Re: [fxtf-drafts] Luminance masking is slower than alpha masking

@thebabydino I don't think `mask-mode: luminance` will help in the way you are thinking, but `mask-composite` should.

@jrmuizel 
In SVG masking, all the shapes inside the `<mask>` are composited together using normal painting operations, before converting to a mask.  That means that you can add a black shape on top of a white shape to create a donut effect (among many possibilities).  You obviously couldn't do that with a transparent shape on top of an opaque shape.  

But with `mask-composite` and mask layers, you would use two solid shapes (images or CSS gradient layers) and a subtract or exclude composite mode.  Of course, in order to support `mask-composite`, all mask layers are converted to alpha masks before compositing, so there is no difference between luminance and alpha mask layers as far as what types of compositing are possible with multiple mask-image layers.

So once `mask-composite` is fully supported, the main benefit of luminance masking within CSS masking layers is that you can use existing image files, and that it is consistent with what designers expect.  As Dirk said, mask layers in GIMP and Photoshop and probably other tools use white/black mask layers.  However, those tools don't yet export to CSS syntax (since there isn't the browser support).  They either export to SVG masks or to semi-transparent PNG images that have the mask and image layers already composited together.



-- 
GitHub Notification of comment by AmeliaBR
Please view or discuss this issue at https://github.com/w3c/fxtf-drafts/issues/137#issuecomment-311856465 using your GitHub account

Received on Thursday, 29 June 2017 04:01:01 UTC