added properties
update inspector
UX improvements
and minor optimizations
deg_to_rad
moved methods
move variables
performance and cleanup
Faster
Compare the angles/smooth of faces instead of vertices, skipping for loop.
Using a LocalVector instead of Vector, this simplifies the code while making it faster, and I don't think we need COW.
all done and working
Optimizations and cleanup
Clarifying and expanding the description of autosmooth and smoothing_angle properties.
Optimizations to skip steps in the loops:
Split loops to be able to skip vertices better and replace division with multiplication.
Suggestions for future changes.
Skip loop if smoothing_angle is lower than 0.1, as a more performant alternative to default smoothing.
Do not compare vertices that belong to the same triangle.
Remove has_smooth. Use a check at the start of the loop instead. This allows us to skip faces.
Perform normalize() at the end of the loop as we don't need the normal of the vertex afterwards, as we use the normal of the face for smoothing. This way we skip normalizing flat faces.
If one vertex of a triangle is connected to the current vertex, the other 2 can never physically connect, so we skip them (break;). This isn't as reliable but it could add a slight performance increase.
I calculate a decrease of at least 80% of calculations in the for loops compared to the previous code.
Co-Authored-By: A Thousand Ships <96648715+AThousandShips@users.noreply.github.com>
This allows removing it from `class_db.h`, significantly reducing the amount of files
that include it transitively.
Also includes some include cleanup in `control.h` and `rich_text_label.h` done while
ensure they don't depend on `callable_mp`.
This will allow decoupling `display_server.h` from a number of headers in the
codebase which only require those enums and not all the DisplayServer API.
A number of headers in the codebase included `rendering_server.h` just for
some enum definitions. This means that any change to `rendering_server.h` or
one of its dependencies would trigger a massive incremental rebuild.
With this change, we decouple a number of classes from `rendering_server.h`,
greatly speeding up incremental rebuilds for that area.
On my machine, this reduces incremental compilation time after an edit of
`rendering_server.h` by 60s (from 2m57s).
