[SOLVED] PC Freezes during games after applying new thermal paste to CPU (i7 2600k)

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.

Soul Sojourner

Prominent
Oct 30, 2019
21
0
510
Hello. I have had my computer for 7 years now, I built it in anticipation for Diablo 3 in 2012. It was a Windows 7 that later "upgraded" to Windows 10. My CPU is an Intel i7 2600k Sandy Bridge and I have only very mildly overclocked it over the years.

I never used the stock heatsink and fan at all, they are still stored in a box. Instead, I installed a closed loop watercooling system for the cpu only. I have only lately noticed higher temperatures on my cpu while running. Reportedly, the 2600K is a model that runs hot and 90c isn't too alarming, but I was seeing 90c-98c often enough while gaming (No Man's Sky) in the last week that I knew there was a problem. On top of that, it began to make a rattling noise I'd never heard before. Also, I was idling anywhere between 63c and 74c.

I bought a tube of Thermal Grizzly Kryonaut thermal paste and received it today. I removed the closed loop water cooler and the processing unit. In doing so, I noticed the screws for the mount were actually a bit loose, which would explain some higher temps and also the rattling noise.. When I did the original build I used Arctic Silver 5 thermal paste. I used Arctic Silver thermal paste remover and surface polisher to clean the surfaces of the watercooling unit and the processor. I was very careful about removing the old thermal paste and did it out of my case, as Arctic Silver 5 is slightly capacitative (Kryonaut is non-conductive and non-capacitative). The surfaces were very clean when I was finished and I also used some 91% Isopropyl Alcohol to clean some mild dust buildup off a few surfaces of the WCS and the chassis where the radiator is mounted. I didn't do anything else, save for unplug the cpu power cord from the motherboard and the WCS fan, as the cpu power cord was in the way and the fan needed to be unplugged to remove the WCS.

To note, I have checked the WCS while running, the fan works as it should and I can feel the water running through the two tubes. One tube is warm, the other is cool. After applying the new thermal solution and spreading it with the included spreader (ugh), I reinserted the processor, locked it in, and remounted the WCS atop it (in the same position as before). I hooked everything back up as I had it before. After a few minutes, I booted up and my pc rebooted once before I had any visuals, and then booted up. It told me I installed a new cpu and wanted me to change settings, including overclocking options. I changed nothing as nothing really changed. When I got into Windows, I used Speccy to check my running temps. My CPU idle temp dropped to 34c. It sits between 35c and 42c most of the time now, just browsing the internet or watching Netflix. If I run No Man's Sky, I have had few chances to check the temp but what I have seen is temps no higher than 72c so far.

I can boot up the game just fine, load my save and begin playing. However, within 10 minutes the game will freeze. I am able to perform one or two tasks before everything becomes unresponsive. Task manager won't open unless already opened, and nothing I try closes the game. One time I waited for about 10 minutes and then it suddenly performed the tasks I tried before, alt tabbing, ctrl alt del, etc. I was able to do things again and check Speccy, all temps were good. I was not able to shut down the computer, log out, or close the game in any way (which prevented the first two things). Every time I have had to hold the power button to shut down.

Since then I have tried setting all bios settings to Asus optimal, default settings. Setting things to Auto. I continue to have this problem.

Specs:
Windows 10 64 bit
Intel Core i7 2600k Sandy Bridge
16 GB RAM Dual Channel DDR3 (G.Skill 4x 4GB)
Asus P8Z68 Deluxe Motherboard
EVGA Nvidia Geforce GTX 760 (4GB Vram)
500GB Samsung SSD
Corsair 750W PSU
Lancool Chassis

I haven't updated my BIOS in years... though everything worked fine before I applied the new paste, besides the cpu running hot, and now I have a new problem I didn't have before.

Also, no blue screens, and no heat related shut downs before or after.
 
Solution
If there is a newer BIOS available, I would install it. I realize you didn't need to have it before, but before, your BIOS didn't think a new CPU was installed either.

It might actually be worth trying to update the BIOS, because unless there is a KNOWN reason for not doing so, such as a bad BIOS release (Which you'd know about most likely, and so would the manufacturer, and it would have been pulled and replaced years ago. Good thing about waiting a long time to do BIOS updates, it's pretty much impossible to get one that isn't tried and true), it is ALMOST always better to have the latest version than to not have it. They don't release updates just for fun. They do it to resolve known issues and problems, so it's a good idea...

Soul Sojourner

Prominent
Oct 30, 2019
21
0
510
Externally, at least, all power heading to the computer should be stable without spikes or fluctuations. Since I would never run a computer without an UPS. I know that's not the same thing, and that a faulty PSU could still cause problems, but it definitely helps at least.

I'll wait until tonight when I go to bed to run the Windows memory test.

I kind of feel like this is a G-card issue because of the graphical issues in-game, but then I am not sure what to think about it freezing up during those stress tests... if the tests were to stress test the cpu, why would the g-card cause a freeze up? Unless it was the cpu not receiving adequate power from the PSU? Then the same could be said for the G-Card issues... I wish I had a decent PSU to swap out and test it with, but I don't.

Once the kids are in bed I'll open the tower up, replug, reseat, and take pics.
 

Soul Sojourner

Prominent
Oct 30, 2019
21
0
510
Windows memory test passed without issues or errors. I chose the extended test. It ran for over 12 hours.

I opened up my tower and pulled cables, blew them out, cleaned a few connectors with alcohol. Took out the Cmos battery, did a hard reset, put it back in. Reseated all cards and RAM, re-plugged cables. I only left the sata cables to the HDD's and DVD drive and SSD alone, really didn't see much point in adjusting those.

After bootup, my mobo label screen popped up a few times when it shouldn't. It'd pop up right away, as normal, then go to it's usual black screen, then pop up for a split second again, then back to it's usual next screen, then pop up for a second again, then go to windows loading screen... kinda odd. I entered setup and loaded optimized default settings, restarted and saw the same. Booted all the way up to windows, logged in, waited about 1 min. and then shut down. Waited another 30 seconds, then turned it back on, saw the same oddity, and booted back up to windows again.

Internal pics:

dqF6XLj.jpg


Naaorgv.jpg


kJG2F5B.jpg
 
I think you need to strip this down to the bare essentials.

Are you overclocking the RAM, like, WAY beyond their XMP profile speed? If not, get rid of that memory cooler. For now at least. Totally not necessary, at all, unless you are seriously overclocking the RAM beyond it's regular XMP profile speed. If those are 1333mhz sticks and you are running them at 1866 or 2133mhz, then by all means cool them, but for now at least, let's NOT overclock the RAM (Which doesn't mean the same as let's not run at XMP. XMP is not what we mean by "overclocking" even though technically most XMP profile ARE an "OC" if they are beyond the default JEDEC specifications). If your memory does not need that cooler because you are not heavily overclocking them memory, take it off. Completely out of the system. For now at least.

Then, disconnect ALL of the storage devices. Disconnect all of the SATA power and data cables from all of the drives, and then boot the system to see if you are still seeing the same behavior with the splash screen or any other other weirdness. Sure, it's not going to boot into windows, but that's ok. Just see what it does.

I would also recommend that you download one of the free Linux bootable distros like Ubuntu, and create bootable flash drive media, and try booting to that to see if you have any of the same problems, with all the drives disconnected. Obviously you'll need to reconnect the drives temporarily to create media, or create it from another system.

Other than that, I'd say to bench test it. Pull the board. Use only the very most minimal hardware, and see what shakes out.

 

Soul Sojourner

Prominent
Oct 30, 2019
21
0
510
No, I'm not currently overclocking anything specifically, because of the issues. Before I was overclocking things pretty minimally. After the hard reset I obviously lost my admin and user pass, so I set those again and my bootup was back to normal. It's possible it used to be like that and I just didn't remember that because I've had the password set since the beginning.

I tried NMS out again today, I played for hours successfully. Although periods of time it was left on pause due to taking care of my kids, it nevertheless lasted for hours without any major issues. I feel like some graphical things take longer to render than normal... it's still quick but not like I think it should be.

That bottom card is an EXTRA sound card I don't even use. I was going to pull it but didn't have the metal deal that goes there on the case. I know I have it, probably at my parents place, so when I get a chance to run out there and grab it I will pull that out, even though having that metal plate isn't wholly necessary, I prefer to prevent insects from being able to make their way in.

I think for now I'm just going to watch it until if/when I have any more issues. If it seems like everything is good in a week or so, I'll just come back and credit you for helping me. If not, then I'll try some more of your suggestions out and see what happens, and come back and let you know what happens/what I find.

Thanks for all your help.
 
You can remove the rear bracket from the sound card, and put that there until you find the PCI slot shield. Usually they are removable from the add in cards anyhow. In any case, if an insect wants inside your case, I imagine there are plenty of other ways for it to get in there. I'd probably remove that thing asap. There's a good chance it could be the root of your problems, or at least A chance anyhow.
 

Soul Sojourner

Prominent
Oct 30, 2019
21
0
510
I definitely plan on pulling that card, though I doubt it's the root of the problem it certainly has no reason to stay.

Tbh, this case really has very little for holes and gaps, most openings have dust shields and filters, which I find myself surprisingly doing a good job keeping clean 90% of the time. That's why you see so little dust in the case. I have blown it out periodically over the years, but every time I do the dust buildup is minimal. I didn't clean it before (or after) those pictures. But there are certainly ways in for smaller bugs, I have surprisingly never seen one in my case, believe it or not. I did in my old computers though, lol.

Btw ALL of the links in the post you linked above about bench testing are broken. Thought you might want to know. It looks like a very thought out, thorough, and informative post, so thank you for that. I've read through some of it already. I am hoping I won't actually need any of that information any time soon, but I'm glad it's there.
 

Soul Sojourner

Prominent
Oct 30, 2019
21
0
510
I've played NMS and other games and not experienced much for issues. Sometimes I still had some issues in NMS again, so I finally decided to look at the game itself, uninstalled it and deleted a ton of super old cache files, and then reinstalled it. I only played it once after and didn't notice any issues, not long enough to know for sure. Other games have had no issues at all, and I am still not noticing any impact outside of games.

I'm not sure what the issue was/is for sure still, but things work adequately enough that I'm satisfied if they stay this way until I can afford a new build.

The "rattling" noise started again at some point too. I checked all the fans and stopped them from moving last time it happened and still swore I heard rattling after, but this time, one of the fans for my memory was clearly the problem. I took those fans out and it's been silent since.

I also removed the sound card, it affected nothing, but it had 0 reason to be in there anyway.

Thanks for everything.