Hi @anoano I'm wondering if your question has been answered. If so, please mark this thread as solved. Cheers, Maxime.

Hello. I have solved it by using SpecialEventAdd and sending a message to dialog's CoreMessage. Thank you.

Hi Petros, in order to have the MATPREVIEW_PREPARE_SCENE being dispatched, you've to set MatPreviewObjectInfo::bNeedsOwnScene to true. At the same time also MatPreviewObjectInfo::bHandlePreview needs to be set to true when the host object is responsible for the rendering. Finally I think that looking at MATPREVIEW enums can be helpful as well. Best, Riccardo

Sorry for the lateness in replying to this. I was busy creating lots of python versions of my COFFEE scripts, for R20. I will now try to start porting my C++ plugins to R20. Thank you for all the links.

Hey Maxime, thank you very much for the detailed explanation. Makes absolute sense now! It works perfectly now. I tried a similar approach, but I was adding the Undo in g_MakeInstanceEditable() per object - that was the main culprit. Cheers, Robert

In your specific example, it could look like this: static maxon::Result<NodeData*> Alloc() { iferr_scope; return NewObj(SoloPrefs) iferr_return; } And when you construct your NodeDataclass, you have to access the actual value of maxon::Result: SoloPrefs::Alloc().GetValue(); Note, if that allocation fails that you can access the error with GetError(). There have changed lots of things, Sebastian already posted the links to the relevant documentation.

Thanks Riccardo, I think we reached the bottom of this issue. I just finished implementing a workaround, getting the tag associated with the shader, it's polygon, points and uvw tag, then paint each face with the average vertex value in a bitmap using GeClipMat. Working really good! Best, Roger

Hi @Hexbob6 I'm wondering if your question has been answered. If so, please mark this thread as solved. Cheers, Maxime.

@m_adam Cool. Thank-you for your time and patience.

Hi @mikeudin, I'm wondering if your question has been answered. If so, please mark this thread as solved. Cheers, Yannick

Thanks Andreas. I will try Rick's trick again today. When I looked into it yesterday it didn't seem to work, but now I am wondering if perhaps it was visual studio locking the folders, and not actually C4D, since I was debugging at the time. I just tested renaming files manually (without running it through VS) and it was possible. I think Ricks dialog was done in Python, so he wouldn't have hit this VS issue. Fingers crossed it all works today.

Thanks very much. (There is so much information to be read!). -Pim

Hi, just want to ask, if can consider this topic as solved. Bye, Andreas

Hi @Swinn are you sure in your script you define in line 19/20 settings[c4d.VOLUMETOMESHSETTINGS_ISO] = -5.51 settings[c4d.VOLUMETOMESHSETTINGS_ADAPTIVE] = 0.0 With the same value, you got in your volume mesher object? The voxel range threshold can't be directly passed as an argument, but you can expand this parameter in the volume mesher and then you get the Use Absolute Value (ISO) check box, which let you choose an ISO value. Cheers, Maxime.

@s_bach sorry, forgot about the tags. I'm using C++ SDK on R19/R20. I try calling ExecutePasses(), thanks.

@m_adam said in R20 Startup script location - python_init.py: This will be fixed within a future release. Well, that's a relief then. Good to know. Thank you Maxime for the clarification. Kind regards, Tom

@m_adam Thanks Maxime, I'll check that out. The customer was using R23.110

Great, thank you.

That's it! Thank you so much, mp5gosu! Cheers, Mark.