Default animation.convert_args to ["-layers", "OptimizePlus"]. #23371
Conversation
|
This can be rebased on main now that #23370 is merged. |
On the simple_anim.py example, adding e.g.
```
plt.rcParams["animation.writer"] = "imagemagick"
plt.rcParams["animation.convert_args"] = []
ani.save("old.gif")
plt.rcParams["animation.convert_args"] = ["-layers", "OptimizePlus"]
ani.save("new.gif")
```
shows that OptimizePlus yields a >2.5x decrease in gif size.
|
rebased. |
|
Do we have a minimum supported ImageMagick version? When was this added? |
|
I don't know when it was added exactly, but https://legacy.imagemagick.org/Usage/anim_opt/ dates back to 2007 and already mentions it. |
|
I'm not really convinced that we should change the defaults like this, but rather users should be choosing these sorts of things depending on their application. This is presumably lossy, and I would guess the lossy-ness depends on the animation. |
|
@jklymak Reading the docs linked above, I see limited mention of lossy transformations--and those cases are explicitly mentioned (LZW lossy) as unsupported. The rest of our animations in h264/mp4 are already using lossy compression, so I don't see it as an issue regardless. We should be providing defaults that make the most sense to the majority of our users. IMO, that means providing out-of-the-box functionality to produce GIF animations that have a good size-quality (image and time to write) tradeoff. So I'm fine with having these enabled since there's an escape hatch. |
On the simple_anim.py example, adding e.g.
shows that OptimizePlus yields a >2.5x decrease in gif size. (Admittedly I haven't tried more examples.)
Goes on top of #23370. Supersedes #15784. Thanks to @erelson for suggesting to use
-layers.PR Summary
PR Checklist
Tests and Styling
pytestpasses).flake8-docstringsand runflake8 --docstring-convention=all).Documentation
doc/users/next_whats_new/(follow instructions in README.rst there).doc/api/next_api_changes/(follow instructions in README.rst there).