Angles Of View

If 1998 marked the achievement in our industry of numerous projectors that at last have become bright enough, 1999 will see the commencement of a new trend which will even more profoundly change the way we look at visual displays. After nearly twenty years, the Video Age is about to become the age of High Definition and the new millennium will usher in a world of wide screen, nine by sixteen displays. The implications of this impending format are numerous and non-trivial. As a start at identifying some of them, let's try

Sizing Up 9:16 - The Shape of Things to Come?

We'll want to make clear from the outset that we're not talking about Oprah. Yes, we understand that the force driving this HD business is TV, not A/V, but we still don't think corporate America is going to reformulate itself just so it can get wide screen soaps into its boardrooms.

But soon enough the executives leaving those boardrooms are going to go home at night and watch whatever it is they watch on HDTV and that by itself is going to be enough to get them hooked on the wide screen format. They really will be seeing a better picture.

Then, there's the computer industry for which, surely, the shift from the currently crowded 3:4 desktop to a hugely more generous 9:16 should be an irresistible excuse to reinvent itself. Can we imagine Windows 2001 ("A Space Odyssey"?) as the new, 9:16 operating system? Of course we can and, probably, we will. So it's coming, this paradigm shift, and it's coming soon.

Right now, however, it may be worth remembering that all of the research that went into the establishment of the 9:16 format was directed toward the entertainment value of such a larger and wider TV screen. Its very purpose has always been to escape the limitations of the NTSC and PAL resolutions with their inescapable dependance on close-ups and the visually static talking heads. HDTV was invented to liberate television as a medium, enabling it to become epic, panoramic and, finally, as cinematic as cinema itself.

Whether HDTV was invented to display a spreadsheet, a computer graphic, or even a PowerPoint presentation is quite another question. All of those media have been developed to be shown inside a 3 x 4 rectangle and, perhaps, they should have been. Let's have a look.

The four rectangles which follow illustrate the spatial conversions between the two aspect ratios. In the upper pair the height remains the same and in the lower pair it is the width that remains unchanged.




As may be plainly seen, the choice between which dimension to keep constant has a significant effect on the consequent image area. Compared to the area of the original 3:4 aspect ratio rectangles on the left, the upper 9:16 is 33% bigger than its partner, while the lower is 25% smaller.

The first conclusion, then, that we should draw is that all of the new screens we'll have to size and specify had better be wider rather than shorter than our clients' present displays. Whatever our customers are writing up onto their current screens can't possibly be more legible when projected onto a new surface that is smaller.

Then there's the question of resolution. The pixel density of an HD image is expected to be 1080 x 1920. Let's give that resolution to both the right rectangles pictured above. Then, let's imbue the two left rectangles with XGA resolutions of 768 x 1024. Now let's load all four screens with an imaginary spreadsheet. On the two XGA screens the cell matrix should be about 10 columns by 30 rows. At the much higher resolution of the two right screens, however, the matrix is likely to become 18 columns by about 40 rows — which are a whole lot of additional data packed into the 9:16 displays and which are not dependent on absolute screen size.

That last point is crucial. Because both right-hand screens have the same aspect ratio and resolution, they will each contain the same data, even if one is a lot smaller in overall area than the other. And while in other contexts that observation is perfectly obvious, it is emphasized here so that we can explore how best we might establish rules for trying to ensure that the 9:16 screens we'll all be specifying are, in fact, big enough.

If a good rule of thumb for commercial, 3:4 screens is that their height should be at least one-sixth of the distance to their Least Favored Viewers, the guideline for sizing 9:16 screens is easy enough to suggest. Instead of one-sixth, make the screen's height equal to one-fourth of the distance to the LFV. This has the effect of reducing the audience's viewing distance which is perfectly appropriate given the enormously higher available resolution. Makes sense, doesn't it?

Unfortunately, there are two serious problems with this dictum and neither can safely be ignored. The first is ceiling height. If the old screen was 6 feet high by 8 feet wide and the edge of the back row was 36 feet back, then we're only barely fine in a room with a nine-foot ceiling. Our screen bottom is only a barely acceptable three feet off the floor (44 inches is much better).

But now when we upgrade to a 9:16 display and we go to our new, ¼ rule we see the new screen's supposed to be nine feet tall. Tilt. Ironically enough, the resultant enlarged width (16 feet) generally won't be a problem. We will have space for it, but all to often we won't have space for the optimum height.

The second problem has to do with what we call bend angles. To illustrate it, let's suppose that you're seated opposite the right hand lower corner of a 6 x 8 screen and you're looking at data being displayed on the upper left hand corner. If your chair is eight feet back from the screen, the angle through which the light containing that upper left data has to be bent to get into your eyes is, roughly, 120º.

Now let's swap out the 72 x 96-inch screen and put in one that measures 72 x 128. To reposition yourself such that you're opposite its right hand lower corner, please note that you have got to get up and move 16 inches to your right. And, the data in the upper left hand corner will also move a corresponding 16 inches to your left. Now, for simplicity's sake, let's assume that the throw distance requirements of your new projector are unchanged from your old and that, therefore, all incident angles to the screen can remain unchanged. The new bend angle still increases to 125º.

You could correct for that shift, of course, by moving farther back from the screen but, if you did that, you would effectively nullify all that extra resolution which your new projector is providing. The point here, then, is that higher resolution and closer viewing distances are inextricably linked. And, thus, if you cannot increase screen height because of too low a ceiling, what you can (should?) do is move your audience closer to your not taller but definitely wider screen.

How do you calculate how much closer? Well, here's one idea. Harkening back to the rectangles above, we now see that it is the upper pair which interests us. At XGA resolution, the number of pixels available per unit area on the 3:4 screen will be 7,282 (768 x 1024 = 786,432 ÷ [9 x 12 = 108] = 7,282). On the 9:16 screen, however, that number is 14,400 or, effectively, doubled. If the unit area was 1 ft², that would mean that each square foot of the right hand screen would contain twice as many pixels as are contained in each square foot of the screen on the left.

If, as viewers, we're going to take advantage of that major increase, obviously we've got to be able to see it. But does that really mean that we should move ourselves twice as close to the new screen as we were to the old? In the example given above, would we really want to end up seated only four feet back? (Incidentally, if we did, guess what that bend angle would be. 140º!)

The problem with all of these speculations is that we are obliged to make them because we are not using the 9:16 aspect ratio and its resolution for what it was designed for. We are not looking to be entertained. We are trying to read information. With that thought firmly in mind, we might ask when's the last time anybody tried to read something from a viewing angle greater than 60º? And who, in fact, would want to?

What we've been taught our whole lives to read generally has an aspect ratio that's a whole lot closer to 8½ by 11 than it is to 9 by 16. Yet there is an inevitability to the new format which is going to prevail. At Da-Lite Screen Company, we have always welcomed change and the impending arrival of this revolutionary aspect ratio will be no exception. But we think that some of its implications are tricky and we invite you to join us in puzzling through them.