I just updated the diff so it works again. But there’s still the same work left to do to make it part of Blender: https://developer.blender.org/D5908
@mathers it’s yours if you want to finish it. Otherwise, I’m no C dev so don’t expect too much, I’ll just try to add the missing pieces with Richard’s guidance if nobody else wants to finish it
It might be interesting to have some users take a snapshot of their VSE setup, and then mock up how they would like the thumbnails to appear over their snapshot image. By comparing the different views ( how wide the thumbnails are, does the text info appear above or overlaid on the image, how thin are the strips, how does the end image appear ), similarities can be used as an end result, and differences can potentially be user options etc.
Maybe even post examples of how other video editors do it well / do it not so well
Sure, if you want to share mockups, that’d be welcome. There will be some technical constraints: memory allocated to thumbnails, background rendering… to start with, I’m going to put 1 thumbnail after every cut, get that working with background rendering and caching, then we can look into improving the system.
Just checking in. Is this on the roadmap, or is there even a feature request for it? This seems like a very important feature for vse. I know it is for me. VSE is very hard to use without it.
To expand on my use-case a bit. I work with still image montages, where i’m loading a lot of images into vse, and synchronising their display with an audio track. For a regular movie clip sliding along the strip and watching the preview window is almost sufficient, but for a montage I’m working with a hundred or so colored squares with no identification.
I have worked on this feature some time ago, I remember that I had to duplicate some code and I wasn’t too happy with that. But main problem was that priorities have shifted.
Right now this isn’t on roadmap in official capacity, but it is something I would like to have in VSE.
I will re-check patch to see how much time it would take to finish basic functionality at least.
Currently I work mainly on major code restructuring and longer term design task T78986 so I don’t want to take on another major project