+Schwarzer Ritter yeah, an infrared tv would be hot!

+AnarchistMetalhead When you say hot do you mean thermally or do you mean "I could see through thin dresses" hot? because yes! but UV light would be a terrible idea Ritter.... people would be going blind!!! that said. the original CRT's produced Xrays too.... soooo glad were at a stage in their development that we have Plasma or LCD and LED XD

Aaron Ameer Beg i"ll leave that up to you

Yeah. I guess so, but what I really want is more in the gamma range. We all have to settle I suppose. :/

+Jay A W I think so.

Scientists have done exactly that. They call them "impossible colors." Stimulate only the red cones in one eye, and only the green cones in the other. You get a new color, one you can only see under specific conditions. Normally your brain "guesstimates" that these colors are white, but by giving inputs to each eye separately, you force the brain to be precise and you see these colors you're normally too imprecise to see It's like watching videos in 480p your whole life, then being shown a movie in 4k: nothing new is there, you're just better able to perceive what's going on.

Batfan1939 I want to test this specific stimulation

You are probably never seeing that, but you can experience a similar sensation: yellowish green (for individually stimulating L cones), green (for M cones), and violet (for S) cones.

@@batfan1939 Not exactly. Impossible colors have nothing to do with individually stimulating your cone cells. I've experienced them, and they are clearly just illusions. Most people just cross their eyes with the illusion on their screen and say "wow, yellow and blue are opposites, and this is looks like a mixture of yellow and blue without looking strange, that's impossible! No, you just haven't played enough with a color selector.

+Aleksandr Berdnikov What would that look like? Just a brighter green? Or a new color?

+gotbread2 I am not quite sure, but from all my experience with this stuff I would say that one would perceive it just as very saturated color, though you hardly could encounter it before. While just the brighter green (I mean, not more whitish, but more intense) excites red or blue cones more as well and hence does not lead to a new hue, the method I suggested provides a really new proportion. The problem is, it is not something amazing, as it is not for, say, hearing a pitch a bit higher than you could hear before: there is no qualitative difference, you may even not notice it...

en.m.wikipedia.org/wiki/Impossible_color These colours sound like they would be interesting to see, I feel like I'm being deprived of part of my vision now! >:(

I think I can program that...

@@gotbread2 I tried it and green does look brighter. And there's the lilac chaser illusion, where you see green on white since your red and blue cones are tired out.

Mateusz Nowik, right, seems it's only in females too. I wonder how true yellow differs from red/green (too bad I'll never know). Here's more : en.m.wikipedia.org/wiki/Tetrachromacy#Humans

nah you can also get another type of cone as either gender through certain mutations that exist plus everyone can get the rod tetrachromacity. lastly, there are some glasses which can allow seeing ultraviolet light as an independent color through some neat tricks. some folks are also born with this ability rarely. similar glasses exist for the colorblind to create a sense of a third color.

But no it doesn't matter even then, because as he pointed out, the video cameras capturing the shows you're watching only record in 3 colors. So tetrachromats always used to see a muddy tv screen with muted colors, and they till would now too.

No, even though they have another cone they can still see the exact same range of light as normal people. The only thing it might help them with is distinguishing between different types of yellows, still not too impactful as normal people can do the same, just with a fraction of a percent more work. I drew up some diagrams in paint to show what I mean: imgur.com/a/4vteMNb

+Antsaboy94 Good point! I hadn't thought of that.

+Frame of Essence The HSV colour model (used by most image and video compression algorithms) could actually take advantage (provided that the sensors in cameras are able to record this as well) of a bit more range in the turquoise range of the spectral lobe where the mismatch between the actual colours and the colour triangle is the largest. On the yellow axis though, I do agree there is little to be gained, and even on the axis between green and blue there would only be much to be gained if you are a tetrachromat (and yes, for women it is possible to have an extra receptor in between the green and blue one). So, keeping that in mind, I'm very doubtful it would actually be worth the cost of adding in extra sub-pixels.

+Bart Kuivenhoven You guys are relying on our perception of what is portrayed in this video as an equilateral triangle being linear. I don't think an objective measure like the difference in area of the achievable colour polygon can illustrate the amount of difference it would make to our experience to gain that area. Also, especially as screens get closer to replicating reality and as the gap decreases, I expect exponentially more value in extra colour information and for phenomena like uncanny valley to set in. I acknowledge that I'm not being particularly eloquent here but I hope you follow.

+Lawrence Job The point you're trying to make is?

Thank you! Your feedback tells me I'm doing something right. ;)

i look at the specs, most people i know look at the specs. which specs people seem to prioritize: screen dimensions, resolution, refresh rate otherwise i agree with this pretty much

I wouldn't think so. There's a lot of overlap with the red and green cones, and video cameras are only trichromatic, so synthesizing that last colour wouldn't make much of a difference. I hear tetrachromats don't even notice that much of a difference in real life anyway.

Energy saving probably? The same reason why we got black cartridges for a printer even though C+M+Y = Black

+toni walter (toni714) The REC2020 colour space for UHDTV does that. It still misses out on a lot of the purples, though.

+Trence Grrawin Those colours aren't a true representation, just a guide.

Frame of Essence okay, good to know XP (You maybe said that in the video but i didn't notice that XD)

+Adeon Writer You mean higher frequency, and lower wavelength.

+vlad saioc Wavelength IS frequency, you're talking about amplitude.

Tim Stahel No. He said that we need lower frequencies than blue to create violet. Violet has higher frequency than blue, not lower. It also has a shorter wavelength (because they're inversely proportional). And wavelength is not the same as frequency. They're directly correlated, but describe different things.

+Tim Stahel (Moustached Viking) Frequency has an inverse relationship with wavelength. Wavelength is the measure of the distance between peaks while frequency is the measure of the number of peaks over a given time period. (usually over one second) In light, quantum mechanics steps in and makes it more complicated. So wavelength is usually explained as the diameter of the photon. The 'frequency' of light is usually given as it's wavelength in nano-metres because it's simply impractical to use hertz in that case.

+Adeon Writer The only reason some people see ultra-blue monochromatic light as violet is that the red cones in our eyes are very slightly bi-modal. There is a tiny "bump" in the R cone transfer function in the near ultraviolet. So people see violet there because the deep blue light is exciting both the blue and red cones. BTW, I do not see violet at that end of the rainbow. If one were going to add a fourth primary color to an existing RGB monitor, the best choice would probably be a vivid cyan.

There are a couple different colour spaces with different triangles: en.wikipedia.org/wiki/Color_space#/media/File:CIE1931xy_gamut_comparison.svg Not sure exactly why they are the way they are, but I suspect it's because the chromaticity diagram doesn't fairly represent what colours usually show up in the world.

An even-later reply :) - the choices of subpixel colors are based on a compromise of what colors a given display tech can reproduce. CRT displays standardized on color choices of red/green/blue which could be easily produced by available formulations of glowing phosphors. These specific choices were incorporated into the SDTV digital standard (Rec601) and were further incorporated (with a slight compromise between PAL/NTSC SMPTE C) into the HDTV standard (Rec709), which was then adopted into sRGB. As we moved-away from CRT displays, other display tech developed which produced different primaries, opening the question of how to take-advantage of a different gamut of available colors while still retaining the ability to display the correct colors within the now-ubiquitous CRT-based color choices made for HDTV. The main thesis of this video - that the Y sub-pixel buys us nothing because we already have R & G that add to any yellow representable in the Rec709 color space - relies on an assumption that these LED panels could produce exactly the Rec709 colors for the RGB primaries, which they did not. In fact, early LED panels struggled to produce complete coverage of the complete sRGB color space, and increasing this coverage was one of the main justifications for the Y sub-pixel. Since then, displays have continued to improve, and now many can produce 100% of the sRGB color space, along with many colors outside of it. This creates the need to consider wider-spaced primaries, like the monochromatic primaries in Rec2020 used for UHDTV, one of the widest spaces where every defined color is visible. You'll also find spaces which stretch outside the CIE diagram, such as ProPhoto RGB, which can specify colors which are outside our visible range. This circles back to your question: why not choose primaries which cover as much of the CIE diagram as possible? It's really just a way to use space efficiently: defining the colors displays can actually reproduce, to avoid devoting data to colors they can't. Rec709 was based on CRT phosphors because that's all that was useful at the time, and wider spaces have been used as data has become cheaper & compression better, along with a growing need to define more colors, to include the wider gamuts of newer devices.

I greatly appreciate this response! Thank you so much :)

What do you mean?

+Bryan Geonzon True black is where instead of letting the subpixels display the colour black, they get turned off.

Yeah, I plan to use a different font in future videos. But for the record, this font is actually Chalkboard, not Comic Sans.

Different flavour, same taste. The thing is we generally associate those kinds of fonts to childlike worlds and although that can be useful in some contexts, it's not in the majority; this included. For these kinds of purposes I'd suggest Avenir, Gotham or Helvetica. Beyond that, good topic, great video!

Thank you! I'll look into these

When you look at a black light lamp, you are not seeing the ultraviolet, you are seeing the visible colors it is also producing. As for objects in the room, you are seeing a combination of those visible colors emitted plus visible colors produced by the UV causing materials to fluoresce with their own light.

+Niohimself 1) I haven't been able to find anything conclusive, so I think it's still an open question. 2) I suppose you could, but your eyes would probably detect reflections from the rest of the room off the screen first. 3) Wouldn't make a difference, since your eyes don't respond to them. 4) I haven't really looked in to brightness, but minutephysics seems to cover a lot of it in this video: kzfaq.info/get/bejne/grGepKh0yM2ad6s.html

+Frame of Essence Thanks!

Red receptors do react to violet. They have a small secondary peak in that range.

A green that has more than 100% saturation, similar to the effect gained by first overstimulating the R and B cells, except to a greater extent.

+storminmormin14 You'd see a green that's greener than is physically possible. You can get a similar sort of effect by staring at magenta for a while then suddenly switching to pure green on a standard monitor. The staring at magenta makes the red and blue receptors tired and under-stimulates the green one, which then gets a massive dose of stimulation when you switch to green.

no white is already in the range of possible colours a pixel can make

+Dan Dart Tetrachromats' 4th color is between Green and Blue, however.

+Ryuu Ainaki Er, as far as I know the 4th colour in human tetrachromats is a variation on the red cone, meaning that the 4th colour is yellow-orangey-red, more or less like the red cones themselves. You may be confusing it with the purkinje effect, where in low-light vision you have increased sensitivity (relatively speaking) in the blue and green range, as that is where the rod cells are most sensitive.

+Ron Maple shweet, wait isn't that ears?

+Kieron George The 3D color space you refer to is just a coordinate transform of the 3D space of (Red,Green,Blue) photoreceptors that the video introduces. The mixing triangle in the video is a constant brightness slice, to represent color independent of intensity. So what you point out doesn't contradict the video

+Robert Magkalas You're just looking to have an argument, aren't you? We all know it's white and gold.

@ its obviously yellow and brown.

I understand that it would increase the span of brightness values, but I don't understand how the format is irrelevant. If the format only contains enough information for colours with brightnesses in RGB, wouldn't an increased range of brightness values be artificially created, rather than details that were captured with the recording device?

I said irrelevant because it's a separate question from the one about monitor colors. If application makers never support 4-color displays because they don't exist and display makers never make one with 4 colors because no applications support it, we are in stagnation. Innovation must be started by somebody :) There exist photographic formats that have more brightness values than monitors can display (16 bit or even more). Making use of this shouldn't be very difficult. Whether adding a 4th color is a good idea is again a separate question.

Oh, that makes sense. I hadn't considered that :)

+Duong Pham This point was tackled in the video. Watch the part about the chromaticity diagram again.

Add a mod for microwaves. Good for popcorn.

Not comic sans, it's chalkboard... Edit: Though to be fair, it does look like comic sans.

Fair points. I don't think I fully understand C though. If you convert RGB to RGBY, how do you get the yellows that weren't there before?

+shadfurman I based the claims off this page: www.sharp-world.com/aquos/en/product/4_color_innovation.html In all fairness, they do talk about increasing colour range. Now that I've re-read it, they do also mention brightness as you said, which I admittedly didn't notice before.

Frame of Essence +1000 internets points for honesty! For that I'm subscribing to your channel.

+shadfurman Yay :D Thank you

+shadfurman Wawawhat? Did I just read a mature intellectual discussion in a youtube comments feed? Brownie points to you both

Mas0o0n mmmm... Brownies