Print Page
The Next Step in HDTV: Super Hi-Vision

Those who have been involved in the television industry for a while might remember that the first HDTV system to be commercialized was the Hi-Vision system.

Hi-Vision was pioneered by the Japanese broadcasting organization NHK, in cooperation with Japanese television equipment industries, and we first saw it in the United States around 1980. Hi-Vision began life as an analog HDTV system, but we are now well into the digital television era, and HDTV has come a long way since 1980.


Now that HDTV has become a matter of daily routine for broadcasters, NHK's Science and Technical Research Laboratories has taken another leap by developing an experimental ultrahigh-definition video system called, logically, Super Hi-Vision.

The experimental Super Hi-Vision system was demonstrated in a Super Hi-Vision Theater over the course of several months at the 2005 World Exposition in Aichi, Japan. Let's see what Super Hi-Vision is all about.

The original Hi-Vision video system was a 60 field per second (30 fps) interlaced system that scanned 1,035 active lines and had an aspect ratio of 15:9. This scanning structure evolved into the digital 1920x1080 scanning format which has an aspect ratio of 16:9, and which we encounter frequently today in broadcast as 1920x1080i/59.94 fps, and in production and post production as 1920x1080p/24 (23.89) fps. The Super Hi-Vision system has 4,320 active lines, with 7,680 horizontal pixels per line, progressively scanned at 60 fps.

The astute will notice that this is exactly four times as many horizontal pixels and four times as many scan lines as 1920 x 1080, a total of 16 times as many pixels per frame, maintaining the standard 16:9 or 1.78:1 aspect ratio of HDTV. The 1920x1080 frame contains about two megapixels, while the Super Hi-Vision frame contains 16 times that many, or a little more than 33 megapixels.

The design goal for HDTV was to accommodate an ideal viewing distance of three picture heights from the screen, which provides a horizontal viewing angle of 30 degrees.

While this was a significant improvement over the ideal NTSC viewing distance of five to seven picture heights, the Super Hi-Vision system is designed for a viewing distance of 0.75 of a picture height, affording a horizontal viewing angle of more than 100 degrees. This could truly be called total immersion TV.


The demonstration theater in Japan used a 13-by-7 meter screen (about 42.5 feet wide by 22 feet high), with a diagonal dimension of about 600 inches, and a gain of 1.5. A 0.75 picture height for a screen of that size is about 16.5 feet.

The projector provided 800 lumens of illumination, yielding a peak luminance from the screen of about 40 cd/m2. This is reported by NHK to be about equivalent to the peak luminance from a standard cinema screen.

Two cameras have been developed for the format. Both are four-channel devices using four sensors. The first camera, developed in 2002, uses 2.5-inch CCD sensors, while the second, developed in 2004, uses 1.25-inch CMOS sensors. Each individual sensor is two megapixels, the total pixel count for the four sensors in a camera being eight megapixels.

The details of the scanning system are unclear from the information that is readily available, but the camera splits the incoming light into four channels: red, blue, and dual greens. The resulting 24 Gbps video signal is transported from the CCU to a hard disk recording array, and from the disk array to the projector over 16-channel HD-SDI links.

The camera is not an RGB device, but dual green channels implies that twice as many green pixels as red or blue pixels are sampled. This approach, which is used because the eye is more sensitive to green than any other color, is often used in single-chip cameras, and in digital still cameras.

A front-firing LCOS microdisplay projector was used. The projection engine contained four, 1.7-inch, 3840x2048 pixel LCOS microdisplay chips. It goes without saying that a Super Hi-Vision system needs a super audio system to complement it. The demonstration used what NHK calls 22.2 Multichannel Surround Sound.

This system uses a three-tiered loudspeaker arrangement, with nine loudspeaker channels situated near the top of the theater space, 10 channels about midway from floor to ceiling, and three channels (including two LFE channels) near the floor.

In addition to the recorded program material, the 2005 demonstration included a live transmission of uncompressed Super Hi-Vision video and 22.2 Multichannel Sound signals to the theater over fiber-optic links.

It is safe to say that we will not be seeing this kind of resolution, with its huge bandwidth requirements, being transmitted over the air for some time. This was truly an experimental demonstration, and the data rates involved are extremely high. But it serves as an interesting demonstration of the current state of the art in video resolution.   Print Page