable at 1333MHz. I'll report back later today to confirm. I will not be able to get similar sticks for a couple of days most likely, if at all.

            If indeed it is stable at 1333MHz, what would the next step be to get it working at the rated 1600MHz? Or would the conclusion simply be that these were bad/poorly binned/whatever sticks and that I have to RMA?

            Edit:
            I forgot something in my OP: while none of the memtest86's that I ran this year threw up any errors or alarms, it did lockup once. No reboot, just freeze needing hard reset. I had dismissed it at the time because I hadn't yet tried swapping CPU or Mobo to eliminate those (which I have since done.) Will start leaving it running again whenever computer is idle and see what happens.
            Last edited by Tobywoby; 07-16-2012, 10:07 AM. Reason: Forgot something

            Comment


            • #7
              I normally try raising the memory voltage and the memory controller voltage, if it runs stable by adding some voltage and doesn't get to hot, may be a minor flaw somewhere

              Comment


              • #8
                Try maybe DRAM + .05 and raise VCCIO up about + .07


                Pls offer comments on support I provide, HERE, in order to help me do a better job here:

                Tman

                Comment


                • #9
                  Update:

                  Tradesman: the increased voltage still resulted in games crashing when the ram is run at 1600MHz and 9-9-9-24-2T timings. 1333MHz, on the other hand, has still yet to fail under any circumstance.

                  Was able to get 2x4GB identical sticks (had a friend who built a pc a couple months after mine and has near identical specs, with identical motherboard and memory. His pc is rock solid by the way, with zero crashes or freezing in a year of pretty heavy use.)

                  I've started using Linpack testing with IntelBurnTest, with memory usage set to maximum. This has proven to be the fastest and most reliable way to sort out which configurations work and which don't, and is certainly faster and less random than playing video games for hours. Minimum test is 5 passes, overnight 'final confirmation' test is 25 passes. Results so far are as follows:

                  With my memory:
                  -1333MHz/stock timings/voltage: Pass
                  -1333MHz/stock timings/voltage, overnight 25 trial test: Pass
                  -1600MHz/stock XMP: fail
                  -1600MHz/XMP + 1.6V DIMM voltage, 1.13V VTT: fail
                  -1600MHz/XMP + 1.58V DIMM, 1.13V VTT: fail
                  -1600MHz/XMP + 1.56V DIMM, 1.13V VTT: fail
                  -1333MHz/stock timings/voltage: Pass (repeated this test again just as a sanity check)

                  With friend's memory:
                  -1600MHz/XMP, stock timings/voltage: pass
                  -1600MHz/XMP, stock timings/voltage, overnight 25 trial test: pass
                  -1600MHz/XMP, stock timings/voltage, 25 trial test, alternate memory banks (slots 2/4 instead of 1/3): pass

                  So what's next?
                  Last edited by Tobywoby; 07-18-2012, 04:00 PM. Reason: Results updated

                  Comment


                  • #10
                    May be time for a RMA or if within the time limit a defective exchange with your re/e-tailer


                    Pls offer comments on support I provide, HERE, in order to help me do a better job here:

                    Tman

                    Comment


                    • #11
                      Updated results, alternate sticks passed the overnight test too. Bought the memory a year ago so well outside retailer exchange period... Looks like I'll have to RMA then...

                      Can you give me any tips on speeding the process along as quickly as possible? And, do I send back the whole kit(s) (I bought 2 8GB kits) or do I need to figure out exactly which stick(s) are defective?

                      Comment


                      • #12
                        Think I'd contact the RMA folks and see what they suggest, rma@gskillusa.com . Cross ship might be a possibility if you provide a credit card # to ensure the return of the bad sticks, but I just don't know, some companies do it that way, but I've never had an RMA to GSkill where I needed it ASAP (in fact only had 2 over the years and I've bought well over 150 of their products), so never looked into it


                        Pls offer comments on support I provide, HERE, in order to help me do a better job here:

                        Tman

                        Comment


                        • #13
                          Alrighty, thanks Tradesman for your help, has sped along the testing process.

                          After further testing stick by stick, it appears that just one out of the four is causing the problem.

                          Off to RMA!

                          Comment


                          • #14
                            Keep in touch with your progress, I'll be around


                            Pls offer comments on support I provide, HERE, in order to help me do a better job here:

                            Tman

                            Comment


                            • #15
                              Sending off the RMA today, strange twist though: not 1 but 2 sticks are bad. Other 2 appear fine.

                              Comment

                              Working...
                              X