Gary Oberbrunner

I'm having the same issue, just FYI. If there's a workaround or suggestion for where to look to start fixing it, I'd be willing to help.

Yes, you're right -- I have a pending PR to fix this at least partially. I'm not sure it uses prepublish; I'll have to check.

Interesting. The original code deleted both the structure and the view; that change switched to disposing them both. I expect your fix won't cause any problems directly, but it would...

Hi 0gust1; welcome! Contributions of time and code are always appreciated. You could look into some of the open issues, or find things that you think could be improved.

I did try moving some of this into the trackball code, but since my main application for this is animated motion tracks for the molecules, having it there makes that...

The sample encoders/decoders on the page you linked, @ChrisMoth , look like they parse ints just fine as well as the extended formats, so they could be drop-in replacements for...

Thanks -- DX looks pretty simple and there's a python/numpy parser/writer for it too. That said, here's a test DX file I created that doesn't import very well in NGL....

This took me ages to get to, but here's a demo of color-managed NGL viewer: http://oberbrunner.com/ngl/examples/webapp.html The main difference you should see is that the current method darkens the darks...

My changes are on my feature/color-management branch. Diffs here: https://github.com/nglviewer/ngl/compare/master...garyo:feature/color-management

Yes, you're right about some precision loss in the darks. It doesn't seem bad to me, especially since we're just converting & storing the colormaker values, not entire textures. Also...