Arch on a Gigabyte 990FXA-UD3 mobo - 3 [SOLVED], 1 given up on

Yesterday I was able to upgrade my main machine from an old 2-core AMD processor to a Gigabyte 990FXA-UD3 with an AMD FX8350 CPU and 2 sticks of Corsair Vengeance DDR3 1866 MHz 4GB RAM.  I chose that mobo because of my need for both a PCI slot and a Firewire port for various audio gear.  My storage, PSU and GPU were staying the same.
It went pretty well, all things considered, but I ran into 4 issues along the way.  This post is to document my fixes and to see how close I am with one last issue.
1) At first boot, the machine gave me the Syslinux menu but wouldn't boot into Arch.  This was sort of expected.  The fix was straightforward:
  - Boot from USB
  - mount my / and /boot partitions as /mnt and /mnt/boot
  - run "arch-chroot /mnt"
  - run "mkinitcpio -p linux"
  - reboot
Now that I could boot into my own environment, I was ready to.... figure out why my mouse and keyboard weren't responding after getting past the BIOS screen.  It's a USB keyboard and mouse, and though they both worked fine in the BIOS environment, as soon as Syslinux took over, they were unavailable.  I had a PS/2 keyboard available, so that went on to allow me access.  Now that I could log in, I could Google the issue and see....  wait, eth0 is there but not getting a DHCP address.  OK, hard-code an IP and... I still can't ping my router.  That isn't good.  So, onto a laptop to work this one out....
2 & 3)  No network and no USB
This was the hardest to work out.  I found that my USB3.0 ports were fine, but the 1.1/2.0 ports were not.  I also saw that the mobo uses a Realtek RTL8111/8168/8411 Ethernet controller.  All sorts of playing around building the r8168-all package (sneakernetted via USB key from the laptop) from AUR didn't help.
The fixes for these happened simultaneously, so I've grouped the solutions together here.
  - Get into the BIOS, go to Peripherals
  - Change these items from the default:
    EHCI hand-off - ENABLE
    IOMMU controller - ENABLE
    Get to the boot menu, hit TAB to edit the boot options, add iommu=pt and boot up
    Edit your boot configuration file (mine's /boot/syslinux/syslinux.cfg but GRUB users will change /boot/grub/menu.lst) to add the iommu=pt definition.
Now there's only one thing to look at - the RAM speed.  This is where I'm really out of my element.  I'm not an overclocker so I don't know much about latency and timings and bus speed multipliers.  I have 2 sticks, and as recommended these went in alternating slots to enable Dual Channel mode.
I saw in the BIOS that the RAM was defined for 1333MHz, not what I'm looking for.
From the BIOS, I went to MIT -> Advanced Memory Settings, and I fond I could change the Extreme Memory Profile (XMP) from Disabled to Profile1.  (TIL that the XMP is stored on the UDIMM and lets you automatically set timings etc. to the defined profile.  Nice.)  Profile1 sets the RAM to run at 1866MHz, and all looked great at the BIOS level.
When I boot into Arch and run various utilities, I don't see what I'm expecting:
[jh@xtc ~]$ sudo lshw -c memory -short
H/W path                Device     Class       Description
==========================================================
/0/0                               memory      64KiB BIOS
/0/4/5                             memory      384KiB L1 cache
/0/4/6                             memory      8MiB L2 cache
/0/4/7                             memory      8MiB L3 cache
/0/2c                              memory      8GiB System Memory
/0/2c/0                            memory      DIMM Synchronous [empty]
/0/2c/1                            memory      4GiB DIMM DDR3 Synchronous 667 MHz (1.5 ns)
/0/2c/2                            memory      DIMM Synchronous [empty]
/0/2c/3                            memory      4GiB DIMM DDR3 Synchronous 667 MHz (1.5 ns)
[jh@xtc ~]$ sudo dmidecode --type 17
           *** snip ***
Handle 0x0034, DMI type 17, 34 bytes
Memory Device
        Array Handle: 0x002C
        Error Information Handle: Not Provided
        Total Width: 64 bits
        Data Width: 64 bits
        Size: 4096 MB
        Form Factor: DIMM
        Set: None
        Locator: Node0_Dimm3
        Bank Locator: Node0_Bank0
        Type: DDR3
        Type Detail: Synchronous Unbuffered (Unregistered)
        Speed: 667 MHz
        Manufacturer: Corsair           
        Serial Number: 00000000   
        Asset Tag: Dimm3_AssetTag
        Part Number: CMZ8GX3M2A186
        Rank: 1
        Configured Clock Speed: 933 MHz
Now, I know from reading a lot of forums that because of Dual Channel and multipliers, it can be hard to figure out what the actual RAM speed is.  I *might* be there already, since 933*2 = 1866, the number I'm looking for, and I've seen at least one post saying that this would be the expected number, but I'm just not sure.  I tried pulling one stick to disable Dual Channel, but got the same result as above.
I've also gone in and bumped the multiplier from 9.33x to 10.66x, which according to the BIOS gives me 2132MHz.  In that state, lshw still reports the DIMM at 667MHz, while dmidecode gives me "Configured Clock Speed: 1066 MHz" but at least it's different, even if it is 1/2 the number I was expecting.
So now I'm working on trying to find out if dmidecode is reporting numbers incorrectly, or if I'm missing something in correctly configuring my system.  All pointers to knowledge of this are greatly appreciated!
Hopefully this post will at least let another Gigabyte mobo user avoid some of the pain I went through on this.
Last edited by JoeHartley (2014-01-16 13:45:55)

A followup here - I've learned more about this board, and have changed a couple of things in my setup.
I found that even with IOMMU enabled, the network port seemed flaky.  Neither the Realtek driver in Arch nor the one inAUR helped; the net would hang at times for brief moments but there was nothing I could put my finger on.  I happened to have an Intel network card handy so I dropped that in and disabled the onboard LAN.  It's been rock-solid.
I've also tried connecting an eSATA interface to the box to make a set of backup disks, but I never saw the drives in parted and there are errors like this:
Jan 15 21:27:36 xtc kernel: [   12.434714] AMD-Vi: Event logged [IO_PAGE_FAULT device=04:00.1 domain=0x0000 address=0x0000000221720450 flags=0x0070]
Some Googling showed these errors are related to the IOMMU.  Since I'm no longer using the onboard LAN, I disabled IOMMU, which brought up the drives  in the eSATA interface.  Downside is I lose the USB2.0 ports, but the USB3.0 ports work, and two USB ports are enough for now.  The Firewire port still works, so I'm at a stable place with this configuration.
I think this is a great looking mobo and it's screaming along, but I only give it 7/10 for Linux compatibility.

Similar Messages

  • Installing Arch on Asus Eee, Already Read the Wiki [Solved]

    I'm attempting to install Arch on my Asus Eee 701 (4GB SSD model) and I made a mistake that I'm having trouble correcting.  First, I accidentally wrote the image to my Eee's SSD, instead of my USB drive.  I tried to install Arch with the installer on my SSD, set up the partitions (one for /, one for /boot, and one for /home), but when I issued the command, it told me that the disk was in use.  It was then that I realized what I did.  I used an Ubuntu LiveCD on another computer to burn the image file again, this time to my USB drive.  I booted my Eee from the USB drive and tried to partition my SSD again, but this time, it tells me:
    Device or resource busy
    This disk is currently in use - repartitioning is probably a bad idea.
    Unmount all file systems, and swapoff all swap partitions on this disk.
    Use the --no-reread flag to suppress this check.
    Use the --force flag to overrule all checks.
    When I boot from my SSD, I get the same message, so I'm not even 100% certain that I booted from my USB drive before.  My USB drive's activity light was on though.
    Is there any way to fix this?  Also, given that I have an SSD, what partitions should I set up and how big should they be?  I can't find any concrete numbers in the wiki.
    Thanks in advance for the help.
    Last edited by Tyrian (2008-11-17 04:14:23)

    I'll try that, thanks.  What would happen if I dismounted the SSD with -f if it was in use?
    EDIT:  I just realized that when I go into the installer, it lets me set up partitions manually (I run the installer, pick "Prepare Hard Drive", "Partition Hard Drives", "/dev/sda").  But how should I set them up?  When I test it by just making one 4GB partition, it tells me, "Wrote partition table, but re-read table failed.  Reboot to update table."  So I rebooted, and it appears to work.  So I deleted that partition, made a 512MB one, rebooted, and repeated until I had 3 partitions in ext2, a 512MB (sda1, bootable), 512MB (sda2), and one roughly 2.5GB (sda3).  I then went to "Set Filesystem Mountpoints", said "NONE" for the swap (due to the avoiding pitfalls section of the wiki), picked "/dev/sda1" as "/" in ext2, made "/dev/sda2" as ext2 called "/var", and made "/dev/sda3" in ext2 called "/home", then said "DONE".  When I say "Yes" as the confirmation prompt, I get this message:
    Error creating filesystem: mke2fs /dev/sda1
    I hit enter and get dropped back to the "Prepare Hard Drive" menu.  I followed this section of the Beginner's Guide too.
    Is there any way to fix this?
    Last edited by Tyrian (2008-11-15 18:37:10)

  • Linux 3.14.2-1 ARCH config error - will not build with ABS [SOLVED]

    Hi Guys,
    I need to compile my kernels with ABS because I need 16 DVB adaptors instead of 8. This works normally just fine. Today I tried to compile newest kernel and got following error during config. For the record 3.14.1-1 built just fine.
    Makefile:615: Cannot use CONFIG_CC_STACKPROTECTOR_STRONG: -fstack-protector-strong not supported by compiler
    SYSHDR arch/x86/syscalls/../include/generated/uapi/asm/unistd_32.h
    SYSHDR arch/x86/syscalls/../include/generated/uapi/asm/unistd_64.h
    SYSHDR arch/x86/syscalls/../include/generated/uapi/asm/unistd_x32.h
    SYSTBL arch/x86/syscalls/../include/generated/asm/syscalls_32.h
    SYSHDR arch/x86/syscalls/../include/generated/asm/unistd_32_ia32.h
    CHK include/config/kernel.release
    SYSHDR arch/x86/syscalls/../include/generated/asm/unistd_64_x32.h
    UPD include/config/kernel.release
    SYSTBL arch/x86/syscalls/../include/generated/asm/syscalls_64.h
    WRAP arch/x86/include/generated/asm/clkdev.h
    CHK include/generated/uapi/linux/version.h
    UPD include/generated/uapi/linux/version.h
    CHK include/generated/utsrelease.h
    UPD include/generated/utsrelease.h
    HOSTCC arch/x86/tools/relocs_32.o
    HOSTCC arch/x86/tools/relocs_64.o
    HOSTCC arch/x86/tools/relocs_common.o
    HOSTLD arch/x86/tools/relocs
    CC kernel/bounds.s
    gcc: error: unrecognized command line option ‘-fstack-protector-strong’
    /home/server/core/linux/src/linux-3.14/./Kbuild:35: recipe for target 'kernel/bounds.s' failed
    make[1]: *** [kernel/bounds.s] Error 1
    Makefile:884: recipe for target 'prepare0' failed
    make: *** [prepare0] Error 2
    ==> ERROR: A failure occurred in prepare().
    Aborting...
    Thanks for the help
    Last edited by wdirksen (2014-05-02 20:31:32)

    wdirksen wrote:...I just moved to New Zealand...
    Beautiful part of the the world... you just need to watch out for Saruman and his orc army
    wdirksen wrote:... and the packages in the NZ mirror are not all in sync at the same time.
    Use reflector....
    % which upp
    upp: aliased to reflector -c "United States" -a 1 -f 3 --sort rate --save /etc/pacman.d/mirrorlist && cat /etc/pacman.d/mirrorlist && sudo pacman -Syyu
    You need to `chmod youruser:yourgroup /etc/pacman.conf` for that to work if run by non-root.
    Last edited by graysky (2014-05-02 20:33:38)

  • New Arch kernel won't launch OpenShot video editor [SOLVED]

    I just re-installed Arch the other day and now OpenShot Video Editor won't startup.
    [wordstrings@DTR-Arch-Linux ~]$ openshot
    ------------------------- ERROR 1 ------------------------------
    Failed to import 'from openshot import main'
    Error Message: cannot import name main
    (process:24968): Gtk-WARNING **: Locale not supported by C library.
    Using the fallback 'C' locale.
    OpenShot (version 1.4.4~alpha1)
    Process no longer exists: 24908. Creating new pid lock file.
    ------------------------- ERROR 2 ------------------------------
    Failed to import 'from openshot.openshot import main'
    Error Message: unsupported locale setting
    OpenShot has failed to import some of the Python files or libraries
    required for our application to run. Here are some trouble shooting
    tips:
    Tip 1) Check if MLT can be successfully imported in Python. Run the
    following commands, and see if any errors are displayed. If you get
    an error, you need to investigate the correct way to install MLT.
    NOTE: Do not type the $ or >> characters in the examples below.
    $ python
    >> import mlt
    >> mlt.Factory().init()
    Tip 2) If MLT is working from the first example, then the next tip is
    to look at the above error messages very closely, and google for more
    help. It's likely the problem is already reported, and maybe there is
    a simple work-around. Also, you can search for bugs or report a new
    bug at https://bugs.launchpad.net/openshot. Good luck!
    Command running python:
    [wordstrings@DTR-Arch-Linux ~]$ python
    Python 3.3.3 (default, Nov 26 2013, 13:47:45)
    [GCC 4.8.2] on linux
    Type "help", "copyright", "credits" or "license" for more information.
    >>> import mlt
    Traceback (most recent call last):
    File "<stdin>", line 1, in <module>
    ImportError: No module named 'mlt'
    >>> mlt.Factory().init()
    Traceback (most recent call last):
    File "<stdin>", line 1, in <module>
    NameError: name 'mlt' is not defined
    So;
    [wordstrings@DTR-Arch-Linux ~]$ pacman -Ss mlt
    community/mlt 0.9.0-5 [installed]
    An open source multimedia framework
    Also;
    [wordstrings@DTR-Arch-Linux ~]$ export PYTHONPATH=$PYTHONPATH:/usr/lib/python2.7/site-packages
    [wordstrings@DTR-Arch-Linux ~]$ echo $PYTHONPATH
    :/usr/lib/python2.7/site-packages
    I'v read & done everything I can find on Google and in Arch bbs about this. I even installed all the other video editors I could find and none of them work.
    Last edited by customscrollbar (2014-01-03 16:53:02)

    I downloaded the openshot tarball off their website (openshot.com) and checked/installed the list of dependencies that came with it in the README file, (this)
    OpenShot Video Editor is programmed in Python, and thus does not need
    to be compiled to run. However, be sure you have the following
    dependencies in order to run OpenShot successfully:
    1) Python 2.5+ (http://www.python.org)
    2) GTK & pyGTK (http://www.pygtk.org)
    3) GooCanvas & Python bindings (http://live.gnome.org/GooCanvas)
    4) MLT Framework (http://www.mltframework.org)
    5) Sox audio library
    6) Frei0r video effect library
    7) FFmpeg
    Afterwards, I cd into the openshot directory, to run this command:
    $ sudo python setup.py install
    which triggers the setup.py file in it, but the shell gave me this error output
    [wordstrings@DTR-Arch-Linux openshot-1.4.3]$ sudo python setup.py install
    [sudo] password for wordstrings:
    File "setup.py", line 23
    print "Execution path: %s" % os.path.abspath(__file__)
    ^
    SyntaxError: invalid syntax
    Is there anyway I can alter line 23 in setup.py so the install command will work?
    Last edited by customscrollbar (2014-01-03 01:40:18)

  • Arch Linux USB installation does not boot Correctly [SOLVED]

    Hello,
    I created a bootable USB stick with Universal USB Installer on a Windows System.
    When I try to boot it, it shows me a Gummiboot Boot Loader Switch, where I can switch between two EFI shells and ArchLinux installation media.
    When choosing the installation media, the screen flashes and after that, it shows the following lines:
    :: running early hook [udev]
    :: running hook [udev]
    :: Triggering uevents ...
    :: running hook [memdisk]
    :: running hook [archiso]
    :: running hook [archiso_loop_mnt]
    :: running hook [archiso_pxe_common]
    :: running hook [archiso_pxe_nbd]
    :: running hook [archiso_pxe_http]
    :: running hook [archiso-pxe_nfs]
    :: Mounting '/dev/disk/by-label/ARCH_201403' to '/run/archiso/bootmnt'
    Waiting 30 seconds for device /dev/disk/by-label/ARCH_201403 ...
    [     5.060855] sd 6:0:0:0: [sdc] No Caching mode page found
    [     5.060960] sd 6:0:0:0: [sdc] Asuming drive cache: write trough
    [     5.062636] sd 6:0:0:0: [sdc] No Caching mode page found
    [     5.062688] sd 6:0:0:0: [sdc] Asuming drive cache: write through
    [     5.064566] sd 6:0:0:0: [sdc] No Caching mode page found
    [     5.064614] sd 6:0:0:0: [sdc] Asuming drive cache: write through
    ERROR: 'dev/desc/by-label/ARCH_201403' device did not show up after 30 seconds ...
        Falling back to interactive promt
        You can try to fix the problem manually, log out when you are finished
    sh: can't accsess tty; job control turned off
    [rootfs /]#
    In UEFI menu I unabled secure boot and fast boot.
    Can anybody tell me what I can do to boot the istallation media correctly?
    Thanks for your suggestions
    Last edited by outschi (2014-03-20 00:31:18)

    Welcome to arch.  I suspect this is the problem:
    outschi wrote:I created a bootable USB stick with Universal USB Installer on a Windows System.
    The error message you get is the same as those encountered by people who try to use unetbootin (eg here).  I know Unetbootin does not work to create bootable arch media - perhaps the windows tool also doesn't.
    Did you see and follow the tip here?  It seems that tool will create working usbs, but you have to fix the label.
    Last edited by Trilby (2014-03-19 16:52:22)

  • Just installed Arch, can't perform tasks like useradd/pacman [Solved]

    Followed the Beginner's Guide step by step. I'm now on the 'Extra'. However I'm running into problems doing anything. The first thing I tried to do was change user password. I got 'Authentication token lock busy'. Tried to add user and got 'cannot lock /etc/passwd/'
    I figured out that I could go back into the initial setup, remount the drives, and chroot and then change password/add user. Then - as the guide says- I unmounted them and rebooted.
    Now I can't install sudo. I get 'failed to innit transaction: unable to lock database' 'could not lock database: read only file system'
    my /etc/fstab says sda1 rw which I thought meant that it was write-able
    I can't do pacman -Syy for the same reasons, however I can do it when I chroot and mount from the setup
    I followed the Beginners' Guide line by line I believe. But somewhere after unmount and reboot, it quits working.
    Last edited by mmat (2012-10-18 06:19:52)

    DSpider wrote:
    Hi, and welcome to the forum.
    The title is a bit misleading. Is the root read-only, or is useradd/pacman the problem (which both are in the "Install" section, not "Extra") ?
    I'm sorry. This is my very first time ever using linux, so I don't fully get what you're saying about read only root.. I'm not sure what the problem is. I can and did useradd/pacman when I was installing. I finished the install section. I cannot now do the first step on extra.
    DSpider wrote:You didn't remove "ro" from syslinux.cfg, did you? Or do you use GRUB?
    I used GRUB because the guide made it sound better.
    DSpider wrote:Post your fstab.
    I installed it in VirtualBox. Not sure if that matters.
    http://oi46.tinypic.com/21170rb.jpg
    -- mod edit: read the Forum Etiquette and only post thumbnails http://wiki.archlinux.org/index.php/For … s_and_Code [jwr] --
    Last edited by jasonwryan (2012-10-18 06:03:05)

  • 990FXA-GD80 sudden poor performance

    I bought this board less than a month ago.  It is replacing a Gigabyte 990FXA-UD3. 
      I started off with all standard settings and it ran well.  Then using the CPU ratio and voltage I took it to 4.6 ghz and ran OCCT and IBT.  It ran fine. 
      Then I dropped it back to 4.5ghz and took it back to 4.6 ghz using the FSB frequency and NB voltage.  I have been reading some guides and it seems like this is the way to do it.  Anyway IBT showed stable and using this method it also showed better Gflops.  This is in all stress levels from standard to max.  I did not use OCCT in this time.
      Then something happened.  The clock lost over an hour and for some reason game play has taken a big performance hit.  The computer never BSOD or even CTD.  It starts and runs fine with good temps, but seems to be just loafing along.  While playing the most stressful game I have, Bioshock Infinite, the video cards used to be running at 90-99% and overheating.  My solution was to enable vsync which dropped temps well down.  No more!  The VCs are now at 50-80% and running cool without vsync and frame rate is barely at 60fps.  It used to be 120+.
    Also, I tried OCCT Large Data Set again which uses 3584mb of memory and it errors out in 1-2 minutes.  And yet IDT says it is stable and it still runs ok.
    I am hoping it is just a setting in the bios that needs tweeking, but don't know where to look now.

    Ok, I dropped back to 4567mhz and it went 1 hour in OCCT LDS without errors.  Would like to learn how to get it to 4.6ghz stable.
    But most importantly, why is the frame rate in games only half of what it used to be?  50-80 fps is unimpressive when it was 120-180 fps a week ago.

  • Windows 7 - "Cannot complete setup" of Windows XP Mode

    I am running Win 7 OEM 64 bit PRO on new PC:
    AMD FX-8120, Gigabyte GA-990FXA-UD3 MOBO, latest BIOS
    The setup gets to 99% then fails with the above message. So far I have tried:
    Ensuring Virtualisation is on in BIOS
    There are no bootable disks in the drive
    I am running as admin with my own account logon name
    I have run SFC - no conflicts
    Yes I have downloaded the correct XP mode file
    I have re-installed Win 7 and tried to run before installing any drivers or apps  - same problem.
    Now what? I tried vmware but the standalone converter gave me an error message saying it was not configured properly.  I suspect either a policy problem with Win 7 or a MOBO bios incompatibility. Anyone got any ideas?

    @robonline
    I believe I'm seeing this same issue as well, I've got a windows 7 system that was running with an Athlon II 250 on which Windows XP mode installed and worked perfectly.
    However when I upgraded the CPU to an AMD FX-8150 (in an otherwise unchanged system) Windows XP mode no longer starts properly, it seems to hang before the Virtual PC starts to boot the OS (so possibly either during BIOS POST, or right
    at the transition.
    I've discovered that it doesn't ALWAYS fail, and if power of the Virtual PC and Start it again a few times in quick succession it seems to then get past what ever point causes the problem and then boots the XP Virtual PC and
    it works normally until it's next turned off.
    So if you need XP mode (as I do for some work applications) the (crazy) work around might be to install XP mode with a none bulldozer CPU (if the try, try, try again method doesn't get the installer to work) you can probably get it to
    start once you change back to your bulldozer CPU by simply trying to start the installed OS a few times in short succession.
    @Alex
    Many thanks for looking into this!
    Rob

  • [SOLVED] NetworkManager can't use DHCP

    I just installed a fresh Arch Linux on whatever network modem is onboard the Gigabyte 990FXA-UD3 motherboard. The ethernet port works out-of-the-box using dhcpcd on the installer ISO and on my actual installation of Arch. Using KDE, I followed NetworkManager's setup guide in the wiki, but if I try to connect to anything, it instantly fails once I click on it. Then I tried clicking on Manage Connections and making a new Wired connection. It still failed. Then I tried changing that connection from DHCP to manual IPv4 and DNS numbers, and it worked just fine. I've done a lot of Googling, but nothing in the wiki or forums is fixing this issue for me.
    Last edited by Cadeyrn (2013-06-25 22:24:32)

    I'm confused, did someone prevent you from updating the Wiki?
    Yes, I tried, but I admittedly lack the skills to debate technically, with someone who obviously possesses the skills.
    https://wiki.archlinux.org/index.php?title=NetworkManager&action=history
    (cur | prev) 07:09, 3 September 2013 Lahwaacz (Talk | contribs) . . (28,477 bytes) (-253) . . (Undo revision 274198 by Jeff story (talk) completely covered by the note above removed text)
    (cur | prev) 23:18, 2 September 2013 Jeff story (Talk | contribs) . . (28,730 bytes) (+253) . . (clarify and instruct to disable and stop dhcpcd.service)
    Honestly though, at this point, your question is no longer important.
    What would be cool though is for someone with the expertise on this subject to take notice, figure out exactly what is lacking in the wiki and be willing to contribute the info for the good of the community, which I believe should include the Arch newbs or non coders / programmers.
    Then be able to persuade Scimmia and Lahwaacz that in fact, something needs "fixed", even though from their perspective (Linux guru?) there is nothing wrong or missing.
    EDIT: Added second name to clarify situation.
    Last edited by jeff story (2013-12-30 03:53:51)

  • Blue Screen (atikmpag.sys) on MSI R9 280

    Hello, i have one Video Card MSI R9 280 but with drivers problems resulting in failure to enter Windows by a Blue Screen.
    All the components are new, i build the P.C. a few days ago so in the first Windows installing all work fine, i installed the video drivers with the CD of the box. Reboot the pc when installed and all fine. The trouble came after install a couple of games and played it, one freeze and the pc reboot. Since then when i try to start windows and crash it with this:
    Only in Safe Mode i can start Windows, but i disable the AMD drivers and right now i am with the VGA drivers from the motherboard.
    I dont know if is a Software issue (I installed two Windows 7 - Ultimate & Proffesional) or Hardware (because in the first use all work fine even the drivers)
    Please Help, i dont know what to do 

    This is my pc:
    Cpu: AMD fx8320
    Motherboard: gigabyte 990fxa ud3
    RAM: corsair vengeance 8gb 1866 MHz
    Cooler: cooler master seidon 120 v
    Psu: corsair tx750
    Hdd: western digital green 1 tb
    Video card: msi r9 280
    I already installed the last versions of the AMD drivers and nothing. That file (atikmpag) is already in the drivers of windows, i dont know why is making the blue screen. I have the motherboard BIOS updated. Not even in safe mode the pc recongnize the video card :|
    By the way, i have windows with service pack 1 installed.
    The thing is that in the very first configuration and windows installation, al work fine, the problem was i played a game and the screen freezes and reboot, since then i have the blue screen :(
    Another problem is that i live outside EEUU and i cant return it, i cant imagine if i lose the 250$ of the vídeo card, its a lot of money for me.

  • YAHOO SEARCH PAGE OPENS W/NEW TAB

    mistakenly opted in to install YAHOO! toolbar
    uninstalled YAHOO! toolbar
    now YAHOO! search page opens whenever new tab opens
    http://www.google.com/firefox set as home page
    win 7 x 64 PRO SP1 [ updated ]
    Firefox 10.0.2
    Gigabyte 990FXA-UD3
    AMD 8120 3.1 G HZ
    8.0 GB RAM
    600 GB WD BOOT DRIVE WD6401AALS-00J7B0 ATA
    2 TB WD DATA DRIVE WD2002FAEX-007BA0 ATA
    2X 150 GB DATA DRIVE WD1600JS-22MHB0 ATA

    I watched closely as it opened and found this script [url]
    click.w3i.com/?Programid=132&Elementname=TabSearch&Applicationid={5E8E3DF7-A01E-4B94-BC69-BD32867958BB}&Version=3.6.5&Vintage=20120209&Defaultbrowserid=27&Productid=2780&Vendorid=6413&Offerid=6894
    it then went to a yahoo! search page
    http://search.yahoo.com/?fr=freeze&type=W3i_NA,132,3_6,Tab%20Search,20120209,6894,0,27,0
    currently trying to remove freeze.com net assistant
    [that's where w3i.com led me]
    will advise

  • 7970 lightning Crash at stock clocks

    OK i just got my 7970 lightning
    and i wanted to test it, so i completely reinstall my system and fires up furmark.
    It runs for about 1.5 min and then the system reboots at stock clocks 1070/1400... if i crank it up to 1125 it happens at the 1 min mark.
    what is wrong here? do i have a broken card or is furmark broken ?
    BTW my system is stable 100% otherwise. 850w thermaltake psu, i7 2600k, maximus extreme IV.

    I didnt play with furmark once saw temps tend to raise even after 75C mark.
    My issue is with 'driver stopped responding and was restored', and i'm curious about one thing, i can not see 'system' in crashes, it crashed on desktop today, before it usually did when i tried playing games in stereo 3d, its not overheated while it does crash, and oppositely when i managed to get biggest temperature of 73C, in Civ5, i dont remember it crashing.
    Oh, and another thing i'm interested in, my cards ASIC is 75.3%, which is right between 2 suggested voltages 1.112 and 1.175, default voltage i saw through afterburner is 1.112, could it be the problem that chip actually demands 1.175, but due to its marketing posture of premium class it was made to work on lower voltages it can not hold?
    (Phenom x6 @3600Mhz,
    Gigabyte 990fxa-ud3 rev 1.2 FC bios,
    Corsair Vengeance 8gb 1333Mhz,
    7970 Lightning 1070/1400,
    Samsung HD642JJ,
    W7 x64,
    Zalman 850w HP.
              min  max peak
    +12V1 0.5A 18A 22A
    +12V2 0.5A 18A 22A
    +12V3 0.5A 28A 30A
    +12V4 0.5A 28A 30A
    +12V5 0.5A 18A 22A
    +12V6 0.5A 18A 22A)

  • GTX 760 (N760 TF 2GD5/OC ) unstable at stock settings. (Does MSI cross ship?

    I recently purchased a GTX 760 TF
    N760 TF 2GD5/OC
    At stock settings, it is unstable, for example, running unigine heaven 4.0 will cause the card to reset in a way where it gets locked at around 500MHz. Nvidia cards do this when the GPU is unstable.
    if the card is set to the slower silent mode, then it is stable, and does not crash.
    If the stock settings are used 1020MHz, the card videocard will crash after around 40 minutes.
    If the card is set to OC mode, then it will crash anywhere from 1 minutes, to 10 minutes.
    if the card is set to the OS mode speeds, and the voltage is set to +12mv, then it can sometimes go a good 20 minutes+ before crashing.
    Overall, the videocard is unstable at anything above the reference speeds
    Related system specs are
    Phenom II x6 1075T
    12GB DDR3 1600
    Motherboard: gigabyte 990FXA-UD3
    Power supply: thermaltake SP-750PCBUS 750 watt
    All voltage rails are stable during gaming and benchmarking.
    The store I purchased it from no longer has it in stock so an RMA with them will not work in a reasonable amount of time if I want a replacement.
    I don't want to be without the card for an entire month, so I was wondering, does MSI offer any kind of cross shipment or any other method to speed an RMA process.
    gpu-z showing the stock settings of the card

    Same issue happened on both of my computers.
    second PC has a MSI 870A-G54 motherboard and a AMD phenon II x4 965 (underclocked to keep the system quiet for home theater use)
    I am pretty much swapping in my GTX 460 for now.
    When mailing the videocard in, do I use the original box?  also, would it impact the warranty if the UPC section was cut out of the box to be mailed in for the rebate that MSI had going for the card?
    another weird issue I noticed, if I run a game that causes the card to go up to 1019 MHz and not clock higher (e.g., running audiosurf and twitch streaming, the card will crash more quickly than if the card is loaded up enough to jump to 1123MHz at 1.2V
    anyway it looks like this card in only stable at nvidia reference speeds
    At reference speeds, it can handle a full work load, even run unigine and bitcoin mine (useful stability test) at the same time without crashing where at any other speed, you are looking at a crash in under 1 minute.

  • MSI 280X Gaming 3G - Crashing while gaming

    Card is crashing while gaming has been going on since December intermittanly, it was purchased May 2014
    Tested with Furmark and seems ok Hit max temp of 60-70 depending on ambient, same temps while gaming but  randomly reboots.
    Have tried differnt AMD drivers the latest Omega seems more unstable than the lastest beta, but crashing more with this one now.
    Rolled back to 14.4 but the same.
    Is this a faulty card or do I need a Bios update? ( if so please supply lgacy and uefi bios)
    OS : Windows 7 Ultimate 64-bit SP1
    CPU: AMD FX-8350   20 °C
    RAM: 16.0GB Dual-Channel DDR3 @ 937MHz (10-11-10-30)
    MB : Gigabyte 990FXA-UD3 (CPU 1)   25 °C
    GFX: S23B550 (1920x1080@60Hz)
        MSI 280X GAMING 3G - 3072MB ATI AMD Radeon R9 200 Series (MSI)   29 °C
        vga bios : ms-v277 tv277ms.350   from MSI Update
        S/N : 602-v277-31sb1402067937
    Storage
       1863GB Seagate ST2000DM 001-1CH164 SATA Disk Device (SATA)   34 °C
       1863GB Seagate ST2000DM 001-1CH164 SATA Disk Device (SATA)   31 °C
       1863GB Seagate ST2000DM 001-1CH164 SATA Disk Device (SATA)   28 °C
    Optical Drives
       TSSTcorp CDDVDW SH-224DB ATA Device
    Audio
       Realtek High Definition Audio
       PSU Corsair CX750
    AMD Catalyst :
    AMD 14.4
    Tried 14.3
    & Omega driver

    Here you go though I doubt that vbios is the problem:
    TV277MS.351 is legacy
    TV277MH.351 is uefi
    If no help follow >>Posting Guide<<

  • Issue Regarding 990FXA-GD80 BIOS flash

    Hello all
    This is my first post, so please bear with me if i do not include everything needed. Also, this forum looks very informative and helpful from what I've seen. My gratitude extends to all that help out!
    I am having trouble flashing the BIOS on the 990FXA-GD80 mobo. I am currently running v11.12 (7640vBC) and I am trying to upgrade to v11.14 (7640vBE7 or any BE for that matter). Thus far, i have tried BE7 as well as BE6
    For BE7:
    I used the MSI Forum HQ USB Flashing Tool and all goes well with that part of the process and is performed successfully. I clear the CMOS properly (disconnect power cord, hold button for 5 seconds, remove battery, hold down power button)
    After booting back up, I load default settings in the BIOS, yet the BIOS version stays at v11.12 and I have no clue as to why.
    Is there a Boot Sector Protection that I may be overlooking? I didn't see that type of option in the BIOS
    I then tried to update to BE6:
    I haven't actually successfully flashed to this beta whatsoever, because when I do boot up with the usb and try it, this message comes up:
    PI Version in SYS ROM mismatched PI Version in ROM File!
    Force update it may be destroy the system BIOS
    We do not recommend flashing your BIOS
    after seeing this message, I did not go forward with it.
    Please help in this matter, as this is my first attempt at trying to flash my bios
    Thanks in advance
    My system:
    990FXA-GD80
    AMD FX-8350
    G.SKILL Ripjaws X Series 16GB (2 x 8GB) DDR3 1600
    128gb OCZ Vertex 4
    Cooler Master Silent Pro M 850W

    Thanks Svet, I'll try it again and go through with it
    But, why does it only give me the warning message for BE6 and not BE7? I would rather try to solve the issue I'm having with BE7 rather than BE6

Maybe you are looking for

  • Anyone having trouble with guide snapping in Illustrator CS5?

    I was trying to adjust some type and I was pulling guides to adjust widths etc but when I moved a guide to meet the black of the type it would snap a pixel to the left or right of where I was trying to put it.  I tried turning off Smart Guides and sn

  • Scanning... and Mulitple String lines

    My text file that I want to scan looks like this format: Name of Student Test Answers (True or false) Example: Joe Smith TFTFTFFTFFFTFFTFFTFTTFF etc.. Jane Sims FTFTFTTFFTTFFTFFFTFTFTF etc.. The issue is - I know how to scan the data of my textfile.

  • Adobe forms central for team of 60

    hello i have a team of yoga teachers i would like them all to have a dicounted use of this - i see u do a 20 person package could u give me a price for 60 people

  • How to Replicate Standard Alternating Colors using XSL-FO??

    Hi All I am using Apex_3.1, Apache fop and Standard Report Printing to produce PDF reports. We would like to replicate the theme 12 (Blue) standard alternating colors in our PDF output. I do realise that this involves modifying the xsl-fo and have tr

  • Performance degrade

    I want tune my database, it's 10G big and warehouse type. Before tuning, it just use default tablespace. I put it's datafiles to different hard disks to reduce I/O contention the server has 2 CPU and 2G memory, I set max parallel server = 8, I change