No announcement yet.

Extreme 9 (some bugs and tests)

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

  • Extreme 9 (some bugs and tests)

    1. Vertex 2 (with latest FW update) will not work on the X79 SATAIII port (crashes on crystaldiskmark and HDTunePro error scan and generally any intensive disk activity like windows update) displaying a F4 BSOD.
    I found a workaround by using the Intel SATAII port instead. %100 stable after doing that no matter how much disk activity there is.

    2. The USB3 transfer rate is half what it should be. I have a LianLi EX-503 with a WD Green SATAIII drives installed in it and USB3 cable, and on a separate USB3 controller card, was getting transfer rates of around 70/80MBs copying files in explorer.
    Its now transferring at 40MB/sec (copying 446GB / 2hrs) using the motherboards USB3 ports.

    3. Bios P2.30 likes to make USB drives the primary boot device automatically, so if you install a new usb external device, reset the computer, it will try to boot the usb device and fail. you then have to go into bios and remove the usb device manually from boot priority, and the machine will turn off and on again after changing this setting alone, which is rather silly and unnecessary.
    Last edited by GunnzAkimbo; 09-07-2012, 04:54 AM.

  • #2
    Re: Extreme 9 (some bugs and tests)

    Can you verify that the 3.3v rail is at 3.3V? I have a X58 extreme6 that has a problem with the the 3.3v rail. Under load it drops to 2.69 at the PCIe connector. The PSU always measures 3.31V.

    Thanks,

    Ernie

    Comment


    • #3
      Re: Extreme 9 (some bugs and tests)

      Just a few comments, your board has six SATA III ports provided by two different Marvell SATA chipsets, and two SATA III ports provided by the Intel X79 chipset. It's well known that the Marvell SATA chipsets don't perform as well as the Intel chipsets do, particularly when their standard Marvell drivers are used. The Marvell chipset and driver pair is also not known for great compatibility. I've been using the standard Windows 7 AHCI driver, msahci, on Marvell chipsets, which improves their compatibility with no loss in performance. Of course, if you're using the Marvell chipsets in RAID mode, you can't use msahci. Are you sure the Vertex II was on the Intel SATA III ports, and not on the Marvell ports? Also, X79 boards use the Intel RSTe driver, which is not used by any other Intel chipset, which might explain the compatibility problem with the Vertex II.

      Your board has USB 3.0 support provided by two TI chips, which may not work as well as some other chips, or the native Intel USB 3.0 on their other 7 seried chipsets. Your ASR board has the "XFast USB" software feature, do you have that installed? I've read reviews of other ASR boards with this software, and when used without it, their USB 3.0 performance was lower than other boards. The XFast software can be a bit of a pain, but will help if you don't have it installed.

      The 2.30 BIOS thing with USB drives is a goofy bug, I can't see any purpose for that or BIOS setting that would fix it.

      The 3.3V power rail question of studiof, is likely not relevant, since this X79 board is completely different than a X58 board, and years apart in construction.

      Comment


      • #4
        Re: Extreme 9 (some bugs and tests)

        i was distinct at what port i was using. it might be the intel driver, but im using the exact same driver now on the SATAII connectors and its all fine now, but not when using the X79 SATAIII connector. i have 11 HD's and 1 optical. 6 on data drives on the marvell connectors, 2 data drives on the X79 SATAIII connector, 2 SSD's on the SATAII connector and 1 optical and 1 data drive on the other 2 SATAII connector. i'll just stick with the non "Xfast usb" transferring. why cant they optimize their drivers for usb3? voltages are top notch across the board for everything, and the system has been stringently stress tested for stability (this is how i found the F4 BSOD on the vertex - i thought the drive was dead by it's activity on the stress testing results, but installing it into another machine performed flawlessly).

        Comment

        Working...
        X