Aris Silzars Display Technology Consulting


The Goodness Asymptotes...August 2002

As usual, the airport rental-car bus driver dropped me off in front of the "preferred" customer building and told me that my pre-selected car this time was in space E-12. For me, as a "frequent renter," this has become a familiar process. Basically, it goes as follows: get in the car, show your driver's license at the gate, and adjust the seat and mirrors while driving to find the exit from the airport. But on this trip my experience was not quite typical. The car assigned to me was not the usual bland model with cloth interior and cup holder covers that are so flimsy that they most likely broke during the first or second rental. Instead of dull gray cloth, the seats were tan leather, the car was a burgundy red and the layout and quality of the instrument panel was as good as any that I have seen -- even on the most prestigious brands. The subsequent driving experience matched this initial impression of a quality product. The car handled precisely. It was well behaved at all speeds available to me. All the control functions were easy to understand and I could operate them without distraction while driving in dense freeway traffic. I soon realized that I was treating this car with the same care and enthusiasm that I would lavish on one that I had purchased.

As I observed my own behavior over the next two days, it occurred to me that there really was nothing more that I needed or wanted in a car. Yet this was a model that was not one of the well-known luxury brands. Having driven those as well, I would have to conclude that the differences in features and quality had become so small as to be virtually indistinguishable. Oh sure, those of you who are serious car enthusiasts would be able to tell. The specialized models that sell for well over $50K may corner a bit better or accelerate faster, but when and where can I experience such performance in practice? For the enthusiast, simply owning a certain brand or model may be what is really important. But from a practical standpoint, even as a very demanding customer, the differences between this moderately priced rental vehicle and the top-name brands have indeed become very very small.

Over the years, we have seen similar asymptotic evolutions in other product categories. The audio components business is a good example of this. During the late 50's and into the 60's there were major improvements in performance. Harmonic and intermodulation distortion levels dropped from the 5% range down to hundredths of a percent. Amplifier power levels increased from a few watts to hundreds of watts. Turntables for playing 33-rpm records improved, with wow and flutter achieving imperceptible levels. CDs then took us to the next level by getting rid of background noise and increasing dynamic range. But somewhere in this process audio equipment -- with the exception of speakers -- became so close to perfect that we could no longer hear the improvements. This, of course, was seen as a benefit by some enterprising folks who created businesses to sell items such as interconnect cables that presumably allowed the sound to "flow more smoothly" with "clean high frequencies" and "robust base response." And at $100 per foot, that absolutely has to be true! Of course, we shouldn't mention that inside that power amplifier the circuit boards are made of the cheapest material available and the interconnect wiring is similarly economical. Thus, it seems that for audio components we have not only approached the asymptote of goodness but exceeded it by some significant margin.

Some time ago, I wrote a column that suggested that desktop computer speeds in the 1 GHz range should be sufficient for most applications. Many of the current computers exceed that -- but how many of us can really appreciate this increased speed? As a calibration point, consider that I am writing this column on an old laptop computer running with an Intel 486 processor, and using WordPerfect 5.1 as my word processing program. Why? Am I some relic from the past? Well, that may be true, but before you make that your final answer, consider that I have three other computers here in my office. One is a Pentium 4 machine, another a Pentium III, and the third is a two-month-old super-slim laptop with a docking station. So what gives? They can all do this column just fine. What brings me back to this old dog of a machine? The two reasons are that I like the feel of the keyboard and I like the simplicity of the old word processor. That is all that I need to string the words together for this column. I also like that this old processor does not try to "help" me. It lets me make whatever stupid mistakes I chose to make in formatting, spelling, and grammar. Then, when I am ready, I can ask it for some assistance to help me make the fixes that I wish to make.

The question then becomes, what should we do to justify the multi-GHz machines now so readily available? The only significant applications I can think of are for image processing, and for real-time video (for playing virtual reality games?). A few of us may also have work-related needs for computation-intensive problem solving such as electron beam or optical ray tracing. But that is a very small fraction of the desktop user population. It seems to me that in computer speed and software features we too are close to the asymptote of goodness.

What happens when more and more features are added to products? It seems to me that once the asymptote of goodness is crossed, matters can actually get worse instead of better. From cell phones, to car radios, to VCRs, we have all encountered the multi-level menu-driven functionality that only our teen-aged children can unravel. Yesterday, while sitting in an airport, I tried to get my cell phone to tell me how many minutes I have used so far this month. After cycling through the many menus, I finally got to what seemed to be the right one only to be told that I needed a password. Wonderful! I have never needed a password to make a phone call or to access other parts of the menu. Therefore, I have no idea what my password is or even if I have one. A bit of frustration? You bet!

Could this asymptote of goodness also apply to display technologies? I firmly believe that it can and does. I think we are already seeing it today in the major display market segments such as television, desktop and laptop computers, and portable devices such as cell phones and PDAs. For television, at normal viewing distances of about 3 meters, a 35-inch diagonal screen only needs about 500 lines of resolution to be about as good as most viewers can appreciate. For larger screens with 50 to 60-inch diagonals, a resolution of about 700 lines turns out to be "good enough." So is it any wonder why there is no great rush to implement a new and incompatible broadcast standard? The 1080-line HDTV format may turn out to be useful only for electronic cinema. For desktop computers, there seems to be a similar "good enough" situation where most people have been using their computer displays at 600 x 800 resolution settings. There is now a gradual migration to 768 x 1024, but for the typical user there may not be much of a need for anything beyond 1024 x 1280. Higher resolution displays may only find a market in new or specialized applications where customers are willing to pay the premium price. For small displays, we are seeing a similar situation. Here, readability under a wide range of ambient lighting conditions is important and today's displays do not always perform as well as desired. We could benefit from more contrast and brighter images. But overall, the majority of users seem to be getting by without major complaints.

Surprisingly, even price can have an asymptote of goodness. If the price is reduced to the point where customers begin to believe that the product approaches the "cheap junk" category, they also will not buy. We have seen this phenomenon with digital watches and with some audio and video components. Can you really buy a "good" VCR for under $100? Shouldn't I buy a more expensive model to make sure it will work well? In actual fact, I have no idea what is inside either one, but the very low price has created a suspicion that it simply cannot be a quality product.

Once the asymptote of goodness is approached what should product developers do to continue to create interest among consumers? It seems to me that there are several interesting directions that can be taken. One is to emphasize specialization and thereby to create more product variety. For displays this can mean, for example, the exploration of new markets that demand greater color accuracy, unique form factors, or new ways of presenting complex information. Another is to seriously focus on convenience of use and design elegance. This may require a combination of science and art, but then I also believe that the next phase of technology evolution will not be to add more features but to make accessing the ones we already have more convenient. For this we will need displays of ever greater versatility and customized functionality. Are you ready to have your desktop be a display and your active "white-board" coupled to your computer? I sure could use something like that this very minute.

How close do you think displays are to approaching these "asymptotes of goodness?" I welcome your comments. You may contact me by e-mail at Email, by telephone at 425-557-8850, by FAX at 425-557-8983, or by mail at 22513 SE 47th Place, Sammamish, WA 98075.