1. Welcome Guest! In order to create a new topic or reply to an existing one, you must register first. It is easy and free. Click here to sign up now!.

Issues with 2 BSOD crashes on threadripper system that has been stable for months...

Discussion in 'Specific Hardware Issues' started by liberty610, Jun 12, 2018.

  1. liberty610

    liberty610 Registered Members

    Joined:
    Sep 20, 2010
    Messages:
    67
    Location:
    Pa - USA
    Operating System:
    Windows 10
    Computer Brand or Motherboard:
    Gigabyte GA-X99-Ultra Gaming
    CPU:
    Intel Core i7-6800K
    Memory:
    G.Skill Ripjaws 4 series 32GB (4 x 8GB)
    Hard Drive:
    Samsung 950 PRO 512GB M.2
    Graphics Card:
    MSI GTX 1080
    Power Supply:
    Corsair Professional 750W 80+ Gold Certified Semi-Modular ATX Power Supply
    Hey guys,

    Forgive the longer post here, but I wanna make sure I cover everything I have attempted to do so far. Yesterday I started having issues with my Threadripper PC I built a few months ago. I am trying to pin-point possible causes before I have to proceed with a dreaded time consuming RAM test, as I only have my current build to test all of my ram, and I need it (somewhat) operational today to do projects.

    Here is my system via PC Parts Picker:
    https://pcpartpicker.com/b/BqHhP6

    Ok, so I had the Samsung M.2 as my boot drive from day one of the build, and recently added the Western Digital M.2 drive as well as the the Western Digital SATA SSD drive. I am not saying they are specifically the cause of this, but I am hoping maybe I just bumped or jarred something loose when i installed them that could be causing my current and random issue.
    After I installed them (about a week apart from each other with the SATA drive being installed this past weekend), and I believe this incident was right after I installed the SATA SSD. What happened was, I randomly got a beep code error when trying to boot the system up. I am assuming it was 3 LONG beeps, although I am not sure what my board considers a long beep over a short beep. The board's manual only shows what the digital readout codes are on the board's LED display, not audio beep codes. According to some web searches, 3 beeps is ram related, but I didn't find much on it, and what I did fine had conflicting answers.

    Anyways, it did these 3 long beeps, shut itself down, and attempted to reboot itself. When it attempted to do the reboot, it again did the 3 beeps, shut itself down and kicked back on. This time, it posted fine. I thought, 'huh, that was weird' and went along my business with normal operation for several days, including shut downs and reboots.
    Yesterday morning, in my down time, I was playing a game of Hitman on it, and received 2 BSOD messages in back to back play sessions. They did not happen instantly, but they where within at least 10 minuets after rebooting and starting the game up again. The first BSOD stop code was ATTEMPTED_WRITE_TO_READONLY_MEMORY and the second BSOD code was DRIVER_OVERRAN_STACK_BUFFER

    I had the latest Nvidia driver from June 4th installed for my GTX 1080. I am not sure what the exact version was. I have been on the fence about re-installing Windows the last few weeks on a personal preference, so in a nutshell, I said 'Now would be a good time to do it" and just re-installed windows.

    When I did that, I started randomly getting the 3 long-ish beep codes again on one or two of the boot ups, with one of the times the system shutting itself down and trying to reboot, only to get the 3 beeps again (much like my very first encounter with the beeps). It then booted normally. I then figured, I had an official problem of some sort.

    I called Gigabyte and the not-so-english-speaking rep wasn't a huge help, where all he told me to do was things I planned on anyways. Update to the latest bios, make sure the latest NVIDIA driver was the latest, and to test my ram one stick at a time. With threadripper/Ryzen being pretty new, I figured a bios update might not be a horrible idea.

    I have a 4 Ram stick kit from G.Skill as my parts list says 32gigs. I pulled all of them out, put one stick back in the first main slot. As soon as I powered it on, I got the 3 beeps again. It powered itself off, booted up fine the next power cycle. I then shut down again, took the ram stick back out, pulled the video card, and blew all the slots out with an electric air blower. I then cleaned the graphics card out real good (wasn't that dusty, I do regular cleans out on my builds), and re-seeded the card as well as that first stick of ram I had in when I got the 3 beeps. Powered on, no issues. At this point, I also updated the BIOS to the latest version, F3j. No issues so far on any further boot ups or re-starts yet, but then again, the ones I have been getting have been very random.

    I then ran memtest86 on the one inserted Ram stick for 4 and a half hours and saw no errors. It was 3 passes, 6 Iterations. I know the golden rule is to run it longer on each stick, but I have to try and put my system to use for the day before I can go back to testing ram over night.

    Is it possible that there was something jarred loose when installing my new drives? Specifically the video card? My system has been rock solid the last several months since I put it together the day Threadripper was released. Is there any other steps I can take to see what might be causing this to happen? Any suggestions would be helpful! Thanks.
     
  2. DSTM (Dougie)

    DSTM (Dougie) Registered Members

    Joined:
    May 3, 2009
    Messages:
    8,270
    Location:
    SYDNEY AUSTRALIA
    Operating System:
    Windows 7
    Hi. Are you running DirectX 11 or 12?
     
  3. Tony D

    Tony D Super-Moderator Super Moderators

    Joined:
    Sep 25, 2009
    Messages:
    2,758
    Location:
    SE Pennsylvania, USA
    Operating System:
    Windows XP Professional
    I was thinking RAM or power supply. Maybe reseating the RAM fixed it.
     
  4. liberty610

    liberty610 Registered Members

    Joined:
    Sep 20, 2010
    Messages:
    67
    Location:
    Pa - USA
    Operating System:
    Windows 10
    Computer Brand or Motherboard:
    Gigabyte GA-X99-Ultra Gaming
    CPU:
    Intel Core i7-6800K
    Memory:
    G.Skill Ripjaws 4 series 32GB (4 x 8GB)
    Hard Drive:
    Samsung 950 PRO 512GB M.2
    Graphics Card:
    MSI GTX 1080
    Power Supply:
    Corsair Professional 750W 80+ Gold Certified Semi-Modular ATX Power Supply
    Sorry guys, I never got e-mail alerts that you replied to this. Thank you for the replies.

    I believe I am running Direct X 12. I know some games have the options to choose between 11 or 12, but I honestly just use Steam to install all my games and whatever settings Steam sets up are usually the ones I use, other then the resolution and full screen type options. I have ran the PC a lot the last 2 days doing various tasks. Audio production, ran a few hours of the game Rocket League the last couple days... no issues so far at all. I even fired up Hitman and ran it from the same save file that had the crashes the other day. Didn't get to run it long, but no issues.

    I am assuming there was loose ram. But I am not 100% sure. I am going to avoid the timely ram testing until I get another crash, if I even do....
     
  5. Seth Anthony

    Seth Anthony Registered Members

    Joined:
    Mar 31, 2017
    Messages:
    646
    Operating System:
    Linux Based
    Computer Brand or Motherboard:
    Altaire 8800
    CPU:
    Modified Texas Instruments calculator
    Memory:
    2 transistor tubes
    Hard Drive:
    pen and paper
    Graphics Card:
    TV ready
    Power Supply:
    Mouse on a wheel
    Memory testers are 100% accurate if it shows a fail, but memory testers aren't 100% accurate on a pass.

    It's way faster and more accurate just to try one stick at a time.
     
    Tony D likes this.
  6. liberty610

    liberty610 Registered Members

    Joined:
    Sep 20, 2010
    Messages:
    67
    Location:
    Pa - USA
    Operating System:
    Windows 10
    Computer Brand or Motherboard:
    Gigabyte GA-X99-Ultra Gaming
    CPU:
    Intel Core i7-6800K
    Memory:
    G.Skill Ripjaws 4 series 32GB (4 x 8GB)
    Hard Drive:
    Samsung 950 PRO 512GB M.2
    Graphics Card:
    MSI GTX 1080
    Power Supply:
    Corsair Professional 750W 80+ Gold Certified Semi-Modular ATX Power Supply
    You mean to try one stick at a time under normal OS use? Or test one stick at a time with a tester like memtester86?
     
  7. Tony D

    Tony D Super-Moderator Super Moderators

    Joined:
    Sep 25, 2009
    Messages:
    2,758
    Location:
    SE Pennsylvania, USA
    Operating System:
    Windows XP Professional
    He wants you to try one stick at a time. You'll have less RAM installed, but it may uncover a faulty RAM module.
     
  8. liberty610

    liberty610 Registered Members

    Joined:
    Sep 20, 2010
    Messages:
    67
    Location:
    Pa - USA
    Operating System:
    Windows 10
    Computer Brand or Motherboard:
    Gigabyte GA-X99-Ultra Gaming
    CPU:
    Intel Core i7-6800K
    Memory:
    G.Skill Ripjaws 4 series 32GB (4 x 8GB)
    Hard Drive:
    Samsung 950 PRO 512GB M.2
    Graphics Card:
    MSI GTX 1080
    Power Supply:
    Corsair Professional 750W 80+ Gold Certified Semi-Modular ATX Power Supply
    Right, but is the suggestion to use one stick at a time in memtest86 or in my normal OS operation?
     
  9. DSTM (Dougie)

    DSTM (Dougie) Registered Members

    Joined:
    May 3, 2009
    Messages:
    8,270
    Location:
    SYDNEY AUSTRALIA
    Operating System:
    Windows 7
    Try one stick at a time in every Ram slot on your Motherboard, and reboot each time.
     

Share This Page