Comp Alpha Over Counter Intuitive?

Support Compositing and Post Processing
#1

Hello,

I’m not new to compositing, but I am new to Blender’s compositor. I was doing a tutorial, and reached a point where I had to put one element over another using the Apha Over node. Then oddly enough, I was told to put the above element under the bottom element in the Alpha Over node.

Immediately, this seemed counter intuitive. Why would my compositing tool visually want me to put something that’s going over… under? I’m sure there’s a mathematical reason for it, but again… to me… when you put something on top… it should go on top. A user shouldn’t have to do mental gymnastics around the software just to get a job done.

In the very least, can someone help set me straight by explaining the logic behind this work flow?

Thank you,
M

#2

Hello,

I’m not new to compositing, but I am new to Blender’s compositor. I was doing a tutorial, and reached a point where I had to put one element over another using the Apha Over node. Then oddly enough, I was told to put the above element under the bottom element in the Alpha Over node.

Immediately, this seemed counter intuitive. Why would my compositing tool visually want me to put something that’s going over… under? I’m sure there’s a mathematical reason for it, but again… to me… when you put something on top… it should go on top. A user shouldn’t have to do mental gymnastics around the software just to get a job done.

In the very least, can someone help set me straight by explaining the logic behind this work flow?

Thank you,
M

#3

No logic, it just is this way. Only reason I can think for this is that for node trees built left to right, the trunk is most logically left on top, so additional elements are added from bottom. It conforms with the logic of reading top down.

Personally I prefer top-bottom trees and nodes that don’t have fixes socket locations.

#4

It is consistent with other operations in blender like the modifier stack, they are applied in order from top to bottom

For alpha over you add the first component (top input) then you add the second component (bottom input). The second component will of course be on top of the first component

#5

This is a convention in blender that you can find nearly everywhere. Indeed it can be quite confusing.
In blender internal texture stack work that way (very confusing) , Modifiers and constraints are also evaluated from top to bottom, but I found that less confusing.
In the VSE layers are stacked as we expect.
At least it’s quite consistent, and maybe that would be harder if it wasn’t .

Why stay like this ? I guess it’s because of backward compatibility of blend files and tutorials , I haven’t heard about an official answer.