Angles Of View
Even though the elements of a complete visual display system are varied and complex, they can nevertheless be generalized as a series of progressive stages which begins with a light source and ends with recognition activity behind our eyes. Sometimes this sequential chain of events is called the
The figure above depicts an idealized display system which begins with a power source and ends at the brain of an observer. In between this engine (a) and its caboose (i) is a minimum number of cars which make up the rest of the train.
All projection systems need a light source which we'll loosely call a "lamp" and which is located at (b). This is the device which turns electricity (those electrons zipping through the power cord) into light (photons flying through the air). Now the thing about light emanating from a lamp is that its rays will scatter in every conceivable direction and we would much prefer that they travel more or less along the tracks of our train.
To accomplish this a reflector (c) is usually inserted behind the lamp so that light rays striking it will be reflected back toward their source. But even though we've got light from (b) and (c) now going in the right direction, the rays are still spread over quite a broad front.
In the projection system we're looking at, a simplified slide projector, this broad packet of light rays is passed through a condensing lens (d) which, through the refractive angles of its two surfaces, bends the rays such that they will pass through the plane (e) - in this case the aperture of a slide. The "contents" of the slide, that is, the information contained in its emulsion, is the first instance in the train where we can declare that there is an image.
It is important to stress, however, that this plane (e) need not be a slide at all. It could also, for instance, be the face of a CRT. To be sure, with a video projector we don't need a condensing lens and the "lamp" where electricity is turned into light is actually the surface of the tube itself.
The plane (e) could also be said to represent the surface of a LCD. Here too, the pixels within its face will be varyingly illuminated by a light source behind them and will combine to form an image.
If the final purpose of an optical system is to actualize a comprehensibly viewable display, then there must always be a point where pictorial information is added to the light path. Exactly which sort of tickets, analog or digital, the data present is not important - what is important is that they board the train.
Now, even though the packet of light rays moving down the tracks has acquired significant additional structure by virtue of its new information content, it is still not organized enough for us to decode it. To get the visual data coherently spread across an adequately large visual field, we need next to pass the bundle of light rays through a lensing system at (f).
Depending on the type of projector the number and kinds of lenses contained in (f) will vary greatly but all such systems are designed to capture as much light as possible from the elements in the train preceding them. Then, by a series of consecutive refractions, the lenses shape the light so that it exits out into the air in a conical shape which will diverge according to a particular aspect ratio.
Furthermore, the lensing system has to arrange that all of the light rays it manages are focused such that at some exact distance out from its frontmost element, a plane surface (g) set perpendicular to the projection axis can exhibit all of the information imparted at (e) in a correct and error free way.
As we learned previously (Vol I, No.4), the insertion of a projection screen (g) into the projection beam can only discover a focused image - it cannot create it. All of that work has been done within (f) and is quite independent of a screen.
What the projection screen does do, of course, is to scatter all of those incoming light rays so that they are dispersed over a large enough solid angle that each of us in its audience can comfortably detect the information arrayed across its surface from any vantage point within that included angle.
The device shown at (g) is depicted as a rear projection screen just to keep the diagram straightforward. It could as easily be a front projection display and everything drawn to the right of it would remain unchanged. But if it doesn't make much difference to the optical train whether (g) is front or rear, what about this other word that we use to distinguish one screen from another. What about the word Gain? Since gain implies an increase in brightness it will be interesting to look at the optical train strictly in terms of energy.
We begin at (a) with a theoretically boundless source of power which (ignoring impedance questions) flows through wiring until it reaches (b) where its electrical nature is transformed into electromagnetic radiation - light. And let us say that from out of our plug in the wall we have lots and lots of energy - let's assign it a unitless value of 1,000.
The moment that those 1,000 units of electricity begin to course through the filament of that lamp two products promptly emerge. One is the visible light that we're looking for but the other, distinctly less welcome, is heat. How much of our initial 1,000 energy units is lost to heat? At least 500. Fully 50% of our available energy is taken away from us and our train has barely left the station.
But wait. If we really would like to have 1,000 units of light, why don't we just increase the output through the plug to 2,0000 incoming units? Then, after we pay the 50% heat toll, we'd still be left with the 1,000 units of light we want. Considerations of the electrical bill aside, it seems like a good idea.
Except that in order for the lamp to produce 1,000 units of light (twice as much as before) it's got to be able to tolerate and dissipate twice as much heat as before and twice as much heat is certain to be more than enough to burn it out. If only projectors didn't have to worry about heat, they could theoretically be as bright as we wish. Since the laws of physics have decreed otherwise, however, we must proceed with just 500 initial units of light radiating out from our lamp.
What happens next depends on what kind of projector we're considering but this much is certain: each and every time our light rays are manipulated by any portion of an optical system they will inexorably and inescapably lose energy.
In the generalized figure above the ability of (c) to be a perfectly efficient reflector is of course limited. Some amount of light will be absorbed at (c), and not all of the light reflected by it will bounce off at an angle suitable to reach (d) where at least 4% of the incoming energy will be reflected off its back surface and another 4% will be lost to reflectivity off its front surface (with, one might as well add, an extra little bit lost to absorptive phenomena in between).
But let's be optimistic and say that all of that adds up to only 10% and so even if we're down to 450 units, we're now ready for our information pickup at (e). Does information cost energy? It certainly does.
If the page you are reading were blank, it could reflect more light than it presently does. All these squiggly marks of black ink add up to a fair amount of surface area which, because it is black, essentially doesn't reflect at all; it absorbs. Thus by the very process of adding information to a packet of white light we must reduce some portions of its energy if we are to have any contrast between its dark and light elements.
If the display is not monochromatic, if it is to include color, then all wavelengths other than, say, that of the required red (or green, or blue) will need to be discarded from the total wave packet with the result that its overall intensity (its amplitude) will be grossly reduced. Still, let's be generous; let's say adding information only costs 50%. We're down to 225 units and we're now ready to go through the lensing system.
It's hard to generalize how many different pieces of glass or plastic will go into the (f) stage of an optical train. But it's easy to state that each and every piece will extract a price similar to the ones paid at (d) above. So if there were, for example, four lenses within some (f), then losses at its eight surfaces would reduce our 225 brightness units down to something like 160.
Finally, our light beam has escaped out of the projector, and with its 160 units of brightness in tact is streaming toward the projection screen, the place where, at last, we get to see it. And, since our screen has a "Gain" of at least 1, we can assume there won't be any more losses, right?
Wrong. If the screen is a rear projection screen at least 50% of our 160 units will be thrown away to back scatter and reflectivity. But it is true that the other 50%, a mere 80 units and less than 10% of our original supply, will be left over for distribution into the screen's field-of-view. A front projection screen will conserve nearly all of the incoming 160 units, but it will scatter them over such a large area that only about the same 80 of them end up illuminating a usable field-of-view. (This is how a 1-gain rear projection screen can look exactly as bright as a 1-gain front projection screen.) Choosing a higher gain screen only decreases the size of that field; it will never, never add energy to the system.
At last some portion of the remaining 80 units reaches our eyes - how much each of us is allotted depends exclusively on our viewing position within the established field-of-view. Since our 80 units must be spread throughout that field each of us will be lucky to get as much as 1% of it into our eyes.
Yet our eyes, comprising only a tiny fraction of the overall field (much much less than, say, our shirtfronts), still manage at (h) efficiently to re-image light rays from all portions of the screen onto their retinas. And at (i) those light rays are converted back into electricity and the optical train pulls into its final destination.