FORUMS: list search recent posts

# Sharp RGBY

VIEW ALL
 Sharp RGBY on Mar 31, 2010 at 9:43:19 pm

Hi Bram....

Would love to know your take on this technology - no one I know explains this kind of stuff better.....

PW

 Re: Sharp RGBYon Apr 1, 2010 at 2:52:51 pm

Hi Peter. A couple of disclaimers before giving you my take on this:
1. I have NOT seen a Quatron display, yet.
2. It is NOT something I have spent much time researching because Quad Pixel displays are only available in 40" + sizes at this point and FSI has no immediate plans to pursue large format monitoring solutions.

So now for my 2 cents on this, though I certainly stand to be corrected by those that have spent more time looking into this technology:

I find this to be an intriguing development. I've seen a lot of talk about how this has a huge impact on achievable color gamut, but I've seen little in the way of facts/figures to back this up. However, what I find to be most intriguing about this technology is that without having to increase sub pixel bit depth the number or achievable distinct colors shoots up exponentially. Now granted a 4th sub pixel (yellow) is added and this is probably no insignificant technological accomplishment, but basically we will now have two types of 10 bit panels on the market, those capable of producing over 1 Billion colors (traditional RGB displays) and those capable of producing over 1 Trillion Colors (Quad Pixel displays). This is a real and provable performance benefit of the Quad Pixel technology, at least mathematically.

Just to break this down in detail in case the above is a bit unclear:

Traditional 10 Bit RGB displays give you the following:
2^10 (1024) steps per sub pixel. With 3 sub pixels (RGB) per pixel you get over 1.07 Billion possible colors (1024*1024*1024).

Quad Pixel 10 Bit displays give you the following:
2^10 (1024) steps per sub pixel. With 4 sub pixel (RGBY) per pixel you get over 1.099 Trillion possible colors (1024*1024*1024*1024).

So the result is that while the per sub pixel bit depth does not increase the overall number of achievable distinct colors does increase as a function of having 1 additional sub pixel. To put the ramifications of this in context think about this, a 12 bit panel based on traditional RGB sub pixel technology will still come nowhere close to being capable of producing the number of colors on a 10 bit Quad Pixel based system. Though let me be clear that I am not convinced that a 10 bit Quad Pixel system would actually outperform a 12 bit traditional system.

All sounds good right? I'm not so sure. There are some fundamental issues to consider:
1. What use is a Quad Pixel 10 bit display, or even a traditional 10 bit display, for home users when they are watching broadcast TV at the 8 bit level? Maybe with image enhancement techniques and a powerful interpolation engine the display can 'invent' colors not really in the signal to reduce banding and produce stunning images, or at least what the average untrained eye might consider stunning.
2. For professional use there is a similar concern. If you feed such a consumer TV a professional 10 bit signal how does it map the 1.07 Billion possible colors within the signal to the 1.099 Trillion available on the TV? Will the display invent colors through an interpolation engine, will it limit the display to producing only 1.07 Billion Colors, etc.
3. Are you willing to pay a premium to have 1.099 Trillion Colors as opposed to 1.07 Billion Colors? How many distinct colors do you think you can see and how many colors does it take to completely eliminate perceivable banding (hint: you don't need 1 Trillion!)
4. If, as many reviews and PR pieces imply, the main real advantage is extending the achievable color gamut of the display (to show you colors not previously possible to show on a TV) does that translate into a real advantage if you are monitoring in REC 709, EBU, or other defined colors spaces? That is to say, are most of the gained colors outside of the color space the user should be viewing in?

One thing is clear, from this point forward bit depth and quoted number of achievable colors is going to become an even more confusing topic. Quoted bit depth is going to have to be qualified by many other specifications: is dithering used and if so what type, is this a traditional or Quad Pixel system, how much bit depth is used to manage color temperature or is color temperature managed exclusively by the backlight, and probably many other qualifiers I haven't even thought of yet...

The technology behind this is exciting and it hints at other possible developments down the road with 5, 6, or 7 sub pixel per pixel display technologies that will be able to create virtually all real world colors. Of how much real world value this is now I do not know, but my guess would be very little. There are wide gamut applications, both in the professional and consumer worlds, that could certainly benefit from this type of display technology, but the reality is that for the majority of professional and consumer uses out there this is likely overkill. In the professional world especially there are serious technical caveats to consider/evaluate with respect to this type of technology. On the consumer side though it gives companies something new to hype and ultimately if it makes the image prettier, though maybe not 'accurate', then it will likely reap these companies rewards for their admittedly innovative work.

Bram Desmet
FSI (Flanders Scientific, Inc.)

 Re: Sharp RGBYby Tim Kolbon Apr 19, 2010 at 3:20:02 pm

Hi Bram,

I desperately wanted to get to your booth at NAB and simply didn't get there.

On the addition of the yellow "valve" (I assume the yellow is in the LCD matrix and is transmissive with the LED emitter being nothing much different than what is out there presently)...

Since a typical television color space based on Y'CbCr would create most of the colors that would be out of gamut for RGB in the yellow (ish) neighborhood, do you think it might be based on helping an RGB-based display to better carry those shades?

...just a guess. I've been watching Mr. Sulu pitch this myself and have been thinking about it a bit.

TimK,
Director, Consultant
Kolb Productions,

 Re: Sharp RGBYon Apr 20, 2010 at 3:26:34 pm

As far as I can tell the technology claims to have two main benefits:

1. Producing colors not within the achievable gamut of traditional RGB displays. Not really very interesting if your goal is to work within a defined color space such as Rec 709 that is already completely covered by traditional display technologies.

2. Producing more colors within a defined space, over 1 trillion compared to around 1 billion for RGB 10bit displays. Again, this is what I find most intriguing, but the question is where does the display get the information from? It must interpolate colors from a 10 bit RGB based color system only capable of defining around 1 billion colors to the display's native 1 trillion color capability. I have no objection to this for consumer devices as it is just another display enhancement technology to tack onto the long list of other consumer image enhancement techniques. However, this is not something I expect to see being accepted for professional use anytime soon. In short: I'd love to have one in my living room, but I'd hate to try and sell it to a colorist, which of course is a moot point because this a consumer technology being marketed for consumer applications and not professional environments.

Bram Desmet
FSI (Flanders Scientific, Inc.)