Jump to content
Search In
  • More options...
Find results that contain...
Find results in...
featherwolf

System shuts off as soon as games load

Recommended Posts

Posted · Original PosterOP

So, I feel this may be a foregone conclusion, but wanted to see if anyone has any other tips.

 

I recently switched GPUs and Mobo in my PC. Bought 2 used RX 580s and have them set up in crossfire on a B450 Gigabyte Aorus Pro Wifi board. Ever since then, I have experienced total system shut downs every time I try to launch an intensive game. No blue screen, just completely shuts down. All other functions and even some other casual games are fine. I'm trying to avoid having to return the Motherboard, but it's starting to seem like it may be the only option. Here's what I've tried:

 

-Removed GPUs and tested them on a different board. They were fine separately, but could not test them together on that board.

-Switched to known good PSU.

-Tried each GPU separately in Gigabyte board (thought maybe the board just had problems with multi-gpu setups.)

-Reinstalled board drivers 

-Used AMD Cleanup Tool to completely remove display drivers and reinstalled them (multiple times).

-Reset windows on my PC.

-Changed the power strip my PC was plugged into.

-checked for BIOS updates (there weren't any).

 

Let me know if you think there's something I should try before returning the mobo. Thanks!

 

Link to post
Share on other sites
4 minutes ago, featherwolf said:

So, I feel this may be a foregone conclusion, but wanted to see if anyone has any other tips.

 

I recently switched GPUs and Mobo in my PC. Bought 2 used RX 580s and have them set up in crossfire on a B450 Gigabyte Aorus Pro Wifi board. Ever since then, I have experienced total system shut downs every time I try to launch an intensive game. No blue screen, just completely shuts down. All other functions and even some other casual games are fine. I'm trying to avoid having to return the Motherboard, but it's starting to seem like it may be the only option. Here's what I've tried:

 

-Removed GPUs and tested them on a different board. They were fine separately, but could not test them together on that board.

-Switched to known good PSU.

-Tried each GPU separately in Gigabyte board (thought maybe the board just had problems with multi-gpu setups.)

-Reinstalled board drivers 

-Used AMD Cleanup Tool to completely remove display drivers and reinstalled them (multiple times).

-Reset windows on my PC.

-Changed the power strip my PC was plugged into.

-checked for BIOS updates (there weren't any).

 

Let me know if you think there's something I should try before returning the mobo. Thanks!

 

Crossfire and sli are s%"t. It's more likely that the problem is just with that. Many many games have problems with multigpus as they depend on the drivers and the actual per-game implementation of that configuration. Plus, you mentioned USED 580s. Most 580s on the used market come from mining systems, which might have affected them at least partially in some situations. Have you tried using them as multigpu but not crossfire on a DX12 supported game?

Does this happen with EVERY game or just one? Are you overclocking?

Link to post
Share on other sites
Posted · Original PosterOP
9 minutes ago, 3rrant said:

Crossfire and sli are s%"t. It's more likely that the problem is just with that. Many many games have problems with multigpus as they depend on the drivers and the actual per-game implementation of that configuration. Plus, you mentioned USED 580s. Most 580s on the used market come from mining systems, which might have affected them at least partially in some situations. Have you tried using them as multigpu but not crossfire on a DX12 supported game?

Does this happen with EVERY game or just one? Are you overclocking?

I thought that might be the issue, so I tried putting just a single gpu in and reinstalling the drivers, but same issue. I had the GPUs running on a different motherboard just fine though I was using them separately as it was an matx board and couldn't fit both at the same time.

 

It only crashes with more intense games like Shadow of War, Borderlands 3, Far Cry 4. I've played a few casual games and it was fine.

 

Also, if I try to run a gpu stress test on AIDA64, it crashes.

Link to post
Share on other sites
Just now, featherwolf said:

I thought that might be the issue, so I tried putting just a single gpu in and reinstalling the drivers, but same issue. I had the GPUs running on a different motherboard just fine though I was using them separately as it was an matx board and couldn't fit both at the same time.

Either GPUs used in single mode on the same motherboard gives you this issue? Does it happen with ALL games? Have you tried using that single GPU is a different PCI-E slot, even at reduced speed?

Link to post
Share on other sites
Posted · Original PosterOP
Just now, 3rrant said:

Either GPUs used in single mode on the same motherboard gives you this issue? Does it happen with ALL games? Have you tried using that single GPU is a different PCI-E slot, even at reduced speed?

It happens with graphically intensive games, like Middle Earth: Shadow of War, but not with less demanding ones. I have not tried a different PCIE slot. I will try that next.

Link to post
Share on other sites
9 minutes ago, featherwolf said:

It happens with graphically intensive games, like Middle Earth: Shadow of War, but not with less demanding ones. I have not tried a different PCIE slot. I will try that next.

If it doesn't always happen, it's much more likely that the problem is GPU related than motherboard related. One thing that might be happening could be related to BIOS. If the used card were BIOS-flashed for various reasons, they might have different memory configurations that make them unreliable at gaming. Mining cards usually get memory overcloked and tuned directly from bios. 

 

Or they might be overheating the core/memory modules. Check that the models you have are running within specs all the time (check a review or the official specs on the manifacturer site).

 

If nothing seems out of the ordinary, also check the PSU. Can it sustain the entire system or are you close to its limit?

If all seems fine then yeah you might have a problematic board (or CPU, since PCI-E is managed by it)

Link to post
Share on other sites
Posted · Original PosterOP

Also, just checked System Event Viewer and all the crashes were logged with the source being "Kernel-Power"

 

Event ID 41

Level: Critical. 

The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.

Link to post
Share on other sites
Posted · Original PosterOP
1 minute ago, 3rrant said:

If it doesn't always happen, it's much more likely that the problem is GPU related than motherboard related. One thing that might be happening could be related to BIOS. If the used card were BIOS-flashed for various reasons, they might have different memory configurations that make them unreliable at gaming. Mining cards usually get memory overcloked and tuned directly from bios. 

 

Or they might be overheating the core/memory modules. Check that the models you have are running within specs all the time (check a review or the official specs on the manifacturer site).

That's a good point. I should look into the BIOS it is running. Do you know how I can check that?

Link to post
Share on other sites
1 minute ago, featherwolf said:

That's a good point. I should look into the BIOS it is running. Do you know how I can check that?

Using GPU-Z, go to Advanced tab then from the dropdown select AMD BIOS and you should see all the info needed. Then you need to google which bios version your cards should be running (depending on the manifacturer) and also which specs (clock, mem clock etc) are they supposed to be rated for and which are currently running (which can also be checked with GPU-Z). Do the clock test both at idle and during a stress test.

Link to post
Share on other sites
6 minutes ago, featherwolf said:

Also, just checked System Event Viewer and all the crashes were logged with the source being "Kernel-Power"

 

Event ID 41

Level: Critical. 

The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.

Might be random but also PSU/MOBO related.

 

Could you please list all your components to check that out?

Link to post
Share on other sites
Posted · Original PosterOP
4 minutes ago, 3rrant said:

Might be random but also PSU/MOBO related.

 

Could you please list all your components to check that out?

Gigabyte B450 Aorus Pro Wifi Rev. 1

Asus ROG Strix Gaming OC RX 580 (x2)

G.Skill Ripjaws V DDR4 3200 MHZ (4x4GB)

AMD Ryzen 5 2600 

Micro Center Brand 500 GB NVME SSD

Seagate Barracuda 4 TB HDD

EVGA Supernova G+ 80 Plus Gold 750w PSU

Link to post
Share on other sites
Posted · Original PosterOP

As far as I can tell everything seems stock in GPU-Z. Here's screenshots fom GPU-Z and Techpowerups bios database:

 

677568879_Annotation2020-05-27103038.png.475f0ab55dc3caa7c97141c998fe24a5.pnggpuz.gif.3517d22ac51bd32dbfa2df1abc2614cb.gif

Link to post
Share on other sites
15 hours ago, featherwolf said:

Gigabyte B450 Aorus Pro Wifi Rev. 1

Asus ROG Strix Gaming OC RX 580 (x2)

G.Skill Ripjaws V DDR4 3200 MHZ (4x4GB)

AMD Ryzen 5 2600 

Micro Center Brand 500 GB NVME SSD

Seagate Barracuda 4 TB HDD

EVGA Supernova G+ 80 Plus Gold 750w PSU

Bios seems fine.

 

While also the total power consumption of the system should be fine (600-700W) it's very near your PSU capacity and more importantly it might trigger the overcurrent protection on the 12V if the cards are pulling too many amps on the same rail. According to this article https://www.evga.com/articles/01185/evga-g-plus-power-supplies/ at least, the G+ series has a single 12V rail. One thing you might try to exclude this from the problem would be heavily downclock and tune the voltage down on both of your cards (like 300-600Mhz as a clock) and try to run them on a stress test with these settings. This will lower the amperage by a lot on the PSU rail.  Don't overclock the CPU as well for the test.

 

For stability you should also try to use only 2 sticks of ram, as Ryzen is picky about 4 sticks.

Link to post
Share on other sites
Posted · Original PosterOP
15 hours ago, 3rrant said:

Bios seems fine.

 

While also the total power consumption of the system should be fine (600-700W) it's very near your PSU capacity and more importantly it might trigger the overcurrent protection on the 12V if the cards are pulling too many amps on the same rail. According to this article https://www.evga.com/articles/01185/evga-g-plus-power-supplies/ at least, the G+ series has a single 12V rail. One thing you might try to exclude this from the problem would be heavily downclock and tune the voltage down on both of your cards (like 300-600Mhz as a clock) and try to run them on a stress test with these settings. This will lower the amperage by a lot on the PSU rail.  Don't overclock the CPU as well for the test.

 

For stability you should also try to use only 2 sticks of ram, as Ryzen is picky about 4 sticks.

Thanks for the info. I'll give it a shot!

Link to post
Share on other sites
Posted · Original PosterOP

So, I ended up returning the motherboard and getting an X370 board which has support for Crossfire at PCIE 3 x8/x8. Since then, I have had no more crashes. I think there may have simply been an issue with the power delivery through the PCIE on the previous motherboard or something like that especially since the GPUs were constantly shooting up to high 80's in temp as soon as I loaded a game, making the fans sound like a jet engine. This does not happen on the new Mobo either as the GPUs now hover around 61 under load on the hottest card (granted I did replace the thermal compound when I switched, but that's still a huge difference). Very strange. Didn't have huge expectations for that b450 Gigabyte board, but this was pretty frustrating and disappointing nonetheless. Thanks for the help along the way! I'm glad the issue has been resolved.

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×