My Predator Gaming PC Frequently Crashes - G3-710

For a couple of weeks now my Predator Gaming PC has been frequently crashing. Even if I do something simple like Open the Twitter app or send a message through Discord there's a tiny chance of it crashing. Sometimes I might get off my pc and come back to find it frozen or restarted. Its been crashing a few times EVERYDAY and I finally checked the reliability monitor today. There seems to be a common pattern where Hermes and Radeon Settings: Host Application stops working followed by windows having a hardware error and occasionally shutting down unexpectedly. I've tried updating my graphics card but that didn't fix anything.

Accepted answers

All comments

Balatekie

Hi!!

@TeleBlue,

I would like to know the CPU/GPU temperatures of your computer. Moreover, does the laptop crashes while adapter plugged in or without?

TeleBlue

Balatekie said:

Hi!!

@TeleBlue,

I would like to know the CPU/GPU temperatures of your computer. Moreover, does the laptop crashes while adapter plugged in or without?

My CPU is an intel i7-7700 and is at 65 degrees celsius.

My GPU is a Radeon RX 480 and its at around 60 degrees celsius

Obviously these aren't the temperatures before my pc crashes and I dont have an adaptor because its a desktop

TeleBlue

TeleBlue said:

Balatekie said:

Hi!!

@TeleBlue,

I would like to know the CPU/GPU temperatures of your computer. Moreover, does the laptop crashes while adapter plugged in or without?

My CPU is an intel i7-7700 and is at 65 degrees celsius.

My GPU is a Radeon RX 480 and its at around 60 degrees celsius

Obviously these aren't the temperatures before my pc crashes and I dont have an adaptor because its a desktop

Btw the gpu went down to 52 degrees celsuis, it was probably higher because my pc was still starting up

billsey

Any chance it's actually an app that's crashing, when it's doing something bad with the GPU? Your temperatures seem fine. I'd likely start with a clean boot and wander around trying to crash in that mode. If that seems stable try enabling one startup app after another until you can cause the crash again.

TeleBlue

It crashed again today, this time when I turned the pc on it came on saying "Scanning and repairing drive (C:)

TeleBlue

billsey said:

Any chance it's actually an app that's crashing, when it's doing something bad with the GPU? Your temperatures seem fine. I'd likely start with a clean boot and wander around trying to crash in that mode. If that seems stable try enabling one startup app after another until you can cause the crash again.

I'll begin trying that, hopefully, I find the cause quickly

TeleBlue

TeleBlue said:

billsey said:

Any chance it's actually an app that's crashing, when it's doing something bad with the GPU? Your temperatures seem fine. I'd likely start with a clean boot and wander around trying to crash in that mode. If that seems stable try enabling one startup app after another until you can cause the crash again.

I'll begin trying that, hopefully, I find the cause quickly

I seem to still be getting the hermes and Radeon errors on the event viewer and reliability monitor but no crashes yet

billsey

If you look at the details in Event Viewer it might tell you which exe is calling the failing function.

TeleBlue

billsey said:

If you look at the details in Event Viewer it might tell you which exe is calling the failing function.

While there are failed Hermes and Radeon Settings errors scattered around, there are no failures that were recorded during the crashes. I've clean booted my PC. After clean booting the PC I opened Chrome and Discord. Very soon after I noticed an error on the reliability monitor so I decided to close Discord. Ever since then it's been running for around 20 hours with no issues popping up. Despite that, the stability of my PC is still shown to be extremely low on the reliability monitor. I'm going to begin opening a couple of startup apps each day and see if anything changes.

billsey

Yeah, please keep us up to date on it...

TeleBlue

billsey said:

Yeah, please keep us up to date on it...

Okay, so I looked through the event viewer's history and found that the Hermes and Radeon Settings errors have been occurring as far back as the 6th of December 2018 (The event viewer doesn't go back further than December). I've also restarted my PC three times and learned that the Hermes and Radeon Settings errors occur on startup no matter what startup apps are opened. I still have no way of telling if these errors are related to the crashes since they've been occurring before the frequent crashes begun unless the errors have been slowly destroying my hardware.

Looking into the system tab of the event viewer I found A LOT of this

"A corrected hardware error has occurred.

Component: PCI Express Root Port

Error Source: Advanced Error Reporting (PCI Express)

Bus:Device:Function: 0x0:0x1C:0x5
Vendor ID:Device ID: 0x8086:0xA115
Class Code: 0x30400

The details view of this entry contains further information.

This warning is shown constantly but the event viewer only shows events back to the 14th of April 2019.

My PC's stability has also just taken a massive dip this morning.

TeleBlue

Okay so I rebooted my PC and opened some of the more essential startup apps, Discord and possibly Battle.net. Then I decided to open Minecraft Windows 10 because my PC has crashed quite often while playing that (But a lot of crashes occur when not playing the game). I set my character up on a machine that would continuously move him forwards to see if my PC could handle the game continuously generating more land. It eventually crashed after some time causing my PC to restart. Once the PC began to boot up it crashed again and restarted. This time it booted up successfully. I clean booted my PC again and opened just Minecraft, did the same things and didn't find any issues.

This morning I booted up my PC with just the essential startup apps along with my anti-virus, Wallpaper Engine, Battle.net and Origin. The PC crashed when it booted up but successfully restarted itself. According to the reliability monitor, hermes.exe stopped working before it was able to fully boot up. Radeon Settings: Host Application then stopped working after it successfully booted up.

I'm going to begin restarting and playing around with my PC while one of those startup apps are open at a time and see if that crashes my PC.

billsey

The PCI Express Root Port error is likely going to turn out to be he culprit. Hopefully it's a problem in the GPU and not the motherboard. I don't think you've ever told us your full model number, so I don't know if you've got onboard video you can use if you pull the GPU card. If you do, run pretty much the same types of tests after pulling it and verify the problem is cleared up, albeit with reduced performance. If that does it we have two options, swap in a new graphics card and hope it's not the motherboard or find a graphics card a very good friend can loan you. Either way if it's stable with the onboard graphics and stable with a new GPU you're golden. If a new GPU starts crashing again then it's likely time to swap the motherboard...

TeleBlue

billsey said:

The PCI Express Root Port error is likely going to turn out to be he culprit. Hopefully it's a problem in the GPU and not the motherboard. I don't think you've ever told us your full model number, so I don't know if you've got onboard video you can use if you pull the GPU card. If you do, run pretty much the same types of tests after pulling it and verify the problem is cleared up, albeit with reduced performance. If that does it we have two options, swap in a new graphics card and hope it's not the motherboard or find a graphics card a very good friend can loan you. Either way if it's stable with the onboard graphics and stable with a new GPU you're golden. If a new GPU starts crashing again then it's likely time to swap the motherboard...

Oh yeah, I completely forgot to mention the model! It's a G3-710

billsey

Yeh, you should have a couple of video ports you can use as soon the the RX480 is out. I don't think you have to select the onboard video in the BIOS, it should just recognize that the card is no longer there.

TeleBlue

billsey said:

Yeh, you should have a couple of video ports you can use as soon the the RX480 is out. I don't think you have to select the onboard video in the BIOS, it should just recognize that the card is no longer there.

So I wasn't able to pull out the graphics card physically but I disconnected the power cable and started up my PC. The hermes error still occurred and I'm still getting these warnings

Image: https://us.v-cdn.net/6029997/uploads/editor/kg/cp5teg4bsaos.png

TeleBlue

I wanted to leave my pc on over night to see if it crashes but I'm kinda worried if leaving the gpu in there, unplugged might damage it. Would it be safe to do so?

TeleBlue

Okaaayyy, so I think I stopped the PCI Express Root Port error but turning off my Wifi... The hermes and Radeon Settings still occur when booting up my PC and I realized I got these system errors whenever I boot up my PC, with the last two appearing twice.

Image: https://us.v-cdn.net/6029997/uploads/editor/h3/za9nx0nnhjvt.png

Image: https://us.v-cdn.net/6029997/uploads/editor/js/qeyn3bse1gxr.png

Image: https://us.v-cdn.net/6029997/uploads/editor/vq/5c13uddky7lj.png

billsey

Those last three are all pretty normal. The COM error is well known and caused by permission issues with the default Windows 10 build. Both are retried with different permissions during startup and the succeed the second time around. ZeroConfigService is the wireless drivers from Intel and usually means it''s trying to find a network resource that isn't there.

Now, why weren't you able to get the card out? Running with the card partially unplugged could cause all kinds of problems, some fatal!

TeleBlue

billsey said:

Those last three are all pretty normal. The COM error is well known and caused by permission issues with the default Windows 10 build. Both are retried with different permissions during startup and the succeed the second time around. ZeroConfigService is the wireless drivers from Intel and usually means it''s trying to find a network resource that isn't there.

Now, why weren't you able to get the card out? Running with the card partially unplugged could cause all kinds of problems, some fatal!

I have no idea. There was only one screw I had to undo at the back of the case but it still seems to be stuck. The clip at the pci slot also seems to be really hard to unclip. I was able to loosen the gpu from the clip's end and partially unclip it as I pull it out but the gpu was still stuck from the back of the case, halting my progress.

Quick Links

Top Members

Weekly

Puraw 175

Diya1811 130

Jack22 75

roberthansjorg 62

Brian7 60

Swift Series Intel^® Evo™

Game Pass

Acer Corner

Gaming on Chromebook

Swift Series Intel® Evo™

Game Pass

Acer Corner

Gaming on Chromebook

Swift Series Intel^® Evo™