Interview with David Wiswell, Panasonic Group Manager, Advanced Digital Television Products

Interview...

---

HDTV NEWS ONLINE: We're looking for the problem solving that a videographer might find with HDTV.

We see that there are really two parts to this HDTV business. There is a broadcast business, that may include some HDTV, and then there is another part, which is the kind of higher-end, post-production, and digital cinema business. It's forming into two business areas, and how you resolve for digital cinema, where you might be doing something in High Definition that might be projected up on a 40 foot screen involves some equipment that would be quite different from what a broadcaster would use for local news, or what someone might use for producing local advertising.

Starting with the higher end way of dealing with things, we still think at Panasonic that the best way to shoot cinema or video for digital cinema is film. There really isn't any substitute for the resolution of 35 millimeter film that's available electronically today. The technology simply doesn't exist. We believe that resolution does matter, and that saying that you can accept a lesser resolution, a lesser quality product, simply because that's all that the manufacturers have available today, is not really true. In reality, 35 millimeter film is significantly better than electronic shooting is today, in many characteristics, as well as resolution. We think that for cinematographers, continuing to shoot in 35 millimeter is a good way to do the actual gathering, and then once they've done the gathering process, what they need to have is a production and post-production system that's capable of dealing with the full resolution, or as close to the full resolution of the film as possible.

What are you considering full resolution of film at this juncture?

I think resolution is considered at slightly over 2k by 4k. That's way above the ability of High Definition television equipment or magnetic tape recording systems. Physically we just can't do that today. Probably not for awhile. What we can do, essentially, is a 2k by 1k. If you do that in a progressive image, that's what we can do today. And that has pretty much been proven in the Lucas tests with Star Wars, that I saw side by side, split screen, with film. Resolution-wise, it compared very acceptably to a release print.

For release, because it allows you to deal with an early version of the film, with few film copies being made, and now you can make a very clean copy that stays the same all the way through. We appear to have now reached a point where we have technology that's capable of creating digital cinema, but that technology requires full resolution...1920 by 1080...so our position is that, for instance D5 in quarters [?][085] and systems that can deal with full resolution that is required for the high end (so that's digital cinema)(?). We also believe that those systems, for the sake of production houses and production business, have to be capable of dealing not only with digital cinema, but also with high definition production. For post environments, that system can be the same system (be equipped), but it requires that it be able to deal with 1080 by 1924 in a [?] progressive image. Because there is no electronic cinematography equipment today that is full resolution.

There is one competitor who has delivered a camera to Lucas that's a 24 frame progressive.

Well, it's 24 frame progressive, but it's not full resolution, and Larry Thorpe makes a big thing of that in his papers.

Is this because it's running at horizontal at 12, or something like that?

Yes, and the reason for that, is that you can't get it on a video tape. So it's not that we can't make an image imager that's capable of 1920 by 1080, but the problem is we can't get that on a video tape. So Panasonic has DVC pro HD or 100 Mb per second. That's very equivalent resolution. It's a greater chromanance resolution, a slightly less luminance resolution than HD, but we're not proposing that that should be used for digital cinema. Or another thing is an 8 bit recording system just like the competitor's system is the 8 bit recording system, and for digital cinema, a 10 bit recording system is required. On a 40 foot screen, you can definitely see the difference between 10 bit and 8 bit video.

So with those requirements that's more bandwidth than technology is capable of delivering today. We could do that with high compression ratios, but recording with high compression ratios is going to be quite damaging to the video, so nobody's capable of doing it at the quality level.

Are you calling the compression that you do use lossless?

That's way above the ability of High Definition television equipment or magnetic tape recording systems.

No, there is no such thing as lossless compression, but under strenuous tests by multiple movie studios, the D5HD compression has been proven to be essentially lossless. I don't want to be quoted as saying that it is lossless, because there is no such thing. The only thing I would say is that under strenuous test by multiple studios . . . Warner Bros., Disney, all these studios . . . have done strenuous testing on it, and have accepted it as the standard that they want to project on. Even Lucas used it for Star Wars.

What is the compression scheme for that?

It's a field-based, or frame-based (depending on whether it's field or frame-based video) but pixel [?] seen picture by picture [?] [128] sign transform, and I've been trying to get people away from compression ratios, because in reality, compression ratios vary by scene content, and the type of quantizing tables that are used to reduce the bandwidth on a scene by scene basis are really the secret to making compression performance work properly. Four or five to one area compression ratio, but the only real way to test it is to do multiple generation testing, and then put it up on a 40 foot screen, and see if you see any artifacts or any reduction in quality.

To divert a little from the technical discussion, what has been your experience, personally with the studios? Are they, at this juncture, in the investigative stage, or are they in some form of application or implementation stage of employing high def in their various departments?

I think they're in between investigating and investment in actual activity. Almost every studio is actually doing some high definition work now. It's the only thing that people are talking about, frankly, in video, and we're receiving a lot of talk about that from film people also. The only thing they're talking about is High Definition Television and the application of HDTV.

You know there's been a lot of discussion, chatter behind the scenes, that the HDTV movement hasn't been going so well at the consumer side. We have a hundred stations operating, and I think even yesterday, we had three networks, all with some kind of prime time HDTV programming on them.

Yes, myself, I think that's pretty good.

I do too, but I'm just wondering, you must be looking at it from quite a stretch, from both the consumer and production side, do you see the movement going along with some degree of happiness, or are we struggling.

As a manufacturer, who has made really significant investments in this, of course we would like to see it move along faster, for obvious reasons, but the reality of it, just as you said, I think is really quite significant . . . the progress that has been made. I think that many of the problems that people have thought to be stumbling blocks, have passed by the wayside. For instance, the cost of High Definition equipment, like DVCpro HD, VTR's and so on, is coming down to where it is within the envelope where you can actually do business as a broadcaster, or as a post house. D5 HD has also come down within that range. The ability to go back and forth between image formats has come down within the range that's reasonable. Consumer products, I think, are within much greater range. I think there are some set-top boxes now for like $750.

Are you at all concerned about the Sinclair Initiative?

Well, we officially don't have a position on that, but I am very hopeful, because of the work that I've seen going on with the next-wave chips and with other competitors' chips, and there's some work of our own going on, that we're going to be able to resolve these issues.

One of the issues that seems to be a moving issue, literally, is the mobile issue, and people have now said, "well, that should be one of the options that we have". Are you confident that a single carrier system is going to be functional in a mobile world, or is a mobile world something we really shouldn't be addressing in a television system?

Well, again, as I said, we don't have an official position on that. I don't know, myself; I don't think that in the original investigations of the NTSC system, anybody was considering mobile.

Absolutely not ...it was strictly a stationary system. Some people do say "yes, well that's fine, but the environments do change, and now we're heading for a highly mobile, wireless universe, and do we want to leave one industry without an arm reaching out to that group?"

It's a hard question, and it's a very complex question. The actual engineering complexities of calculating channel capacities and bit error rates, and modulation schemes, are really best left to the people who are really good at that. I don't consider myself to be in that class. You really should have a Ph.D. in electrical engineering before you start conjecturing about which one is best. It's a very complex issue.

I just wanted to feel out your personal view. You're obviously closely involved with this whole industry, and opinions are what drives this industry as much as facts.

So, as you said, maybe you're going to have consumer products that are sub-samples, essentially like some of the professional products.

Well, certainly, and I think when you have people who challenge the technology, the response to that challenge has been very encouraging. When Sinclair challenged the technology, we got an almost instantaneous response from certain chip manufacturers. That sometimes is the way you speed up progress.

I suggested to Nat Ostroff that he wins, either way, either he provokes a decent response, or if the response fails, he can step into the multi-carrier system. He wins either way.

I think the major concerns...there are many major concerns in this whole transition...the technology...in the post environment that's been a major concern for post production facilities...how to do this without going broke ...main concern for broadcasters . . .

There are many major concerns, but as you said earlier, there are some among us who aren't very vocal, who are very positive about this, who think that we are making very significant progress.

What I'm very concerned about...obviously Panasonic is making a financial contribution to ABC's sports programming, and MIA is making a significant contribution to CBS. Sony's made a contribution to NBC, and not so much as NBC, but both you and MIA could pull out, if you ever felt that your investment wasn't paying off, and therefore either leaving them with paying that bill, and if they decided there weren't enough sets out there, they could simply go into a multi-channel proposition. I think that my concern is a little less today than it was six months ago, but still, let's face it, CBS has been subsidized all along on this. They have a 50 million dollar tax break they got for being in HDTV, as did HBO. So that's a big incentive, and then you add on to that free transfers, and everything else, why it looks like the networks are heroes, but in fact, they're making money all the way.

I think from my perspective, if all I saw was the network activity in the bigger picture, then I might still have my doubts, but frankly, I don't think any production facilities are considering making any significant investment, that isn't high definition. It's because of that what you see is that the ground swell is a background that is producing the material that everybody is going to be putting on the air. It's all moving into high definition. That means that that's all going to find it's way through the industry, and as more and more people see high definition, and more and more consumer sets are sold, then there will be competitive pressure to take advantage of the fact that the production is being done in high definition, and that will filter it's way through to consumers.

It's amazing, because three or four years ago, you would have never heard this conversation.

No, it would have been pure conjecture.

Conjecture, and it would have been a small group of hard-core advocates who had less business associations, and more just cheering. It's amazing that we've come this far.

It is amazing. I might have been in that first group, also, but it's been very rewarding to see the way the industry has moved along, as you said, despite the naysayers.

Truly a testimony to the strength of the whole movement, because I think most people intuitively see that this is simply that much more attractive, and attraction is what we sell.

Yes, it is. Entertainment is what we sell, basically.

Getting back to the specifics. You're talking about the post areas, especially in areas where you want to have a system that fundamentally works for the motion picture side of things, and for the television side of things. I think we've talked about that. Tell me a little bit about your 720 philosophy, why you're moving in that direction.

We really don't have a 720 philosophy, or a 1080i, or a 24p philosophy. Our philosophy is multiple formats. We, having watched the industry develop, and talking to customers for some time now, realize that you can't change people's opinions about which image format is correct for them.

What we have adopted is a philosophy of not telling the customer what to do, but to listen to the customer tell us what to do, and what they've clearly told us with 720 progressive, is that there are many people who are very intelligent at Disney imaging and ABC, who have decided that there are very valid applications for 720 progressive signals, and that they think they are very high quality pictures, and for a variety of reasons, have selected that image format. Panasonic, as a manufacturer, really shouldn't be telling them what to do. What we should be doing is responding to them.

What we started to do was respond both to folks who were in the 720p camp, and in the 1080i camp with products that were switchable between those two image formats. Now, as we're moving along, we're creating products that work in multiple image formats . . . in 601, as well as 1080i, 1080p, 720p, all of these different image formats where frame rates are built into the equipment. a new D5HD product is coming available in 2000, and a DVCproHD product is coming available in 2000, and the current Universal Format Converter, which allows you to get back and forth between these formats. Our philosophy is a multiple format philosophy, in providing very high quality images that can be used either as native image formats, or as converted image formats.

We think by doing that, and doing it intelligently, we can take the confusion out of the image format question.

Some people say "well, that sounds great, but you also spent a lot of money supporting ABC in a facility". Is that not more than enabling?

It is, but we also support CBS. If you look at the CBS network release facility, you're going to find D5HD in there. D5HD is their selected format, their digital format of choice for video equipment.

Of course, there wouldn't be any reason for all of the manufacturers to go to one broadcaster. We're the only ones who are capable of supporting 720 progressive. I think it's a matter of them being available. We think that NFL football is something that is a very good way to promote high definition for consumers. It's not so much a matter that they are 720 progressive, but that they are available, and they have the type of programming that we felt would be a good investment for us, that would pay off in having viewers want to watch it. We think that building experience in how to do football, which is what we've been doing up to now, is leading up to the super bowl game in high definition. It's really the key, and being 720 progressive is ABC's choice. We're about the only manufacturer who is capable of supporting in a substantial way, both image formats

Speaking of image formats, will you also then have a 1080p 24?

The new VTR's are capable of 1080p 24. The cameras, I don't know, but we don't think that is the correct way to go, frankly. We still believe that 35 mm film, as an acquisition format is far superior to electronic acquisition.

What about going back to the idea that you're supplying what the customer wants?

Well, yes, we are investigating that, and we do have a plan to create field acquisition with 24 frame progressive on DVCproHD, which we will develop, but as early on as two years ago, as you said, many conversations would have been a lot of conjecture...at this point, with 24p as a replacement for film, that might be a lot of conjecture. We really need to wait and see what kind of experience we build. We are going to respond to that. But we think at the moment, if you want to create digital cinema for projecting on a screen that's 40 feet across, which is the primary purpose for shooting in 24 frame per, that if you're going to do that, you really need to have full resolution, and right now, video tape recorders don't deliver full resolution. It doesn't matter what the camera does, it matters what the VTR does.

A lot of people like this TI DLP projector, and I don't believe it has 1920 across.

Not presently, no.

And I haven't heard too many complaints about the resolution on that baby.

But the point is, also, that part of the reason for arguing 24 frame progressive is to say we may want to go back to film with it, splice electronically originated video and write it back to film as a low-cost replacement for shooting rather than on film. If that's the case, then you're archiving images, and those images that you're archiving certainly shouldn't suffer in the coming years as being visibly electronic.

There are times when creatively it is wonderful to use visibly electronic images, but what you don't want to do is to say at this point, that you can archive those images as though they were film, and nobody's going to know the difference, because, while the TI system today does not go clear out; you know it goes to 1280, it doesn't go to 1920, they do have plans, and they are working on 1920. If you invest in a lot of product, you may find out that before that product is even half amortized, that all of a sudden, here is the new technology that does do full resolution.

So our philosophy, and we want to say very emphatically, we differ from our competitors, is that resolution does matter. When you're investing in equipment at this level, you should be understanding that resolution matters.

When you say this level, are we talking a lot of money to make this happen?

Yes, it is a lot of money when you're covering production facilities.

But I mean from the standpoint of the development.

Well, yes, it is. It's a great deal of money. That's why you're seeing the progress that's being made is being made primarily by the big manufacturers, because the capitol investment is phenomenal. Unless you have research laboratories, and all of that at your disposal, there's really no way that you would survive that transition.

Is it cannibalizing your NTSC business with no greater margins for the HDTV business, or have you found there is an incentive?

The incentive is that nobody today, in any production facility is making any significant investment in...

I was referring actually to you as a company. A lot of people have said this is great, but we have this resistance to market, which is putting downward price pressure on us; we're killing off our NTSC business...

For every opinion that you can see in the industry, you can find that at Panasonic. To answer that question, the evidence of what we're actually doing answers it. Yes, we do see a significant motivation for investing in this. We are moving forward with caution, but we are definitely moving forward.

When do you think you will move forward with just raw courage?

Well, I think we are! Having been on the bleeding edge of this, I think we have, or I have, anyway! We certainly have been leading, but we're not foolish. There's a difference between leading, and moving forward boldly, and being foolish. Panasonic has not been foolish with it's investment. The investments in the technology that we have [?] are proving to be very sound ideas. We are very fortunate to have some really good laboratories, who are able to help us to do the type of things that are necessary. For instance, the D5 format is proving to be an extremely sound basis for forming both digital cinema products, and post-production products. The cost is coming down, we're integrating 601 into that process, so what we're doing now, because of the multiple format philosophy is not that we're driving 601 away, or standard definition away, we're actually integrating it in with the HD solution, and making overall, a very cost-effective method of dealing with the fact that the reality that you have to deal with standard definition and high definition.

Can you buy still, or will you be continuing to supply the market with just 601 devices for those who don't want the additional cost?

Certainly, in the over-all business strategy, our largest market place is still 601, and will continue to be 601 for a number of years. If you look at the DVT pro product line for news and so on, what we're really doing in that area, in the volume product area, is replacing NTSC composite equipment with 601 equipment.

Somebody said to me the other day, that the 525 world, low cost ENG type cameras are low-cost because there has been a development of a consumer camera that takes the rejects of the professionals...

There isn't enough volume in the professional business to do that. We couldn't possibly meet the consumer price points if we took that philosophy.

The thought was the use of the consumer cameras actually led to lower cost on the professional side, and I postulate that people are going to like very much, the idea of having their own personal HDTV cameras in due time.

Right, and what you're going to see is a clarification in the professional's mind. The difference between an image format, and resolution-- because you can have 1920 by 1080 as an image format, but we can have within that, a variety of resolutions that are based on price performance. So, as you said, maybe you're going to have consumer products that are sub-samples, essentially like some of the professional products. 601 is just standard definition video technology. It has developed out to a point where you can make very high quality product at very reasonable cost. If you compare where we are with that technology today with a few years ago, it's also quite phenomenal. The performance you get from a video tape recorder that costs $20,000 today, was up at the $100,000 area 5 or 6 years ago.

I think it was Larry Bloomfield that said to me: "I don't know why these people have all these damn formats. All we need is a bit bucket". Is that true?

No, it's not true. You can call a VTR a bit bucket if you want to, but VTR's are still required to do many things that a data recorder is not required to do, and one of them is real-time recording, slow motion, all of these editing and production features are still required in VTR's. We don't have any let up of people pressing us for production values to this equipment. That you cannot do at the data recorder. You have to have it formatted so that the heads going on tape are related to frame rate of the video, and that allows you to make a production. Once you've done that, the difference between the data recorder and digital video recorder is no difference. It really is a data recorder, it's just a matter of what information you're putting in. Production values are very important.

What innovations do you have currently and on the drawing board?

Being innovative is our whole multiple format philosophy. The idea of our universal format converter which allows you to convert in very high quality ways--to adjust images size-wise, and very cost effectively. This is a technology that doesn't exist in any other competitor. This ability to deal with multiple image formats in a very easy and convenient way in our VTRs is something that should excite everybody. You don't have to make choice when you are in a production facility and say "I am only going to do 720p," or "I am only going to do 1080i". You don't have to make that choice. You can have equipment with universal format conversion technology built in to it throughout your facility that allows you to work in native or converted image formats.

Are you the only one that's providing that level of flexibility?

I think so. I think if you checked other manufacturers, and we're talking about sub-sets of that, but nobody has actually come out and proven their ability to do that.

You can have equipment with universal format conversion technology built in to it throughout your facility that allows you to work in native or converted image formats.

I'm going to presume that everything is competitively priced. Is anybody actually going to go broke moving in this direction?

No, I think this is one of the surprises, that the products that are available today in high definition are within the price range of what 601 video was just a few years ago.

It's interesting that a few years ago we were in the half a million dollar camera stage, and the few people that were in the early days are kind of swamped andnow see companies with 8 and 10 and 14 cameras. Do you know what the total install base of HDTV cameras might be?

No, I really don't. It's in the hundreds, not thousands. Less than 500. Between all manufacturing.

What would be the total market? Of cameras in this country alone?

It's so segmented. Of high-end production cameras--the total market is somewhere in the area of 6 or 7 hundred cameras--the very high-end cameras. But if you go into the broadcast cameras, then it's a much larger market--somewhere in the thousands.

The news market is surprisingly large. If you got some statistics on how many ENG cameras are out there. Potentially all of those ENG cameras are HD cameras.

One of these days I think the industry is going to have to stop and look at Joe Flaherty and say "you know, Joe, we're really going to have to hand it to you", because he created the ENG for you and now HD, I think most of credit and blame can be laid at his doorstep.

Your right, he gets the heat for a lot of things, but he was a visionary, and a lot of what he envisioned has come true.


Thank you