Aris Silzars Display Technology Consulting


The Display Continuum

Do We Need a Number?…October 2004

The digital camera makers have it easy! All they have to do is specify one number -- the number of pixels -- and everyone instantly “knows” how good the camera is going to be at taking pictures. One megapixel used to be considered a good number -- but no more. Now the number has to be closer to ten megapixels before anyone considers it to be a serious product. Even what we call point-and-shoot cameras are now typically in the three megapixel range. But what about all the other important features, like the lens, the storage capability, the waiting time between shots, the time between when the button is pressed and when the picture is actually taken, battery life, and yet many other important parameters? If you really want to know, you can most likely find them somewhere in the product brochures, but how many buyers do this careful analysis? Surely a five-megapixel camera must be better than one rated at only four! So, why bother with all those other details, especially since there are likely to be more “features” than the typical user will ever comprehend, let alone put to good use. So isn’t it nice to have just one number that seems to say it all?

Should we, therefore, have a number that could similarly be conveniently applied to express the “goodness” of displays? Could it be resolution, contrast ratio, brightness, or size? Well, we do seem to arrange televisions by size and we do promote computer monitors and laptop computers based on the size and resolution of their displays. But how much does that say about the quality of the images that we will see? Resolution does give us a reasonable starting point for the selection of a computer monitor, but for television, the various standards ranging from NTSC up to 1080-line interlaced HDTV seem to create more confusion than help. What good is a display that is capable of 720 lines of progressive scan when used with an NTSC signal, or even with a 1080-line interlaced signal? One would expect it to be better than a 480-line VGA display, but by how much?

Over the last couple of years, with the introduction of the new plasma panels and larger LCD TVs, there was a surge of promotional information touting contrast ratio as a highly significant specification. The numbers started out at reasonable levels but then over time escalated into the thousands. It is not unusual to see advertisements that claim 3,000:1 contrast. Given that a typical printed page is only about 10:1, and it looks quite good to most of us, one has to wonder how these incredibly high numbers are achieved. Was the black level measured in a dark room with the power switch turned off? I hope not. But lately the contrast ratio race seems to have run its course. I don’t see nearly as many advertisements touting such numbers. But what do we have instead?

In a recent visit to a Circuit City store, I tried to see if there was some way to evaluate displays, especially for the various types of TVs, other than by just looking at them. I began my survey by studying the specification tags placed by the products on the showroom floor. Here are a few of the interesting features that I found listed: Three-line comb filter, velocity scan modulation, Invar shadow mask. I think those of us who have worked with these products might know what these are, but how about the average shopper? And these are not examples that I specifically selected for their uniqueness. The other items listed didn’t provide information that I would consider any more helpful. How does it help to know that a Sony TV has velocity scan modulation while a Panasonic set has an Invar shadow mask? In the rear-projection TV area, the descriptions did not help any more than the examples just listed. The most prominent promotion was for rear-projection sets that were based on LC technology – they didn’t say which one. However, in observing these products, I was surprised and a bit puzzled to note that they were noticeably dimmer than either the CRT-based ones or those with a DLP logo on them.

Even as a reasonably knowledgeable buyer, I quickly became frustrated with this confusing and disparate set of claimed features that have at best an indirect connection to the observed image quality of the display. On the other hand, perhaps having no number at all is better than having a number that does not tell the full story. We have already noted the limitation of megapixel counts in digital cameras. Similarly, in the audio section of the store, it seems that in today’s world the only number that is used for comparison is the watts of output power. Is it as simple as that? Some years ago we worried about specifications such as harmonic and intermodulation distortion, and quiescent noise. Then we created incredibly inefficient speakers to get more base notes out of small volumes and along the way seemed to have forgotten such “minor” details as the quality of the sound. I guess when you are driving down the street making loud “boom-boom” sounds, low distortion is the furthest thought from your pounding head.
In the computer monitor and front projector part of the store, the product descriptions became more logical and more useful. Most of the computer monitors were clearly specified with screen resolution and brightness in nits. That was a pleasant surprise. I could actually walk down the aisle and compare screen-to-screen performance. The visual appearance of the displays seemed to match what one would expect from the numbers listed. The front projectors were also easier to evaluate based on resolution, and in this case, the ANSI lumens for brightness. Since the projectors were not operating, I could not see if the lumen numbers would actually match what would be observed. I really don’t know what would happen if a serious shopper asked the sales staff to turn on a few of these to try to compare the actual display brightness.

Perhaps the best overall answer is that displays are meant to be viewed and we should be glad that there is no single number like the pixels in a digital camera or the watts of power from an audio amplifier. For television, at least for now, this may also be our only answer. While the computer monitor and front projector specifications of resolution and brightness seem to be good indicators of display performance, the same approach does not work for TV. There are too many other variables that determine picture quality. By the way, in a side-by-side comparison, I could actually see that the “three line comb filter” did produce a more stable image. But that’s just one of perhaps a hundred such items that could be evaluated, many of them depending on a particular manufacturer’s design approach.

Therefore, when you are ready to go shopping for your next large screen television or home theater projector, don’t rush the process. You will need to spend a good number of hours looking and appreciating – and mostly ignoring what the sales people and the promotional placards tell you. And while the price may provide some indication it too should mostly be ignored as a measure of quality when looking at two fairly similar products. Some years ago, I spent an entire day listening to speaker systems because I felt there was no other way to really sort out the good from the really good. The same applies now to displays for television. You have to look and then look some more. After a few hours or a few days, you will begin to appreciate the really good ones. Then one of them will simply insist that you take it home with you.

Have you done some serious display shopping lately? If you would like to share your experiences, please contact me via this site, directly by e-mail at, by phone at 425-898-9117, or by fax at 425-898-1727.