Interview with CableLabs' Dick Green

What are some of the key things CableLabs has been doing to help the industry prepare for the next generation of high-definition television?
We are focused on finding technologies that advantage the cable industry and, in terms of HD, one of the major areas is, of course, 3-D, which we've been working on now for about 18 months. We're looking at it on all three levels -- capture, transmission and display.

It is a very interesting project, because there is kind of chaos right now. People are proposing dozens of different transmission schemes, which reminds me of the earlier proposals for high-def.

The transmission side is particularly interesting to us because we would like to be one of the first carriers of 3-D. Having a transmission standard would be very useful because then the set manufacturers can build displays.

We're also looking at production, which is quite complicated. There are lots of things you have to think about when you're doing capture for 3-D and it is a difficult process. There are only a handful of people who really know how to do this, and they mostly come from the film world and animation, which has really been the leading 3-D product.

The truth is that there is good 3-D and bad 3-D, and producing good 3-D is hard. It takes a lot of skill, and we'd like the cable industry and our programmers to be have some leadership role in that. Right now, those who are connected to the film industry have a leg up on us.

How do you see some of the transmission efforts developing for 3D?
There is a whole bunch of schemes. One is, you put the picture in the middle and you put the left eye on the left side and the right on the right side and when it gets to the TV sets, the set knows to expand those into an 3-D image. You can also do that vertically, with the left eye at the top and the right picture.

If you do it this way, it can be processed by high-definition gear. It has the advantage of being the least expensive way because it doesn't change your production processing equipment. But it has the disadvantage of sacrificing resolution because you are only getting half the resolution for each eye.

So you have one camp that says, "Let's make it compatible with our existing capture chain," and another camp that says, "We need to find a way to do this with full resolution for each eye."

The extreme position for doing that is with two channels, one for each eye, which uses a lot of bandwidth. But you don't really need to do that because there is a lot of redundancy between those two pictures. If you use a compression scheme that compares the two eyes, you can take advantage of the redundancies, much like you do with MPEG, and you can get the bandwidth down to about 1.5 times the size of a regular channel.

I think the first standards for home use will probably be based on Blu-ray. All the studios are very interested in getting 3D products in the home on Blu-ray. It would require a special player and special TV sets and how soon they will be in the market will depend a lot on the set manufacturers; but a lot of them are rushing to get something out, not for Christmas this year, but maybe Christmas next year.

Obviously, we haven't decided which way to go and we working with our members to pick the best path.

How do you think cable will find the additional bandwidth needed for 3-D and for higher-quality HD programming?
The knight on a white horse for doing more HD is going more and more digital because that frees up the bandwidth. As the analog reclamation goes forward, it gives us the capacity to carry these things. There are also a lot of technologies beyond analog reclamation that we could deploy -- MPEG-4, variable rate encoding, etc. -- to recover more bandwidth as we need it, which will come in stages.

The first generation of 3-D for example, could be done in a single HD channel. It doesn't require a lot of bandwidth but as we move to higher and higher standards, they will require bandwidth.

The next generation is 4K, which is what they use to electronically capture movies. It has four times the resolution of existing high-def.

But there are a lot of things you can do before you move to 4K that will improve quality. You can get more out of the existing MPEG standards as well, and that will make it possible deliver a higher-quality picture using extant standards. That is very interesting and I'm sure that will come before the 4K. People are very sensitive to quality and that is an area we want to lead.

In the U.K., cable operator Virgin Media plans to test 3-D and higher-resolution HD content over a new Data Over Cable Service Interface Specification 3.0 system that has speeds of 200 Mpbs. Does DOCSIS 3.0 offer a way to deliver more HD content?
Sure. One of the great strengths of the cable industry is that we have multiple transmission paths. There are strengths and weaknesses to either path. But the good thing about cable is that you have multiple choices about how you go about delivering the product, and people are certainly thinking about using the DOCSIS channel as well.

By George Winslow, Multichannel News