What is the problem with my new build? I have no idea.

[Cryose]

On October 23 I assembled my first gaming computer (I'll post the specs below), after getting help on what to buy here. I put the parts together, installed the OS and drivers and everything ran great. I could run Crysis at high settings very smoothly at 1280x1024, and I went on to purchase TF2 + Gmod, which also ran wonderfully. However, I now seem to have a bit of an issue. Occasionally, when playing games my computer will completely lock up (except for music, which continues). When this happens (again, it's pretty random as to exactly when it happens), lots of bars of mostly-solid color (there is a bit of a gradient to the side of each bar) run the length of my screen.

The part that concerns me the most is that this did not initially occur within the first few days of me building the computer. The Crysis demo ran great from start to finish, but now I can't play for more than about thirty seconds. I had attributed this to memory, since I'm usually sitting at about 30% load on idle, but I cannot be certain.

Another issue is that I did get the BSoD once, but the message flashed so quick that I couldn't read the error, and my computer successfully rebooted. Another time I got a message that my graphics driver stopped responding after my screen turned white for a few seconds while using Chrome, but it recovered (my system let me know as well).

I frequently come across the first problem in Crysis, SCII, and Gmod, but not in WoW or TF2. Any help would be appreciated, especially since I need to send in rebates soon.

Specs, in no particular order (if you want links I'll try to find them):
CPU: AMD Phenom II X4 965 Black Edition Deneb 3.4GHz Socket AM3 125W Quad-Core Processor
Mobo: GIGABYTE GA-890GPA-UD3H AM3 AMD 890GX SATA 6Gb/s USB 3.0 HDMI ATX AMD
Case: Antec Three Hundred Illusion Black Steel ATX Mid Tower
OS: Windows 7 Home Premium (Family Upgrade Pack)
RAM: Mushkin Enhanced Blackline 4GB (2 x 2GB) 240-Pin DDR3 SDRAM DDR3 1333 (PC3 10666)
HDD: Western Digital Caviar Black WD6401AALS 640GB 7200 RPM SATA 3.0Gb/s 3.5" Internal Hard Drive -Bare Drive
GPU: XFX HD-585X-ZAFC Radeon HD 5850 1GB 256-bit GDDR5 PCI Express 2.1 x16 HDCP Ready CrossFireX Support Video Card w/ Eyefinity
PSU: OCZ OCZ750FTY 750W ATX12V / EPS12V SLI Ready 80 PLUS BRONZE Certified
DVD: LITE-ON 24X DVD Writer Black SATA

And I apologize if this is not the right board, but it seemed like the right place for this message.

 

[davewolfgang]

Just wondering if you have any web pages open in the background or on a 2nd screen. I started having some video issues after Flash updated a few times, and if I would go to a page with heavy flash use, it would cause the video driver to crash.

But mine would always recover after a screen reset, I would never have to reboot. I'd get the error that the video driver stopped responding. Never happens if I'm not surfing.

 

[Cryose]

I've made sure to close out as many applications as I can when I play games (barring Valve games, where Steam needs to be open). The only time my video driver crashed was that one experience I explained above. Every other event was in a game. I am also currently not running a second screen.

Also, my computer is making a strange humming sound as well. Perhaps I should open up the case to see what that could be. To my knowledge, all of the fans are working fine.

 

[davewolfgang]

Yeah, humming isn't good. Sounds like maybe your video card might be going out. Do you have RivaTuner or another program to monitor temps of it? And the CPU temps too?

 

[Cryose]

I just opened up my case, switched the 140 mm fan to high, and moved a couple cables around. The humming seems to have stopped (at least for now), but I will continue to listen for it.

But no, aside from checking my PC Health Status in the BIOS, I do not have a way to monitor temps. What would be a good program for that? While I do not think my current build will overheat that easily, especially with the case I selected, I suppose it is possible.

 

[rolli59]

My suspicion would go to a bad GPU possible failing vram on it.

 

[Cryose]

I was kind of thinking the same thing too. Unfortunately, I do not know how to really test to see if it truly is a bad GPU.

If it does turn out to be the GPU, will I be able to RMA it in some way if I send in for the rebate?

 

[mosox]

It's the video card (hardware problem) most likely but have you tried some other drivers?

 

[rolli59]

If you send in the rebate you will have to RMA to manufacturer, otherwise you can go to the Vendor if the package is intact.

 

[Cryose]

Well, I'm certain that my GPU drivers are up to date, but now that I think of it, I don't think I updated my drivers for my motherboard (besides what came on the disc). Are there any other drivers that I don't know about that could be causing the issue?

EDIT: Sigh, I had already cut out the UPC to send out for the rebate, but I haven't mailed it yet. In either case, it looks like my warranty with Newegg is done, so I would have to go through XFX.

 

[mosox]

Try some other video card drivers, not mobo drivers but it looks like it's hardware related. Maybe you can get a 6xxx series instead.

 

[Cryose]

I don't have the experience to make this statement with confidence, but I wouldn't think I would be able to exchange my 5850 for a 6xxx card. That being said, my 5850 should be able to handle these games (except Crysis, perhaps) maxed out without any issues, so it could just be a bad card.

But going back to that post above, could anyone recommend any temperature-monitoring programs?

 

[davewolfgang]

I use RivaTuner with a Windows 7 Gadget.


RivaTuner

GPU Monitor Gadget

 

[Cryose]

That program combo is telling me that RivaTuner does not support my current graphics driver. I'll look into that more when I get the chance.

However, I did attempt to check temperatures using GPU-Z and SpeedFan. After about ten minutes of gameplay (of SCII), I closed out to my desktop and quickly checked my temperatures. My graphics card was at about 30% and around 60 degrees Celsius. My CPU was at about 40-50 C. Would these temperatures be considered high for how long I was playing? I suppose overheating could be an issue; my case has a lot of fans, but because of its relatively small size, cable management is not optimal.

Or it could be because of a GPU malfunction, but I have no idea how I would find out more information about this.

 

[aluminiumframe]

check around for loose connections, then with a temperature monitoring program run a benchmark program you have that is GPU intensive. if you get any image problems, it would be your GPU. if there is a noise, isolate it, try the previous setting that caused a noise. it could be the PSU giving out.

during the benchmark it is important you monitor the temperature so that it does not exceed the manufacturer max temp for the GPU. A failed GPU fan could be involved.

try to disconnect your psu and remove it from your pc, only if the cables are long enough to reach the components from OUTSIDE the box. then if you do hear a noise you can isolate if its the PSU or a GPU fan.


reccomended programs: everest for temperature monitoring.

 

[Cryose]

First, whatever caused the noise no longer seems to be an issue.

Actually, I am now certain that the entire issue originates from the GPU. After looking around online, I saw that others have had the same exact issue (it looks like it's the GSoD, for those who know what that is). However, after downloading an official driver hotfix, my card now fails to send any signals to my monitor, even after removing all drivers (which I had to do by removing the card from my system). While I appreciate all the help you guys have given me, I think I need to take this up with XFX customer support now.

 

[rolli59]

New card working fine then suddenly starting to have graphics glitches and hangs usually points that way!