Tech Support banner

Status
Not open for further replies.
1 - 16 of 16 Posts

·
Registered
Joined
·
11 Posts
Discussion Starter · #1 ·
While playing some games like Left 4 Dead 2, Fallout New Vegas and Age of Empires III Asian Dynasties, the computer crashes and I have to press and hold the power button for it to turn off. It happens frequently in AoE III Asian Dynasties and never happens in Team Fortress 2 for example.

The screen goes black and the monitor light starts blinking, and the speakers make a really annoying loud buzzing sound.

I've tried everything, various video card drivers, disconnected all USB devices, clean OS installs with no extra software on the background...

The PC is ~3 years old.
Windows 7 SP1 64 bits
I've had BSODs before but not since the last format.

I found this thread using Google, looks similar.
 

Attachments

·
Registered
Joined
·
2,210 Posts
Your symptoms reek of hardware failure. My initial suspicion is your video card. Though let's run through some testing.

First thing's first: some initial troubleshooting. Check inside your PC for any dust or slow/stopped fans. Ensure that proper airflow is taking place (cold outside air in, hot inside air out). While you're inside, go ahead and re-seat your video card, as well as your memory sticks, and reconnect your drives' power and data cables (both drive and motherboard ends). Check things after all of this, and see if the issue has been resolved or has been very noticeably reduced. If not resolved, continue onward.

Before the testing, we'll want a couple HWinfo logs from ya. Set it to Sensors only at startup, then have it log two instances: one for idle, and one for high load (gaming, etc.). Make sure each log records a decent amount of time (~15 minutes). So we can get some good solid averages and discover trends going on.

Once that's done, on to the hardware diagnostic tests:

GPU: MemtestG80/CL - If your video card supports either of the tests, run twice.
RAM: Memtest86+ - Run 7+ passes.
CPU: Prime95 - Torture Test, Large FFTs, 9+ hours.
Disk: Seatools - All basic tests. No Fix all or advanced ones.

Note that these tests (besides MemtestG80/CL) are available on the UBCD. It's also preferable you run them on there as it offers a light, reliable environment to run them on. Note that Prime95 is called Mersenne Prime Test in the UBCD.
 

·
Registered
Joined
·
11 Posts
Discussion Starter · #3 ·
Thanks for responding!
Since I know very little about hardware I did my best to clean it up. All the fans are working and almost all of the dust is gone.

Now, the logs. What happens if during the high load test the PC crashes? Does it still log? I'm in the middle of the idle test (I'm typing from my laptop) and after it ends I'm going to install Age of Empires III: Asian Dynasties and play it during the high load test.

Again, thanks! :smile:
 

·
Registered
Joined
·
2,210 Posts
I do notice that your CPU cores (especially core 0) are getting pretty hot in the latest log you provided. The first core is getting up to 88C. This is pretty hot for a CPU, and it's most likely there's some instability involved as it starts to reach critical temperatures.

However, the previous logs show lower temperatures, and if you are correct in that these logs stopped when the game (and system) crashed, then we're most certainly dealing with permanent hardware malfunctioning that only seems to be worsened with higher CPU temperatures.

If you wish, you may continue to perform the hardware tests to try and determine the exact hardware that's failing.
 

·
Registered
Joined
·
11 Posts
Discussion Starter · #9 · (Edited)
The PC only crashed on the last log (crash.zip).
In the first logs (Idle and High Load) the PC was fine, and I played until I got bored. :p

EDIT: Will it help if I apply some heat sink paste? Since I don't know how, I've never done it (the PC is +3 years old). Also I have a stock cooler on the CPU and 0 fans on the case.
 

·
Registered
Joined
·
2,210 Posts
It sounds like your CPU may just be reaching critical temps and is therefore becoming unstable. Given that you stated you cleaned things up, the only other thing I can think of is to re-seat the CPU heatsink, which means removing it, eliminating any existing thermal paste, and applying an appropriately uniform layer of new thermal paste before re-seating it back onto the CPU. Since cleaning the inside still didn't seem to reduce temperatures all that well, you'll want to do this to ensure that the temperature is as good nominal levels.
 

·
Registered
Joined
·
2,210 Posts
Wait... so you're saying you've had the CPU running this whole time without thermal paste applied between the heatsink and the CPU? Or you're saying that it was already there when you bought the PC and you just never reapplied a new layer?
 

·
Registered
Joined
·
11 Posts
Discussion Starter · #13 ·
I never reapplied a new layer. This isn't a custom PC. When you asked me to clean the dust, I removed the CPU cooler and noticed the thermal paste was missing in some spots.

Thanks for the help! I'll ask a friend to apply a new layer.

Just one last thing. When I restart after a crash, the Windows comes up with a message saying the graphics card has failed (or something like that). On the last log (the crash one), are the GPU temperatures okay?
 

·
Registered
Joined
·
2,210 Posts
The thermal paste can dry over a long, long period of time, but pulling the heatsink off will cause it to stick in some places on the heatsink and others on the CPU, giving the appearance that it's dried up (try it with toothpaste or peanut butter on two surfaces, same effect). Still, it doesn't hurt to reapply a new layer, especially given that there is in fact some heating issues. At most, you'll need to buy a new CPU heatsink/fan combo.

The GPU temps looked very stable and fairly cool (~60C range). GPUs also have a better tendency of handling heat.

Though now that you mention the GPU error message, can you write down next time the exact error message it provides you? It sounds like me like something called TDR kicked in, in which case you may very well be having a bad GPU issue, as was my initial diagnosis. Again, don't forget the hardware tests I provided in my first response!
 

·
Registered
Joined
·
11 Posts
Discussion Starter · #15 · (Edited)
I'll do those hardware tests in a moment.

About that TDR link you posted. Now I remember that sometimes, while playing Team Fortress 2, the game freezes and I have to Ctrl Alt Del, and if I'm quick enough to get to the desktop I'll see a balloon tip on the system tray (it doesn't stay for long) saying something like "the display driver has stopped working and it has been recovered" (I'm not sure if these are the exact words, but it's something like that). Also, if I don't Ctrl Alt Del and just wait, the game will eventually resume.

It's hard to reproduce this error and it's not the same as the original system crash I talk about in this thread. Maybe they are related. I don't know. :p

EDIT: English is not my first language, so if you don't understand something I say please say it. :smile:

EDIT 2: I was looking in the reliability history and found one
 

·
Registered
Joined
·
2,210 Posts
Ouch, I'm afraid your GPU is not looking too hot. I highly, highly doubt we're dealing with a bad driver here, especially since you mentioned you already tried variously dated drivers. Looks like yer gonna have to replace her.
 
1 - 16 of 16 Posts
Status
Not open for further replies.
Top