SPEAKER: DAVID ELLIOT

MR ENOCH: Our next speaker is David Elliot, who's president of Elliot Technologies, a consulting firm which is specializing in advanced television and digital, audio and visual systems. Mr. Elliot was previously vice president of engineering services for the ABC television network, spending over 30 years at ABC-TV. He's served in virtually every area of the broadcast operations and engineering divisions, and retired from ABC just this year, in 1997. He's a two-time Emmy award winner, is active in many industry committees, he's served on the FCC Advisory Committee of Advanced Television Services, the board of the Advanced Television Test Center, or ATSC, and on the executive committee of the ATSC. He serves as chair of the Advanced Television Implementation Subcommittee of the ATSC, and he's standing right by me now [LAUGHTER], helping to implement his own digital reality.

MR. ELLIOT: Thank you for that introduction, and good afternoon. And what we've just seen is a good example of all the fun of computer interoperability, and of course we won't have any of that problem with digital television. It's only software. And good afternoon.

If you remember Paddy Chayefsky's Oscar-winning film Network about 20 years ago, one of his main characters, Peter Finch, says at one point, "You are messing with the primordial soup of the universe." He wasn't talking about digital television, but he might have been. And you just heard Steve speak of the myriad changes in digital television. I'm here to discuss some of the details behind those.

But one of the things that's true is that digital TV will open a myriad of new doors, and close some. The intricacies of the technological and business changes in digital television have been debated for the last eight years. Nine, now. And we could spend the next eight months here trying to focus in on those changes and not have all the answers, but I'll try to focus in on some of the key issues in the next few minutes.

While the difference between black and white and color TV may be more immediately apparent to many people than the difference between digital and analog, although you just heard, we have an analog projector back there, the change is really revolutionary. We are indeed messing with the primordial soup. This is revolution, not an evolution. The last major change to affect broadcasting this way occurred in 1952, with the invention of video tape. But one of the things that's been true ever since then is that every television program made in North America can be viewed on any television set made in North America since that time. Film, video tape, live, it doesn't matter, it was all made to the same standard, it is all interchangeable.

With the new standards of digital television, that will no longer be true. That means that historical programs will have to be converted, that conversion itself will not be complex, only expensive, but the issues of integrating legacy material into programs and commercials with new aspect ratios, and I'll talk about that in a minute, will be complex. If you have any reliance on previously produced programs, that's just going to be the tip of the iceberg, and it's not going to be a minor inconvenience or inexpensive. Add to that the much more significant options provided by digital, and I don't think it's an overstatement to say that we are in fact messing with the primordial soup. So let's talk about some of the details.

What is DTV? Well, DTV is the FCC term for the digital bit stream that Steve spoke of. It consists of, in terms of what is broadcast on the television end, as opposed to some of the other ancillary material that can be transmitted, either standard definition or high definition television program streams. Standard definition is programs that are defined to have roughly the definition of today's television programs, roughly what you would see today on a studio monitor. Although, as Steve spoke of, received in the home in the digital format, it will look much better than the typical home picture of today, the difference is that standard definition programs embrace both current aspect ratios, four by three, and we'll see in a minute, and sixteen by nine aspect ratios. High definition television, by definition, is television that's about twice the resolution of what is on a studio monitor of today, and is wide screen.

This is what I call the programming pyramid, and this talks about, or sort of illustrates, as in a menu kind of way, what can be in one television channel, one of these ATSC bit stream television channels for digital television. And you can choose from column A and column B, and package C. Basically it says that I could have one very high quality, high definition program, taking the full channel bandwidth, or could have two HDTV films, I could have two standard definition programs, or as I reduce the quality of each one of those programs, I can put more and more program streams into each television channel, until at the bottom, there are some broadcasters that are talking about broadcasting sub-VHS quality television programs, saying that the consumer would be very happy to accept these, in which case you could put eight, ten, twelve program streams in a single television channel.

Here are the trade-offs that made that possible. Bandwidth is in the middle. That's what we're talking about. There's a fixed bandwidth, one television channel, and for a given period of time, the trade-offs are in quality, and the number of services. So therefore, you want to offer more services, the quality of any or all of those services has to decrease. You want to increase the quality, the number of services you're providing decrease.

That's somewhat of a moving target; technology gets better, but basically those trade-offs don't change. You're trading quality and number of services that fit in the given bandwidth in unit time. And the reason I say unit time, Steve talked about the fact that you could deliver, in the background, small chunks of data that are stored on some device, a VCR or a hard disk, that give you very high quality programs, but it might take all day to deliver them in the background to give you a five or ten minute or one-hour program.

Here's the timetable for this happening. Basically, this is what's been published in numerous places. Some of the top ten markets, the O and O's in the top ten markets have to broadcast DTV November of 1998. All the affiliates of the top four networks have to begin DTV broadcasting the top ten markets by May of 1999. Markets 11 through 30 by November of '99, and then, this is the big question mark, the target date for the give back of the MTSC spectrum. Is it 2006? Well, we'll have to see what Congress and the FCC do about that.

But what are the consumers doing while all this is happening? Well, I think there's a lot of confusion in the consumer marketplace, but we'll talk about that. Here's what CEMA, the Consumer Electronics Manufacturing Association, estimates, their latest, that I got from them about a month ago. Their estimates for costs of a digital television receiver. They say in 1998, the digital television receiver will cost twenty-six, twenty-five. Now, I put in there $5000 to $13,000, because I think in fact it is more likely that the general run of sets will cost more like five, somewhere between $3000 and $13,000. There's going to be some issues of competitive, some competitive issues dealing with present MTSC sets not being devalued by new digital sets, so there's going to want to be some plateau that the consumer manufacturers are going to want to hit, and not price their digital TV sets too low. But that's what they're saying. By 1999, $2510, and in the year 2000, it breaks the $2000 price mark.

Here's what the estimate for market penetration. And for those of you who are researchers, these start to be more meaningful numbers, I think. For those of us who have been involved in planning and broadcasting and how this is going to happen, these are also meaningful. These are estimates. This is from CEMA, these are the number of households. In 1999, zero percent, 2000, one percent, 2006, the year that some legislation says the analog spectrum will be given back, thirty percent. Now this curve pretty much follows a very aggressive roll-out of satellite television, so it sort of follows a, the satellite television growth curve, which has been a very successful roll-out. We'll have to see how that proceeds.

Network and station costs. Basically, for a broadcast station or a network, these are the kind of dollar figures that each station or each network is going to have to invest in order just to do HDTV, somewhere between one and a half and five million dollars for each broadcast station. That's without doing production. If they want to produce in digital television, you can see the dollars, it's somewhere between five and fifteen million, the network somewhere between fifteen million and thirty million dollars just to replicate existing service in digital, to feed to their affiliates, and then if they really want to start producing in high definition television or some of these digital formats, we're talking somewhere between fifteen and a hundred and a fifty, fifty million and a hundred and fifty million dollars.

But here are the production issues that are going to have to be dealt with by everyone. First of all, aspect ratio. Here's the current aspect ratio, four by three; here's what happens when you go to sixteen by nine. But certain things are going to have to change. If you're producing, remember, you're going to be producing for both analog NTSC broadcasts, because as we saw, a number of people are still going to have analog TV sets in the first five, ten, fifteen years of the roll-out, plus you have the digital broadcasting to deal with. So if you produce and you try to do lower thirds for a sixteen by nine aspect ratio, here's what happens when that picture, when that image is displayed on a four by three TV set. Conversely, if you put a logo in for sixteen by nine, and you try to display it on four by three, oops. If you put the logo in for the four by three image, and now display it in sixteen by nine, it's not exactly where you wanted it.

So these are all production issues. Windowing, that talks about what happens when you have one of these new sixteen by nine TV sets when you're watching legacy programming. That's in a four by three ratio, you have windows on the side, letter boxing, we're pretty much familiar with, but that's what's going to happen when you try to display a sixteen by nine image on a four by three set.

But there are production opportunities to go along with the problems. Additional imaging. Steve showed some of that before, but I can envision this: here's the main program, and three other programs, all going on, all being viewable, in one example, we don't know how the TV set manufacturers are going to make this happen, but that's one possibility. So if the family is actually watching this, question for the research people is, what are they watching? What's this viewing experience, really, and how do you measure it?

Data transmission. Well, here you go. While you're watching TV, I can be contacted, I'm, I'd be happy to sell that space on the side for you.

And then there's the idea of delivering, in the data broadcasting spectrum. That's one of the things I showed in the program pyramid before, that data package could be substituted for any one of those program services, or it could travel along with a number of programming services. This is digital broadcasting; it's all just bits, whether it's audio, video or data, it all goes out as bits, and so while you're transmitting your program services, you can also be transmitting to PC TVs, to the mobile, to the home, to the office.

There's a lot of technical issues that I won't cover here. I'll be glad to talk about them if questions come up. These are technical issues that still have to be resolved before we really are able to do digital TV the way we do today's analog TV.

Digital television: it's not really about technology. It's about business. This is the relationships, these are some of the relationships that exist within the broadcast community and the cable community. And all these relationships are going to change as digital television moves into place. So the business issues really become those of partnership versus competition, between broadcast and cable. Is broadcast going to HDTV or multi-channel? What will cable do? Cable and DBS, same question. Consumer products in the computer industry. They are both struggling to get shares of the market. What services will be offered to differentiate them from each other or to give one competitive advantage over the other; what kinds of partnerships will be created?

Networks and stations: there are exclusivity issues. For instance, if you're a station owner, might you sell a piece of your spectrum independent of network, so even if the network's broadcasting an HDTV program, you say, "Sorry, I've sold a piece of my network stream off to some, I don't know, I'll make up a name, Microsoft, and sold a piece of the spectrum off to them, so I can only broadcast standard definition network programs, I'm sorry, but I've got this other business going on the side." Or a station might decide on their own to do multi-casting. So there's big exclusivity issues, and as I said, there are spectrum hunters out there who are looking to go out and buy chunks of spectrum, and create ad hoc distribution services for various things, not only programs, but data.

Cost of conversion is a big business issue, ATV production is a big issue. When and how much we'll actually be out there to drive some of this market. Cable carriage: will cable carry it? What's the wide screen versus four by three? Multiplex issues: HDTV, SDTV, multiplex?

The consumer environment: what's the consumer going to do with cost and features of sets, the availability of programming? Are set-top boxes going to be available to convert present TVs?

Research and ratings. What are meaningful ratings in a multiplex scenario? How are computers and TVs viewed? This is where the researchers, and many of you here in this room, will be the key to much of the future. In choosing the proper model for the future, it's necessary to understand how the consumer will respond to the changes, the options, and the costs of this new environment? Now, I'm just a technoid, and you folks are the experts in research and consumers. But even I can see the enormous potential for your skills and expertise in this evolving industry. Some of the challenges I envision for research start with measuring this complex environment. Then there will be likely to be even greater fractionalization of the audience. That means that understanding more about who is using each aspect of the range of information and entertainment options that are available through the television, and how they use it, will be even more important.

Until now, the majority of the decisions have been in the hands of engineers. I'm not aware of much rigorous consumer attitude research done to date; maybe you can tell me. But understanding how the public wants to use television in the future, and how much they're willing to pay for the choices, will really shape the future of the medium.

Some people thing that these are the proposed market drivers: high quality pictures and sound, digital television multiplex, the enhanced viewing experience, ancillary data services, the satisfaction of early adopters. The bottom line is, what's the quality of the programs, what does the viewer think of all this? Unless there is some better definition in the marketplace, and some more concentration on production, it's likely that you people will eventually tell the rest of us, another line from Paddy Chayefsky's film, "I'm mad as hell, and I'm not going to take it anymore."

Thank you. [APPLAUSE]