I've found that if I set the environment varriable `CA_LAYER_SURFACE=0` then my core animation app gets better fps. My application is scroll heavy, and scrolling involves adding/removing many CALayers. The ftp difference that I see is between Quartz Debug reporting 55+ fps to 60fps.
I discovered this key through this post: http://jwilling.com/blog/debugging-core-animation-on-osx/
It mentions that the key switches between in process vrs out of process rendering. I'm wondering if there's more info on that key. Or more important tradeoffs that I should know about. For my app setting to CA_LAYER_SURFACE=0 really helps avoid judder/jank and so I'm planning to include this setting in my LSEnvironment key. But I'm wondering why the seeminly faster render path isn't the default?
Thanks,
Jesse
