Application Error 56; ACPI 2 - hard freezes

Hi OCOWers

Got a real conundrum here. Freshly built 10700K + Z490i Aorus Ultra + RTX 2070 rig, with an SX650-G PSU. sitting in a lian li q58 which runs a pci-e gen4 riser cable.

Randomly - sometimes minutes after powering up, sometimes hours or days after - it'll hard lock up, in a strange way. The mouse will keep moving but windows will otherwise become unresponsive - then after jiggling the mouse around enough, that freezes too, necessitating a hard reboot (it doesnt BSOD by itself though).

Checking event viewer after rebooting shows a red X error that reads "Application error 56....ACPI 2".

Had turned "power loading" on and off in the BIOS but it doesn't seem to make a difference. Have also swapped out RAM, same issue.

I am guessing it's got to be the motherboard, CPU or the PSU (although i guess it could be the GPU, i have had it for ages and its never caused issues in any other rig ive had it in).

anyone come across this? have googled extensively but just found a bunch of threads on other tech forums that were left open-ended...

 

Have you tried turning it off and on again?

 

What OS? and have you got anything else plugged? (PCI, USB etc)

-nav.

 

windows 10 pro - only things plugged in are a keyboard (a standard ducky one that i have used for over a decade on lots of PCs) and a razer deathadder mouse (similar story to the keyboard, i know its a working unit - have also been using it with a logitech wireless unifying mouse and it does the same thing).

 

Hmmm, my guess would be beefing up the PSU.
I'm a bit old school and tend to over spec on power and always prefer a nice heavy unit.

but good luck with it, pretty hard to track down issues like this

-nav.

 

ive got a heavily OCed 6700K and a 1070 running in another SFX rig with a 450w PSU and its been fine for over a year, i definitely dont subscribe to the club of people needing over-specced power supplies - 650w is more than enough for this build!

but ive just bought an SF600 off the forums and will try that. going to be a huge pain dismantling this system, i was so proud of myself when i built it and managed to actually cable manage it properly

 

Yea, I hear ya.

Hopefully it is the PSU, otherwise mobo or cpu will be a massive PIA.

-nav.

 

if its the CPU then its just bad luck cos its the ultra-cheapie delidded one i got off the forums - but ive got a receipt for the mobo so can RMA it. PSU is kinda worst case scenario tbh cos theres no way to swap in a new one without dismantling the whole thing (have threaded various cables between the two "halves" of the case).

 

oh i see, hmmmmm then i'd be leanding toward cpu.
you might be able to get away with disconnecting the existing psu and just placing the new one outside the case then connecting it all with the side panel off?

-nav.

 

Have you tried stress testing cpu-RAM?

 

Have stress tested the CPU for a while - ran for a couple of hours with no crash. It only seems to happen when the pc is under a little bit of load (like watching a YouTube video). Haven't error tested the ram but have tried three separate known-good kits.

 

I had something very similiar with my faulty X570 ITX sanjay, you know which one I mean, the ASROCK!

I know it's different because mine happened at idle. But try change the idle power load settings in the BIOS. It could be dropping the idle power too load on a downspin of effort on the PC and causing crashes.

In the end I did have to RMA the board and new ones been perfect ever since.

 

Yeah the bios only has an option for power loading, but I could fiddle with the c states too (maybe turn off c8)

This problem wouldn't be happening if I had a perfect beautiful ASRock board

 

the plot gets...slightly harder to stir? I think that's the saying.

ran a 12 hour fireplace video overnight with a different PSU hooked in, and when i came back to it this morning, the video had frozen, as it did with the old PSU. but when i moved the mouse and tried to click things, the whole screen went black...then everything came back to normal a few seconds later. Event Viewer this time said that nvlddmkm.sys crashed but recovered (Nvidia driver).

i ran the fireplace video again and went out...back a couple of hours later, the video had frozen again - moved the mouse - screen went black - came back - same error in Event Viewer.

so, same symptoms, but with the different PSU, it manages to recover, and generates a different error.

will re-seat the GPU, DDU the driver and reinstall the very latest nvidia drivers...and test again!

 

Then try a different gpu

 

i'll do these fixes first and if it's still buggered, time to put my spare 1080Ti in. i am 99.99% sure this GPU is rock solid though, had it running perfectly fine in a ryzen rig for ages...

 

so put my 3060 Ti in, with the same silverstone PSU, and it gave me the ACPI 2 error again. ordered a 750w SFX CM PSU from scorptec, that'll be the next swap step.

 

put in the SF600 that i got from the forums, still gives the Error 56, ACPI 2 error in event viewer every few hours but it no longer seems to make the system hang/freeze. will do a gaming session on it over the weekend to see what it does. googling seems to suggest this is a very common problem with gigabyte motherboards and 10th/11th gen intel...

 

With the nvlddmkm.sys errors, did you try clean reinstall of drivers? I've had Nvidia problems from crappy drivers that needed two reboots to remove cleanly.

-nav.

 

yep i did a DDU in safe mode, reboot, reinstall drivers, reboot

weirdly though with the SF600 i havent seen another nvlddmkm.sys error...

i do have another rig with a Z590 board and a 10105F in it, which is all i would need to troubleshoot both the board and CPU, but ughhh don't wanna remove the bloody CPU from this rig