Application blocked from accessing Graphics hardware

[Murkdweller]

Hi.

I have a new built PC and have been struggling with crashes for a month. I have tried more or less every "how to fix" suggestion found out there. I think. But it still crashes.

It started with me getting the message:
--------
Event 4101 , Display
Display driver nvlddmkm stopped responding and has successfully recovered.
--------

I don't get this message any longer, but it still crashes with another message

---------
Application has been blocked from accessing Graphics hardware.
---------

I am not sure what removed the first error message, because it happened around the last release of both windows 10 patch and Nvidia driver update.

The strange thing it doesn't crash in the more graphic demanding games like BF1. It has never crashed in BF1, but games like Civilization 6 and Fortnite crashes all the time.

I have done the following "how to fix":
- DDU to remove everythig and clean install of drivers
- Set TDRDelay to 8 or something
- sfc /scannow - no issues
- DISM.exe /Online /Cleanup-image /Scanhealth
- DISM.exe /Online /Cleanup-image /Restorehealth
and tried other dodgy fixes that messed with the registry so i had to roll back.

I am more or less wondering. Is this a GPU problem or a Windows 10 problem, or possibly a PSU problem. I din't buy a new monitor yet. Can a old 60hz monitor cause any of the problems?

Any help would be very much appreciated. Realy depressing having a new PC that is useless.

Thanks in advance.

 

[eidairaman1]

Thats the gpu driver crashing.

Have you installed the latest chipset drivers from the AMD site?

 

[Murkdweller]

Now i have. Mine was from november last year. Now i installet chipset drivers from mid january

 

[John Naylor]

There are multitude of causes of this error, most solved by driver updates, but on the nVidia site you should be able to find a thread which passes this nvlddmkm error off to a Windows timeout error ... from following the thread, the recommendations have helped some but most not.

 

[Mussels]

i had similar issues when my OC wasnt totally stable, and my specs arent that different to yours

go back and double check long term stability of your CPU OC and RAM OC (remember that ryzen doesnt officially support 3200Mhz)

 

[Murkdweller]

I haven't done any kind of overclocking.

Now i am on top of it all getting BlueScreen with error message VIDEO_SCHEDULER_INTERNAL_ERROR.

Can the RAM be the issue? It is RAM made for Ryzen, but i HWMonitor recognizes the RAM as something else than it is.

 

[StefanM]

... It is RAM made for Ryzen, but i HWMonitor recognizes the RAM as something else than it is.

Double-check with HWiNFO64 or AIDA64 if you suspect false infos.

Regarding the crashes: kernel debugging is some hard stuff.
I would complain in the offical NVIDIA driver forum.
Keep the file C:\Windows\MEMORY.DMP handy along with info about the driver version used at time of crash (GPU-Z screenshot), just in case an NVIDIA engineer wants to do a "crash dump analysis".

 

[Murkdweller]

AIDA64 recognizes the correct RAM it looks like

 

[Mussels]

if the rams running above 2133Mhz, its overclocked on ryzen

 

[DRDNA]

I'm going to side on it is an OS issue and I would recommend reinstalling windows then installing All your MOBO drivers starting with the Chipset first with a reboot before the rest of the MOBO drivers then running windows update and then retest and see whats still going on.

 

[RatusNatus]

Add the software folder to windows defender exceptions.
Just that. As you can read in the message, the app is being blocked...by Windows.
If the folder is already there, delete it and add again.

 

[er557]

Had all of those issues; the reasons vary from buggy gpu driver, gpu overclocking not stable enough, gpu overheating, cpu overheating, a game incompatibility with windows 10(scheduler internal error bsod), even graphics card being in the wrong pcie slot, bios settings related to northbridge, general system instability. Also i would give the gpu more headroom by raising power target to 120% with precisionXOC

 

[W1zzard]

Also try underclocking the GPU a bit, to rule out that the GPU is not fully stable at stock speeds

 

[Murkdweller]

I actually opened up the pc and removed all connectors for power, GPU, ram etc..
Blew it clean with pressurized air and put it together. I added the folder to exceptions and will try for a while.

Thanks for several good suggestions. I will try them one by one and report back. Hoping i wont have to fully reinstall. A bit of a hassle.

 

[BCN1000]

I actually opened up the pc and removed all connectors for power, GPU, ram etc..
Blew it clean with pressurized air and put it together. I added the folder to exceptions and will try for a while.

Thanks for several good suggestions. I will try them one by one and report back. Hoping i wont have to fully reinstall. A bit of a hassle.

How did this go in the end? I have the same problem.

 

[eidairaman1]

An


How did this go in the end? I have the same problem.

Mega necro, try the solutions above or change gpu chip maker.

 

[BCN1000]

I am aware of the date of the thread. The problem and it's solution are still relevant in the year 2021 (as indicated by the fact that I am replying to it).

Switching to a different chip manufacturer for future purchases is not going to help myself or others with the problem described by the OP. Neither does it provide the resolution the OP found (if any). Hence the reason I addressed the OP in the reply.

OP (or anyone able to supply some information here about the OP's solution, or help in some other way), thanks for your help.

 

[bogmali]

I am aware of the date of the thread. The problem and it's solution are still relevant in the year 2021 (as indicated by the fact that I am replying to it).

Although I agree with your above sentiments, I would have to suggest that starting your own thread is best going forward, closing this.