The colorful world of tomatoes and lasers – wrap-up: keep it practical
A walk in a forest
Let’s go back to my story about the black and white TV set. Representing colors on a TV definitely adds a dimension – but are the colors accurate? A clever guy named Pointer measured a lot of colors found in ‘everyday’ nature, and derived what’s today called ‘Pointers real-world color gamut’. Refer to Figure 1 below (and remember that in fact it’s actually a color volume). It doesn’t mean that colors beyond these don’t exist – think for a moment of lasers – pure wavelengths. It’s just that we don’t really bump into lasers when we go for a walk in a forest… The holy grail of display technology would be the ability to faithfully represent each and every color that is in nature (provided it can be captured correctly by a camera – but that’s a different story).
The color gamuts of early television were defined by the practicality of the phosphors used in the CRT television sets. So we have color gamuts such as ‘Rec.601’ or ‘SMPTE’ for TV.
Then came HDTV and with it a newer color gamut standard (a kind of mix between the EU and US standards for standard definition (SD) television, still driven by CRT phosphor availability). This became the now famous ‘Rec.709’ color gamut – which is the same as RGB used in the graphics industry.
Next came the age of digitization of film projectors. Because of the properties of the film, one could not express all the colors simply by three points. But all digital projectors worked only with three primaries – so a sufficiently wide color gamut was needed to represent most of the colors captured by film – and again, economical, practical etc. to make. Next up was the ‘DCI P3’ color gamut. This is currently the standard for digital cinema projectors.
And recently, with the advent of new technologies such as OLEDs, quantum dots and laser projectors, an even wider color space is being proposed – the so-called Rec.2020 color space.
The chart in Figure 1 shows the three most current color gamuts in the industry today, together with the Pointer color gamut of real-life colors. The Rec.2020 target primary gamut includes most of the Pointer colors – so in theory, such a display would be capable of faithfully producing them, which was probably a driving force in defining this gamut in the first place.
Figure 1 – A (u’,v’) representation of three common color gamuts: Rec.709 for HDTV (the smallest triangle), DCI P3 for digital cinema projectors (the middle triangle) and Rec.2020 for next generation UHDTV sets (widest triangle). The Pointer gamut of real life colors is contained by the wobbly line.
The price of color
That’s it for the theory. Let’s get practical now.
One of the key things we have learned is that not all colors (wavelengths) have equal brightness at the same optical power: 555nm green produces the most lumens per watt, blue and red produce far fewer. So we need to strike a balance in terms of going ‘crazy’ in terms of wide color gamut, and frying the projector. The closer the green is to 555nm, the more lumens it makes. The shorter the red wavelength (the closer to orange) and the longer the blue wavelength (the higher up on the chart), the more lumens the projector will have as well.
For example, to achieve the DCI P3 color gamut, we need a green wavelength of 547nm or shorter –547nm would be ideal in terms of lumen/watts. The red needs to be 616nm – or longer but preferably not as that decreases the lumens – and blue needs to be between 455-465nm – 465nm thus – to fall within the DCI tolerances. This is all summarized in Table 1 and Figure 2.
Table 1 – A summary of required laser wavelengths and resulting power to make 60.000 lumens with the chosen color gamut and white point.
When we calculate the optical power we need for the Rec.2020 primaries, we see that we need about 27% more laser power (and cooling) than if we would make a laser projector only for DCI P3 gamut.
That’s some expensive color!
Figure 2 – The produced watts per lumen for the wavelengths required for Rec.2020 and for DCI P3 color spaces. It can be seen that the DCI wavelengths are more efficient.
Secondly, and this is where we get a bit technical – finally! – using a single laser wavelength as a primary is kind of not done in projector design. The reason is a phenomenon called ‘speckle’. Speckle is a sparkly effect on screen, most prominent when produced by a single wavelength, ‘narrowband’ RGB sources on silver screens. It gives a kind of a ‘sparkly dot’ curtain over the image, and can distract you from being immersed in the story. In order to ‘despeckle’ the image properly, we also need – amongst other things – many, many laser wavelengths that are mixed well. Not all lasers come with this spread, so we need to choose wisely. That’s another tradeoff we need to make: color purity or luminous efficacy or both.
Thirdly, there is 3D. You can read all about it here. One of the great things about RGB laser projection is that we can make efficient ‘6P’ 3D (six primary, color separation 3D) without requiring internal or external filters or additional 3D equipment. The takeaway is that, in order to produce color separation 6P 3D without crosstalk, we need to choose sets of two red, two green, and two blue laser primaries that are sufficiently separated – by say 15-20nm at least – so that the color filter glasses can do their job properly. This implicitly means that lumen efficacy goes down, and also means that you deviate from the Rec.2020 ideal single wavelength colors. But you do get crystal clear and bright 3D – typical cinema audiences would know what to choose!
Finally, you have the economics of different lasers choices, something that the makers of the Rec.2020 color gamut specification did not consider at all. Some laser devices and wavelengths are easy to make or obtain, some are not. Some are expensive, and others even more so. Some give sufficient brightness per laser – suitable for high brightness projectors because the étendue is still limited even with lasers – some don’t. This will hugely impact the cost effectiveness of the laser projector.
And on top of this are market requirements – projector brightness needs to be as high as possible; TCO needs to be as low as possible. Color gamut needs to be as wide as possible. Lifetime needs to be as long as possible. Nice job for a product manager… Luckily we also get to write a blog item or two, to bring some joy in our lives...
Where did Pointer go?
Let’s put ourselves in the shoes of a creative person that makes movies and wants to push technology boundaries in order to bring a new visual experience to the audiences. Whether it’s a frame rate increase, HDR, or wide color gamut – people have experimented with it all. And this is great as the content industry needs such visionaries, and this is the only way to keep on improving the visual experience. But what good is all this, if the chosen technology has no market viability due to something as profane as… high cost?
Cinemas need to make a living and the equipment needs to pay off. So making a new technology very expensive will be prohibitive to all but a rich and happy few. As a result, very few people will adopt it and as a consequence – very few people will see, or be touched by, the new exciting visual experience that the creative has envisioned.
You see? These are the two sides of this coin. A visual experience that’s attractive enough and an affordable technology are both needed, and one cannot exist without the other.
So are we willing to depart slightly from the dream of showing every color exactly as it is in nature?
Taken all the other benefits laser projection can bring? You bet we are!
Smart people in Barco – but not as famous as Pointer – have figured it all out and have made the right RGB laser projector concept selection for us:
Brightest projector? Check.
Wide color gamut (wider than P3 and still looking awesome)? Check.
6P 3D with the lowest cross-talk (no ghosting) and highest efficiency from a single projector? Check.
Most despeckled laser projector on the market? Check.
Most economical solution for your PLF screen? Check!
To wrap up: it takes quite a lot of knowledge and judgment to make the right engineering choices, but also courage, faith in your decisions, and a tremendous amount of work, of course. Having seen the results in real life, I for one am ready to forget (or bury) the nostalgia for my black and white TV set and enjoy the spectacular and accurate color of Barco RGB laser projection. Now you can too!
About the author
Goran Stojmenovik is Senior Product Manager within Barco’s projection division and is currently working on laser projection for the cinema and other Barco markets. With focus on image quality as well as user experience, Goran has managed different products in Barco since early 2005. Initially he was responsible for professional LCD monitors and software solutions for various Barco professional markets (control rooms, broadcast and post-production). In September 2011 Goran started at Barco digital cinema where he worked on introducing dedicated projectors for post-production as well as on remote service solutions for cinema (CineCare Web). Before joining Barco, Goran Stojmenovik acquired a PhD degree in Engineering Physics at the Ghent University, Belgium. He is based in Belgium.