Better colour rendition thanks to the 4th pixel

[Andrew Smith]

While the rest of the TV industry goes romping off into the third dimension, Sharp has decided to do its own thing by focusing instead on the "fourth pixel". The "Quattron" technology in its new range of Aquos-brand LCD sets gives them yellow pixels in addition to the standard red, green and blue primary colours.

Read the rest of the glowing review of the very real model here.

Given the expanded colour gamut that is in HD video (he says without referring back to the manual for his Sony camera to double-check), this should be quite an interesting development.

Andrew

[Jad Meouchy]

No offense to Sharp, but there are many technical reasons that no other manufacturers are spending time on that side of the lab. Frankly it's just a big marketing gimmick..

[Jon Fairhurst]

Disclaimer: I work here: Sharp Laboratories of America

When I first saw our yellow pixels (Quattron) TVs at CES in January, I wasn't so sure about it. The rushed prototypes on the show floor showed aliasing, and they were adjusted to be overly bright and saturated. More recently, I got to evaluate some of the first production models - and the images look great. There's no hint of aliasing - in fact, the edges and contours are extremely smooth. And the color adjustments out of the box were ideal: bright and saturated, without being too aggressive.

One place where I would differ with the reviewer was about "Standard Mode". (They thought it was a bit dark and dull.) I tried a photo that I shot myself on the TV from a USB stick. I thought that Standard looked the most natural and had the most depth.

FWIW, here is the photo:
http://www.dvinfo.net/forum/eos-5d-m...ml#post1520525

There are a few things to note about the technology. You can't just add a 4th subpixel to any old display without some sacrifice. That extra subpixel steals some real estate from R, G & B. If the dots have a lot of metal (slits and ribs) blocking the light, shrinking the dot size will really lower the efficiency.

Sharp solves that with the X-Gen panel that uses UV to control the liquid crystal material. That means we don't need metal ribs to spread an electromagnetic charge. With the larger aperture dots, we can add the fourth subpixel and keep the brightness high. In turn, we can turn down the backlight (and the power), because we can pass light so efficiently. And, of course, yellow was chosen since it has the highest luminance of any color. We gain both more color control and higher brightness. (A white subpixel would give even more brightness, but would tend to desaturate bright images.)

The other advantage is that while most 1080p TVs have about six million subpixels, the Quattrons have eight million. With proper phase alignments on every sub-pixel, that effectively gives 33% more resolution than we get with an RGB TV.

Some people have claimed that the 4th color is a scam, since broadcasters only deliver RGB. But keep in mind that high-end printers have twelve or more inks in order to reproduce colors that cannot be achieved with only three color pigments. It's the same story with the yellow sub-pixel. Also, with the yellow pixel placed beyond the gamut triangle formed by RGB, we can push the green color further up and toward cyan. We can now create green with a combination of blue and yellow or with our new, shifted green color. So, not only do we have more control over yellows and golds, but we have more control over cyans and greens as well.

Comparing the Quattron to our own models, we get sharper edges, a wider, more vibrant yet natural range of colors, and more subtlety on smooth color gradients. Where previous models showed some contour lines, the new model shows smooth gradations. With more control between yellow and red, face tones are natural looking as well. Faces don't seem to teeter between too red or too green so easily.

Anyway, there's more to the technology than just an additional color. Many people are used to working in RGB and dismiss the idea without considering the difference between signals and physical pigments.

The bottom line is that the implementation looks great to my eyes - against both our older models and against the competition. The next time you're in an electronics store, take a look and see what you think.

Nuff said. Disclaimer mode off...

[Andrew Smith]

Hi Jon,

So glad you popped in with your answer direct from the horses mouth (as much as possible), so to speak. If I'm in the market for a new TV and the Sharp doesn't cost too much extra, I'd definitely be swung towards the new tech.

Adding on 3D abilities is comparatively straight forward, whereas this sort of thing (disclaimer: haven't seen one myself yet) is genuine nutting-it-out innovation. Colour me impressed!

Andrew

[Thomas Barthle Jr.]

Originally Posted by Jad Meouchy

No offense to Sharp, but there are many technical reasons that no other manufacturers are spending time on that side of the lab. Frankly it's just a big marketing gimmick..

Its funny to say its a gimmick when the rest of the companies are doing 3d with those ridiculous shutter glasses. I have not looked at this TV in the stores yet but I have looked at the 3d TVs and don't like them at all. There is even talk that after prolonged use of these glasses, REAL depth perception could be affected, possibly permanently.

[Erik Phairas]

Just for giggles here are the pixels of my RGB X-gen panel from my Sony 52EX700. Not the RGBY panel but you get the idea.

http://www.angelfire.com/al/metalali...701_pixel4.jpg

[Jon Fairhurst]

Andrew,

Definitely take a look in person before making your decision. Hopefully, it will look as good to your eyes as it does to mine.

I'm not sure about the Australian models, but in the US, the TVs include a "demo mode" that's worth seeing. If the people in the store haven't lost the remote control, they should be able to activate it.

One thing that I believe is unique to the US is that we get TV ad with George Takei (Star Trek's "Mr. Sulu") and his "oh my" catchphrase. YouTube - Sharp Electronics Quattron quad pixel technology w George Takei

The 4th subpixel isn't just a marketing gimmick. But hiring George Takei certainly is! :)

[Andrew Smith]

I agree with you on George Takei. I'm not right in the middle of American culture, but is there a real benefit to hiring him to be on a commercial? I know that he's been on Star Trek and all, but it's just not working for me. I just don't get it.

Andrew

[Jon Fairhurst]

I think it works in the US. Older people remember him from the original Star Trek, people in the middle know him from the Star Trek movies, and young people know him from his role in Heroes. My sons (18 and 22) and their nerdy friends think he rocks. :)

[Bill Davis]

Isn't the essential core issue ADDITIVE color verses SUBTRACTIVE color?

In printing, more ink varieties work better because you're essentially BLOCKING the reflectance of certain color wavelengths of light with every print path. More color ink variations allow you to block specific light frequencies better e.g. using LIGHT BLUE to block reflectance rather than a pattern of primary BLUE). Creating some colors by printing mixtures of other multiple colors LOWERS efficiency. The more colors you print, the denser and less colorful/bright the image becomes.

In TV, with ADDITIVE color, you derive Yellow simply by firing both blue and green, and as you do so, you INCREASE brightness toward white. A process TOTALLY different from printing.

I think George Takai is very cool, and I hope he gets paid a LOT for the ads. I also admire his risking damage to his career coming out so publicly and I applaud Sharp for hiring him.

But the ads make me immediately think "snake oil for the poorly educated." Sorry, George.

(Either that or I just don't understand the underlying technology, which I freely admit is quite possible.)

[Jon Fairhurst]

Promoting a fourth subpixel has been a real challenge. The quote: "snake oil for the poorly educated" sums up the reaction of many people. The difficulty that we've had is with the moderately educated. I know a number of imaging experts with doctorates here at the lab who have to noodle it out and are only convinced after seeing the diagrams, the math, and the implementation. The cool thing is that once they get it, they start to think about ways to push the concept to the next level. People who understand how RGB is used, but don't have doctorates in color theory are a tougher sell!

The best way to explain it is to show the CIE Color System chart.

The one below shows the limits of color reproduction with three additive pigments. Though you can see the whole range of colors, the reproduction system is only capable of rendering a gamut within the triangle formed by the three points.

If you shift the green pigment to the right, you can capture more yellow, but it comes at the expense of the cyans. Shift green to the left and you can reproduce more cyans, but at the expense of the yellows and golds. You could simply push the green pixel up, but we are limited by the physical pigments that are available.

With Quattron TVs, we add the yellow pixel to the right of the triangle. We then shift the green pixel to the left and slightly toward the curved peak on the CIE chart. The result is a larger gamut as defined by the four points.

And, yes, we lose a small corner right at the current green dot. But we gain a much larger amount in the cyans and yellows. Overall, we can reproduce a wider range of green tones. In the real world, green isn't a single 100% saturated green point, but a whole range of tones that might lean slightly toward yellow or cyan. The gain outweighs the slight loss many times over with real pictures.

We could have added a cyan pixel and shifted the green toward the yellow. That would have also extended the gamut, but we would have lost efficiency. And with an additional white pixel (many have barked up this tree), you gain efficiency, but lose saturation. Because yellow is the brightest color, it allows the best balance between brightness and gamut.

The secret sauce, of course, is in how the RGB signals are mapped to an RGBY (with green pushed toward cyan) space. And that's where the technical tour ends. It's also where one needs to look at the results and judge it with their own eyes.

To summarize the technology:
* It extends the available gamut.
* It uses UV for a wide aperture, giving high efficiency so we can add yellow and maintain brightness.
* Yellow is used because it is both bright and colorful.
* By adding yellow we can shift green, providing more gamut in the cyan area as well.
* It provides 33% more subpixels, which allow more detail to be reproduced.
* It provides more available gradations, since the color space is more finely defined - especially in the shorter path between yellow and red; yellow and green.

Also, note that extended gamut has been a hot topic for some time in the display community. xvYCC was developed specifically for displays with extended gamut: xvYCC - Wikipedia, the free encyclopedia And xvYCC is part of the HDMI spec. HDMI - Wikipedia, the free encyclopedia

But the theory only counts if the implementation is good. That's why the George Takei ad offers the call to action: "You have to see it to see it."

[Bill Davis]

OK, I'm willing to keep an open mind.

But I have to say I'm still confused. Particularly by your use of "pigment" in your explanation. My brain is struggling to separate that from the idea of a SUBSTANCE that works subtractively with light waves.

If you're telling me that adding a 4th generator of LIGHT in the yellow spectrum allows you to generate more colors of light, a more appealing spectrum of colored light, or some other demonstrable qualitative improvement in picture, then I'll continue to reserve judgement until I experience it for myself.

But right now it's kinda like the razor blade wars to me. We're up to what? SIX separate blades in the damn things? And sure it's a fine shave. But will a 10 blade system be "better"? What about 25 blades? Will we someday have a single razor cartridge the size of my entire cheek so I can shave half my face in a single one inch stroke? At what cost?

At heart, while the comfort of my daily shave has certainly improved somewhat over the years. The COST of the razor blades has increased at a FAR faster rate. If someone had told me I'd be paying $10-15 for a pack of razor blades at some point in my life I'd have laughed at them.

I don't laugh, but I certainly resent the fact that I can't go to the drug store and easily buy anything with fewer than 4 blades these days - and at a MUCH inflated price point compared to the older system.

Still I'll reserve my opinion until I learn more.

Thanks for your time, Jon to help us all understand the tech.

[Frank Vrionis]

ah...that razor blade anaolgy sounds terrible.

not all pixels are created equal - they have four different colours in fact. where as razor blades are all the same and they do the exact same job.

[Andrew Smith]

Suffice to say that this wouldn't have been able to be achieved in an analogue world.

It's only thanks to digital processing that the complex colour re-mapping can be achieved in order to make the most of this extra pixel colour source.

Andrew

[Erik Phairas]

I haven't even upgraded to Blu Ray. I am perfectly happy with DVD and you can really see the difference between the two.

This kind of tech is totally meaningless to me. It's cool they are trying new things but it will be VERY hard to make me care.