CurtPalme.com Home Theater Forum

[skippyar]

Has anyone seen the advertisment for Sharp's new innovation and what are your thoughts.

SharpEurope Video Explanation: http://www.youtube.com/watch?v=eiKNzFTTsHo

SharpElectronicsUSA Commercial: http://www.youtube.com/watch?v=4hSnAxvRdmY

Paul.

[dturco]

[skippyar]

Ah right, I did use the forum search option parameters before posting but found nothing under Quattron.

[cmjohnson]

I can see some merit to the concept. It gives a wider color gamut BUT in order to fully exploit it, that wider color gamut has to be used in the production of the signal to be displayed. A matching four-color camera system would be required in order to fully exploit the fourth color.


I think that there is certainly some room for the idea of extending the color gamut. Look at anything which is a fluorescent or day-glow color. Photograph it, record it with a video camera. Print it out or display it on any display device. The result is nothing like what
you see with your own eyes when looking at the original item.

If the color gamut of the record-reproduce system was wide enough, fluorescent colors WOULD be seen with their fluorescent effect intact.


CJ

[skippyar]

I did read this http://www.rgby.org/ though it doesn't really go into great detail.

I wondered what use it would be when source content is derived from RGB colorspace.

Some queries to my questions were reflected here: http://answers.yahoo.com/question/index?qid=20100426223651AAzX2XT



What I do find odd is how Sharp refer to the individual R G B Y diode elements as quad pixels. I thought pixels were collectively derived from the elements, seems an erroneous use of terminology.


Paul.

[winduptoy]

If the red and green were accurate, the yellow wouldn't be needed.

[cmjohnson]

It can also be a matter of efficient use of the colors on hand.

If you filter out yellow as its own separate channel, it allows you to make more efficient use of red and green individually.

You wouldn't have to mix red and green to get yellow, so the red and green channels would not have to work hard when the signal
calls for just yellow.

You could easily extend the concept to a six color system, red, green, blue, cyan, magenta, yellow. Now complete the visual spectrum
with two more colors: A deep violet bordering on long UV, and a super deep red bordering on near IR. With that setup you should be
able to reproduce every color the eye can perceive.


It's probably been done, in a laboratory.

CJ

[skippyar]

[AnalogRocks]

What ever happened to Mitsubishi and the 6 color system?

[AnalogRocks]

What ever happened to Mitsubishi and the 6 color system?

[cmjohnson]

The entire RGB concept is based on how our eyes work, which are RGB devices. But the specific responses of our eyes in the R, G, and B ranges are due to the specific differences between how the cone pigments in our retinas absorb light. Retinal is the prosthetic group for each pigment. Differences in the amino acid sequence of their opsins accounts for the differences in absorption.

So, each specific type of opsin (retinal pigment) has a specific chromatic response characteristic. In order for our TVs and film to give
"perfect" color response, allowing us to see the full range of colors on the screen that we see in real life, the cameras, films, and display
devices would have to use sensors that have matching response characteristics. We don't have that. What we see as red, green, and
blue in our displays is an approximation of the pure red, green, and blue that the eye actually detects but it's not exact. It's reasonably close, though.


Here's a chart showing the actual response curves of the pigments in the eye of someone with normal vision.




Note that the red pigment actually has a very broad response curve and its peak sensitivity is closer to yellow than red.


[img]http://www.ncbi.nlm.nih.gov/bookshelf/picrender.fcgi?book=mcb&part=A6255&blobname=ch21f49.jpg[/img]

http://www.ncbi.nlm.nih.gov/bookshelf/picrender.fcgi?book=mcb&part=A6255&blobname=ch21f49.jpg

You might have to click or copy and paste to see this chart.


CJ

[HogPilot]

I've always wondered why the lowest frequency opsin is referred to as the "red" pigment when it's clearly most sensitive at the yellow wavelength.

[cmjohnson]

Because the lowest frequency it has significant response at is in the red area of the spectrum.

The blue pigment is called blue because it has significant response in the blue area of the spectrum.

Both those colors represent the extremes of our vision's color range. Green falls in the middle, but is most sensitive in the yellow-green range.

The RGB we use in video is a good compromise as there are only so many easily engineered color response curves available for film,
video cameras, and display devices, but it's far from an accurate representation of our vision system and there are some things you can see that our video systems simply can't reproduce, like fluorescent colors. I think that would require extended blue and red range.


CJ

[HogPilot]

[cmjohnson]

It'd be primarily because you can't create colors that are outside of your color gamut.

With red being the lowest range of frequencies that the eye has substantial sensitivity to, and blue being the highest, it makes sense
to choose red and blue as your lower and upper range colors. This allows all colors between red and blue to be created by mixing
them (along with green) in appropriate ratios.

If you were to choose the strongest response colors of the eye, then our TV red would be closer to yellow and it's not possible to create red with yellow, green, and blue light in any combination. This would be a very limited color gamut with no reds.


But, our eyes have a wider color acuity range than the gamut covered by television as we've known it. A truly accurate color
system would have equal or wider total spectral response than our eyes do.


It would be very interesting to obtain a deep violet/near UV camera and mate it up with a CRT or digital projection device outputting similar spectra, and do the same for deep red/near IR and converge the images along with a standard projector. This should be
able to accurately portray every color your eyes can see, including day-glow fluorescent colors.


CJ

[HogPilot]

[cmjohnson]

But..violet isn't indigo, or near-UV.

Look at a Compact Disc (or better yet, if you have one, a LaserDisk) in sunlight. Look at the rainbows you can see in it.

The purple you'll see outside of the blue zone is NOT the same purple you get by mixing red and blue light! Though it's good enough
for television purposes, real spectral purple (Better termed Indigo) is a color that can not be ACCURATELY displayed or simulated with
an RGB system.

Neither does the purple you see by mixing red and blue (typo corrected) actually occur within a contiguous spectrum. But it's not the only color that
isn't in the spectrum. This particular "purple" is really a dark Magenta. Magenta does not occur in a natural spectrum.

CJ

[HogPilot]

[skippyar]

Thanks for the sharing your views and the informative feedback.

Paul.

[macgyver655]

If I had some friends over and we were watching a movie and one of them said, "hey, that violet doesn't look right"....I'd throw him out with a do not return kick......

[HogPilot]