Doug Kerr
Well-known member
This is today's entry in the "Kerr's least useful Photoshop fact" contest.
*************
In Photoshop, a light gray checkerboard pattern (that's the default - we can redesign it if we wish) is said to show "transparent pixels". That's actually accurate, but in a way that may be a little more complicated than we recognize. In some situations, it may be a little difficult to figure out what the thing we see means (especially when we see a combination of a faded checkerboard and a faded actual color, and in multi-layer situations with masks having semitransparency and various blending modes).
There's another, completely-compatible outlook on the meaning of the checkerboard that can help us unscramble some of the difficult situations. (Experienced Photostat wallopers will of course be able to grasp as much as they need to know intuitively, but we telephone engineers sometimes need to have a block diagram.)
I'll lead up to it.
If we have a file with two or more image layers, we see the checkerboard where the composite image is transparent. That's hardly strange; we see red where the composite image is red.
If we look at the mechanism, we realize that the composite image is transparent only where every one of the component layers is transparent. (We will assume Normal blend mode throughout.)
But we can look at what happens a different way. Lets imagine we have two layers, Upper and Lower. (By the way, no masks are involved here.)
Where the Upper layer content is transparent, we "see right through it". What do we see? The content of the Lower layer there.
Suppose that in all or part of that region, the Lower layer content is also transparent. What happens there? Again, we "see right through it". What do we see now?
There being no further layer below, we see right through into the "great empty beyond"?
What does that look like? Well, not like anything. But Photoshop helps us out by making the "great empty beyond" look like a gray checkerboard (on screen).
Said another way, it is as if, for visual purposes only (that is, with respect to what we see on the screen), Photoshop has placed below the lowest layer of the stack a ground sheet covered with the gray checkerboard.
And in this outlook, that's what we really see, not the "transparent-ness" of the pixels on either of the layers as such.
Now, suppose we have only a single layer, with a colored square in its center and the remainder transparent. We set the layer opacity to 40%. What do we see?
In the center, a faded colored square "mixed with" a faded checkerboard; around that, a checkerboard.
How can we explain this in terms of our "great empty beyond" metaphor? In the center, we "40%" see the colored pixels, but we "60%" see through them. What do we see? The great empty beyond, which looks like a checkerboard. So we see the blend of those "two things seen".
Suppose we save this setup as a JPG file? Then, Photoshop makes the "great empty beyond" look white. If we examine that file in our Photoshop, we see a faded colored square (it has a different color than as we drew it) surrounded by white. Nothing is transparent or semitransparent anymore.
Supposed we save it as a GIF file, with transparency preserved? Then, Photoshop makes the colored area semitransparent (alpha=0.4)) and makes the surround transparent (alpha=0).
Now if we examine this GIF file in our Photoshop, what do we see? Well, it looks just like the file did before. Lets talk about the surround, which appears as a gray checkerboard. We can look at the significance of what we see in two different ways:
a. The pixels there are transparent, and Photoshop signals that to us with the gray checkerboard.
b. The pixels there are transparent; we see right though the into the "great empty beyond". Photoshop makes that look to us like a gray checkerboard.
These are both correct.
Now in this case, we would probably have no reason to think of this, or to describe it, other than per (a).
But in complicated situations, the other outlook can help us figure out what has happened.
Best regards,
Doug
*************
In Photoshop, a light gray checkerboard pattern (that's the default - we can redesign it if we wish) is said to show "transparent pixels". That's actually accurate, but in a way that may be a little more complicated than we recognize. In some situations, it may be a little difficult to figure out what the thing we see means (especially when we see a combination of a faded checkerboard and a faded actual color, and in multi-layer situations with masks having semitransparency and various blending modes).
There's another, completely-compatible outlook on the meaning of the checkerboard that can help us unscramble some of the difficult situations. (Experienced Photostat wallopers will of course be able to grasp as much as they need to know intuitively, but we telephone engineers sometimes need to have a block diagram.)
I'll lead up to it.
If we have a file with two or more image layers, we see the checkerboard where the composite image is transparent. That's hardly strange; we see red where the composite image is red.
If we look at the mechanism, we realize that the composite image is transparent only where every one of the component layers is transparent. (We will assume Normal blend mode throughout.)
But we can look at what happens a different way. Lets imagine we have two layers, Upper and Lower. (By the way, no masks are involved here.)
Where the Upper layer content is transparent, we "see right through it". What do we see? The content of the Lower layer there.
Suppose that in all or part of that region, the Lower layer content is also transparent. What happens there? Again, we "see right through it". What do we see now?
There being no further layer below, we see right through into the "great empty beyond"?
What does that look like? Well, not like anything. But Photoshop helps us out by making the "great empty beyond" look like a gray checkerboard (on screen).
Said another way, it is as if, for visual purposes only (that is, with respect to what we see on the screen), Photoshop has placed below the lowest layer of the stack a ground sheet covered with the gray checkerboard.
And in this outlook, that's what we really see, not the "transparent-ness" of the pixels on either of the layers as such.
Now, suppose we have only a single layer, with a colored square in its center and the remainder transparent. We set the layer opacity to 40%. What do we see?
In the center, a faded colored square "mixed with" a faded checkerboard; around that, a checkerboard.
How can we explain this in terms of our "great empty beyond" metaphor? In the center, we "40%" see the colored pixels, but we "60%" see through them. What do we see? The great empty beyond, which looks like a checkerboard. So we see the blend of those "two things seen".
Suppose we save this setup as a JPG file? Then, Photoshop makes the "great empty beyond" look white. If we examine that file in our Photoshop, we see a faded colored square (it has a different color than as we drew it) surrounded by white. Nothing is transparent or semitransparent anymore.
Supposed we save it as a GIF file, with transparency preserved? Then, Photoshop makes the colored area semitransparent (alpha=0.4)) and makes the surround transparent (alpha=0).
Now if we examine this GIF file in our Photoshop, what do we see? Well, it looks just like the file did before. Lets talk about the surround, which appears as a gray checkerboard. We can look at the significance of what we see in two different ways:
a. The pixels there are transparent, and Photoshop signals that to us with the gray checkerboard.
b. The pixels there are transparent; we see right though the into the "great empty beyond". Photoshop makes that look to us like a gray checkerboard.
These are both correct.
Now in this case, we would probably have no reason to think of this, or to describe it, other than per (a).
But in complicated situations, the other outlook can help us figure out what has happened.
Best regards,
Doug
