Angles Of View

Now that brightness control can reside once and for all where it belongs, in the projector, screens, particularly the rear projection ones, are at last left alone to the job that they do best. But what is that job? And, just how do they do it? Ignoring profiled screens for a moment, what should you look for in a diffused rear projection screen and how can you make sure you get it? The answers to those questions are a little intricate, but they can be seen easily when we come to look at

Da-Lite Diffusion - Making Light Work

Everybody who thinks about it has a pretty good intuitive grasp of how front projection screens work. Basically, you point a light source towards their surface and, one way or another, the light bounces back in a way that lets you see an image. Maybe some of us in the screen business might want to complicate that definition a little (just a little), but even we couldn't argue with it vigorously.

When we turn around and look at rear projection screens, however, our intuition suddenly becomes a good deal less reliable. This is so because we're a bit less sure about what is being done to the light from our source which turns it into the same recognizable image we were just looking at on the front projection screen.

We do know that the process can't involve reflection — there is, after all, no bounce. So maybe it's that other word, refraction; could that be it? Well, no, actually that isn't it, either. Profiled screens rely on refraction, but diffusion screens typically do not. So let's see now, if it's not reflection and if it's not refraction, the only thing left is what we call scatter. But the only trouble with scatter is that we thought it was a process reserved exclusively to matte white front projection screens. And if we just delete the word "front" from that sentence, we have it exactly right.

What is Matte White? It is, well, a diffuser. And what is a diffuser? It's a collection of tiny particles of a white, chalk-like material which has the property of not absorbing light. Considered singly each of these little particles has a three dimensional shape which is microscopically irregular and chunky. Furthermore, the orientation of each particle (for example, whether its longest dimension is pointing up or down) is generally random.

The way we make a diffuser out of a whole collection of them is to cause them to become distributed more or less evenly out over the surface of a piece of substrate. Now, lets train a microscope at some arbitrarily small area of the surface and see what it reveals. What we'll discover is a coarse distribution of our particles which will appear as though they were unceremoniously dumped out of a wheelbarrow. They will be strewn irregularly over each other and whatever area we're looking at and no one will have raked them out flat.

Since none of them absorb light, all incoming rays will be bounced off them in directions which often will careen them into other surfaces of other particles from which they will again bounce in any and every direction whatsoever. When this process is integrated over a very large collection of particles being struck by a very large collection of light rays which are bounced around through a very large collection of successive angles, you end up with a surface that scatters.

To recall what we mean by scatter, we can restate that all light emanating from the surface of such a screen (luminance) will be distributed in a pattern which is completely independent of the incident angle(s) of that light (illuminance). Thus, it doesn't matter what your viewing angle is to a Matte White screen, you'll always see exactly as much light as every other viewer, from every other viewing angle. Pretty neat, eh?

Here's a test question: If the preceding paragraph is true (and it is), can you see why a Matte White screen must have a gain of 1 and only 1?

Here's another interesting question: Will a rear projection screen with a gain of 1 also exhibit full field symmetric scattering? We'll tell you the answer to this one. No.

To see why, we have only to recognize that a diffuser intended to display a rear projected image must be thin enough that, somehow, light can get through it. Should the light not get through, we'd be back to a front projection screen but since it does get through, we're able to see the front projected image from the back.

Looking back to our microscope for a moment, we see that the particles of a rear projection diffuser are also strewn about on the surface in a random jumble but that the density of this jumble is noticeably less. Projected light rays here also do a fair amount of ricocheting around as they carom off one diffusion particle into another and each incoming light ray is resultantly broken up into a large collection of little light rays (raylets?) which flow onwards in a pattern more or less equally distributed about the direction of their original parent ray.

Some of them, of course, do bounce backwards. That's how come someone in a rear projection booth can easily discern the projected image from the screen's back side. We wish that weren't so, of course. It's inefficient and wastes energy. But, so far, we just haven't been able to come up with particles that are only forward thinking….

It would also be nice if we could find particles with enough integrity to be independently upstanding. Then we could dispense altogether with the transparent substrates whose only purpose, by the way, is to hold the diffusion layer up in space. In every other way substrates are unwelcome and unwanted. Their compositions absorb light (bad) and their back surfaces reflect it (worse). Incidentally, having said that, it is definitely worth mentioning that users should always prefer acrylic substrates to glass ones because their transmission (the percentage of light that does get through) is measurably higher.

Ok; now that we've coarsely established how a rear projection screen works, let's refine our view a bit and determine what makes one diffuser better than another.

The criteria one should always use in appraising any particular screen is the accuracy with which it displays the projected image and the absence of any detectible artifacts of its own. For better or for worse, the attributes which will be most prominently affected by diffusion are Uniformity and Resolution. You don't want to buy a screen whose coating has, one way or another, been put on unevenly. If you do, gain levels throughout the screen will vary unacceptably and the top will be too bright, the center ok, and the bottom too dim, for instance.

You also don't want to buy a screen where the materials making up the diffuser can be peeled away from the substrate. If you do, you'll have bought a diffuser that is optically too thick and your resolution will suffer.

Da-Lite has gone to considerable effort to avoid these manufacturing pitfalls by having completely reconstituted its diffusion chemistry. Ten years ago, for instance, the average diameter of our diffusion particles was equal to those of our competitors' today — 10 or more microns. Since a micron is only one-millionth of a meter, 10 microns may not seem to be many until, of course, they are compared with 5 — the size of our current diffusion particle.

Put simply, this reduction means that the thickness of the Da-Lite diffusion layer need be only half as great as the best of its alternatives. That thinness means that the image plane can be thinner which means that the focus of the image can be sharper.

Any image is "in focus" when all of the light rays emanating from its lens or lenses intersect and converge at some exact distance from their source. This may be thought of as a series of XXXXXXX's where the strokes converge from the bottom of the line and intersect and then diverge and separate for ever after toward the top. If the optical system is intended for projection, the line formed by the intersection of the X's should be straight and, generally, perpendicular to the center most light ray. Thus, if we insert a projection screen along that line, we get XXXXXXX and a sharply focused image will be revealed.

If the screen plane bisecting our light rays is allowed to be too thick, however, its depth will extend beyond the exact point of intersection and it will, therefore, include and scatter light rays which are not in focus because they have either not yet reached the intersection point or they have already passed it. The trick, then, is to make a diffuser that's dense enough to scatter a visible image without it being thick enough to be ever so slightly out of focus.

Can you tell the difference if you were critically to compare a Da-Lite diffusion screen with one of our competitors'? From the back row, of course not. But from the front row or from any other vantage point where the system's highest resolution is important? You certainly can. We believe you should. We hope you will.

Rear projection diffusion screens have been relegated to inferior status over the past decade or so. They have had to take, as it were, a backseat to their more illustrious cousins, the Fresnel/Lenticulars. But they need not any longer remain so overshadowed. Projectors, you see, have finally become powerful enough that the brightness advantage and resolution available from profiled screens are just not any longer so clear.