General CMS Question!!!

Do you also have a graph that shows the results if you calibrate at 100%?

In any case, I believe the fast majority of content is between the 30 and 80%, so your display should like good with general content.

First off, probably not. Just cause it has buttons does not mean they will work correctly nor does it mean the display is designed and built to high enough tolerances to be linear.

Secondly, you didn't mention what meter you are using, your results can only be as accurate as the equipment.

What display was this on?

Which colour spaces have you tried this calibration in? I know that the RS4810 has quite a few options compared to my X35, so perhaps one is better than the one you've used? I found Standard was pretty close on my X35, but the blue luminance was about 30% under according to Chromapure (i1D3), I use the Mini3D for the 125 point calibration, so it seems to have taken care of this for me, so I can't help regarding JVC's CMS.

What are the pre-calibration xyY values? It looks like green is undersaturated, so that will not be fixable, unless you can start at another preset that gives you more room to work.The key is to have a good starting point, which is defined by no undersaturated colors and no colors whose luminance is too low. White and black levels need to be set properly and the grayscale and gamma errors should be modest, not huge--in short a decent preset without undersaturated or low luminance colors.

If the conditions TomH mentions cannot all be made correct or if the display just has an inherrent non-linearity, you MIGHT get rid of some of the error by calibrating with one of the new LUT-based solutions that require either CalMAN v.5 and a Lumagen Radiance video processor or LightSpace software and an eeColor box (CalMAN is working on support for the eeColor LUT calibration also, but I don't think it is finished yet). LUT-based calibration takes more measurements at more different places within the 3D color space and creates many more corrections that (help) resolve issues like non-linearity issues your graph shows (larger errors at 100% sat).

It really depends on the CMS. On everything I've used, Samsung comes the closest to being ideal for an internal CMS. JVC has issues, Sony and their RCP has issues if you go too far, BenQ they interact too much so you can't get them right no matter what, etc...

Adding on to what people said about the external LUT tools, I have the Sony VPL-HW50ES here now. With RCP on the Sony, I get the best results using 75% saturation, keeping the controls to +/-5 and not the +/-30 they allow (bad, bad calculations happen if you push them too far). Using a SpectraCal ColorBox, I can do almost nothing to the Sony and use the ColorBox to get it virtually perfect (I don't have a Radiance to test). So really, most CMS systems in displays and projectors are OK to Decent, and can get you much closer than you would have before (and are light years ahead of the CMS systems in computer monitors), but almost always have that trade-off that you're seeing somewhere.

As far as I know with my limited knowledge if your tv doesn't track the colors well across the rest of the gamut there is nothing you can do to make it better if you calibrate to 100% saturation. It's a trade off, having 3/4 of the gamut correct as opposed to having only the outer limits correct.

It would seem that you'd still be within a dE of 3 going slightly above a 75% saturation level.

Then the question becomes how much content falls between that level and 100%.

Does anyone know?

Trying to analyze content is one thing I would love to be able to do, but haven't found a way to do it to this point. I imagine it would be possible to rip a Blu-ray disc and write a tool to analyze it, but that's probably a bit beyond my capabilities as much as I want to know.