add arrow-down arrow-left arrow-right arrow-up authorcheckmark clipboard combo comment delete discord dots drag-handle dropdown-arrow errorfacebook history inbox instagram issuelink lock markup-bbcode markup-html markup-pcpp markup-cyclingbuilder markup-plain-text markup-reddit menu pin radio-button save search settings share star-empty star-full star-half switch successtag twitch twitter user warningwattage weight youtube

New PC build randomly freezing - help please

joejccva71

49 months ago

Here is my build: http://pcpartpicker.com/p/YGTJ6h

I have tried both Windows 10 and Windows 7. It doesn't matter. It will randomly freeze whether I'm playing a game, watching a video on youtube, or just sitting at the desktop. It's very random and I could be playing a game fine for 3 hours and then it will freeze. Or I could be just looking at a website or a video or anything.

Here's what I've done so far:

  • Updated ASRock BIOS to the latest version: 3.0
  • Updated all the drivers from the motherboard website
  • Updated latest graphics driver from Nvidia.
  • Monitored heat with CPU HW Monitor and the CPU and the system are pretty cool. 30deg celsius at idle, and 54-60deg celsius at 100% load.
  • Ran Memtest86 for 7 hours straight with no errors. It completed 6 passes, no errors.
  • Ran SSD health diagnostic on both my Samsung SSD drives and they both checked out 100% healthy.
  • Ran sfc /scannow and found no issues.
  • Ran prime95 for 8 hours with no issues. Temps never got above 60deg celsius.

Like I don't know what else to do. I mean I guess it could be a bad PSU or the motherboard maybe? Could the RAM still be bad after testing it with memtest86 for hours?

Anyone have any ideas? It's very frustrating.

Comments

  • 49 months ago
  • 2 points

Do you overclock? Wrong settings can lead to system instability. Maybe try running your system without any overclocking for a while and see if the issue still persists.

You're doing the right things with the checks and assumptions (faulty components). It's never easy to figure out what's wrong.

If it's possible, try borrowing someone else's PSU or motherboard for a wile. But always swap 1 component and test for a while so you can find the faulty one easily.

I've looked at Intel's web site on i5-6600k and they say the max temperature is 64 Celsius. If overheating causes you problems (60 is getting close to 64) then first your machine starts getting slow, laggy, etc before completely shutting down... I'd consider dusting off the machine or getting a better cooling as well. Is there a proper airflow available in the case?

  • 49 months ago
  • 2 points

I really don't know why Intel claims that 64 is the max temp. It's just not true. The real max temp is probably more like 80 and they're just trying to cover their backsides with 'official specification'. Thermal shutdown probably doesn't happen until the 90s.

Can someone with a skylake processor and its stock cooler please tell me what they get under full load because I have a feeling the stock cooler may not cut it.

  • 49 months ago
  • 1 point

Not sure. I use the Hyper 212 cooler.

  • 49 months ago
  • 1 point

Appreciate the response.

No I have the system at it's default settings and it freezes. Do I have a quality PSU? It's an EVGA semi-modular 80+ Gold.

Yea the machine would never get above 60 degrees running at stock speeds so I'm not sure there. Between my case and cooling it should be fine. I just bought two industrial Noctua fans for exhaust. They were pretty expensive and they run great. I also have two Cougar Vortex intake fans that run pretty well too. My motherboard temps look good also.

I mean I can run prime95 for hours upon hours without issues. I can play a game that is graphic intensive like Witcher 3 for hours and be fine. Then I could exit the game, browse the internet and boom...freeze.

It's very strange and I'm not sure what could be causing this.

  • 49 months ago
  • 1 point

I did some testing. I manually turned down my pump and fan speed whilst under full load to increase the temperature. At 64 degrees nothing happened.
Or 70...
Or 80...
Or 85... at this point I decide; "That'll do!"

I was using intels Xtreme Tuning Utility to measure temps and thermal throttling.
Or rather just temps because even at 85 degrees no thermal throttling was recorded.

If that 64 degrees was true. Why didn't I throttle?

  • 49 months ago
  • 1 point

Hi, Yutang!

Do you have the same CPU as the OP? :)

I have an Intel® Core™2 Extreme Processor QX9770. It's pretty old, but works OK (well, most of the time, some computation heavy games can cause problems).

Intel says, the Tcase is 55.5 °C. When I got overheating issues, my system started to stutter first and completely die a couple of degrees above 55 °C.

So either you have a different CPU with different temperature values than the OP, or the new Intel CPUs handle overheating better than old ones, like mine. :)

  • 49 months ago
  • 2 points

I can guarantee you I have a 6600k. Don't believe me. Look at my completed build.

Yeah. 85°C is way over 64°C so I should have crashed and burnt. but nothing happened. Ask anyone else with a 6th generation processor. They'll probably tell the same story.

Intel CPUs handle overheating better than old ones

That or they have expanded their margin of safety. By a lot...

  • 49 months ago
  • 1 point

I can guarantee you I have a 6600k. Don't believe me. Look at my completed build.

I never questioned your honesty :), just didn't know your build (you didn't mentioned either).

  • 49 months ago
  • 1 point

I was always under the impression that the Skylake processors were overclocking machines. I've seen videos and read forums on overclock.net where people were overclocking the 6600k skylake at minimum 4.4ghz without issues. People would run at 70 degrees celsius under load and were fine.

The highest I've been at was 60 under load.

  • 49 months ago
  • 2 points

If all the BIOS settings are correct and since you have already updated all your drivers, I would suspect the ASRock.

  • 49 months ago
  • 2 points

ASRock is king in Support. I had issues either motherboard, PSU, CPU, or memory. They helpmed me figure out, suggested me what memory kit to buy (try 1.35V instead of 1.5V), etc. They are great guys.

  • 49 months ago
  • 2 points

I concur, my experience with Asrock has only been positive. They have always replied to me within 24-ish hours and were very patient and helpful.

  • 49 months ago
  • 1 point

Ok I'll try emailing them directly. I tried to find a phone number for them but couldn't find it.

When you say try 1.35 volts instead of 1.5....what are you referring to?

I reset the bios to defaults and left everything on AUTO which is how the BIOS originally had it.

  • 49 months ago
  • 2 points

Some memory modules are designed to operate on 1.35 V instead of 1.5 V (usual voltage). Since I was running OC setup on my motherboard, the 1.35 V modules they suggested worked, while the 1.5 V modules I chose on my own did not (same brand, just different product lines).

  • 49 months ago
  • 1 point

Ah ok. When I checked the DRAM voltage in the BIOS for my ram, it showed 1.200 volts I believe it was as a default. Should I change that or leave it?

  • 49 months ago
  • 1 point

1.2 V is the right voltage for your RAM: http://pcpartpicker.com/part/crucial-memory-bls2k4g4d240fsa. Just leave it on the normal values.

Besides, motherboards automatically assign the right (default) voltage to the RAMs (and everything else), so you shouldn't change that. Especially, since your memtest turned out just fine.

  • 49 months ago
  • 1 point

Quick question. Could it be the PSU though?

  • 49 months ago
  • 2 points

Check the power supply voltage readings while you're in the BIOS and see if they are within allowable specifications. If they are, that basically leaves only the ASRock.

  • 49 months ago
  • 1 point

Ok will try that. Thanks.

  • 49 months ago
  • 1 point

Oh, Yay! Other people have had the freeze in bios problem as well! Its really not a problem with me; it happens so irregularly and seems to be when I change settings too quickly as soon as I get to the bios.

  • 49 months ago
  • 2 points

My take on steps to try:

1)Reset your bios settings to default (even if you haven't done anything) and see if you crash.

2) In your bios make sure you have XMP enabled and have selected the profile and you are getting the correct speeds from within Windows.

3) Remove your graphics card and see if you still crash.

4) Try installing and running driver booster 3 by Iobit to update all your drivers.

5) Contact Asrock?

  • 49 months ago
  • 1 point

Ok no problem.

  1. I did reset the bios to default. Still crashed. Also the bios is updated to the latest version.

  2. XMP is enabled for the memory (if that's what you mean). I didn't see any XMP for the cpu in the bios but i may have overlooked it.

  3. I haven't tried removing the Geforce GTX 970 card yet...will try that.

  4. I just sent Asrock an email.

  • 49 months ago
  • 2 points

XMP is enabled for the memory (if that's what you mean). I didn't see any XMP for the cpu in the bios but i may have overlooked it.

Lol, yeah. XMP is only for memory. XD

  • 49 months ago
  • 1 point

I appreciate all your help guys. These are the things I'm going to try next.

  1. I can't believe I haven't looked in the event viewer yet to see what that says. So that's my first thing.

  2. Depending on what that says, I'm going to switch the memory around to different bank and test.

  3. If still freezes, I'm going to pull the board out, place it on the box, and only connect the cpu/cooler, video card, 1 ram stick, 1 hard drive, and the psu obviously and see if i can duplicate issue.

  • 49 months ago
  • 1 point

One more question for you guys.

I'm using a Samsung 841N SSD 128gb drive that I got out of one of my Dell machines as my primary drive in my newly built machine. Then i have a purchased Samsung 850 EVO 250gb drive that I have as a secondary drive.

Do you think that this SSD drive that belonged to a Dell laptop could be the culprit of causing me issues even though I have wiped the partition and formatted this drive before installing Windows 7 on it?

  • 49 months ago
  • 1 point

Doubtful unless you were already experiencing same issues while it was running in your DELL.

  • 49 months ago
  • 1 point

Ok so just an update. These are the things I did last night when I got home from work.

  1. I checked the event viewer and noticed an error that happened whenever i booted into windows. However, this error did not happen at the moment of freezing, only when I booted.

"Driver ACPI returned invalid ID for a child device"

So I uninstalled both my audio and graphics drivers and then reinstalled just the motherboard audio, and just the Nvidia graphics and physx drivers. Played a game for about 30 minutes. Freeze. So that didn't work.

  1. Opened the case. Pulled the memory out, used canned air and dusted out the memory sockets. Then I put the memory back in different slots. So instead of slots 1 and 3, I used 2 and 4. I have 8 gb (2x4gb sticks).

Booted into windows and played a game for 5 hours or so, didn't freeze. But I wasn't convinced.

  1. I downloaded the stability tester, OCCT and I let it run all night.

Woke up and no freeze. People have said that if you can run the CPU: LINPACK tester with that program even just an hour, then you're machine is 100% stable for overclocking. Well, I ran it all night and no freeze, no crash, nothing. Keep in mind I'm not overclocking right now, everything is at stock speeds.

Still not convinced yet.

  1. This morning, I'm letting a Youtube video run all day while I'm work. I've crashed before just playing a Youtube video.

Could it be just switching the memory around fixed this?

Sort

add arrow-down arrow-left arrow-right arrow-up authorcheckmark clipboard combo comment delete discord dots drag-handle dropdown-arrow errorfacebook history inbox instagram issuelink lock markup-bbcode markup-html markup-pcpp markup-cyclingbuilder markup-plain-text markup-reddit menu pin radio-button save search settings share star-empty star-full star-half switch successtag twitch twitter user warningwattage weight youtube