No announcement yet.

Gigabyte F2A88XM-D3H AMD A88X

  • Filter
  • Time
  • Show
Clear All
new posts

  • #11
    ALC887 codec? At least use ALC892.

    Also, an other question: I have a lenovo laptop with kabini APU......but running "pwmconfig" to try to make a config file for fancontrol says: "There are no usable PWM outputs." Is this because the laptop doesn't let you control the fan speed at all? Or is that because linux hasn't added support for fan control for kabini APUs yet?..
    Laptops generally use ACPI to control the fan, so pwmconfig is the wrong approach...


    • #12
      Originally posted by GraysonPeddie View Post
      $80 not being too expensive? Hmm... I'd put $50 or under in a "not too expensive" category. And I bet Gigabyte released the motherboard since last year which did not support Kaveri yet.
      At that level, you're probably cutting cost on quality parts and essential features.


      • #13
        Originally posted by DanL View Post
        ALC887 codec? At least use ALC892.
        The higher end one: Does use ALC892, its just the mini-ATX one that doesn't.


        • #14
          Originally posted by Ericg View Post
          The higher end one: Does use ALC892, its just the mini-ATX one that doesn't.
          It's micro-ATX, and the form factor shouldn't dictate that you use an older, inferior audio codec unless you're shooting for bargain basement pricing in the $50-$60 range...


          • #15
            Such a pity, I cannot find any PCIe 3 boards compatible with fx-8350. So it's all APU now... hmm :/


            • #16
              Originally posted by Sonadow View Post
              So Gigabyte sells a board which is Kaveri compatible, but ships with an outdated UEFI that does not support Kaveri, although Kaveri support can be added by flashing the firmware to the latest version.

              Way to go Gigabyte.
              It's called store/warehouse backstock. The board received by Larabel was shipped for Trinity and Richland, since Kaveri hadn't been released yet there was nothing for Gigabyte to test their BIOS/UEFI against.

              This is why every system builder buys the cheapest of the new socket launch chips., so that during transitional periods to newer processor architectures on the same socket you can put it in the board to update the BIOS.

              You'd know this if you built comps or did comp repair.


              • #17
                Originally posted by schmidtbag View Post
                This is a pretty common issue with AMD boards of any brand - it's basically the side effect of them trying to allow backward compatiility. When AM3+ was released, it required a BIOS update with a non-FX CPU to upgrade it, including some of the black socket boards.

                Personally, I lucked out because I have a 3-year-old white socket motherboard that happens to be AM3+ compatible, and I used my old Athlon II to do the upgrade. I'm not sure what other people do when they buy a new board that doesn't ship with the upgrade and when they don't have an older CPU.
                Yeah, Intel "fixed" this problem by releasing a new socket for every change to their CPU architecture. Everything LGA775 and older had the same issue of mobo chipsets being able to run a new chip design, but the boards already shipped not having the necessary BIOS because they are sitting on a shelf somewhere other then the factory.


                • #18
                  Originally posted by Rexilion View Post
                  Such a pity, I cannot find any PCIe 3 boards compatible with fx-8350. So it's all APU now... hmm :/
                  PCIe3 doesn't help you unless you are doing 3+ card crossfire.

                  No, the FX-9590 is the last CPU and the 990FX the high end for the AM3+ socket, with the Gigabyte GA-990FXA-UD7 the most high end mobo

                  AMD's roadmap says it's all APUs from here on out with the focus on HSA. They've partnered with most all of the ARM SoC manufacturers to ensure the concept gets entrenched.


                  • #19
                    Originally posted by edgar_wibeau View Post
                    Baconmon, you should check the question mark key on your keybord, seems to be broken. If your Lenovo is a ThinkPad, try thinkfan. Works fine for my E145 (Kabini aswell). Oh, and it's all in the article, Kaveri sensor is not in the kernel yet, but will be there in 3.14 which is the next release and not "some time in the future". Accordingt to the article.
                    Oh wow, actually mine is the E145 also......and so I researched how to use thinkfan, and now it is working perfectly I guess, so thanks.. Before, I was just trying to use "fancontrol" instead of "thinkfan".......and I guess fancontrol can't really do much to control this laptop's fan..but thinkfan controls it perfectly.. so yay..


                    • #20

                      Originally posted by Baconmon View Post
                      Does the sensor exist? But just not supported by linux kernel yet?.. Meaning, some time in the future it will be supported?..
                      I havea F2A88X-D3H right now - thats the full atx version and it has an IT8620 monitoring chip that is unsupported under Linux. Now if the XM uses the same chip, its a problem since the lm-sensors guys stated that ITE stopped sharing diagrams with them and thus new ITE chips ( are unlikely to be supported. I sent a mail to the kernel developer in charge of the it87 module and he said he didnt even hear about this chip model.
                      Previously i had a F2A85X-D3H mobo that has a supported ITE 87xx monitoring chip and i supposed this one will use that too but it seems i was mistaken....

                      But anyway, the mobo works fine, i have it for a few months now and had no issues with it (i have a A8-6500 APU with it). Thermal monitoring is still possible for both the CPU and IGP, but the reportred values from the k10temp (CPU) and radeon (GPU) drivers are not real ones especially when idle - instead they are used internally for thermal throttling and they approach real values when the chip is under load for some time (for example i have 3-5C reported when idle).