1

My laptop (runs on Windows 7) recently started to freeze after moderate workload. I can play a game, watch videos, sometimes even put it under heavy workload for extended period of time and it works just fine (of course it gets hot but it's not melting hot, somewhere around 80-90 degrees Celsius, that is quite affordable for a laptop). But shortly after such workload is finished, while it's getting cooler, the system gets frozen. The interesting part of it is that it keeps randomly freezing for a while if try to turn it on after it (before it cools down), and what is more important it may happen before I actually login (it even happened during powering up, before system has even been loaded a couple of times).

Currently I'm trying to collect some memory data and analyze it using built-in Windows tools but to no avail (nothing that consistently happens before freezing event). Also, provided that this happens before system is loaded makes me think that this is rather hardware problem. All tools I have show zero problems with RAM or hard drive and I have zero clue of which exact part causes this.

Is there any advanced technique I can rely on to find out the root cause of the problem or at least the broken piece of the PC?

15
  • 3
    Random freezes are one the hardest to diagnose, sometimes it just plain luck finding the problem. It could be hardware, driver or software, I would start with disabling programs from startup one at a time (reboot each time and test), see if it stops freezing. Use MSconfig/startup tab to do this. Update any drivers you can. Install speedfan and monitor temps. – Moab May 30 '20 at 21:41
  • @Moab Thanks for the input I'll go through these steps. However as title and description suggest it's most likely independent from software problem and I need to somehow examine hardware problems. Ideally I'd like to see a list of possible reasons with exclusion of what it cannot be with given information in mind. I understand it would be a high-effort answer and I greatly appreciate any piece of advice here from anyone – The Dreams Wind May 31 '20 at 8:54
  • Backup now all that you can, then start to investigate. What do you exactly mean with the system get frozen? (Really you should backup often) Does it turn completely off? (Remember that is important to Backup) Turn black? (Backup Now) You can just move the mouse but no other interactions? (Did you already do the backup? Go) No mouse/keyboard/led activity but still visible last screen? (Backup, now) Do you listen if fan are still working? (Backup) HDD? (Backup) Eventual USB? (Backup) Did I remember to say you to do a full backup? Do it and after add some details. Good Luck. – Hastur Jun 2 '20 at 21:50
  • @Hastur by "frozen" i mean it does not respond to any kind of input (keyboard or mouse) and just shows "last screen". Any sound/playback also stops immediately, just last visible state on the screen without any artifacts on it, completely frozen. I also noticed what during this state my mouse, which has some rgb flashing stops flashing. That's about it, i believe. Fan is still working. I tend to think that this might be caused by GPU/CPU somehow, as these are one of the hottests parts after any workload, but it's just a wild guess.. – The Dreams Wind Jun 2 '20 at 22:07
  • 1
    Did it ever restart by itself? Since it is random we can only guess.Then you can test each guess :-). Give it a try with an intensive memory test (you can start with something like i.e. a usb with linux and memtest86). I had a similar pseudo-random problem with the second memory bank (old laptop). Only after a memory test of about one hour I was able to find it. BTW to exclude software/drivers problem you may test the computer with a Linux Distribution from an external usb. Log the thermal behavior too.(Seariously do a backup) – Hastur Jun 3 '20 at 0:00
2
+50

The usual suspects

You experienced random freeze, unfortunately a not reproducible one.
In cases like yours, there is a bunch of usual suspect:

  • Software (programs, drivers, OS ...)
  • Hardware (fan, thermal control, graphic card, HDD, memory)

The procedure is to formulate all the hypotheses, to order them by causes, and then to verify their correctness.

It is important not to be discouraged if you are not immediately successful: indeed, meanwhile you proceed you rule out candidates, increasing the probability to find the one you are looking for.

Your specific case

You stated:

  1. I can play a game, watch videos, sometimes even put it under heavy workload for extended period of time and it works just fine

    • So the fans work it can be their control hardware or software

    • Graphic card works too (video, games...)

    • heavy load --> So it should work even the thermal control hardware/software

  2. Shortly after such workload is finished, while it's getting cooler, the system gets frozen.

What we can rule out.

OS related. The use of another OS loaded on a CD/DVD or nowadays on a USB pen, may rule out all the OS/Software/Driver/Configuration related causes (except rare programming errors...).

Critical temperature. Now if it was a thermal problem you should experience an abrupt interruption and probably you should be not able to start again the computer for a while (protection).

Drive failure. If it was the HDD (or SSD) you should have a somehow working system that cannot write on the HDD or even read (this happens in cases like this with Linux, maybe with windows you should have a blue screen...)

What remains?

Memory. A memory failure instead may hang the system leaving the graphic card and the motherboard working and allowing you to see the last screen refreshed...

The memory failures may occur for a specific operation (not only the simples read/write operation), and they can occur even only when the temperature is higher than a threshold but not so high to rise an alarm.

With Linux it comes memtest86+ that can run to test your memory before loading the OS. It is often needed a long test (even of the order of some hours) to check all the type of operation a memory can do and at a different temperatures.

Further readings

2

I see three possibilities for your problem.

  1. I once had a similar problem with my Clevo Notebook: it wouldn't even boot the system when it got past some break point that'd restart the system right away, which was rather unclear to me what it was, but I knew it was getting more and more frequent. As the notebook was rather expensive, I started to feel a bit dismayed thinking about an expensive system board replacement, because such kind of problem that worsens over time shows it's about something that's degrading, i.e. hardware. It turns out I had to send it back to the vendor for maintenance and the culprit was single a component in the main board that energized the GTX 1070 video, which was easily replaceable by them, in their workshop (i.e. I wouldn't be able to mess around with soldering components to a delicate motherboard). It was a cheap fix and the computer is good some two years later, but I had to ask for support.

  2. If temperature and workload doesn't seem to make any difference between running and freezing, it might point out for two other possibilities. One is RAM, in which a faulty position might be used just every now and then. This one is easy to debug if you have more than one bank, just remove one of them and see if the system works. If it does, you can try to reproduce the issue by swapping the banks so to confirm the other is defective. If the issue still happens with the first bank, insert back the removed one, removing the defective-candidate one (swap them). If just one of the two+ banks solve the problem, you know you've got faulty RAM. The other option would be...

  3. ...a cracked component or weld you your board, which is pretty much compatible with a system that hangs when it cools down: things expand and welds tend to become softer as they heat up, and when it cools back down, things shrink back while they're still soft. After enough time powered down has passed, contacts come back together again and things seem normal for a while. That'd require substituting the motherboard in the worst case, and I'm not sure it's something that can be diagnosed without a proper worlshop/lab. Additionally, such behaviour would become worse and worse over time, as such cracks/welds degrade. I take this worsening as a tell-tale signal of components degrading as described. It can happen either with welds or be internal to the delicate components.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.