This is not a bug.

Hi Omar, please, can you close this bug report? Thank you.

Here on
Operating system: Windows-10-10.0.19041-SP0 64 Bits
Graphics card: NVIDIA GeForce GTX 1070/PCIe/SSE2 NVIDIA Corporation 4.5.0 NVIDIA 512.59

It does happen with version 3.1 too but, please, note that in the __init__.py file of the CloudRig feature set it states clearly that the minimum compatible version of Blender is 3.2, so I wouldn't expect it to work with 3.1 version. Anyway, the error messages are the same.

Refractive index of a transparent medium is always greater than 1 (it cannot be less than 1) because speed of light in any medium is always less than that in vacuum. IOR 1,00028 (approx) which is for Air should be the lowest value to be used.

Is this really a bug, or simply a misuse of this parameter in the Principled BSDF, being the Specular value a trick to play with the IOR within plausible values for dielectric materials? The Pixar paper on Principled BSDF, v3, states this for the Specular parameter: The specular parameter may be pushed beyond one to reach higher IOR values but should be done with caution.
A value of 12.5 seems far from being plausibile and within cautious values. Am I wrong?

I cannot tell you if this is the cause of the CUDA synchronization errors, because it's a quite dated solution to CUDA TDR error, and I usually set and forget it. This is where I found a solution the first time I got this problem: http://artificialflight.org/blog/2013/cycles-crash-cuda-tdr-error/

Reducing the tile size might have an impact on the memory used at render time, but I doubt it might "fix" the TDR warning/issue. Have you tried? Did it make any difference? Possibly, a developer might provide you with a more detailed answer, if needed. I still recommend to follow the given instructions and make some tests.

Sometimes, it happens that when new NVIDIA drivers are installed the registry key where the TDR variable is set (if set) it is being reset to default values, and that variable is deleted.
You have to follow the instructions given by Robert about to the TDR setup in the Windows registry, as described at the given link.
This doesn't seam to be a Blender software bug.

Ivy Generator places the ivy where the cursor is, so you first have to select an object and place the 3D cursor where you wish the Ivy to be created.

Have you tried with the most current version 2.93.8? https://www.blender.org/download/lts/2-93/
There have been so many bug fixes that you might give it a try.

Just a brief update. While this issue is still present in 2.93.7 LTS (and possibly 2.93.8), it's no longer happening in version 3.0.1, nor in 3.1 Alpha, and neither in 3.2 Alpha.

Here on:
Operating system: Windows-10-10.0.19041-SP0 64 Bits
Graphics card: NVIDIA GeForce GTX 1070/PCIe/SSE2 NVIDIA Corporation 4.5.0 NVIDIA 472.47
Blender version: 2.93.5, branch: master, commit date: 2021-10-05 12:04, hash: rBa791bdabd0b2

Here on:

I see what do you mean, and I double checked that behaviour (it works as expected only when the mesh is over or below the cursor) in all previous version of Blender and it works that way back to version 2.79b, so it seems to be working that way "by design" or it has never corrected if this is a bug.

You should carefully read the manual and its instructions. The operator is view dependent and if you are NOT in Ortographic view but, instead, in Perspective view, like in the first screenshot you posted, you get a different result. This is not a bug.

Here on:
System Information
Operating system: Windows-10-10.0.19041-SP0 64 Bits
Graphics card: GeForce GTX 1070/PCIe/SSE2 NVIDIA Corporation 4.5.0 NVIDIA 462.59

Brilliant! I tested your patched script and it works fine with all these versions of Blender, without breaking the relative paths of linked objects and materials:

Ciao Antonio,

Hi Jason, thank you very much for having a look at this. I look forward to read about your discoveries on this add-on's behaviour.

Hello devs. I've noticed a recent update to this useful add-on (ea945ab58d85). Are there any plans to tackle this quite annoying problem with linked libraries path being messed up? I would like to use it in production but I am not confident to try it in this state. Thank you for your attention.

The add-on is made only for the Principled BSDF shader node, nothing else. If you run it and 1) you haven't selected anything or 2) you have selected a node which is NOT the Principled BSDF, the add-on will show a message on the screen or in the bottom bar that you HAVE to select a Principled BSDF node for the add-on to work.

I put your meriva.blend file inside a project folder, added a \textures folder inside it and copied there the image textures from your reference. The Node Wrangler add-on worked fine adding the set of images to both materials' Principled BSDF shader node. So, the issue is not caused by the images. I tested your file (created in Blender 2.83) in Blender 2.83.13, as well as in Blender 2.92.0 stable release, with the same results.

I don't know how to tell you without hurting your nerves, but even in Blender 2.92.0 the Node Wrangler works fine here. So, I suggest you to open the Blender Console from the menu Window | Toggle System Console and watch there for any error or message related to the Node Wrangler add-on.

Here on:
Operating system: Windows-10-10.0.18362-SP0 64 Bits
Graphics card: GeForce GTX 1070/PCIe/SSE2 NVIDIA Corporation 4.5.0 NVIDIA 462.31

The folder name has two underscores before "pycache" and two other underscores after it. I don't know why they are being stripped off the text. Anyway, you cannot see these directories possibly because you've just installed Blender but not yet run it. Run Blender and look inside its directory structure, by example inside the scripts folders, and you'll see those Python cache folders. That is what I was talking about. I use a file browser with recursive search capabilities to search for those files.

Here on:

Hi, just for my curiosity, why you didn't change the default value of the MAX_INPUT_NUMBER variable at line 43, instead of that single variable at line 252? Thank you.

Here on

Thank you for the explanation.

Hello, just for my curiosity, isn't Blender 2.92 as well as 2.93 based on Python 3.7.7? Why are you compiling for Python 10, not yet officially supported for the Blender development? I read there are plans to move to Python 3.9 but that's all I know about. I repeat it, just for my curiosity. Thank you.

Here on

Sorry, but version 2.90.1 has reached end-of-life and the currently supported version is 2.91.2. Please, download the latter and verify if the issue is still there: https://www.blender.org/download/

Hi Paul, I simply verified that your issue was replicable with and without your sample files. At this point it's better to leave this into the good hands of the developers for further investigations.

Here on

Thanks for the report, but this add-on is not developed by Blender, please report the bug to the original author.

How is it that the versioning has switched directly to 2.91.2? This version numbering is both in the file names (https://download.blender.org/release/Blender2.91/) and in Blender itself (as shown in the splash screen). Probably, this was intentional but, just in case, I let you notice it. Thanks.

I confirm that after applying the above mentioned correction, changing from dummy.select to dummy.select_get(), the add-on works perfectly in Blender versions 2.83.10, 2.91.0 and 2.92.0(alpha). Thank you.

This is not a bug. Some zipped add-ons are not properly set for direct installation from the zip file. If the add-on is just a single xxxx.py file, it should be packed singularly. In the contrary, if it is made of more that a single python script, these should be placed inside a dedicated folder, and the main script should be named

__init__.py

(with two underscores before and after init).
If you check the content of the zip file you attached, the "singular file script" is placed inside a folder but there is no "init__.py" and Blender don't know how to handle it after you install it. You have two solutions: 1) extract the single xxxx.py file somewhere, then run the installation procedure in Blender, choosing this single python script; 2) go to your add-ons folder and look for the CircularArray-master directory that has been created when you installed the add-on; inside it you will find the CircularArray.py script, rename it to init__.py. These steps should solve your problem.

Here on

This is perfectly normal and not a bug. You have to add both a Domain and a Smoke Emitter object. If you want to test the smoke effect you can use the Quick Effects option available from the Object menu, after adding and selecting the smoke emitter object in your scene.

So, is this really a "bug"? Anyway, I do agree with you that a check for validation, and a proper warning message displayed to the user, would be a more user friendly behaviour for the add-on.

The issue is that you shouldn't change topology of the mesh deleting single vertices. If you, instead, delete an entire row of vertices, it works when you click on Landscape Eroder. This applies to all current versions: 2.83.8, 2.90.1, 2.91.0, and 2.92.0.

Hello. Here on

Please, write a proper bug report explaining what's wrong, provide required information, and step by step instruction on how to reproduce the problem, to the developers. Please, watch this video instructions by Pablo Vazquez: https://www.youtube.com/watch?v=JTD0OJq_rF4

Possibly because if that path is pointing to a different drive than the one where Blender is installed, you cannot use relative paths but only absolute.

Here on

Wow! "Bug" fixed in 14 minutes. You're amazing. Grazie!

The above behaviour happens also on Windows 10, and I can confirm that the second time you try to activate the add-on the \matlib folder is created again and everything go on smoothly.

When the add-on is activated in the Preferences, and if you haven't previously set a path in the add-on's preferences, a new folder will be created inside the Blender's root installation directory, named \matlib, and new libraries will be added there. That might be a reason for the user not being able to create a new library, because probably he has no rights to do it. The user should try to set a path for the materials folder outside of the Windows programs directory and check if he has rights to write inside it. The following is a screenshot from my portable 2.83LTS version where as soon as I activated the add-on the folder \matlib was created and new libraries are stored there.

I was reviewing this bug ticket and noticed that:

I confirm the above mentioned behaviour in 2.90 here on

Yes @Leroy (Leroy), the last images I published were taken from the 3D Viewport, not a render.

I did a clean install (portable) of version 2.83.0 stable, vanilla, no external add-ons (just in case they might interfere), and I confirm that here the issue raises in the 3D Viewport with Standard (formerly, sRGB) setup for Color Management's View Transform, as per screenshots below:

Please check that Filmic is set correctly in the Color Management section of the Render Properties. To me they appear to work fine both in Eevee and Cycles, both in 2.82a and 2.83.0. I can get your second image's result only if I set the Color Management to sRGB. Hope this helps.

I can confirm the issue.

I've tested Blender 2.82a official release on Windows 10 64bit (portable version), which is installed out of the Windows control on a separate hard disk drive, and I've been able to create a new library (see the new .blend file inside the add-on's folder in the screenshot below), to add/remove materials, categories and assign categories to materials. So, might your problem be caused by insufficient user rights on the Blender installation folder (and therefore the add-ons folders)? Try to use a portable version and unpack it outside of the Windows Programs directory.

Please follow the submission template and guidelines and make a complete, valid bug report, with required info, precise description of the issue, precise steps to reproduce it, small and simple .blend and/or other files to do so if needed, etc.

Here on :

Great! Thank you. ?

This issue seems related to T67111. Perhaps they could be merged.

So, this is not a bug therefore this ticket could be closed.

Hi @Brendon Murphy (meta-androcto), I went through the samples_scene.py, trying to fix the errors and commented the original lines, and now it seems to work in the latest RC2. Here is the modified script, hope it helps.