Jump to content


Photo

Help with a freezing computer


  • Please log in to reply
6 replies to this topic

#1

KaZi61
  • KaZi61
  • Member

  • PipPip
  • posts 118

Posted 06 March 2019 - 08:38 AM

Hi everyone,

 

I have a PC that's freezing when I use 100% CPU and high levels of ram. It looks like things die when the ram gets close to 100%. It's happened when using at least two different programs that work well on other PCs. I have also run a memtest which passed and done a clean reinstall of windows.

 

When it does freeze and I restart I don't get an error reported so I have no idea what the cause of the problem is.

 

Any ideas on how I can get a better idea of what the problem is?

 

Edit: Just did a stress test, 98% cpu usage and 30% memory usage running at 69 deg cel. Froze after 2:42

 

Cheers

 

Specs:

 

Ryzen 5 2600

8gb (4x2) 2400 DDR4 ram

GTX 1050 ti

Asus Prime B350M

samsung 860 evo 250gb ssd

1tb hdd

corsair cx450w psu


Edited by KaZi61, 06 March 2019 - 09:03 AM.


#2

LinuxUser
  • LinuxUser
  • PriceSpy

  • posts 8,999

Posted 06 March 2019 - 10:48 AM

That's really strange. Check the Error Log to see if that has anything pertinent at the relevant times.

 

Is that temperature from immediately before the crash, or soon after starting the test? Have you overclocked, even slightly? Is the RAM set to the default speed or the XMP or similar profile?

 

It looks to be related to the CPU load, otherwise the stress test shouldn't have frozen or the memtest should have failed, but it could be anything...


  • KaZi61 likes this

Recommended Power Supplies Recommended review sites Why stores shouldn't be removed RAM voltage

i5 7400, Gigabyte GA-B250M-D3H, 16 GB DDR4, integrated graphics, Adata XPG SX8200 480 GB (NVMe), 3x Kingston SSDNow KC380 120 GB, Corsair MX100 128 GB, Spinpoint F3 1TB HDD, DVD writer, all inside a wood case I built, Dell U2412M 24", U2311H 23" IPS and Philips 150B4 15" monitors, wood-bodied mouse, Dvorak keyboard, openSUSE Tumbleweed, Windows 8.1 Pro in a VB VM inside Linux

I may occasionally give advice regarding (consumer) law, but it is only my opinion based on my reading – I am not a lawyer.

Romans 3:23; 6:23; 5:8; 10:9-10,13


#3

KaZi61
  • KaZi61
  • Member

  • PipPip
  • posts 118

Posted 07 March 2019 - 09:44 AM

That's really strange. Check the Error Log to see if that has anything pertinent at the relevant times.

 

Is that temperature from immediately before the crash, or soon after starting the test? Have you overclocked, even slightly? Is the RAM set to the default speed or the XMP or similar profile?

 

It looks to be related to the CPU load, otherwise the stress test shouldn't have frozen or the memtest should have failed, but it could be anything...

 

You fixed it! (maybe?)

 

The mobo was in some kind of self auto overclocking mode in the bios. I put it into the normal mode and it seems to be going well so far. I will keep an eye on it and see how long it can go without dying. 

 

Thanks a lot!

 

Edit: Temp was the reading when the screen froze.


Edited by KaZi61, 07 March 2019 - 09:44 AM.


#4

LinuxUser
  • LinuxUser
  • PriceSpy

  • posts 8,999

Posted 07 March 2019 - 12:53 PM

That sounds extremely likely. It can't have been giving it enough voltage for the clock speed. With that kind of temperature when it froze it definitely isn't a thermal issue, so other than overclocking it leaves the CPU (which is very unlikely), or a faulty motherboard (unlikely, but possible, e.g. VRMs).


  • KaZi61 likes this

Recommended Power Supplies Recommended review sites Why stores shouldn't be removed RAM voltage

i5 7400, Gigabyte GA-B250M-D3H, 16 GB DDR4, integrated graphics, Adata XPG SX8200 480 GB (NVMe), 3x Kingston SSDNow KC380 120 GB, Corsair MX100 128 GB, Spinpoint F3 1TB HDD, DVD writer, all inside a wood case I built, Dell U2412M 24", U2311H 23" IPS and Philips 150B4 15" monitors, wood-bodied mouse, Dvorak keyboard, openSUSE Tumbleweed, Windows 8.1 Pro in a VB VM inside Linux

I may occasionally give advice regarding (consumer) law, but it is only my opinion based on my reading – I am not a lawyer.

Romans 3:23; 6:23; 5:8; 10:9-10,13


#5

KaZi61
  • KaZi61
  • Member

  • PipPip
  • posts 118

Posted 07 March 2019 - 01:13 PM

It went for a significantly longer period of time while at 100%, I had a quick look at the cpu temperature and it was 95 deg cel. I figured this was too high based on a quick google search so I opted to buy a cooler. I must have done something wrong during the installation so I figured starting fresh would be the way to go. I got an AiO cooler and will give the CPU a good clean up before the installation. 

 

I know it might be overkill but since the PC will be running at 100% for a week or so at a time I would rather be safe than sorry.

 

Hopefully that will be the last of it!



#6

LinuxUser
  • LinuxUser
  • PriceSpy

  • posts 8,999

Posted 07 March 2019 - 01:30 PM

If it was hitting 95 then, as it isn't overclocked, you've definitely got a cooling problem somewhere. The AMD specs give that as the maximum temperature, so it was probably throttling too, which would hurt performance a lot. I'd guess the most likely cause is heatsink installation.

 

If it is running for that long at 100% you definitely want to keep the temperatures down, so getting a decent cooler is a good idea.


  • KaZi61 likes this

Recommended Power Supplies Recommended review sites Why stores shouldn't be removed RAM voltage

i5 7400, Gigabyte GA-B250M-D3H, 16 GB DDR4, integrated graphics, Adata XPG SX8200 480 GB (NVMe), 3x Kingston SSDNow KC380 120 GB, Corsair MX100 128 GB, Spinpoint F3 1TB HDD, DVD writer, all inside a wood case I built, Dell U2412M 24", U2311H 23" IPS and Philips 150B4 15" monitors, wood-bodied mouse, Dvorak keyboard, openSUSE Tumbleweed, Windows 8.1 Pro in a VB VM inside Linux

I may occasionally give advice regarding (consumer) law, but it is only my opinion based on my reading – I am not a lawyer.

Romans 3:23; 6:23; 5:8; 10:9-10,13


#7

KaZi61
  • KaZi61
  • Member

  • PipPip
  • posts 118

Posted 13 March 2019 - 03:53 PM

Well things are running smoothly and pretty cool. Currently at 100% for a few hours and I haven't gone over 31deg cel according to the corsair software and 45deg cel according to coretemp. Not sure which one is correct but either way its pretty good! 

 

Thanks for the help, always appreciated!


  • LinuxUser likes this




1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users