There was a time when broadcast standards were current for decades. The sheer pace of modern technology has changed that, especially for recent developments like HDR. Trying to predict the future is always difficult, but there are certainly a few ideas which might begin to emerge, both to solve technical problems and to allow domestic display manufacturers to promote the next big thing.

Domestic Displays

If recent developments have given us anything that meets with universal approval, the capability of home user displays is it. In the bad old days, TV showrooms regularly exhibited sports with a dozen different shades of green grass. That’s become far less true, especially given things like creator (meaning standards-compliant) mode - assuming people know to activate it.

OLED was a difficult technology, taking a long time to reach market, and early versions were less than ideal. Even those built for professional markets were sometimes short-lived when driven to high brightness. Intermediate consumer devices added a white subpixel to the red, green and blue, achieving higher brightness but compromised saturation at high brightness. They were and are some of the best consumer displays ever made, but that compromise is exactly what HDR and wide color gamut was intended to avoid.

Recent developments include displays where all the OLED elements are blue, illuminating red- and green-emitting quantum-dot phosphors. The result is brighter, avoiding the white subpixel and improving ability to render bright, saturated colors. Properly calibrated, the best domestic displays can approach the performance of broadcast reference monitors, a longed-for nirvana since the earliest days of television.

The future is likely to hold more of the same, with brightness perhaps becoming a point of competition among showroom specification cards. The limits are audience comfort and ecology-oriented power consumption concerns. Overall, the future promises much more consistent consumer displays than ever - at least if they can be persuaded to make a standards-compliant configuration the default.

Meanwhile, LCD is the display technology which refuses to lay down and die.  It beat CRT and plasma, and early OLED took years to better it. With any new technology, though, new entrants must outdo the incumbent when the incumbent has enjoyed decades more development. LCD developments such as dual-layer panels can challenge OLED on contrast, but have not been seen outside professional devices. LCD manufacturers (who may also be OLED manufacturers) are often enthusiastic about development and dual-layer domestic displays might become plausible.

More Color

No display with three primary colors can display every color the human visual system can perceive. The CIE 1931 diagram shows the area of visible colors, which is not a triangle, and the area described by any three primaries is a triangle. In theory, since the visible area on the CIE 1931 has curved edges, an infinite number of primaries would be required to fully describe it, although in practice adding just one more primary would significantly improve color gamut, even beyond what wide color gamut systems such as that recommended by the ITU in BT.2020.

Systems with up to six primaries have been demonstrated but improvements are clearest in blue-cyan-green colors. Existing three-primary systems which generally choose a yellowish green in pursuit of better yellow-orange-red colors common in human skin, compromising emerald and turquoise. Adding one more cyan primary visibly improves things. The difficulty is that most video systems are built in the expectation of three-primary images.

Electronic systems could remain the same, perhaps using a system similar to the ACES color system intended for feature film production which uses three virtual primaries outside the CIE 1931 visible area to describe all possible colors. Displays and perhaps cameras, though, need to implement things with real-world colors, requiring significant reengineering. A longer-term goal, perhaps.

Better Cameras

Many cameras already have enough performance for both HDR and wide color gamut. The color performance of single-chip cameras with Bayer filter arrays, meaning most cinema cameras, can be measurably (though usually not obviously) less good than traditional three-chip broadcast cameras, but most cameras handle most subjects well.

This is less true in the case of special-purpose cameras built for, say, high-speed work, which often suffer a performance penalty in exchange for high frame rates. Miniature designs intended for cramped conditions, such as vehicle interiors, might also give something away in return for compactness. Even where a camera implements HDR or wide color gamut standards, there might still be a visible performance offset.

Change here is likely to be less disruptive, more anti-disruptive. Many of these problems are solvable as standards are developed, become more widely implemented, and advances in sensor technology allow manufacturers to make more capable cameras of all kinds.

More Is More - To A Point

HDR and wide color gamut are widely adored, at least in principle. Cameras with huge dynamic range, however, may create grading and vision engineering challenges of their own. At first glance, it’s hard to question the ability to capture more of the contrast in the scene without clipping highlights or crushing shadows. The risk is that this might encourage creatives to present cameras with very high contrast scenes, safe in the knowledge that the camera has the capability to handle it.

Everything seems fine until it becomes necessary to take that very high dynamic range image and rework it for practical displays.

This is recapitulation of the problem with taking an HDR broadcast and creating an SDR version. Ideally, the camera should have more dynamic range than the HDR broadcast, in the same way the HDR broadcast has more dynamic range than the SDR broadcast. Again, in principle, a simple look-up table could be used, though that might simply discard the extra dynamic range the camera was designed to capture. Doing it by hand becomes more difficult and potentially risks strange-looking results if a scene, depending on its content, must keep extremes of contrast visible for all viewers.

As such, there is a point beyond which more dynamic range is less useful, both in displays (for audience comfort) and in cameras (given the display’s limitations). Extra dynamic range as an internal camera engineering option can help create the sort of gentle highlight handling that’s often preferred, but in terms of broadcasting, given hugely capable modern cameras we now have, the inflection point might not be far off.

Artificial Intelligence

In 2024, it’s almost a cliche to mention AI, which might be better called machine learning in the contexts most relevant to broadcast production. AI is already capable of a modest degree of camera operation and editing. It has mostly been used to create coverage of events which might not otherwise have attracted enough viewers to justify a broadcast at all, which is hard to dislike.

In almost any broadcast, though, the subjective nature of vision engineering potentially makes AI a valuable help, potentially being capable of exactly the sort of situational awareness and interpretation which traditionally requires a human. It seems unlikely to replace vision engineers in the immediate term, but it might make for better HDR to SDR conversions - and even SDR to HDR.

In many situations it is necessary to train AIs on the specific event type in question. Doing that can require a lot of high-powered computer time and access to training data. In the case of TV broadcasting, training data may be recordings of preexisting TV broadcasts, presenting rights issues which may be both expensive and subject to controversy in AI and the world in general.

Better Standardization

Finally, and perhaps most crucially, standardization (whether official or de-facto) is likely to make life easier.

In delivery, most broadcasts target perhaps two of three or four common delivery standards. These are largely provoked by the interest of domestic television manufacturers (and the patent holders from which they license technology). The installed base of equipment means rapid change here is likely to be unacceptable to the audience.

What might change, though, are the standards used in production. As so often in standardization, it is perhaps more important that a standard is widely adopted than that it is technically ideal. There is a parallel with single-camera drama, where cameras often offer several proprietary approaches to color and brightness. It is difficult to show how much this helps without access to the internals of the camera, but it is at least possible (and certainly desirable) for more commonality to emerge, reducing the opportunities for error and making knowledge and experience more transferable.

The Future

If there must be a single critical realization about HDR broadcasting, it is that it need not be any more complicated than SDR broadcasting - in principle. SDR always had a brightness and color encoding standard; HDR has a different one. What complicates things is not the technology. The challenge arises from the need to do both at once and the variability between manufacturers.

Better standardization has perhaps the most potential to make HDR easier.  In principle, the technology to do it largely exists, especially considering the guidelines in the ITU’s Recommendation BT.2100 regarding camera setup, graphic white levels and exposure, which involve creativity and can be difficult to describe formally. As so often, technical discussions around the trucks and studios of current production are likely to be the place from which future standards actually arise.