Display driver crash caused by DCOM

[HeavensEtherian]

I've recently got a new GPU,but realised that while it works in some games it crashes at boot or very quickly after entering [stuff like titanfall/overwatch works while "easy" stuff like minecraft/CSGO crash].It appears everytime in the event viewer,2 warnings saying that the driver crashed and recovered successfuly and 2 DCOM errors about local activation.
I've worked with it for a while,and did all i found on the web even closely related to this:I've downgraded the drivers,i've reinstalled with DDU,I've did some permission tweaking for the registry keys and component services [gave my user,admin group and SYSTEM permissions yet it doesn't work]
The name of the regkey in DCOM was "RuntimeBroker",i did everything I could find online but didn't get any fix.
EDIT: Added eventviewer logs

 

[BigFeet]

Exactly what card? Brand\model. Have you tried stress testing the card with unegine heaven benchmark while monitoring temperatures?

 

[HeavensEtherian]

My GPU is a RX 460 Asus 2GB (dual fan one) and yes I've tried furmark,after 10 mins temperature peaked at about 70C (which is pretty low considering room temp).Also it turns out pretty much any game crashes after some time,but it differs a lot,for ex titanfall crashes after about 2 hours (DXGI_DEVICE_HUNG) yet Minecraft crashes in ~15 secs.I've tried multiple drivers too

 

[BigFeet]

If it's still under warranty, I'd RMA it. There's most likely something physically wrong with the card. If you've tried multiple drivers, and clean installed them, and ruled out temperature, then I can't think of many other things which would cause those issues. The only ones I can think of is an inadequate power supply (although an rx460 isn't very power demanding), or, if you bought the card used, I wouldn't rule out a modified bios, such as some use for mining.

 

[HeavensEtherian]

I've bought it used,but it might be the power supply too (I have a "generic" weak PSU),I'll look into it

 

[BigFeet]

Requirements are 350 watts. But if I were building a system for someone with that card, I wouldn't go under a good quality 400 watt PSU. PSU wattage doesn't tell the entire story. Having enough amps on the 12 volt rail is also very important.

 

[HeavensEtherian]

In theory the PSU can deliver 500W [SPS-ATX-500],although i do belive that it is the problem here since after lowering clocks/powerlimit in afterburner,it seems more stable [although not crash-less],but it does make me question why none of the syntethic benchmarks do not crash it.
Also as a sidenote my PC has a FX 6300,which also eats a lot of power.
EDIT: I think i'm gonna buy a Njoy Ayrus 450W 80 plus,i've had one of those before in a older PC and it did the job [also was very quiet,my current PSU sounds more like a helicopter]

 

[HeavensEtherian]

Now after a few tweaks i've started getting "Application has been blocked from accessing Graphics hardware"

 

[BigFeet]

What gpu did you have in there before? I'm still leaning towards something being wrong with the video card itself. There really is no reason even a no name 500 watt psu shouldn't run it without errors.

 

[HeavensEtherian]

It ran on a (Asus?) R7 240 2 GB DDR3,it was pretty weak but it never did stuff like this. Right now I'm thinking of reinstalling windows because not even dism/SFC could fix the "corruptions"

 

[BigFeet]

That would be my last resort. If that doesn't fix it, then I would be fairly certain it would be a faulty card. I'd be surprised if it was a windows issue given the r7 ran fine. But I've seen stranger things. Good luck.

 

[Saltgrass]

The event log shows the statement below. Have you looked for updated drivers?

Display driver amdkmdap stopped responding and has successfully recovered. Some errors/warnings may be the result, not the cause of a problem.

Sorry I can't help with AMD video cards. But you might check the version and date of the driver.

 

[HeavensEtherian]

UPDATE:After downgrading to a ancient driver [16] and turning down memory clocks [from 1750 to 1400] ,I belive it might've been fixed. I've been able to play CS:GO for 2 hours already without any problems at all.I belive the memory chips are a little unstable.
EDIT:I guess not,it was just luck...

 

[lrman97]

UPDATE:After downgrading to a ancient driver [16] and turning down memory clocks [from 1750 to 1400] ,I belive it might've been fixed. I've been able to play CS:GO for 2 hours already without any problems at all.I belive the memory chips are a little unstable.
EDIT:I guess not,it was just luck...

Try going into NVIDIA Control Panel and clicking on the help menu at the top right and then checking debug mode, this helped me but its not a permanent fix as you have to do it every time you turn on the computer. Let me know if you find a real fix.

 

[HeavensEtherian]

Try going into NVIDIA Control Panel and clicking on the help menu at the top right and then checking debug mode, this helped me but its not a permanent fix as you have to do it every time you turn on the computer. Let me know if you find a real fix.

I'm on amd..

 

[lrman97]

I'm on amd..

Oh damn, did you try to downclock your gpu?

 

[HeavensEtherian]

Oh damn, did you try to downclock your gpu?

Yes I've did,it "helped" [it crashed after way more time,about 30 mins instread of 20 secs] but it's not a good fix,however low you take the voltages/core clock/memory clock/power limit it always happens,going below ~900 mhz on core clock didn't even help a bit,no ideea how this can happen.

 

[Comp Cmndo]

I'm on amd..

That's the problem...AMD.
I bought a new video card & monitor won't return from sleep. Tried older drivers & nothing works.
The old nVidia has no problems.