Strange behaviour, could be motherboard?

Toggle sidebar Toggle sidebar
A

alek73

Honorable
Jun 27, 2013
12
0
10,510
Hi guys, I have a strange problem and my average hardware knowledge is not enough. I have been looking around the net for a month but I could not find anything specific. If anyone can help me to identify the problem, it would really mean a lot. Here is what happens.

My Windows 7 started freezing up a month ago. For a while, it was like once in 4 hours of use. Then, one day, it was freezing up after 5 minutes (of intensive use). I would restart and continue with the intensive use (game) and it would hold for another 5 mins, then it freezes again. There was no change in this behaviour, so I have started to investigate the issue.

Scanning the RAM showed no errors. Then I noticed something with the BIOS. Whenever the PC turns on, the first message appears on the screen, stating the manufacturer (Award) and then it would wait for like 30 secs or a minute. I remember this state to be taking just a second in the past. Then, it would start with the memory test, and other events at the normal speed. So, the gap is just after the BIOS version report and before any hardware report/tests. I have checked the BIOS setup but no change in options resolved this issue. Any option involving HDD or RAM quick/skip checks did not remove this gap. Next, I have downloaded the latest BIOS and updated it. The first time after the update when I have restarted the PC, there was no issue and it looked like being resolved. But, after a while I have realised that this BIOS start issue was back and it would happen more and more until it is now back with every start.

Anyway, I went with the reinstall of Windows. I made another copy on a another hard disk and it worked just fine for a week. But, today - the same freeze happened with the new windows, while browsing. It happened just once, but it looks like the problem is back. I believe that this copy of Windows is pretty fresh and clean, so now I don't think it's a software issue.

I suspect the motherboard, but I am not sure if that is the case. Since, the new Windows were just fine for a week. I have scanned for viruses, but the reports were clean. I could repeat some hardware/software tests with better tools if anyone would suggest that.
 
Solution
KnowItAll
Thank you, that makes a lot more sense.

I've never heard of a cpu running 99c or overheating for that matter because of a faulty psu, but It is possible I guess.

The psu would likely NOT cause the fan's on your GTX460 to fail.

At this point, the first think I would do is replace the power supply, thermaltake is not known for making good power supplies, and it's 4-5yrs old so it's likely failing or will fail soon.

Make sure you get a quality unit. PSU is the most important part of your computer. 500w is more than enough if it's a good unit.

If it's not at LEAST 80+ Bronze rated, don't buy it.

Once your replace the psu, then you can be in a better position to know what's going on. Hope this helps and sorry I got a bit...
Sort by date Sort by votes
KnowItAll

KnowItAll

Honorable
May 13, 2013
397
0
10,810
What is the cpu temperature when the crash occurs? Use Hardware Monitor.

This could be cpu overheating, memory issue or power supply failing.

Are you saying the computer automatically starts testing ram on boot?

There should be a fast boot option or something in the bios where you can skip the bios boot screen altogether, see what that does.
 
Upvote 0 Downvote
A

alek73

Honorable
Jun 27, 2013
12
0
10,510
Since I am using my newly installed Windows, the freeze has happened just once for now. So, I can't really tell the monitored values before the issue becomes more common (hopefully, it will not). But, I have used the Heavy Load software to do a stress test and everything looked really good - CPU went to 99C with Fan @2500, Mem still fine with 97% full and no crashes. Everything looks stable at the moment. Aside from BIOS, which is still hanging. I have looked once again to make sure that the Quick Boot is on, but it does not help. Quick Boot, I believe, skips some tests (like the memory) but this hanging is happening before that step. I've measured, sometimes it holds for almost 2 minutes before proceeding. I am sure it's not supposed to be like that. If my Windows starts freezing up again, I will see to update here with the monitored values.
 
Upvote 0 Downvote
A

alek73

Honorable
Jun 27, 2013
12
0
10,510
Sorry, KnowItAll, I think you are missing a point. Yes, it went up to 99C in a *stress test*. It is what the stress test does - it puts a load on all the parts (CPU/GPU/RAM/HDD) until their breaking point. The point is to see when the system breaks - and it is a good result that the system did not break at any point - even with the CPU on 99C it was still stable. So, for one I know that the CPU will not cause an issue even if it goes that high. During regular use my CPU is around 40C. So, I don't think it's CPU, that's what I am saying. Or, RAM/GPU/HDD. It has to be MB or PSU, but I don't know how to prove it. And, the strange thing is that it did not happen during the first week of using new Windows, so it looks a bit like a software issue. Also, BIOS behaviour may be related to it, or maybe not.
 
Upvote 0 Downvote
KnowItAll

KnowItAll

Honorable
May 13, 2013
397
0
10,810
Dude, YOUR MISSING THE POINT. You cannot run your cpu at 99c. IT WILL FRY, IT CAN ALSO FRY THE SOCKET OF YOUR MOTHERBOARD. It's supposed to shut down way before 99c, not sure why it's not.

You seem oblivious to what I'm telling you. YOUR PROBLEM IS YOUR CPU IS OVERHEATING. One would think this is common sense, but in your case it's obviously not.

I can run a stress test and my cpu never goes over 65c.

 
Upvote 0 Downvote
A

alek73

Honorable
Jun 27, 2013
12
0
10,510
OK, let me confirm if I now understand what you mean. Otherwise, please, correct me and see if you can explain to me what I'm missing.

1) One thing that we seem to agree about is that the stress test will heat up the CPU. So, once the temperature reaches a certain point, what should happen? The CPU will stop following the tests in order to prevent overheating. Otherwise, it would overheat and freeze/shut down. Do we agree on this?

2) If so, then my CPU follows the tests until it reaches 99C, while you think that it should be lower, say, 65C when it should decide not to follow the tests any further? If so, I could probably set it up in my BIOS that 65C should be the top point.

3) The other aspect can be that my CPU is heating up too soon and that's why my Windows were freezing up after 5 mins of use? The stress test shows that my fan is giving it's best and goes up to 2500 RPM when needed (so, I don't think that one is worn out). The case I have is a good quality, but could this indicate that some of the fans other that the CPU cooler are worn out and that the inside temperature is too high? I will try and check about that.
 
Upvote 0 Downvote
KnowItAll

KnowItAll

Honorable
May 13, 2013
397
0
10,810
You haven't even told us what cpu were talking about, so I couldn't possibly answer your questions, I'm not a psychic.

The cpu will throttle itself to keep cool.

Have you looked at your heatsink fan to see if it's clogged with dust? Maybe it came unclipped or it's been mounted incorrectly?

 
Upvote 0 Downvote
A

alek73

Honorable
Jun 27, 2013
12
0
10,510
Sure, here are some other specs and please ask if anything else can help:

CPU: Intel i7 870 (2 years old).
MB: Gigabyte P55-UD3L (2 years old)
PSU: Thermaltake Toughpower 600w (4-5 years old)
Case: Thermaltake Aquila (4-5 years old)

GPU is currently an old nVidia GeForce 8800GT, because I just got another one (MSI Hawk nVidia 460GTX) died on me recently. That may be indicative too, what happened with my GPU. It died prematurely (not even 3 years old and still in the warrant period). Last year, I have noticed one of two GPU fans had an issue but it didn't look serious. Finally, both fans stopped working recently. Then I got a new GPU (Gigabyte nVidia GTX 660) few weeks ago and it died within days. Now, while the previous one may have been worn out and the new one may just be bad luck, I am now thinking that it might be related to my PSU or motherboard?

I have checked all the fans in my case and they look pretty strong to me. I will now check the CPU heatsink for dust. I do clean it thoroughly on a 3-6 month period.
 
Upvote 0 Downvote
KnowItAll

KnowItAll

Honorable
May 13, 2013
397
0
10,810
Idle temp is irrelevant, since the processor is not running full speed at idle.

Intel lists the Tcase as 72.7, so I would assume it starts throttling somewhere around there or a little higher.

I can't make any sense of your post. You have an 8800GT, because you just got another one, (msi hawk died). I can't make any sense of that sentence.

Secondly if your 460gtx was under warranty, why didn't you get a warranty replacement?

One of two gpu fans had an issue but it didnt look serious? How does a fan look like it has an issue but it's not serious? What does that mean exactly?....... Then both fans stopped working......Then you got a new 660 and it died too.

Sounds like your power supply is to blame. I would replace that piece of junk regardless of any of your other problems.

No matter, you can't continue to run your cpu at 99c. I wouldn't even recommend running it at 70c.
 
Upvote 0 Downvote
A

alek73

Honorable
Jun 27, 2013
12
0
10,510
:) Sorry, I must sound funny trying to explain everything that is happening. Also, English is not my first language. Let me try and put some more structure to it.

But first, I have just checked for the dust around the CPU heatsink. It is squeaky clean (I've dusted it recently when replacing my GPU) and it is firmly attached. Looking good, as far as I can tell.

About the GPU (old ones and new ones), here is the sequence of events.

1) I bought a 460GTX three years ago. It had two fans. After two years, I have noticed that one of the fans stops at times. At other times it was running and too loud. There were no drops in performance or any other effects, so I did not worry too much. After one more year, the second fan started to play funny and GPU would overheat. I went back to the dealer and they have confirmed the problem. This was a month ago.

2) While waiting for the replacement I have been using an old 8800GT. During that time, I have first noticed the freezing up of my computer. It would happen after 3-4 hours of use.

3) The dealer offered me a replacement, but I went for an upgrade to 660GTX. It worked, but there were some strange things with it. At times, I would hear a clicking sound - like a sound that you can hear when restarting a computer. But the computer would not freeze up or restart. After a couple of days the GPU stopped working correctly. The display was not blank, but deformed and with junk (artefacts).

4) I am now using my old 8800GT again. I plan to replace my 660GTX with a new one.

The more I think about this, the more it sounds like a PSU issue to me.
 
Upvote 0 Downvote
A

alek73

Honorable
Jun 27, 2013
12
0
10,510
Ah, there are news and I feel like a total idiot now.

I plugged in the 660GTX again just now. And it runs fine now. So, the issues (deformed display, artefacts) were coming from elsewhere it seems. So, for everything that was going around would it be PSU? Could the PSU be blamed for the CPU overheating too?
 
Upvote 0 Downvote
KnowItAll

KnowItAll

Honorable
May 13, 2013
397
0
10,810
Thank you, that makes a lot more sense.

I've never heard of a cpu running 99c or overheating for that matter because of a faulty psu, but It is possible I guess.

The psu would likely NOT cause the fan's on your GTX460 to fail.

At this point, the first think I would do is replace the power supply, thermaltake is not known for making good power supplies, and it's 4-5yrs old so it's likely failing or will fail soon.

Make sure you get a quality unit. PSU is the most important part of your computer. 500w is more than enough if it's a good unit.

If it's not at LEAST 80+ Bronze rated, don't buy it.

Once your replace the psu, then you can be in a better position to know what's going on. Hope this helps and sorry I got a bit agitated, I did not know English was your second language.

 
Upvote 0 Downvote
Solution
A

alek73

Honorable
Jun 27, 2013
12
0
10,510
Thanks. I got back today with the Seasonic X-series 850W. It is 80+Gold and working nicely so far. I hope that resolved my issues.

BTW, Thermaltake was my only option at a time and it was about 100 euros. So, not too cheap. They have claimed the average efficiency of 81% and I was already happy with their cases. There were no 80+ certificated PSUs at my market at that time. Now, the choice looks better so I can take advantage of it.
 
Upvote 0 Downvote
A

alek73

Honorable
Jun 27, 2013
12
0
10,510
Unfortunately, the problem is still present. Here is what happens:

1) I have placed the new Seasonic PSU.

2) I was using the 660GTX GPU. After 30 minutes the junk artefacts appeared on screen and the computer resets. After resetting, the junk is still there, appearing during BIOS and getting worse with the Windows up.

3) I have plugged in the old 8800GT that I have and the screen looks good so far.

I plan to test the 660GTX on another PC.
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom