Computer is Dying. Can't figure out what's the problem.

[rehpresent]

I believe my computer is dying. I've been trying the past few days (when the problem arose) to diagnose what's wrong. I'm not too sure if it's a software issue, driver issue, or even a hardware. I'm leaning towards hardware issue after my efforts to figure out what's wrong (likely PSU, or maybe motherboard).

So, I'll try to write this chronologically to the best of my abilities.

First off: I got my homebuilt computer from my friend. He wiped the hard drive, installed a bootlegged windows (windows 7 home premium 64 bit, that could never update), and gave it to me. Link to my rig: http://www.3dmark.com/3dm/7105107

Starting 2 days ago, my computer started having bsod. I tried to figure out what was causing the crashes, so I looked at bluescreenview to see what caused the problems. The results were "kmode exception not handled" (pacer.sys) and "attempted execute of noexecute memory"(hal.dll). As a result, I thought these were driver issues. I tried to update the drivers for the graphics card/chipset. Afterwards, things were fine for the next day and half.

However, starting this morning, more problems came. My computer started crashing again and failed to boot multiple times. Sometimes the monitors would light up, displaying that the computer failed to boot. Other times, it wouldn't display anything and instead would begin beeping (4 consecutive short beeps. Based on AMI, signified that I had a system timer failure). I turned off the power and turned it back on. The computer would then properly boot up. However, it relapsed once more to the previous problem of failure to boot. But I did the exact same thing, and now it seems to be "fine" the past few hours ( as in it would boot up successfully every time). Things keep getting worse and the comp would sometimes turn off and boot up again (as though the power had gone out. This has been a reoccurring theme throughout the past few days, and would scarcely happen over the past 6 months).

Some more blue screens occurred throughout the day, but my memory doesn't serve me well enough to remember between which events each bsod happened. Throughout the day, I've had bsod for the following reasons ( in chronological order): System PTE misuse, System service exception, page fault in nonpaged area, system thread exception not handled, and finally bad pool header. I've decided to system restore today, to see if there would be any progress made, but it's stuck on "preparing to restore your system" for the past hour.

I've tried using msdiag to determine if there's any problem with any problem with the RAM, and the results didn't state anything was wrong with them. I can only think that there's a problem with the software (since it's bootlegged), Motherboard, or possibly the PSU (there might be a small hint of problems lying within the RAM).

Any suggestions? I'm seriously pulling out my hair trying to figure out what's the problem. The inconsistency with the boot up and the fickle reboots aren't helping me conclude the problems.

*I'll continue to update. I know my writing is poor, because my brain is scattered all over the place.

*Edit 1: Just got 2 bsod's while preparing to restore. First was "System thread exception not handled" (due to wdf01000.sys) and "System PTE misuse" (again).

*Edit 2: Reinstalled windows 7 (with a legit copy). Still have bsod, but with "memory management" errors.

 

[Chicano]

One hard drive had 162 errors occur while data transferring. It was while I was playing a game (to test out how my computer is doing). Since then, no more errors occurred during data transfers. From what I understand, Hard Disk Sentinel suggests the problem is likely from wires of the hard drive.

I was going to test out each stick of ram in a minute, but you say "So, as RAM has checked ok till now," should I not bother checking it anymore? If that's the case, then I'll likely do what I did before to cause the crash and see what happens. If it occurs again, I'm going to assume it's either a PSU or Motherboard problem.
But how do I figure out if it's the PSU or motherboard afterwards (if it's not the RAM)? Seems like both things don't have an actual test to determine which one might be failing. Do I just take a blind shot and buy a replacement to the cheaper one?

Those 162 errors as you said, are data transfer errors... the type of errors I meant are HD platter surface errors or (bad sectors)... HD Sentinel tells you how many if any errors it detected (in red, orange, yellow or green background) See this example image:

The "RAM has checked ok til now" comment was based on your software checks... the suggested physical hardware tests are still recommended (test each module one at a time on every slot).

To test the PSU you need to try it in another computer, or try another PSU in your computer. Your PSU working without problems in the other computer would indicate it's not the casue, and the opposite if it does. And if a different PSU in your computer makes it run without problems, your PSU is bad... with problems would mean it's not the PSU.

All tests leading to conclude you have HDDs, RAM and PSU in good condition, would leave the Motherboard as the culprit.

 

[rehpresent]

Honestly, I had no problems yesterday. I ran 2 intensive games and a stream yesterday. Not a single crash occurred and the computer was running smoothly (for the most part). I'd like to close this thread and just blame it on the drivers + bootlegged OS. I checked the voltages on the BIOS and everything is within the 10% deviation. However, one thing that slightly bothers me is the how hot my gear gets. I'm not sure how long it's been like this, and it just happens to be a normal occurrence. But, I noticed the temperatures rising borderline high (for instance the gpu would go up to 93-94 degrees, when the AMD suggested safety temperature goes up to 95). I was thinking maybe I should get some thermal paste and figure out how to apply it properly. My friend stated that it wouldn't really be necessary, as the computer would just throttle more often.
(Also, I believe my CPU isn't overclocked but instead on Turbo, as it only reaches a maximum of 3.7ghz).

*Edit: Forgot to respond regarding the HDD. I can't run any surface tests on Hard Disk Sentinel because it requires the paid version. Is there any freeware that would accomplish the same thing?

 

[shep611]

Honestly, I had no problems yesterday. I ran 2 intensive games and a stream yesterday. Not a single crash occurred and the computer was running smoothly (for the most part). I'd like to close this thread and just blame it on the drivers + bootlegged OS. I checked the voltages on the BIOS and everything is within the 10% deviation. However, one thing that slightly bothers me is the how hot my gear gets. I'm not sure how long it's been like this, and it just happens to be a normal occurrence. But, I noticed the temperatures rising borderline high (for instance the gpu would go up to 93-94 degrees, when the AMD suggested safety temperature goes up to 95). I was thinking maybe I should get some thermal paste and figure out how to apply it properly. My friend stated that it wouldn't really be necessary, as the computer would just throttle more often.
(Also, I believe my CPU isn't overclocked but instead on Turbo, as it only reaches a maximum of 3.7ghz).

*Edit: Forgot to respond regarding the HDD. I can't run any surface tests on Hard Disk Sentinel because it requires the paid version. Is there any freeware that would accomplish the same thing?

glad to see things have been fine for you and hope they continue . as for your temperature getting very high could be a very good reason why you system crashes . wish i knew about this at the start lol . most systems will shut down for safety purpose . recommend to you artic silver thermal pastes if you use some. also clean your fans maybe even add some new ones and make sure they are venting heat correctly . many videos on line that can show you how to apply thermal and set up fans in your case . and as for your question about freeware i personally haven't ever used any but if you google for Alternatives to sentinel . im sure you may find one .

good luck to you

 

[rehpresent]

I honestly feel like my computer is just like an episode of House M.D.. Diagnose one thing, and then another symptom appears.
After yesterday's great day, my computer is currently consistently freezing and looping sound. Also, it previously booted up without the monitors on (I turned off the power and restarted the comp, and the monitors were working fine). However, the most recent Freeze & audio loop resulted in the same old System Timer error (4 beeps. Failed to restart the boot up about 3 times, then booted up perfectly fine).

I guess it's time to call it a motherboard issue?

*Edit: also, i don't think the heat correlates to any of the crashing any of the failure. Because none of the crashing occurred when the system was heated up. In fact, mostly when it's been idle or about 50% load.

**Edit 2: Loaded bios to default settings. Attempting to reproduce same problem.

 

[Chicano]

You don't have to scan with HD Sentinel.. the scan I meant is the initial scan wich is done automatically to show you the disk condition.. and you can see the results in the main screen as per image... and you can select each HDD on the left, to see the results for each HDD... so, if damaged sectors are not show for any HDD, then they don't exist yet.

That your CPU may just throttle down to protect itself from high temps is a fact but that would cause damage in the long run.. so, it would be better to keep temps down and allow the CPU to perform better... so get the best thermal paste you can, and if within your budget add an aftermarket CPU cooler.

On your last comment: If your Power Options settings are enabled (Windows and/or BIOS), fans and CPU work at full speed when you're playing games, editing video, etc. and so heats up and cools the most at the same time... but as you stop those tasks, the CPU will throttle down as well as the fans, to save power, and if the CPU is not dissipating heat well enough, residual heat could build up and cause all types of errors. And if CPU cooling is inefficient, the CPU may heat up even a while after stoping demmanding tasks. This is just theory so install HWiNFO to monitor system temps and fan speeds both when playing games and at idle soon after stopping gaming and a while after to see how temp readings and performance behave.

 

[rehpresent]

Alright, do you have any particular suggestion on which thermal paste (much like Shep611's suggestion of Arctic Silver's)? Also, can you elaborate as to "an aftermarket CPU cooler?" As stated earlier, I have a liquid cooling (Antec Kuhler H2O 620) and additional fan to heat sink. Lastly, I already installed HWInfo. I posted some of my results earlier from playing a graphics intensive game for about 30 minutes. It never left the "safety zone" (so to speak) even after closing out of the game, but the heating does approach the borderline.


Regarding my crashes that I previously posted today: I can no longer reproduce the crash that occurred earlier. Maybe it's just inconsistent and will come back tomorrow. I'll keep it posted.

 

[Chicano]

Arctic Silver is good and a liquid cooler is more than adequate cooling. I missed that part and thought you had stock cooling.

With that much CPU cooling you should be getting very safe CPU temps… it’s said liquid cooling should control max temps ranging from 60 to 80ºC… but the Intel Core i5-2500K has a Maximum operating temperature of 72.6ºC so cooler than this would be desireable… so if it’s getting higher than that, you really need to renew the thermal paste.

Also check your BIOS and/or Windows power options setting and disable them in the BIOS, the PDF user manual should indicate the setting to configure, and in Windows set the High Performance Plan for better cooling and CPU performance/stability, at least for a while to see if the issues dissapear.

You may also want to ask about safe overclocking settings to know if your's aren't affecting CPU stability and temps.