Jump to content

Welcome to Geeks to Go - Register now for FREE

Need help with your computer or device? Want to learn new tech skills? You're in the right place!
Geeks to Go is a friendly community of tech experts who can solve any problem you have. Just create a free account and post your question. Our volunteers will reply quickly and guide you through the steps. Don't let tech troubles stop you. Join Geeks to Go now and get the support you need!

How it Works Create Account
Photo

Mystery Issue Diagnosis


  • Please log in to reply

#1
Magic 8 Ball

Magic 8 Ball

    New Member

  • Member
  • Pip
  • 3 posts
So, I've been having a complex computer issue for a while now (it started around May 2nd, or about 48 days before I posted here), and I've had little luck tracking down the precise nature of the issue. Before I talk about that, though, I'll say first some specs for my compy:

OS: Windows Vista Home Premium SP1 64-bit

--Main hardware--
CPU: Intel Core2Quad Q6600 (OC'd to 3.04ghz/core, up from default of 2.4)
Motherboard: Asus P5Q Deluxe
PSU: Antec TruePower Quattro 850w
RAM: 2x Corsair XMS2 DDR2-800 2gb sticks (OC'd to run at 4-4-4-12 timings, 807mhz, and 2.1 volts, up from defaults of 5-5-5-15, 800mhz, 1.8v; also worth noting that according to the information on Corsair's website and the Newegg page from which I purchased them, this specific model is designed and tested to run at 4-4-4-12 timings)
Video Card: Sapphire HD Radeon 3850 Toxic Vapor-X
Hard Drive: Western Digital WD6400AAKS-00A780 640gb

--Additional hardware--
Tower: Antec Nine Hundred
CPU Cooler: Some huge Thermaltake V1 with craploads of copper and heatpipes, and a roughly 110mm fan.
Optical Drive: A Lite-On DVD-R/RW and CD-R/RW, with Lightscribe, connected by SATA and I think the speed was 22x...
Fans: A total of four 120x25mm case fans, as well as an Antec VCool blowing on my video card and an Antec SpotCool blowing on my RAM, and the ~80mm fan in the back of my PSU. Due to some damage to my case from furniture moving a while back, I have removed the Nine Hundred's signature 200x30mm top fan. The built-in fan on my video card is usually malfunctioning, otherwise I stop it myself because it makes an awful clicking noise.

--Peripherals--
Monitor: Samsung 2343BWX 23" LCD (model code LS23MYZKFV/XAA)
Mouse: Logitech MX518
Keyboard: Logitech G11
Headset: Logitech G330 (mic doesn't work)
Standalone Microphone: Also some Logitech gizmo.
Printer: HP Officejet J6450 All-In-One
Speakers: Harman/Kardon SoundSticks II 2.1 Channel

Okay, now for a bit of history on my issue. When it started (circa May 2nd), I was playing TES3 Skyrim and my compy froze for a bit, with a slightly distorted image. The audio continued for about 5 seconds before the screen went black, and then came back to a BSOD. At the time I didn't record that BSOD code because I figured it for one of those intermittent crashes I get every once in a great while. So I rebooted, and tried playing some more Skyrim. That got me another BSOD, and after just about the same amount of time into Skyrim as well (I tried doing the exact same thing that I was during the first crash). Curious, I dug around for a while...

Since I could type a book about everything I did, I won't, and I'll try to summarize instead.
-Tried other programs (TES4 Oblivion, Sid Meyer's Civ5, Just Cause 2) and found that these also crashed with identical symptoms; pretty much anything graphically taxing would BSOD.
-Tried changing my graphics drivers; I had been running a driver version that was roughly 14 months more up-to-date than the drivers on the disk accompanying the video card when I bought it. I first tried the latest update for my 3850 (as of about May 20th; I haven't checked since then), which did nothing. Then I tried the ones that came with the video card; at first this enabled me to play Just Cause 2 for ~2.5 hours without crashing at all, but also stopped some of my OpenGL-dependent games from functioning (Minecraft, Dwarf Fortress). At that point I intended to establish a cycle of 'New drivers for OpenGL games' -> 'Old drivers for DirectX games' -> 'New drivers' etc., but after I switched back to the recent update and once again to the old one, it crashed within 2-30 minutes of play for Just Cause 2 and slightly quicker for Skyrim.
-Tried more wacky driver things; reinstalling the drivers I had when the issue began (from an old installer package that I never deleted), tried uninstalling and reinstalling the AMD Install Manager, tried uninstalling drivers THEN manually deleting/renaming/moving every AMD/ATI driver file I could find on my computer (around 56 of them I think; I compiled the list based on dependent drivers listed for devices in Device Manager) THEN resetting the compy and reinstalling drivers...nothing helped.
-Tried using System Restore to a point a day or two before the issue began, to no avail.
-Tried running ChkDsk on all my drives/partitions (I only have the one HDD with three partitions). It came up with no errors.
-Tried running Memtest from the disk I made when I built the computer about 4 years ago. No errors from that either, although I don't remember how many passes I did.
-Spent hours mulling over my Event Viewer logs. But, with the combination of my sloppy style of software maintenance and the fact that I've been running Vista for roughly four years now without a reinstall, that was almost entirely a waste of time.
-Tried using CCleaner to get rid of some of the digital dust and mostly-dead malware that I never fully dealt with. Also tried messing with the Registry Editor myself, but I botched that pretty hard and used another System Restore to fix it, and then did some of the CCleaner stuff again.
-Tried making my pagefile larger, even though I had already allotted 11gb of hard drive space for it. Why? Desperation, I suppose. =P
-Tried some stuff in Safe Mode, although all the programs that crashed my computer wouldn't run in Safe Mode regardless.

By the time I did all of this, I had come up with some data that might help troubleshoot the issue:
--Almost every crash had a BSOD. In the beginning, these were either 0x000000D3 or 0x0000001E (most of the 0x1E errors were due to a bad driver install I got at one point, but not all of them), but as I did driver maintenance they gradually shifted to a consistent 0x00000116, which is now the only thing I get. The file cited in the BSOD as the culprit was always either "___kmdap" or "___kmdag", with the '___' being 'ati' if the drivers I had running at the time were from before ATI Corporation was purchased by AMD, or the '___' was 'amd' if the drivers were more recent than the corporate merger.
--Out of the 30-40+ crashes I've gotten from this issue so far, three of them didn't actually crash. I got the standard 'screen freeze w/ visual distortion', but the audio never ceased, and after the screen went black for a bit, it came back with unfrozen video, and a popup bubble from my Notification Bar saying "Display driver stopped responding and was reset" or something very similar. One of these three times, I had only ~15 seconds before it froze again and did crash, but I think that was because I kept trying to play Just Cause 2 when the video came back.
--The most valuable data I found up to this point, at least IMO, was that if I crash the computer frequently enough, then I have massive graphical distortions on my screen when I reboot (random formations of colored pixels (static and 'animated') based on what else is on my screen, green bars/blocks, repeating patterns of columns/rows of white dots, random white text characters (usually parentheses or an 'a' or 'e' with an accent symbol of some sort), etc.). Any attempt to use the computer in this state will crash with another BSOD, often even before I can log into the OS. Safe Mode seems to keep it from crashing. However, to end this state once I have reached it, I need to wait quite a bit of time without the computer crashing at all (the first time I needed to do this, I waited around 3 days; since then I have done it in around 40 hours); it may also go away if I reset the computer, but that doesn't happen often, and it will come back if I restart/crash it again a couple of times.

At this point I began hardware troubleshooting. My primary hypothesis was that the issue was somehow capacitor-related, due to the aforementioned "waiting time" to disperse the persistent graphics distortions; I thought that the wait was required to discharge some overloaded capacitor or something, which typically saw little/no use or was used with very small amounts of power, but was used much more in graphically-intensive programs, which resulted in both the crashes and a gradual stockpiling of electric charge, eventually creating the persistent distortions.
-I checked all the cable connections, and everything looked normal.
-I checked to make sure my DIMMs and my video card were properly seated. Everything fine there.
-I examined all the capacitors on my motherboard and video card for bloating/bulging or leakage, and found nothing suspicious.
-I checked all of the jumpers on my motherboard and video card, consulting with manuals and such to ensure they were where I wanted them. No issues found.

After that I left it alone for a couple of weeks while using a mediocre laptop (what I'm using to type this), and then went out and purchased a new video card - an XFX HD-687A-ZHFC (HD Radeon 6870 with 1gb GDDR5). I uninstalled all the AMD/ATI-related software from my computer, rebooted, shut down again, unplugged all the cables (including power), grounded myself, and carefully removed the old 3850 and replaced it with the massive 6870. Seated the card in the PCIe slot, secured it to the expansion slot section with a screw, connected the two 6-pin supplemental power cables from my PSU. Reconnected the side panel fan that I disconnected in the process, closed up the case, and ran it...No dice. I've tried using DVI output, VGA output (with a DVI-VGA adapter, since the 6870 has no native VGA ports), HDMI output (with an HDMI-DVI adapter monitor-side), tried changing ports, tried changing cables, tried changing monitors. No video at all.

I'm not sure if the new card is bad or what, but I'm open to suggestions. There's still more testing I can do with the 6870 to see if I can get video (both the DVI-VGA adapter I'm using and the DVI cable were the ones plugged into the 3850, and I have extras; I can try some different options with regards to the DVI ports, since one is dual-link and the other is single-link, although I have no idea which, etc.). Or I could go back to the 3850 and try something there. My P5Q Deluxe mobo has no onboard video, but I think there's a mostly-functional HD Radeon 2600XT laying around somewhere that I could stick in for kicks and giggles. Any ideas would be appreciated. =)

--Note: My internet access is inconsistent, so try not to abandon me if I'm gone for a couple of days.

Edited by Magic 8 Ball, 19 June 2012 - 08:27 AM.

  • 0

Advertisements


#2
Alzeimer

Alzeimer

    Member 1K

  • Member
  • PipPipPipPip
  • 1,331 posts
For a possible reason why your XFX HD-687A-ZHFC might not be working on your ASUS P5Q Deluxe is that your motherboard is a 2.0 pcie xompliant while your video cars is 2.1 and even if most 2.1 cards should be backward compatible with 2.0 boards some boards may need a bios update to work with the 2.1 versions.

Maybe an avenue to check.
  • 0

#3
Magic 8 Ball

Magic 8 Ball

    New Member

  • Topic Starter
  • Member
  • Pip
  • 3 posts
I did do some research into the differences between PCIe 2.0 and PCIe 2.1 prior to purchasing the XFX, and I got the impression that graphics cards based on 2.1 lost some or all of their backwards compatibility with 1.0/1.1 motherboards, but that 2.0 <-> 2.1 was still functional. Granted, that research was done via Wikipedia. I'll check the ASUS website for any BIOS updates, although I'll need to plug in the 3850 again in order to get video output and install them.
  • 0






Similar Topics

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

As Featured On:

Microsoft Yahoo BBC MSN PC Magazine Washington Post HP