If a stereoscopic television set has a low manufacturing delta – in other words, if it costs very little more to make a stereoscopic TV – that’s a best case. People can buy stereo-ready TVs today, and as stereo signals come online they’ll be able to watch them. But is that what is going to happen?
But first I want to talk about autostereoscopic television, which recently suffered a major blow. A few weeks ago Philips bowed out of the autostereoscopic television arena. After something like 10 years, and the last five or six years quite intensively – and I guess after spending many tens of millions of dollars and having something like 60 people employed in Eindhoven working not only on improving the autostereoscopic optics but also signal processing and content creation means – Philips called it a day. They blamed it on the downturn in the economy. But they really could have blamed it on the downturn in pixels. I think they finally became convinced, as I eventually did, that autostereoscopic television requires a hell of a lot of pixels.
Therefore, I believe that autostereoscopic television is a long way off in the future, because in order to get a decent quality image – at least something approaching high-definition – the display is going to need a tremendous number of pixels. It doesn’t mean that the signal has to have a tremendous amount of information, but it means that the set has to be able to derive and manipulate the signal to produce the required multiplicity of images.
I should also say that I spent many years and a lot of my company’s (StereoGraphics) dollars on developing autostereoscopic displays, so I greet the failure of Philips in the marketplace and their throwing in the towel with a feeling of loss. When Real D took over StereoGraphics, they stopped the autostereoscopic project (the SynthaGram) I had worked on for many years. Who knows? Maybe they were right. If it was too much for Philips to bite off, maybe it was too much for a much smaller company to handle. Curiously the last display I developed at StereoGraphics, based on an Apple Cinema Monitor, had quality equivalent to the last Philips 4K monitor. I beat them by four years, but to no avail.
Assuming that we’re a very long way away from that (and a very long way away could be 10 years, let’s say, and possibly more), we’re going to live with stereoscopic images that require the use of glasses in the home. Will people accept the use of eyewear in a living room or in other environments away from a theater for viewing stereoscopic images? That’s something I honestly don’t know. I don’t think anybody knows, and if they’re telling you they know they’re guessing, because nobody knows. We won’t know until people actually start using stereoscopic television. Will people accept wearing eyewear all the time or maybe just some of the time? Obviously autostereoscopic television would solve that problem, and the conditions in someone’s living room, television viewing room, or den are a lot different from those in a theater making foregoing eyewear a big plus.
Hoping that everybody is going to be happy to wear a pair, there are only two types of eyewear that are possible, and what they are relates to the display itself. Liquid crystal displays are the dominant market place display used in TV sets and they can be made to produce a stereoscopic image in two ways. They can run at a high field rate and produce time multiplexed or field-sequential images, as has been done for decades using CRT monitors and this is the same methodology used for the DLP cinema and a handful of DLP rear projection television sets that are on the market. When I say “handful,” it may be a few million, but it’s a breed that’s rapidly disappearing. The quest for a high refresh rate for liquid crystal display has been prolonged. For some time the designers and manufacturers of sets of this kind have attempted to make liquid crystal displays perform like CRT displays. Liquid crystal displays are a hold-type display in which the image remains after pixel excitation: when a pixel is excited it takes awhile for it to be extinguished. That’s not good for a field-sequential stereoscopic television, because it means that it will either not work at all or there will be leakage from one perspective view into the other. In their quest to reduce motion artifacts display manufacturers have begun to ship units that are satisfactory for 3D TV. When operating in the field-sequential mode the stereoscopic television must use shuttering eyewear – shuttering eyewear of the type I invented, i.e. CrystalEyes.
The first shuttering eyewear product that appeared on the market was by StereoGraphics Corporation. It was a product I designed called CrystalEyes. The descendents of CrystalEyes, far less costly and less bulky, are now arriving in the marketplace. A notable example is one unit that is made for gaming by NVIDIA. It’s a good product, lightweight, with good shutters and a good form factor, well designed, and eyewear of this type are bound to become commodities as stereoscopic television gains a foothold. I’m not going to go into the principle of how this technology works; it’s pretty well known and has been described elsewhere.