AMD R5 3600 dying again?

UkayJ · 17 Sep 2021 at 13:34

rah^ said:
Odd one

You discount the gpu but I wouldn't for now

I have an rx580 that does exactly the same and the fix so far was disabling 0 RPM fan mode on it (assume the 480 has a similar setting)

Actually I don't have that setting on mine, thanks though

Cyber-Mav · 17 Sep 2021 at 14:32

Is it fixed now?

UkayJ · 17 Sep 2021 at 14:35

I finally have something to go on other than a freeze - BSOD!

While playing Insurgency Sandstorm for about half an hour I got : KERNEL_SECURITY_CHECK_FAILURE

I ran Windbg preview and debugged the Minidump files from the Windows Minidump folder, then copied and pasted the paragraph of information that was the least confusing out of it all :

File1
DRIVER_POWER_STATE_FAILURE (9f)
A driver has failed to complete a power IRP within a specific time.
Arguments:
Arg1: 0000000000000003, A device object has been blocking an IRP for too long a time
Arg2: ffffbd051eba1060, Physical Device Object of the stack
Arg3: fffff80539690850, nt!TRIAGE_9F_POWER on Win7 and higher, otherwise the Functional Device Object of the stack
Arg4: ffffbd051ec159e0, The blocked IRP
File2
DRIVER_POWER_STATE_FAILURE (9f)
A driver has failed to complete a power IRP within a specific time.
Arguments:
Arg1: 0000000000000003, A device object has been blocking an IRP for too long a time
Arg2: ffff9a87e8631060, Physical Device Object of the stack
Arg3: fffff80639c90850, nt!TRIAGE_9F_POWER on Win7 and higher, otherwise the Functional Device Object of the stack
Arg4: ffff9a87e8685960, The blocked IRP
File3
KERNEL_SECURITY_CHECK_FAILURE (139)
A kernel component has corrupted a critical data structure. The corruption
could potentially allow a malicious user to gain control of this machine.
Arguments:
Arg1: 0000000000000003, A LIST_ENTRY has been corrupted (i.e. double remove).
Arg2: ffffe88b2e1cd810, Address of the trap frame for the exception that caused the BugCheck
Arg3: ffffe88b2e1cd768, Address of the exception record for the exception that caused the BugCheck
Arg4: 0000000000000000, Reserved

Does any of that mean anything to anyone?

Cyber-Mav · 17 Sep 2021 at 23:29

Driver issue it says but could still be dodgy hardware causing issue. Seeing as how you get no issues with ram at 2400mhz speed it does sound like its ram issue.

UkayJ · 18 Sep 2021 at 17:28

Cyber-Mav said:
Driver issue it says but could still be dodgy hardware causing issue. Seeing as how you get no issues with ram at 2400mhz speed it does sound like its ram issue.

I don't recall saying I get no freezes with RAM at 2400MHz and I ran a memtest all night and got 0 errors.

I'll still try the other GPU from 2nd PC when I get chance and hope for no freezes/BSODs.

InfiniteNoise · 18 Sep 2021 at 18:21

I had a similar issue with my Ryzen 5 1600, I used to get random freezes, with no warnings or anything, what solved it was leaving RAM at stock (no XMP, no overclocking, etc) and disabling low power/idle states and using high performance power profile in Windows 10, it's been months since I made these changes and I've had no issues at all. I still left the CPU overclocked though, so that's not the issue.

So yeah, it's most likely the RAM.

UkayJ · 19 Sep 2021 at 18:55

InfiniteNoise said:
I had a similar issue with my Ryzen 5 1600, I used to get random freezes, with no warnings or anything, what solved it was leaving RAM at stock (no XMP, no overclocking, etc) and disabling low power/idle states and using high performance power profile in Windows 10, it's been months since I made these changes and I've had no issues at all. I still left the CPU overclocked though, so that's not the issue.

So yeah, it's most likely the RAM.

I had the Ryzen Balanced profile selected, I now have the Performance profile Active.

I've tried RAM on Auto/stock and PC still froze.

I'm really hoping it's not the GPU because prices are still stupid high! But hopefully I'll find the culprit soon enough, thank you.

InfiniteNoise · 19 Sep 2021 at 19:10

Hmm, you could try disabling ULPS if you suspect a GPU issue if you have AMD graphics card, you can do it via registry or via tools like MSI Afterburner, it's one of the things I did along with the changes I made I mentioned in the previous post here.

AMD seems to have some issues relating to low power states on either their CPUs and GPUs in certain circumstances that results in crashing/freezing, hence why I disabled all these things and haven't had any issues ever since, worth checking out if you've tried everything else, though I'm not sure if it'll help you in your case, in my case, most of the freezing happened when the PC was idle.

UkayJ · 20 Sep 2021 at 07:15

I might try that, I try might disabling Cool N Quiet too to see if that's the cause.

I can now tell if it's going to happen soon because the game is laggy but I can be getting over 100fps and lower than 50ms ping.

However sometimes it appears to be running smoothly then freezes anyway.

Cyber-Mav · 20 Sep 2021 at 17:20

Did you try different gpu?

UkayJ · 20 Sep 2021 at 17:41

Cyber-Mav said:
Did you try different gpu?

Doing it right now

UkayJ · 20 Sep 2021 at 19:02

So I've gamed for about an hour and there was no stuttering and no freeze.

Upon taking the gpu out that I'm having problems with I noticed the contacts that plug into the motherboard have what I think is thermal paste on them.

How do I clean that off safely?

VaderDSL · 20 Sep 2021 at 19:09

UkayJ said:
So I've gamed for about an hour and there was no stuttering and no freeze.

Upon taking the gpu out that I'm having problems with I noticed the contacts that plug into the motherboard have what I think is thermal paste on them.

How do I clean that off safely?

just something like isopropyl alcohol will do

Cyber-Mav · 20 Sep 2021 at 20:44

UkayJ said:
So I've gamed for about an hour and there was no stuttering and no freeze.

Upon taking the gpu out that I'm having problems with I noticed the contacts that plug into the motherboard have what I think is thermal paste on them.

How do I clean that off safely?

You got a pic of it?

UkayJ · 21 Sep 2021 at 10:05

I got a pic of the side I could get to easier, that's just 3 gold contact strips thats cleaned up really good, but the pic isn't very good.

The other side was half of the contact strips, it looked as though it had been slid through the slot and scratched but again, it's cleaned up quite well.

Further testing required but going well so far.. .

Cyber-Mav · 21 Sep 2021 at 20:50

did that goop get into the slot itself?

UkayJ · 21 Sep 2021 at 21:55

Cyber-Mav said:
did that goop get into the slot itself?

I'm not sure but if I have that problem again I'll know to check it.

UkayJ · 21 Sep 2021 at 21:59

Well the problem is solved.

Thank you very much for everyones help, apollogies that it was such a noob cause but I can't remember the last time I had the computer apart and could have got the thermal paste on the GPU contacts, its worked fine since then!

Also worth noting though that the exact same symptoms resulted in a dead CPU - very odd!

Cyber-Mav · 21 Sep 2021 at 22:45

UkayJ said:
Well the problem is solved.

Thank you very much for everyones help, apollogies that it was such a noob cause but I can't remember the last time I had the computer apart and could have got the thermal paste on the GPU contacts, its worked fine since then!

Also worth noting though that the exact same symptoms resulted in a dead CPU - very odd!

must be some skill you got there applying thermal paste to cpu ending up getting it on gpu contacts lol

Trig · 22 Sep 2021 at 10:35

A lesson learnt..
I've seen a CPU work fine for everything before with a dGPU, then when it come to using the iGPU it just didn't work, was a single bent pin in the socket..

Things are so bloody tiny these days you have to be very careful building stuff, personally, I always put the CPU into the board and goop and heatsink it before it goes in the case just so you have everything flat when building up and it just gives you room to move things around and test before you case up.