Hi Wes,
Thanks for the response.
Yeah that's what I've got, but as you can see in that example, the alpha is written both into the layer mask AND the layer opacity. So if the user opens that psd and tries to fill the layer mask with white, it can't reveal a whole opaque layer because the layer opacity is cutting off the transparent pixels.
Do you see what I'm getting at? It's the same problem loading a png with transparency into Photoshop and attempting to restore the transparent pixels. You can't, they don't exist.
Thanks
I was just going to write a huge post about the PSD exporter, and this one was definitely one of my bigger feature requests.
I couldn't agree more.
It's ok to eventually use a Channel Mixer to read the alpha as a mask, but pixels in the image become transparent too! The layer should be opaque, with the alpha stored in the mask.
I think it would be MUCH better to read the mask from arbitrary grayscale nodes, as yes, the layer could contain transparent pixels, but they don't need to match the mask.
I hope that doesn't sound confusing, but basically it should be like this:
[LAYER] [MASK]
| |
RGBA node gray node
The layer should be a RGBA node as it may contain transparent pixels, while the mask should just be grayscale node.
Sorry for the ASCII art
