When I come back from a shoot, I normally run my images through darktable-generate-cache before culling/editing, since I like the absence of a delay when I later browse through all the thumbnails and previews in darktable. I often use full screen previews, so I set --max-mip 8 (i.e. the maximum value) when running darktable-generate-cache. My question is, what is the relationship between mip size and thumbnail size? I’ve been using the largest value for max-mip since I want large previews, but not understanding this relationship I’m not sure if that’s necessary.
Go to RFC: Add larger 8-bit mipmaps to handle larger displays and higher resolution cameras · Issue #20178 · darktable-org/darktable · GitHub and there is a section “current” that shows the mipmap sizes.
On my 4K screen I use 3 for lighttable, and 6 for the whole image on the screen. 8 is the full size image resolution
My 2K screen was 2 and 5 and my HD screen was 2 and 4 (IIRC).
Thanks! That’s exactly what I was looking for. It would be nice if that info was more easily accessible than in a GitHub thread, such as in the darktable-generate-cache man page or the darktable manual.
The other place it’s available is buried in the source code 
.
You could raise an issue at GitHub - darktable-org/dtdocs: darktable user manual and ask to have it added to the manual. You should be able to copy the table from the github thread and paste it into the issue so that someone could take it from the issue to the manual.