Question Random BSODs and crashing games ?

[matbmx]

BUILD
MOBO: B450 Tomahawk
CPU: Ryzen 5 3600
Ram: 2x8 Gb Corsair Vegance 36090mhz
GPU: Rx5700xt Sapphire Pulse
SSD Nvme Express XPG 512 GB
HDD: Credo un sata da 1TB
PS: Seasonic 650W Gold Plus

When I first did the build (in 2019) I didn't have any problem. In the last couple of months it became worst as now it's difficult even playing games as they keep crashing (sometimes with BSOD sometimes without)
Crashes are completely random..they can occur as soon as I turn on my pc or after few hours.


What I have done so far:
Format my SSD and install again win10
Teesting ram with MEMTest86--> no errors
testing SSD e HDD con Crystal --> no errors

Please see minidump on the following link:
https://drive.google.com/drive/folders/1JyHbpSE6ZDxtKlvM69uHkniPF1rGgeA7?usp=sharing

 

[matbmx]

I have increased it to 1.4 V but Prime is not working anyway (when it was set on auto value was around 1.2V)

https://drive.google.com/file/d/1M7CF5E0w6oqYOyz8ETvI4uBfu9sloq4a/view?usp=sharing

 

[Colif]

hmmm

Try this, run it with 1 stick:

It's pretty tough to determine the source of errors like this. Even when the memory passes memtest, you could still have a bad stick. Since memory is also the easiest thing to physically remove and replace, I suggest you start there first.
Run on just one stick of memory, run the P95 test again. If it passes, swap to the other single stick, repeat.

If both those tests pass (or both fail), then it's likely not the memory- it would be well and truly rare to get two bad memory sticks!

https://www.mersenneforum.org/showthread.php?t=25106

 

[matbmx]

tried with one stick each time and I have got the same error..
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected running 960K FFT size, consult stress.txt file.
FATAL ERROR: Rounding was 0.4999890703, expected less than 0.4
Hardware failure detected running 960K FFT size, consult stress.txt file.


I also had 2 new crashes when one stick was plugged.
https://drive.google.com/drive/folders/1JyHbpSE6ZDxtKlvM69uHkniPF1rGgeA7?usp=sharing

 

[Colif]

okay, so both sticks fail?
What temperatures are you getting running these tests?
it seems its possible the cause could be heat, have you cleaned out inside PC recently?
might need to reapply thermal paste.

that was the fix for the guy in the post I quoted above

So I've had the chance to re-apply some new thermal paste and think I've made some (limited) progress at least. In order to bring temps down I:

• Cleaned the existing thermal paste and added new paste (Arctic Silver 5)
• Increased the CPU fan profile to run at a slightly higher speed
• Changed the CPU settings from "Default" to "Eco-Mode" in Ryzen Master

The final result was the CPU staying rock steady at 50°C at idle and steady at 85°C while running P95. Lastly I think I've managed to get the system somewhat stable, It was able to run P95 for around 12 hours yesterday with no errors.

No point showing dumps while you getting rounding errors. They will cause them.

 

[matbmx]

GPU temp is around 50 C. I have an AIO Capellix Elite 280 which keeps Temp very low. Also, BSOD happened sometimes as soon as I start PC.

 

[Colif]

I think you need to take PC to repair store.

Its possible cause is Motherboard or CPU, I cannot tell as no tests for Motherboard and only one for AMD CPU is Prime95 which won't finish.
The shop might be able to test everything else though. Or swap in known working cpu/ram to test if you still get the prime errors.
I have a 240mm H100i AIO myself, its unlikely to be heat on a 3600 with a 280mm on it, even if start up is when CPU is the hottest. Mine was 66.2c at startup, doesn't stay there for long though once pump starts circulating water. Average temp at idle is just under 40c (I only use quiet plan)

 

[matbmx]

Little update:

tried one ram stick for 1 week = crashed
tried the other one for 1 week= crashed

tried another GPU= crashed

How can I proceed?

is there any program which can test SSD , Mobo and CPU in order to understand which component could be faulty?

Thanks

 

[Colif]

I believe you could run AIDA64, it might help
OCCT as well - https://www.ocbase.com/

XPG Software - https://www.adata.com/us/support/consumer?tab=downloads
not sure if it has any tests.

Already tried Prime95 and it doesn't finish?

no tests for motherboard, have to test everything else and if it passes, it has to be MB. This isn't a perfect test.

this is why I suggested a repair store. One way to test if its motherboard is put everything on another MB and see if it still crashes, shops may have a spare they can use. I don't expect people to buy a new MB just on a guess.

 

[matbmx]

I believe you could run AIDA64, it might help
OCCT as well - https://www.ocbase.com/

XPG Software - https://www.adata.com/us/support/consumer?tab=downloads
not sure if it has any tests.

Already tried Prime95 and it doesn't finish?

no tests for motherboard, have to test everything else and if it passes, it has to be MB. This isn't a perfect test.

this is why I suggested a repair store. One way to test if its motherboard is put everything on another MB and see if it still crashes, shops may have a spare they can use. I don't expect people to buy a new MB just on a guess.

tested the XPG software and health is good, so I think no problem from here.

Tried a stress test with OCCT and CPU test failed. I've had the same problem with prime test.
I am thinking CPU could be the problem than?

I do not want to bring my PC to a shop. They will charge me for lots of money and nobody can assure me they will fix the problem. My problem is very hard to detect and I don't trust shops at all.

I prefer to "waste money" on new CPU and Mb.

 

[matbmx]

tried again OCCT and I am not able to do CPU test as I get lots of errors. On 2 occasions PC crashed right after I have started the test.

Could be the CPU faulty?

 

[Colif]

Since we can't get to end of the tests, you could assume that. Or its the motherboard.

given it keeps failing CPU tests, it probably is the CPU. Would explain why both sticks fail if problem is CPU.

 

[matbmx]

Since we can't get to end of the tests, you could assume that. Or its the motherboard.

given it keeps failing CPU tests, it probably is the CPU. Would explain why both sticks fail if problem is CPU.

I have ordered from Amazon a Ryzen 5600x which I can give back for free within 30 days.
I will see if test will fails with the new CPU and it keep crashing during games and normal usage. If no issues, than it must be the CPU. Otherwise, could be the MB the problem but - if it was the MB - I don't think a CPU test would fail.

Correct?

 

[Colif]

i was going to suggest a 5600 so if it is CPU, you get an upgrade anyway.

let us know how you go

 

[matbmx]

i was going to suggest a 5600 so if it is CPU, you get an upgrade anyway.

let us know how you go

Can I ask you something?

If you have the possibility to spend a little bit of money and get something better, will you stay on AMD or intel?

I am not sure if I can go for example with the 12400F by changing only MB.

In both cases, if the CPU will be the problem, I will change the MB.
Mine - B450 tomahawk - has a beta version for the 5000 series and it's quite old now.. I really want to try Intel this time but - if it's not worth the money - than I will stay with AMD.

 

[Colif]

Its not just the Motherboard, need new ram too if you go intel as I think they have DDR 5 now.

I would start a new question and ask about that as I am not sure. I know Intel better at games now, but I haven't looked into it a lot. I can't afford a new PC and my 3600xt is good enough for me still.

 

[matbmx]

Just wanna say a Huge thanks to everyone who tried to help me.
In the end, 99,9%, the problem was the CPU.

I have tried as soon as I've installed ryzen 5600x to make a strees test on the CPU with the program suddgested me "OCCT" and the test didn't fail. I have tried to play for 1-2 hours and no crashes occurs...For be 100% sure that problem has been solved I have to use it for 1 week not only half day but so far I am satisfied.

I don't know how it could be damaged as I didn't do anything with it (either OC)..maybe few times during 3 years when lights went off it could damaged the CPU? hard to say..

 

[Colif]

note: i like that CPU appears to have fixed it.
I don't like fact you had to replace CPU but its not first time.
At least its not a downgrade and you don't have to buy all new things yet