Tweet
Broadcast & Distribution – At Display Summit in June, one of the sessions was titled “Pushing Color, Dynamic Range and Frame Rates.” For anyone interested in improving the viewer experience of UHD displays, the three presentations in this session were eye opening. In addition, there was a short session titled “Improving the Projected Image,” also important with people concerned with image quality.
This Display Daily will cover the first of the papers in this session, presented by Joe Kane, with a focus on his discussion of colorimetry. The remainder of the session will be discussed in a separate article, to be published on Display Central and in the upcoming issue of Large Display Report.
The Ultimate UHD TV System
Joe Kane (Joe Kane Productions) presented a talk titled “The Ultimate UHD TV System.” Mr. Kane specializes in the sciences of electronic imaging and accurately reproducing video signals on display devices. He’s a very standards-oriented person, and believes, for example, the colorimetry of a display should match the colorimetry of the video standard of the content being shown. If the content is based on Rec. 709 (HDTV) colorimetry, the display should have Rec. 709 color primaries.
While this approach sounds obvious, it is not always accepted and hardly accepted at all in the world of mass-market consumer electronics. Part of the problem is Rec. 709 colorimetry, which is used in HDTV worldwide, actually has a relatively small color gamut. It not only cannot reproduce all the colors in the real world, it cannot reproduce the added colors animators, special effects people and consumers would like to see on TV.
Display Color Gamuts. Left: 1953 NTSC gamut (white), with real surface colors (red). Center: EBU (green) SMPTE “C” (red) Rec. 709 (white) Right: Rec. 2020 Color Gamut
While Mr. Kane supports the use of displays with Rec. 709 for displaying content encoded with Rec. 709, he is well aware of its shortcomings and spent a large part of his talk discussing both those shortcomings and color gamuts for future display systems and transmission standards.
In the left hand figure, Kane shows the original NTSC color gamut, based on CRT phosphors available in 1953, compared to the gamut of real surface colors, which are the reflective colors you see on real objects under white light. Even this relatively large gamut doesn’t cover all surface colors, let alone all the colors of emissive objects (he used photos of flowers and signage of Las Vegas as examples) and artificial colors that special effects people would like to put into content.
The three very similar color gamuts in the center are also based on CRT phosphors, but ones that are more efficient than the original NTSC phosphors, especially in the green. EBU is used in Europe for SD, SMPTE “C” is used in the US for SD and Rec. 709 is used worldwide for HD (1080p and 720p).
Colors that cannot be reproduced by a HDTV
Rec. 2020 was approved by the International Telecommunication Union (ITU) for UHD or 2160p (Kane’s preferred term) in 2012 for broadcast applications. It includes more than just colorimetry, but the colorimetry aspect of Rec. 2020 is what Kane focused on.
Kane sees two problems with the Rec. 2020 colorimetry standard. (Note: no current UHD TV or home theater projector has a Rec. 2020 color gamut.) First, it is still a limited color gamut and cannot produce all colors people would like to put into content. This is true even though the Rec. 2020 color gamut is, in fact, larger than the digital cinema color gamut.
The other problem Kane pointed out is that it is not realistic now or in the future to actually build a display with Rec. 2020 primary colors. The only way these primary colors can be created is with lasers, and narrow band lasers at that. This, of course, will lead to speckle and, in home theater projectors, laser safety issues. Cost may be an issue as well. More importantly, from Kane’s point of view, was the issue of viewer-dependent metameric color problems. This metameric color problem will lead different viewers to see different colors on the screen. This problem could be particularly severe on desaturated colors, for example, flesh tones. The narrower the bandwidth of a primary color, the more severe the problem with metameric colors. And, the ultimate in narrow-band is narrow band lasers so the problem can be expected to be much more severe than it is with primaries based on phosphor emission.
Some advocates of Rec. 2020 colorimetry say they don’t expect displays to actually use this color system, it will just be used in encoding the video. Kane’s and others’ (including me) answer to that is to ask, “why create an artificial colorimetry system just for video encoding?” Digital cinema uses an XYZ encoding system that covers the entire color gamut and can encode any possible color, not just a restricted subset. Then multi-primary displays with broader band primary colors, which are less prone to metameric issues, can be built by the TV set makers. If the decoding from XYZ to the primary colors of the display is done correctly by the manufacturers, every viewer, regardless of his display type or manufacturer, would see the colors on his screen as intended by the content creator. This, in Joe Kane’s opinion, should be the ultimate goal of a TV color system.
The complete proceedings of Display Summit 2014, including Joe Kane’s presentation, is available for $495. –Matthew Brennesholtz
This Display Daily will cover the first of the papers in this session, presented by Joe Kane, with a focus on his discussion of colorimetry. The remainder of the session will be discussed in a separate article, to be published on Display Central and in the upcoming issue of Large Display Report.
The Ultimate UHD TV System
Joe Kane (Joe Kane Productions) presented a talk titled “The Ultimate UHD TV System.” Mr. Kane specializes in the sciences of electronic imaging and accurately reproducing video signals on display devices. He’s a very standards-oriented person, and believes, for example, the colorimetry of a display should match the colorimetry of the video standard of the content being shown. If the content is based on Rec. 709 (HDTV) colorimetry, the display should have Rec. 709 color primaries.
While this approach sounds obvious, it is not always accepted and hardly accepted at all in the world of mass-market consumer electronics. Part of the problem is Rec. 709 colorimetry, which is used in HDTV worldwide, actually has a relatively small color gamut. It not only cannot reproduce all the colors in the real world, it cannot reproduce the added colors animators, special effects people and consumers would like to see on TV.
Display Color Gamuts. Left: 1953 NTSC gamut (white), with real surface colors (red). Center: EBU (green) SMPTE “C” (red) Rec. 709 (white) Right: Rec. 2020 Color Gamut
While Mr. Kane supports the use of displays with Rec. 709 for displaying content encoded with Rec. 709, he is well aware of its shortcomings and spent a large part of his talk discussing both those shortcomings and color gamuts for future display systems and transmission standards.
In the left hand figure, Kane shows the original NTSC color gamut, based on CRT phosphors available in 1953, compared to the gamut of real surface colors, which are the reflective colors you see on real objects under white light. Even this relatively large gamut doesn’t cover all surface colors, let alone all the colors of emissive objects (he used photos of flowers and signage of Las Vegas as examples) and artificial colors that special effects people would like to put into content.
The three very similar color gamuts in the center are also based on CRT phosphors, but ones that are more efficient than the original NTSC phosphors, especially in the green. EBU is used in Europe for SD, SMPTE “C” is used in the US for SD and Rec. 709 is used worldwide for HD (1080p and 720p).
Colors that cannot be reproduced by a HDTV
Rec. 2020 was approved by the International Telecommunication Union (ITU) for UHD or 2160p (Kane’s preferred term) in 2012 for broadcast applications. It includes more than just colorimetry, but the colorimetry aspect of Rec. 2020 is what Kane focused on.
Kane sees two problems with the Rec. 2020 colorimetry standard. (Note: no current UHD TV or home theater projector has a Rec. 2020 color gamut.) First, it is still a limited color gamut and cannot produce all colors people would like to put into content. This is true even though the Rec. 2020 color gamut is, in fact, larger than the digital cinema color gamut.
The other problem Kane pointed out is that it is not realistic now or in the future to actually build a display with Rec. 2020 primary colors. The only way these primary colors can be created is with lasers, and narrow band lasers at that. This, of course, will lead to speckle and, in home theater projectors, laser safety issues. Cost may be an issue as well. More importantly, from Kane’s point of view, was the issue of viewer-dependent metameric color problems. This metameric color problem will lead different viewers to see different colors on the screen. This problem could be particularly severe on desaturated colors, for example, flesh tones. The narrower the bandwidth of a primary color, the more severe the problem with metameric colors. And, the ultimate in narrow-band is narrow band lasers so the problem can be expected to be much more severe than it is with primaries based on phosphor emission.
Some advocates of Rec. 2020 colorimetry say they don’t expect displays to actually use this color system, it will just be used in encoding the video. Kane’s and others’ (including me) answer to that is to ask, “why create an artificial colorimetry system just for video encoding?” Digital cinema uses an XYZ encoding system that covers the entire color gamut and can encode any possible color, not just a restricted subset. Then multi-primary displays with broader band primary colors, which are less prone to metameric issues, can be built by the TV set makers. If the decoding from XYZ to the primary colors of the display is done correctly by the manufacturers, every viewer, regardless of his display type or manufacturer, would see the colors on his screen as intended by the content creator. This, in Joe Kane’s opinion, should be the ultimate goal of a TV color system.
The complete proceedings of Display Summit 2014, including Joe Kane’s presentation, is available for $495. –Matthew Brennesholtz