Random Computer Crashes

[UncensoredMango]

Hi all,

 

This is my first time posting in the Linus forums so forgive me for anything I'm missing or doing wrong.

 

I recently built my own computer (first build) with the help of a friend. I'm currently running a GeForce 1060, 16GB of ram, Ryzen 5 1600 CPU, 500GB SSD, 2TB HDD, 550W PSU and an AsRock AB350M Motherboard.

 

Over the last few days I've encountered these random inexplicable crashes/freezes, where my monitor will go dead and my computer will become unresponsive. If there's audio playing, the last second or so can be heard on repeat. The only way I've found of fixing it is to either force shutdown or press the restart button.

 

I've tried reinstalling my graphics drivers and uninstalling my anti-virus so far. A friend of mine said to disable Nvidia audio drivers and I've done that, but it's too soon to tell if that will fix it. My theory is that it's something to do with my PSU, as I have a 4+4 CPU input to a 6+2 pcie output. This is because the stock cable that came with my PSU (6 pin to 6+2 pin cable) was several centimetres too short despite my best efforts.

 

Sorry for the wordy post! I'm very concerned as this is my first build I've done and have little clue about what I'm doing.

[Saieshan Dadi]

Are you running an OC on your GPU or CPU ??

 

 What load and idle temps and you getting from your GPU and CPU respectively ?? 

[Ace McPlane]

17 minutes ago, UncensoredMango said:

Hi all,

 

This is my first time posting in the Linus forums so forgive me for anything I'm missing or doing wrong.

 

I recently built my own computer (first build) with the help of a friend. I'm currently running a GeForce 1060, 16GB of ram, Ryzen 5 1600 CPU, 500GB SSD, 2TB HDD, 550W PSU and an AsRock AB350M Motherboard.

 

Over the last few days I've encountered these random inexplicable crashes/freezes, where my monitor will go dead and my computer will become unresponsive. If there's audio playing, the last second or so can be heard on repeat. The only way I've found of fixing it is to either force shutdown or press the restart button.

 

I've tried reinstalling my graphics drivers and uninstalling my anti-virus so far. A friend of mine said to disable Nvidia audio drivers and I've done that, but it's too soon to tell if that will fix it. My theory is that it's something to do with my PSU, as I have a 4+4 CPU input to a 6+2 pcie output. This is because the stock cable that came with my PSU (6 pin to 6+2 pin cable) was several centimetres too short despite my best efforts.

 

Sorry for the wordy post! I'm very concerned as this is my first build I've done and have little clue about what I'm doing.

Well, the first thing I'd do is download and fire up some monitoring software like hwinfo64 and check those CPU temps since there is a possibility it's overheating like if the cooler isn't seated properly and that's the sort of thing you want to nip in the butt right away before it causes damage to the CPU.

 

Let us know what your idle temps are and what those temps are under load like in a demanding game.

If anything goes over 80 degrees, shut her down immediately.

 

That's probably not it though.

Just what you want to check first.

 

Oh and as for your cable problem, you can use cable extensions to solve that issue with the CPU power cable. Check out cablemod for that.

 

 

[UncensoredMango]

 

15 minutes ago, Saieshan Dadi said:

Are you running an OC on your GPU or CPU ??

 

 What load and idle temps and you getting from your GPU and CPU respectively ?? 

Both CPU and GPU are running at their stock clock speeds. HWMonitor shows a GPU idle temp of 30 degrees (most likely a couple degree more as I've just turned my computer on) and CPU idle temp around the same. I'll try run some games and see what temps I get.

[UncensoredMango]

15 minutes ago, stateofpsychosis said:

Well, the first thing I'd do is download and fire up some monitoring software like hwinfo64 and check those CPU temps since there is a possibility it's overheating like if the cooler isn't seated properly and that's the sort of thing you want to nip in the butt right away before it causes damage to the CPU.

 

Let us know what your idle temps are and what those temps are under load like in a demanding game.

If anything goes over 80 degrees, shut her down immediately.

 

That's probably not it though.

Just what you want to check first.

 

Oh and as for your cable problem, you can use cable extensions to solve that issue with the CPU power cable. Check out cablemod for that.

 

 

Thanks for the advice with the cables! I have downloaded HWMonitor and if I'm remembering correctly definitely did not get anywhere near 80 degrees while playing PUBG earlier (although I'm not sure exactly what temp).

[Saieshan Dadi]

MSI afterburner might be a better utility to use to record temps in game

[Ace McPlane]

6 minutes ago, UncensoredMango said:

Thanks for the advice with the cables! I have downloaded HWMonitor and if I'm remembering correctly definitely did not get anywhere near 80 degrees while playing PUBG earlier (although I'm not sure exactly what temp).

With no overclock, around 60 degrees is normal under load.

 About those cables though:

 

6+2 sounds like a graphics card PCIe cable to me not a CPU one.

I'm still on haswell and am a little out of date on this stuff so I could be wrong,

but double check that you're using the right cable.

Even if it's tight, you should still be able to reach with that cable if you're willing to leave the cable in the front and not hide it behind the motherboard tray like you should for now.

[UncensoredMango]

12 minutes ago, stateofpsychosis said:

With no overclock, around 60 degrees is normal under load.

 About those cables though:

 

6+2 sounds like a graphics card PCIe cable to me not a CPU one.

I'm still on haswell and am a little out of date on this stuff so I could be wrong,

but double check that you're using the right cable.

Even if it's tight, you should still be able to reach with that cable if you're willing to leave the cable in the front and not hide it behind the motherboard tray like you should for now.

Essentially my PSU has 4 output slots. Two are for my HDD/SSD and the other two are labelled 6+2 PCIe and 4+4 CPU. The PCIe power input is 6 pins, with a 6+2 pin output for a GPU (I think this is done so you can connect a GPU that has 6 pins or 8 pins). I can try again but I'm almost certain that even with the cable running direct, it's just short of connecting.

 

To sum it up: I'm using the wrong cable and I think that's the issue but I'm not 100% sure. I might take some pictures to make it easier to see what I mean.

[UncensoredMango]

24 minutes ago, Saieshan Dadi said:

MSI afterburner might be a better utility to use to record temps in game

I'll check it out!

 

[Ace McPlane]

5 minutes ago, UncensoredMango said:

Essentially my PSU has 4 output slots. Two are for my HDD/SSD and the other two are labelled 6+2 PCIe and 4+4 CPU. The PCIe power input is 6 pins, with a 6+2 pin output for a GPU (I think this is done so you can connect a GPU that has 6 pins or 8 pins). I can try again but I'm almost certain that even with the cable running direct, it's just short of connecting.

 

To sum it up: I'm using the wrong cable and I think that's the issue but I'm not 100% sure. I might take some pictures to make it easier to see what I mean.

The outputs on the power supply are fine. It's the cable itself I was worried about. Power supply outputs are 1 of two things. Sata, molex and other peripherals on one side usually and PCIe stuff on the other. Both the GPU cables and CPU cables plug into the same type: PCIe but the cables are the part that's different. The CPU should use the 4+4 one.  If it doesn't reach, yea get a cable extension. FYI, I usually get them to reach when they're tight like that by running them behind the motherboard tray, but you have to take the motherboard out to do that so it's a bit of a pain. They usually reach pretty easily that way though. That's if there isn't a hole cut out up in the top left to run that cable through.

[UncensoredMango]

6 minutes ago, stateofpsychosis said:

The outputs on the power supply are fine. It's the cable itself I was worried about. Power supply outputs are 1 of two things. Sata, molex and other peripherals on one side usually and PCIe stuff on the other. Both the GPU cables and CPU cables plug into the same type: PCIe but the cables are the part that's different. The CPU should use the 4+4 one.  If it doesn't reach, yea get a cable extension. FYI, I usually get them to reach when they're tight like that by running them behind the motherboard tray, but you have to take the motherboard out to do that so it's a bit of a pain. They usually reach pretty easily that way though. That's if there isn't a hole cut out up in the top left to run that cable through.

Here's some pictures to better illustrate what I mean. My PSU is semi modular and the power connector to the CPU is further to the right.

[UncensoredMango]

12 minutes ago, UncensoredMango said:

Here's some pictures to better illustrate what I mean. My PSU is semi modular and the power connector to the CPU is further to the right.

Essentially I've got my GPU plugged in where a CPU can be plugged in?

[Ace McPlane]

1 minute ago, UncensoredMango said:

Essentially I've got my GPU plugged in where a CPU can be plugged in?

GPU and CPU cables both plug into the same type of PCIe connector so that's completely fine.

 

It's using the wrong cable not where it plugs into that I'm concerned about.

[UncensoredMango]

4 minutes ago, stateofpsychosis said:

GPU and CPU cables both plug into the same type of PCIe connector so that's completely fine.

 

It's using the wrong cable not where it plugs into that I'm concerned about.

Surprisingly, it's the same cable. I can't quite get a picture of it but the cable I'm using has two 6+2 connectors and one 4+4.

[Ace McPlane]

6 minutes ago, UncensoredMango said:

Surprisingly, it's the same cable. I can't quite get a picture of it but the cable I'm using has two 6+2 connectors and one 4+4.

Okay, I misunderstood then. I thought you said you were using the wrong cable because the right one couldn't reach?

 

Oh, I'm guessing you were using the wrong one that split off of that same cable.

Yea, get an extension for that for sure.

 

You should stop using the computer immediately if you are using a GPU cable in a CPU power socket. They are not the same. I don't know if that could cause damage, but I surely wouldn't risk it anymore.

 

The first 3 pins are the same 12v, but the rest aren't.

I know that.

I would not screw around with using a GPU cable in a CPU cable slot.

 

If that's not it and I'm misunderstanding again,

Hit the windows key, type reliability history, open that up and see what error messages you're getting when this happens.

[UncensoredMango]

3 minutes ago, stateofpsychosis said:

Okay, I misunderstood then. I thought you said you were using the wrong cable because the right one couldn't reach?

Ahh sorry for the confusion! As there are two 6+2 connectors I thought of plugging one into the 6 pin output on the PSU and one into the 6+2 pin input on the GPU. I realise now the cable is designed so that you can plug those both into GPU's and the 4+4 pin into the PSU. This sounds waaaayy too confusing and I'm so sorry for the misunderstanding. Essentially what I wanted to know was if connecting my GPU to the 4+4 pin output would cause any issues.

[Ace McPlane]

1 minute ago, UncensoredMango said:

Ahh sorry for the confusion! As there are two 6+2 connectors I thought of plugging one into the 6 pin output on the PSU and one into the 6+2 pin input on the GPU. I realise now the cable is designed so that you can plug those both into GPU's and the 4+4 pin into the PSU. This sounds waaaayy too confusing and I'm so sorry for the misunderstanding. Essentially what I wanted to know was if connecting my GPU to the 4+4 pin output would cause any issues.

Oh... you mean  the output on the power supply.

No, that's fine.

That's why it 6+2 PCIe & 4x4 CPU

Because one of those slots covers both. It's just the cables you plug into those slots that have to match up right with the GPU and CPU.

So, check your reliability history next. 

It might point to the error or problem you're having. List anything marked in red or yellow for the day that it happened.

[UncensoredMango]

4 minutes ago, stateofpsychosis said:

Oh... you mean  the output on the power supply.

No, that's fine.

That's why it 6+2 PCIe & 4x4 CPU

Because one of those slots covers both. It's just the cables you plug into those slots that have to match up right with the GPU and CPU.

So, check your reliability history next. 

It might point to the error or problem you're having. List anything marked in red or yellow for the day that it happened.

Looking at the history, I can't see any system errors. Only application and misc. Reoccurring errors are "SCP Service stopped working" which relates to the emulator I use to play with a PS3 controller (rocket league haha). And one other error reading "bad module info" there's another unique error "Windows Shell Experience Host" with the line "stopped responding and was closed" aside from that all other errors are "windows was not properly shut down" which I think have occurred because I tend to just press the power button as opposed to shutdown through windows menu (whoops). 

 

Oh! There's one more set of warnings very early on to do with and Nvidia driver, some MicrosoftVCLibs and a cumulative update for windows.

[Ace McPlane]

2 minutes ago, UncensoredMango said:

Looking at the history, I can't see any system errors. Only application and misc. Reoccurring errors are "SCP Service stopped working" which relates to the emulator I use to play with a PS3 controller (rocket league haha). And one other error reading "bad module info" there's another unique error "Windows Shell Experience Host" with the line "stopped responding and was closed" aside from that all other errors are "windows was not properly shut down" which I think have occurred because I tend to just press the power button as opposed to shutdown through windows menu (whoops). 

 

Oh! There's one more set of warnings very early on to do with and Nvidia driver, some MicrosoftVCLibs and a cumulative update for windows.

Every windows install I've done in the last 5 years has started out with tons of warnings like that. I'm more concerned with the recent ones. 

 

This is going to sound weird, but judging by one of those errors, try running the things that are crashing in windowed mode instead of fullscreen.

 

 

[Ace McPlane]

7 minutes ago, UncensoredMango said:

Looking at the history, I can't see any system errors. Only application and misc. Reoccurring errors are "SCP Service stopped working" which relates to the emulator I use to play with a PS3 controller (rocket league haha). And one other error reading "bad module info" there's another unique error "Windows Shell Experience Host" with the line "stopped responding and was closed" aside from that all other errors are "windows was not properly shut down" which I think have occurred because I tend to just press the power button as opposed to shutdown through windows menu (whoops). 

 

Oh! There's one more set of warnings very early on to do with and Nvidia driver, some MicrosoftVCLibs and a cumulative update for windows.

Honestly, at this point, I would backup what's important to you and reinstall windows from scratch and by that I mean, deleting every partition for your system drive even the recovery partition at the beginning of the windows installer and then just highlight the unallocated space and click next. Not some windows repair or other easier method that'll leave stuff behind. 

 

(FYI, most of your game saves will be in a hidden folder called appdata C:\users\<yourname>\appdata in the folder local, locallow, and roaming). Just make sure you only copy your game saves back after reinstalling windows as there is more in those folders than just game saves.

[UncensoredMango]

1 minute ago, stateofpsychosis said:

Honestly, at this point, I would backup what's important to you and reinstall windows from scratch and by that I mean, deleting every partition for your system drive even the recovery partition at the beginning of the windows installer and then just highlight the unallocated space and click next. Not some windows repair or other easier method that'll leave stuff behind. 

 

(FYI, most of your game saves will be in a hidden folder called appdata C:\users\<yourname>\appdata in the folder local, locallow, and roaming). Just make sure you only copy your game saves back after reinstalling windows as there is more in those folders than just game saves.

Thanks for the advice! I was given a tip to disable an Nvidia audio driver that could potentially be causing issues and haven't had any issues since, albeit that wasn't very long ago. If I experience another freeze... I guess I'll have to go with a fresh install.

[Ace McPlane]

15 hours ago, UncensoredMango said:

Thanks for the advice! I was given a tip to disable an Nvidia audio driver that could potentially be causing issues and haven't had any issues since, albeit that wasn't very long ago. If I experience another freeze... I guess I'll have to go with a fresh install.

Cool, make sure to stay up to date with your nvidia drivers as they release updates in the coming weeks. Whatever problem you had will probably be sorted in the next update if that's what was causing it.