Blend file reduced to just the floor, a table and a light. World HDRI removed. Still has same issue.

Thanks for the investigation. This makes sense.

Have you tried following any of the Blender tablet setup videos on youtube to see if they help? I remember watching these two previously...
How to Set Up a Graphics Tablet | Blender 2.8 & 2.79
Blender Tablet Settings for Sculpting
If they don't help, what tablet have you got?

@Peter Goodman (badbunny) Thanks for instructions, I can reproduce the issue.

is inside the BLEND file.

To keep you in touch, I did 4 graphic card test :

• my rtx 3070 : crash
• an rtx 3060 : don't crash
• an gtx 960 : don't crash
• an rtx 3070 + gtx 960 : don't crash if i use the gtx 960 to run the opengl graphic engine

I don't think that outdated video is big concern as many concepts would be probably still valid.

Can't reproduce on
Operating system: Windows-10-10.0.19041-SP0 64 Bits
Graphics card: NVIDIA GeForce RTX 3060/PCIe/SSE2 NVIDIA Corporation 4.5.0 NVIDIA 511.65
Tested 3.1.2 and 3.3

Operating system: Windows 10 Enterprise (10.0.19042 Build 19042)
Graphics card: Nvidia Quadro RTX 4000 (driver 472.88)
Tested version: 3.1.0

@Denis (GMXArtist) I don't see anything incorrect in your file. Seems like a bug really.

Please click on Help > Save System Info and upload saved file.

I think it would be best to edit report with sweep file and description by @Zijun Zhou (Eary) then. I am not sure if difference is intentional though.

In T98417#1365037, @Richard Antalik (ISS) wrote:

Please upload .blend file

I don't know when this happened. I setup the nodes in the Compositing tab. Switched back to the Layout tab. Then, start rendering with Cycles. I did not wait until the rendering is done and stopped the rendering at the Layout tab because I wanted to see the effect. Then, I switched back to the Compositing tab and realized that the connection lines were missing. I knew that the nodes were not disconnected because I saw the result. I have tried few times but I cannot reproduce the issue now.
From the messages at the console, I found the line 729 at https://developer.blender.org/diffusion/B/browse/master/intern/cycles/blender/display_driver.cpp but I have no clue why glMapBuffer(GL_PIXEL_UNPACK_BUFFER, GL_WRITE_ONLY) returned a NULL pointer in this case. From https://www.khronos.org/registry/OpenGL-Refpages/gl4/html/glMapBuffer.xhtml, which states "If an error is generated, a NULL pointer is returned." Based on the description of the "Errors" section of that page. But without the error code, I suspected there are 2 possible reasons:

Please upload .blend file

Image is same in all languages.

Can you upload sample .blend file?

@Dixon osecha (Dxn) can you upload .vdb file as well?

there's a problem with sorting like this though, this changes the triangle order, and when I tried it with custom normals, bad things happended ...
so maybe I need to use sorted values for eqaulity and hash, but keep the original values unsorted, which mean I need to store double ammount of memory for each triangle

@Sérgio Merêces (smereces) I see very large amount of addons in blender_system_info.txt I would guess problem could be in one of them.
Download latest alpha build from https://builder.blender.org/download/daily/, run file blender_factory_startup.cmd and check if crash happens.

Expand the "metallic_sep" node.

Well I had a closer look and the duplicate code fragments are a known issue. I didn't notice that they're inside an ifndef. I'll keep looking into this but I'll wait a while longer next time before jumping to wrong conclusions.

If your GPU usage before rendering is 5.6GB than this would fail as rendering process takes around 4GB.

In that case I can't reproduce this issue on:
Operating system: Windows-10-10.0.19041-SP0 64 Bits
Graphics card: NVIDIA GeForce RTX 3060/PCIe/SSE2 NVIDIA Corporation 4.5.0 NVIDIA 511.65

@Andrew Walsh (andywalshart) Sure feel free to share data when you can. Just a note that policy on tracker is that reports should be no longer than week without activity, so this may be closed, in which case feel free to open new report. Or we can close now if you think this will take some time.

But I can reproduce difference in color strip animation when scene strip is enabled/disabled, so I guess this would be enough to investigate.

My apologies for the edits. I wanted to present all details of information on this.
Note the photo of the Material nodes is for the Fuselage material, so you can see what all is in it. One is a closeup. Same material, close up and far away.

@forceengine Yes I can check on Linux, but not really debug at least easily. But since you report different OS to be working correctly which implies different settings, can you check if you have disk cache enabled on Linux installation and if issue will happen when you disable it?

In T97910#1364098, @Omar Emara (OmarSquircleArt) wrote:

Just a poke regarding my inquiry above, otherwise we will have to archive this for now.

Ah right that's correct. Not sure what would be good solution. Sounds to me that in case when property is animated, cache entry should be created only when keyframe is inserted, but even that is not quite enough - it would have to check if all properties that were changed were keyed. Looks quite complicated to handle this.

In T98405#1364943, @tempdevnova (tempdevnova) wrote:

it returned gray for a Blue value of 40 which I think definetely is way to high to be middle gray.