Announcement

Collapse
No announcement yet.

ASRock 870 Extreme3 + Ripjaws F3-12800CL9D-4GBRL

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

  • ASRock 870 Extreme3 + Ripjaws F3-12800CL9D-4GBRL

    Edit: Make sure you go to AMD Overdrive to get memory timings for the individual timings (not just the regular 4 timings), check in your BIOS if the auto clocks are too tight. If they are, loosen them to AMD Overdrive's timings.

    I am testing a budget gaming build I'm making for a friend and I was having problems with getting these RAM to stock speeds (1600mhz 9-9-9-24-33-2T @ 1.5v), but after a few hours of reading and testing, I was able to get it to work. I just wanted to report this to the forums in case people who have this RAM and/or board are having difficulties reaching rated settings.

    For owners of this board, the RAM voltage in the BIOS jumps from 1.477v to 1.544v, so there was no way for me to get that rated 1.5v. What I did instead was use the higher of the two settings and (later on) even higher voltages (1.544v-1.648v). Using these voltages resulted in getting errors in memtest or even BSoDs during boot up.

    One suggestion was turning off the USB3 controller and that did seem to help a bit, memtest would stabilize more, but would still error. Another thread in a different forum said to use the lower voltage (1.477v) and that seemed to do the trick for these RAM. Windows Memory Diagnostic had no reported errors and memtest cleared each stick separately and both sticks together when tested with 200% clearance runs. I did use the outer white slots (B1/B2), so it's worth switching RAM slots if there are any complications with the blue slots (A1/A2).

    Here are my system parts and settings, tested in an open bench environment:

    ASRock 870 Extreme3 (770chipset) | BIOS ver: 1.60 | NB freq @ 2600mhz | NB-CPU voltage at 1.30v
    Phenom II x4 955 C2 stepping 3.6ghz @ stock voltage (1.35v)
    4GB G.SKILL Ripjaws F3-12800CL9D-4GBRL (1600mhz 9-9-9-24-33-2T @ 1.477v) It's important you manually set the ALL the timings, if your motherboard's auto timings are too tight.
    Gigabyte HD4870 512mb (flashed with Asus TOP BIOS)
    WD Caviar Black 1TB (32MB cache version)

    Note: This motherboard and RAM were not in the compatibility list of either part, but were still able to work.

    I've also tested this board with my own parts and everything worked great:

    4GB Mushkin Blackline RAM (rated for 1600mhz 7-7-7-20-27-1T @ 1.95v)
    Sapphire HD4890 1GB Toxic

    When I get my MSI 790FX-GD70 mobo back from RMA, I will test this RAM with that and see if I can get better results. My mobo is on the approved motherboard list for this RAM, so I have high hopes for better compatibility without the hours of troubleshooting. I'll make a second post about this when that happens.

    Cheers and good luck!
    Last edited by Kokin; 12-07-2010, 11:45 PM.

  • #2
    So when it worked ok, was USB3.0 disabled or did you re-enable it?
    Cheers.

    Comment


    • #3
      I'm not sure why you want to run a high voltage.

      I have 1.5v spec. ram and let the bios run at <auto>. I can run it at 1600 no worries and I see CPUID hardware monitor reports 1.55v.

      If it works at 1.5v, I'd be happy as!

      Comment


      • #4
        I have left USB 3.0 as disabled, I will try to test it with it enabled later today, after I come back from a final at school.

        For the voltage issue, when leaving it on auto, it automatically goes to 1.55v and the RAM isn't stable with that voltage or higher, whereas if I manually set it to 1.48v, it will stabilize and will work without error. This motherboard only allows for 1.48v and lower and then jumps up to 1.55v and higher without any middle ground inbetween, so I couldn't just leave it at auto.

        Comment


        • #5
          Let us know how it goes, the USB 3 controller issue is rather interesting


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

          Tman

          Comment


          • #6
            I've owned several ASRock boards in the past, but never, never again.
            And the main reason: the voltage control of the boards are absolutely terrible.
            And yes, you are forced to over volt because the voltage fluctuates like crazy.
            Coming off a 790GX ASRock board to my Crosshair IV is like night and day difference.
            AMD Phenom II X6 1090T@ 4.2GHz
            Corsair H50 Hydro (push/pull intake fans)
            ASUS Crosshair V Formula
            2x4GB G.Skill RipjawsX@ 1975MHz, 9-10-9-28 (2T)
            SLI: 2x EVGA GTX 570's@ 902/1804/2032
            Creative SB X-Fi Xtreme Gamer
            Corsair HX850W Modular
            Cooler Master HAF 922 (200mm side fan)
            2x Win 7 Home Premium (x64)

            Comment


            • #7
              Memtest with 2x2GB (9-9-9-24-33-2T @ 1.477v) at the outer slots has 0 errors so far, with 60% clearance. This is with USB 3.0 enabled, so the problem might have just been the voltages.

              Originally posted by YourNemesis007 View Post
              I've owned several ASRock boards in the past, but never, never again.
              And the main reason: the voltage control of the boards are absolutely terrible.
              And yes, you are forced to over volt because the voltage fluctuates like crazy.
              Coming off a 790GX ASRock board to my Crosshair IV is like night and day difference.
              Yeah, I set my voltage for 1.35V in BIOS and during idle, it reports 1.35V, but at load it goes up to 1.4v! It is the total opposite of my MSI 790FX-GD70. My board's idle at stock voltage would be 1.347V and load would go down to 1.33ishV.

              I would have bought something else, but the board fit the budget during Black Friday ($80) and it was one of the very few AM3 boards that supported dual gpus @ 8x/8x speed vs 16x/4x. My friend isn't very technical with the computer hardware and his main purpose will be gaming, not benchmarking, folding, or going for that max overclock, so as long as this motherboard is stable at slightly OCed settings, both he and I would be satisfied.
              Last edited by Kokin; 12-06-2010, 02:25 PM.

              Comment


              • #8
                Well memtest was going great for a few hours, but errored once at 57% and errored twice at 64%, so I'm not going to go any further. Looks like the USB 3.0 controller does destabilize RAM.

                Edit: Hmm... redid memtest with both sticks with USB 3 disabled and I errored right away at 5%, but I can do blend testing in Prime95 without errors... Also Windows Memory Diagnostic clears the memory of having any problems. This just might be a case of incompatible RAM since it isn't in the motherboard's QVL. Going to try downclocking to 1333mhz with 8-8-8-22-2T timings and see if that works.
                Last edited by Kokin; 12-06-2010, 02:50 PM.

                Comment


                • #9
                  Might try raising NB to 1.3


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

                  Tman

                  Comment


                  • #10
                    Got it stable at 1600mhz CL9 at 1.48v! The problem was not the voltages or USB3, it was the Auto timings. Most of them were a click or two quicker than it should be.

                    I used AOD to find the CL9 timings for TRP, TRRD, TRCD, TRAS, TWR, TWTR, TRTP, TRC, TCL and manually set it in the BIOS menu. Tested it with USB3 enabled and disabled and it was able to clear 300% clearance in memtest. Had it tested overnight and during school, to fully make sure it was stable. I hope this solves peoples problems.

                    Btw, I did not have to touch NB voltages, since this mobo only has CPU-NB voltages. I had NB frequency at 2600MHz with 1.3v
                    Last edited by Kokin; 12-07-2010, 09:30 PM.

                    Comment


                    • #11
                      Hope so, let us know how it turns out


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

                      Tman

                      Comment


                      • #12
                        gskill

                        I will test it again overnight with USB3 disabled and during school with USB3 enabled to see if there's any problems, if there aren't any, then it's all down to the motherboard just putting the timings way too tight when set on Auto.

                        I normally don't look at what they are set to when in Auto, but I noticed some of them were at the first or second option. i.e. - Auto was set @ 4 or 5 clocks vs 6clocks specified by the RAM's timings table in AMD Overdrive

                        Comment


                        • #13
                          We'll be looking for the results, good luck


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

                          Tman

                          Comment


                          • #14
                            Yep, both tests with USB 3 enabled and disabled cleared 250-300% in memtest. In my case, the auto timings were the problem. Will look forward to it's overclocking potential.

                            Comment


                            • #15
                              Have fun, keep in touch w/ results


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

                              Tman

                              Comment

                              Working...
                              X