Fill in text about performing automatic composition
Manishearth opened this issue · comments
https://immersive-web.github.io/webxr-ar-module/#automatic-composition
We should basically mention that the device will automatically compose the rendered scene on top of the real-world background as perceived by its cameras, alpha blending the two layers on top of each other.
We should also mention that automatic composition is not responsible for dealing with occlusion: the rendered scene will always compose on top of the camera background.
Sounds like a good start. We probably need to call out that this will be different on additive light displays. And explicitly state that developers do not get camera access in either case. WDYT?
Ah, right!
To be clear, do we consider what happens on additive light displays to be "automatic composition" as well? I can tweak the wording to cover both cases.
Personally, I've been considering it to be "implied" automatic composition ^_^
We should be cautious here though. In a world where additive light displays add support for per-pixel dimming... well, we'll probably need to adjust our environment blend mode options, but if possible it'd be nice to have the composition text be flexible enough to handle it without changes.
I'll try to make that work!