Science desk
< March 25	<< Feb \| March \| Apr >>	March 27 >

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

March 26[edit]

Could multi-primary displays be cheaper if only 3 primaries were used at once?[edit]

Could a multi-primary color display be made significantly more affordable than current models, if each pixel were allowed to use only three primaries at once? An example might be an LED pixel with two fixed-wavelength LEDs for red and blue, and a continuously-variable-wavelength "green" that could become yellow or cyan. Neon Merlin 22:54, 26 March 2019 (UTC)[reply]

See Gamut for an explanation of how the colors a display can show are limited by its (usually three) primary source colors. MPC displays can promise only a small expansion of the gamut from a triangular to a quadrilateral area of the Chromaticity space at the expense of adding a fourth source color. There would be associated costs in exploiting any small advantage because existing color coding and transmission standards are based on, and optimized for tristimulus Color models. There does not appear to have been invented a single continuously-variable yellow-green-cyan LED so it's impossible to say how it could be economical in a display; that is unlikely because color-tunable LED products generally consist of multiple fixed-color sources. DroneB (talk) 23:58, 26 March 2019 (UTC)[reply]

Is there some evidence that adding a fourth primary is actually likely to significantly add to the cost anyway? I'm not aware of any. While I'm aware of techniques like PenTile matrix family to reduce cost, and I'm not saying the addition of a fourth primary is going to be completely free, I'm not convinced that a four primary display will actually cost significantly more than a three primary one if anyone actually bothered to properly research and produce them on the scale of current displays.

The problem is the "if". No one is likely to bother because our as DroneB says the entire ecosystem is built around 3 primaries. Sharp Quattron has shown it doesn't seem to offer any significant marketing advantage either. So there's no reason why someone will bother. It makes much more sense to concentrate on other things like the Wide colour gamut and high dynamic range in ITU-R Recommendation BT.2020, all of which are still based on the three primary model.

I mean it's possible someone will decide using more than 3 primaries will be a cheaper or easier way to offer WCG and/or HDR than other alternatives, but that's a complicated question [1]. And more importantly reflects both that as things stand everything is still based around 3 primary colour models and that adding a fourth probably isn't likely to be a big expense in the grand scheme of things.

P.S. To put things a different way, if you want a really cheap display with more than 3 primaries, this will probably involve 4 distinct primaries rather than some fancy tunable primary. And regardless, the 'cheap' part is going to come from some manufacturer producing and selling them in the same manner they do 3 primaries i.e. the same scale and not marketing them as some fancy "super" display. If that happens, I suspect the price difference will be minimal.

Nil Einne (talk) 03:54, 27 March 2019 (UTC)[reply]

The article about Quattron touches on the probably the biggest reason: all sources typical consumers use supply only three-color data, because everything in the chain from the origin camera to the TV uses it. The display cannot show colors the video source doesn't know about. We'll either have to wait for wider-gamut data or come up with a reliable, realtime, automatic process of the sort of colorizing a black & white movie. 93.136.77.149 (talk) 04:03, 27 March 2019 (UTC)[reply]

Isn't that mostly what both me and DroneB just said? (Except for the realtime/automatic bit.) Nil Einne (talk) 04:07, 27 March 2019 (UTC)[reply]