Question New PC build is freezing & crashing ?

Toggle sidebar Toggle sidebar
W

Whitlam

Distinguished
Mar 6, 2013
7
2
18,515
System will freeze / crash when launching or running games. This also rarely occurs when streaming video and browsing web (Firefox / Edge).

Temps seem to be fine on GPU and CPU.

PC BUILD:
OS: Win 11 Home
Motherboard: Gigabyte B760I Aorus Pro
CPU: Intel i5-14400F
GPU: AMD Radeon RX 7800 XT (powered through 2 separate PCIe cables)
RAM: G.Skill Trident Z5 RGB Series (Intel XMP) 32GB (2 x 16GB)
SSD: Samsung SSD 980 1TB
PSU: Corsair SF850L Fully Modular
Cooling: 2x Mobius 120 OC Case Fans
There are 2x monitors one through DP > HDMI and the other HDMI > KVM switch HDMI then KVM HDMI > HDMI. Not sure if this is related.

I have not made any changes through the BIOS menus (thinking there is maybe something there?).

Drivers have all been updated as at 10/01/2025.

I've checked the Event Viewer, but not sure if this actually assists in diagnosing the problem.
Event Viewer:
Critical: Event 41, Kernel-Power - The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.

There are also some non-critical errors that are appearing if required.

Any assistance would be greatly appreciated.
UPDATE

I've re-connected all the power connections and re-inserted RAM, GPU into the Mobo.

BIOS has been updated.

GPU and PSU fans spin up when PC starts, but I've noticed they don't spin up when games have launched and are running. Presume that they don't spin up when under low load.

AMD software is showing that the GPU is utilised and the Device Manager say the device is working properly.

Utilisation at around times of crash:
CPU: 20%
GPU: 20%

Same problems persist, but seems like it's taking longer for the crashes to occur on average.
 
Last edited:
UPDATE

I've re-connected all the power connections and re-inserted RAM, GPU into the Mobo.

BIOS has been updated.

GPU and PSU fans spin up when PC starts, but I've noticed they don't spin up when games have launched and are running. Presume that they don't spin up when under low load.

AMD software is showing that the GPU is utilised and the Device Manager say the device is working properly.

Utilisation at around times of crash:
CPU: 20%
GPU: 20%

Same problems persist, but seems like it's taking longer for the crashes to occur on average.
 
it sounds like a graphics driver issue. . have you tried with 1 monitor only. and see if that is the cause..

as for the gpu fans.. they are designed to only kick in when the gpus get to a certain temperature then they will come to life. .but it could also be a failing in the gpu.. test it below.

do you have gpu-z installed. if so open the app and then click the " pcie render test" ( its the button next to "pcie4 x16 " part.. if it crashes with this. its a graphics card issue.
 
I've installed GPU-Z and run the PCIe render test. It doesn't crash immediately, but it does eventually crash at a similar times as if a game was running (ranging 2 minutes to 45 minutes). I've run this test on both dual monitor and each monitor separately.
 
MEMOFLEX said:
What temperatures are you getting on both gpu and cpu at idle and during the stress test?

Have you tried to run the system without XMP enabled?
Click to expand...
I've just run a FurMark benchmark test:
Test Length: approx 8 minutes (before crash)
GPU Temp: 60 degrees C (140 F)
GPU Hotspot: 79 degrees C (175 F)

PSU and GPU fans did fire up as would be expected.

Subsequent crash after logging back in after stress test was 30 seconds after logging in and opening OneNote and FireFox.
 
Whitlam said:
I've just run a FurMark benchmark test:
Test Length: approx 8 minutes (before crash)
GPU Temp: 60 degrees C (140 F)
GPU Hotspot: 79 degrees C (175 F)

PSU and GPU fans did fire up as would be expected.

Subsequent crash after logging back in after stress test was 30 seconds after logging in and opening OneNote and FireFox.
Click to expand...
Those gpu temps look fine. Doesn't mean the gpu isn't the issue but it doesn't seem temperature related.

What were the CPU temps?

Have you tried the same test without XMP enabled?
 
MEMOFLEX said:
Those gpu temps look fine. Doesn't mean the gpu isn't the issue but it doesn't seem temperature related.

What were the CPU temps?

Have you tried the same test without XMP enabled?
Click to expand...
Sorry mate, build novice here. Just troubleshooting as I go on this one.

What is XMP, and how would I disable it?

Not sure what CPU temps were, going to try another benchmark and keep a closer eye on that this time.
 
  • Like
Reactions: MEMOFLEX
Whitlam said:
Sorry mate, build novice here. Just troubleshooting as I go on this one.

What is XMP, and how would I disable it?

Not sure what CPU temps were, going to try another benchmark and keep a closer eye on that this time.
Click to expand...
No worries buddy. Sometimes these things are slow and steady. Eliminating 1 thing at a time as you go.

XMP is related to your RAM and is something that can be altered in the BIOS on boot. It is basically information that tells the motherboard the settings to run the memory in a pre confirmed overclocked state with the click of a button. Pretty much all memory works this way so do not be put off by the term overclocking.

What is the advertised speed of you memory out of interest. I can see it is the Trident Z5 but there will likely be multiple SKUs with differing speeds.

If you download a program called CPU-Z then this helpful little tool will tell you the key details about your computer. Very helpful. On the memory tab it will tell you what speed you are running etc
 
MEMOFLEX said:
No worries buddy. Sometimes these things are slow and steady. Eliminating 1 thing at a time as you go.

XMP is related to your RAM and is something that can be altered in the BIOS on boot. It is basically information that tells the motherboard the settings to run the memory in a pre confirmed overclocked state with the click of a button. Pretty much all memory works this way so do not be put off by the term overclocking.

What is the advertised speed of you memory out of interest. I can see it is the Trident Z5 but there will likely be multiple SKUs with differing speeds.

If you download a program called CPU-Z then this helpful little tool will tell you the key details about your computer. Very helpful. On the memory tab it will tell you what speed you are running etc
Click to expand...
Mate, you are a legend for sticking with me through this.

I've got HWiNFO and GPU-Z - so let me know if there is anything that might be helpful to you.

RAM:
On the box:
Size: 32 GB
Freq: 2394.1MHz
Timing: 40-40-40-76
tRC: 116

XMP-6400 (not 100% what this means or if it's what you're after):
Freq: 3200 MHz
Timing: 32-39-39-102
tRC: 141

Second test through FurMark:
Test Length: 13 minutes (no crash)
GPU Temp: 51 C (124 F)
GPU Hotspot: 73 C (164 F)
Fans on GPU spun up to 70% this time.

CPU Max: 74 C (165 F)
CPU Avg: 62 C (144 F)
 
  • Like
Reactions: MEMOFLEX
Whitlam said:
Mate, you are a legend for sticking with me through this.

I've got HWiNFO and GPU-Z - so let me know if there is anything that might be helpful to you.

RAM:
On the box:
Size: 32 GB
Freq: 2394.1MHz
Timing: 40-40-40-76
tRC: 116

XMP-6400 (not 100% what this means or if it's what you're after):
Freq: 3200 MHz
Timing: 32-39-39-102
tRC: 141

Second test through FurMark:
Test Length: 13 minutes (no crash)
GPU Temp: 51 C (124 F)
GPU Hotspot: 73 C (164 F)
Fans on GPU spun up to 70% this time.

CPU Max: 74 C (165 F)
CPU Avg: 62 C (144 F)
Click to expand...
No worries at all.

OK so RAM speeds can seem a bit strange when you are learning but DDR stands for Double Data Rate so the 2394mhz is essentially doubled to 4800 which is the base speed of DDR5.

The XMP speed of 3200 is then 6400 accordingly. This is essentially an overclock but as mentioned all RAM works this way.

The other numbers are what are called the timings and without going to far down the rabbit hole then these relate to latency. Basically the lower the better.

As we have been encountering issues what we want to ensure at this stage is that we are running the memory at the default speed to rule a memory issue out of the equation.

Did you download CPU-Z (different to GPU-Z) ?

an alternate way to check is when you boot your computer then keep pressing delete on your keyboard and this will take you into your BIOS. The first screen you see should tell you if XMP is enabled or not.

I would also take that KVM switch out of the equation for now. I doubt its related but just to be sure.

You mentioned that the last test you ran completed without crashing. Is that the first time it has completed?

EDIT : CPU-Z will also tell you what BIOS version you are using. You mentioned you updated it but can you confirm which BIOS you are on?
 
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom