Thank you for being a valued part of the CNET community. As of December 1, 2020, the forums are in read-only format. In early 2021, CNET Forums will no longer be available. We are grateful for the participation and advice you have provided to one another over the years.

Thanks,

CNET Support

Question

Random computer freezing

by Nov 15, 2018 1:41PM PST

Hello everyone,



I am going to provide just a bit of background for this first.

I had originally built this PC back in early 2016 and everything has going fine until July of this year when things would just randomly freeze. After a few days of trial and error I discovered I could get my PC stable If i remove one RAM stick and just keep one in DIMM2. Anytime I would add in a second RAM module It would freeze within 30mins or If I had RAM in any other slot. I tested both sticks separately with memtest64 in slots 2 and 3, multiple times all with passes.

After three weeks of trying different fixes with no resolution I caved and just bought a new MOBO (one currently listed below) this fixed the issue until 10/19 when the same freezes began occurring (first with a BSOD, then just random freezing after), same thing one RAM module in DIMM2 and it runs stable for days but adding in another one results in a freeze. I have since reinstalled windows onto a brand new SSD, I have retested the RAM (pass) cleaned out everything including the PSU and all DIMM slots, updated BIOs and all MOBO drivers,and removed and reinstalled all components to no avail. All systems and components including MOBO are running at default and temperatures are always fine. This issue has been plaguing me with no end in sight for now I appreciate any suggestions.



http://speccy.piriform.com/results/3oH4VXf62XryWh5nClokghC



Additional Specs:



RAM: G.SKILL Ripjaws V Series 16GB (2 x 8GB) (Currently using just one stick due to issue above and both are brand new).
CPU: Intel Core i7-7700K (running stock)
GPU: EVGA GeForce GTX 1080 Ti FTW3 (running stock)
PSU: EVGA SuperNOVA 750W Fully Modular
MOBO: Z270 GAMING M7 (previously used the M5 back in July when this problem started)
SSD: Samsung 860 EVO 500GB
Case: Phanteks Enthoo Pro Series
Heatsink: MasterAir Pro 4 CPU

Discussion is locked

31 - 60 of 41 Posts
- Collapse + Expand Details
- Collapse -
MEM/Speccy
by Owens5565 Nov 19, 2018 1:48PM PST

I use slot 2 and 4 as recommended by the MOBO manual when running two sticks at once and just slot 2 when using one. I am adding a screenshot of my BIOS with relevant information. I have also closely and carefully examined and cleaned and the DIMM slots using compressed air.
Everything shown in BIOS is default.

Also I see speccy is back up so I am including that link as well.

http://speccy.piriform.com/results/I5KeJxUkVgl6jB20DzGrEo7

https://imgur.com/a/14A9etZ

- Collapse -
Speccy
by Bob__B Nov 19, 2018 3:39PM PST

Shows your bios is up to date.

Your screen shot of bios1 shows your vcore and ram voltage is normal.

You ran memtest with no gpu and disk installed.

With sticks in slots 2 + 4 it failed.

Each stick alone works fine.

This is leading me back to a mobo or ram problem.

Contact msi and have a chat.

- Collapse -
Speccy
by Owens5565 Nov 19, 2018 4:11PM PST

Thank you once again for the reply.

I have never actually had to contact a manufacturer for a technical issue before because I have never ran into an issue this damn persistent and annoying. Will they actually take the time and go through the whole thing to try and work it out with you from your experience?

- Collapse -
My experience
by Bob__B Nov 19, 2018 5:15PM PST

With this is very limited.

I once had to contact Dell about a hdd problem.

I explained the problem.... the test I had done and they scheduled someone to come in and replace the unit.

In your case I would expect a Q + A session so get all you ducks in a row before you contact msi.

- Collapse -
Answer
Let's see if Speccy tells us something new.
by Forum moderator Nov 19, 2018 2:09PM PST

Thanks for that. It's a great thing when ghost hunting. Also nice, a reboot prior to the Speccy.

1. Temps look good. Nothing alarming there.
2. There's some USB drive plugged in. Remove during testing except when you need it.
3. If you don't use Skype google how to stop it.
4. Vcore is a tad low. This is a dynamic value but when I've trying to work around a crashing board I like to see that at 1.0V Go get CPU-z to see what it notes for Vcore.
EXAMPLE: http://webcache.googleusercontent.com/search?q=cache:QmT6zz4PoycJ:www.gskill.us/forum/filedata/fetch%3Fid%3D128881+&cd=5&hl=en&ct=clnk&gl=us&client=firefox-b-1
They have theirs over 1.0V and DRAMCH is a tad higher too.
Go get CPUID/CPU-Z to verify. May want to tweak those up a notch and re-test.
-> This should be confirmed with the motherboard maker that the board is fine and you have to tweak to get around this issue.

5. The GPU BIOS looks to be from 2017. Are you sure you ran the Nvidia update?
https://www.nvidia.com/object/nv-uefi-update-x64.html
The GPU can glitch out on CPU RAM issues so I always do this in such situations.

6. CLEANUP possible W10 update sharing. Read
https://www.howtogeek.com/224981/how-to-stop-windows-10-from-uploading-updates-to-other-pcs-over-the-internet/

Most of these are things I do to get to a cleaner PC. Some are tweaks to move to a stable board by slightly upping the Vcore or Vram.

- Collapse -
Speccy
by Owens5565 Nov 19, 2018 2:44PM PST

Thank you for the very quick reply.

-The USB was just to save the MOBO screenshot its never in during any testing.
-I went ahead and uninstalled Skype
-I updated my GPU drivers and downloaded that firmware update tool you listed which prompted me to update my graphics card.
-I turned off windows update sharing via the instructions you provided

Showing a screenshot of my CPU-Z results

https://imgur.com/a/CFtQHug

Are you saying to lower the voltage to 1.0 and up the DRAM frequency and retest with both sticks?

- Collapse -
A few clarifications.
by r. proffitt Forum moderator Nov 19, 2018 3:12PM PST

The 1.0V comment is for the CPU.

The other voltage was for the RAM which I supplied an example with about 0.1V higher than what was in your Speccy. HOWEVER we want to use CPUID or CPU-Z to see what it reports before we tweak those up to see if we can stabilize the board. If it stabilizes I call it a bad board since the owner may reset the BIOS, forget all this and heads to the repair counter. Or sells it and the next owner is unaware of the tweaks.

- Collapse -
Another clarification. Normal isn't cutting it.
by r. proffitt Forum moderator Nov 19, 2018 3:41PM PST

I've worked such machines before and a small boost to Vcore or Vram has helped get a machine to work again. But here's the bad news. It's just a bandaid over the real problem.

- Collapse -
Inexperienced with tweaking settings
by Owens5565 Nov 19, 2018 4:19PM PST

Hello,

I normally don't ever touch any BIOS settings on my machines, is the vcore shown in https://imgur.com/a/14A9etZ for my CPU, take that down to 1.0V?

I am once again half tempted to get a new MOBO but instead this time something a bit more cheaper and a different brand/make then MSI.

After all this time and troubleshooting, its the mystery of what is going that is really bugging me.

- Collapse -
Answer
GPU
by Nov 19, 2018 2:30PM PST

CPU
Intel Core i7 7700K @ 4.20GHz: 29 °C
Kaby Lake 14nm Technology

RAM 8.00GB Single-Channel Unknown @ 1069MHz (15-15-15-36)

Motherboard - MSI Z270 GAMING M7 (MS-7A57) (U3E1): 31 °C

Graphics - ZOWIE XL LCD (2560x1440@144Hz)
3071MB NVIDIA GeForce GTX 1080 Ti (EVGA): 49 °C

What stands out to me is the high temperature differential between your CPU temperature and the Video Card temperature. While running the speccy, it's more likely the CPU temperature would be same or higher than the GPU.

This indicates to me that when you are gaming, the GPU is probably running even hotter than expected, maybe even hotter than the CPU. There's a 60% difference between them. That's unexpected, indicating a problem with cooling on the GPU.

Is it's fan running OK?

11/10/2018 NVIDIA - Display - 10/27/2017 12:00:00 AM - 23.21.13.8813
NVIDIA Display driver update released in October 2017


Is the 2017 driver the latest update for your NVidia?

Since you are running Skype, why keep Discord? If not using Discord, uninstall it. It can cause problems and it's NOT a windows native file .

- Collapse -
GPU
by Owens5565 Nov 19, 2018 3:07PM PST

Thank you for the reply.

My GPU has always idled around that temperature for some reason, I think its because my fans are super slow when its just idling however when its under intense load it still never gets past 70c with fan speeds at 40%, this goes the same for the CPU, I use MSI afterburner to monitor/adjust. Memtest would still freeze even once I removed my GPU from the equation.

I use discord nearly everyday and have been for years, I didn't even know skype was installed I guess windows buts it in by default, I went ahead and removed it.

I always have the Nvidia drivers up to date from here https://www.nvidia.com/Download/index.aspx but there was an older firmware update that I just recently installed at the suggestion of the previous poster.

Back to Computer Help forum

CNET Forums

Operating Systems
Software
Electronics & Gadgets
Hardware
Tablets & Mobile Devices
General Help
Brand Forums
Roadshow Autos
Off Topic

Other Forums

Forum Info