Is there a best practices in adjusting the CMS on a Panasonic 65VT50? - AVS Forum
Display Calibration > Is there a best practices in adjusting the CMS on a Panasonic 65VT50?
JimP 06:47 AM 04-07-2013
I've come across information for adjusting CMS that is somewhat in conflict with each other and would like to know if there is a best practices in adjusting the CMS on a Panasonic 65VT50.

PlasmaPZ80U's Avatar PlasmaPZ80U 09:37 AM 04-07-2013
Quote:
Originally Posted by JimP View Post

I've come across information for adjusting CMS that is somewhat in conflict with each other and would like to know if there is a best practices in adjusting the CMS on a Panasonic 65VT50.

If you're referring to various saturation/luminance combos (like 100/75, 100/100, and 75/75), it would probably be best to start with 100% saturation patterns (75% or 100% stim based on preference/methodology) and then do saturation sweeps to evaluate overall error across various saturation levels.

Keep in mind the errors at 100% saturation (and also 100% stim tend to be more visible with program material than errors at lower saturations/stim).

You can also do luminance sweeps to get a better understanding of how varying stim affects the gamut at 100% saturation.
Wouter73's Avatar Wouter73 10:41 AM 04-07-2013
What's the conflicting information?
JimP 11:06 AM 04-07-2013
Thanks for your response but the question is about adjusting the CMS on this display. FYI, calibrating CMS at the 75% saturation and 75% intensity point works best for this display.
sillysally's Avatar sillysally 12:12 PM 04-07-2013
Quote:
Originally Posted by JimP View Post

I've come across information for adjusting CMS that is somewhat in conflict with each other and would like to know if there is a best practices in adjusting the CMS on a Panasonic 65VT50.

Jim, I think "conflict" is the key word in your question.

And if you are talking about really taming Blue in your VT50, imo it can't be done using Custom mode or ISF mode controls. The 6 point color (R,G,B,C,M,Y) CMS are just a basic control if done right for CMS.

From everything I am seeing, you need to do a LUT cube type "CMS" calibration. In theory the more points the better depending on your display, I am now up to 4096 points for my LUT calibration that I just ran last night using Calman and its defaults for my meter, meaning low light handler 30 reads and five reads for normal mode. I haven't seen or run tests with my S&M disc but just going by the dE's for this calibration it came out pretty nice.

ss
JimP 12:53 PM 04-07-2013
Hi SS,


Maybe I should clarify.

In one tutorial about adjusting CMS, (not display specific) it's recommended that you first get the brightness of your primaries done first. Then it recommends getting the hue of the secondaries adjusted. Then go back to the primaries and adjust hue and saturation and correct any shifts in brightness. Then do brightness and saturation of the secondaries and recorrecting for changes in hue.

Another source just shows you going through each of the primaries and secondaries and getting brightness, saturation and hue done at each color point at the same time..

Just thinking out loud, it would seem that due to the interaction between color points, you might want to start with the ones that's out the most. Correcting those first may get the ones that are out the least to properly fall in place.....but again, this is just thinking out loud.
sillysally's Avatar sillysally 01:03 PM 04-07-2013
Jim, here's the problem they are both correct.
However I will add imo you should keep going back and forth on the primary's and secondary's until they are in harmony with each other, and don't forget about the 75% white (keep a eye on it).
Oh and in doing the above, stay out of the rabbit hole (don't make large changes, keep it baby steps). smile.gif

ss
JimP 01:15 PM 04-07-2013
SS,

Unless I introduced an error somewhere and didn't notice it, I got different settings using the first 2 methods. IMO, they can't both be right.....can they?
airscapes's Avatar airscapes 01:47 PM 04-07-2013
Quote:
Originally Posted by JimP View Post

SS,

Unless I introduced an error somewhere and didn't notice it, I got different settings using the first 2 methods. IMO, they can't both be right.....can they?

Jim, read this article by Michael, it will make things a little clearer
http://www.tlvexp.ca/2013/03/color-management-system-pie-eat-half-or-all/
JimP 02:43 PM 04-07-2013
Quote:
Originally Posted by airscapes View Post

Jim, read this article by Michael, it will make things a little clearer
http://www.tlvexp.ca/2013/03/color-management-system-pie-eat-half-or-all/

Interesting article but still doesn't address the question I'm asking.

When calibrating the CMS of a Panasonic 65VT50, is there a best practices to the routine of adjusting Brightness, Saturation and hue within the CMS for each primary and secondary color.
Wouter73's Avatar Wouter73 03:28 PM 04-07-2013
If you try both, and do a gamut sweep after, you'll know...

For my vt50, I just did brightness saturation and hue for each color rgbycm at 100/100 patterns, just rerun first red, the green, ten blue etc etc as many times as needed (two, three at most). After when I did a sweep, only cyan wasnt totally linear, all other colors pretty much were. And as the article from michael says, you won't see the difference...
Chad B's Avatar Chad B 06:06 PM 04-07-2013
Quote:
Originally Posted by sillysally View Post

And if you are talking about really taming Blue in your VT50, imo it can't be done using Custom mode or ISF mode controls. The 6 point color (R,G,B,C,M,Y) CMS are just a basic control if done right for CMS.

From everything I am seeing, you need to do a LUT cube type "CMS" calibration. In theory the more points the better depending on your display, I am now up to 4096 points for my LUT calibration that I just ran last night using Calman and its defaults for my meter, meaning low light handler 30 reads and five reads for normal mode. I haven't seen or run tests with my S&M disc but just going by the dE's for this calibration it came out pretty nice.

ss
I respectfully disagree. As you can see from the ISF Day and ISF Night modes calibration reports in this post, which shows saturation sweeps and gamut luminance sweeps, taken without the help of a LUT type calibration, the blue result is textbook perfect (ISF Day is a little more perfect in the blue than ISF Night, but even Night leaves little room for complaint).

As to Jim's question, personally I prefer to run through each color in order, doing sat, hue, and lum for each color before moving on to the next. At least 2 passes should be taken, particularly to nail down the red and blue.
sillysally's Avatar sillysally 06:40 PM 04-07-2013
Quote:
Originally Posted by Chad B View Post

I respectfully disagree. As you can see from the ISF Day and ISF Night modes calibration reports in this post, which shows saturation sweeps and gamut luminance sweeps, taken without the help of a LUT type calibration, the blue result is textbook perfect (ISF Day is a little more perfect in the blue than ISF Night, but even Night leaves little room for complaint).

As to Jim's question, personally I prefer to run through each color in order, doing sat, hue, and lum for each color before moving on to the next. At least 2 passes should be taken, particularly to nail down the red and blue.

Well then I must respectfully disagree with you.

Have you ever seen a good 771 color point (or higher) LUT calibration ???

You may want to to offer the advanced LUT cube calibration to your clients. I know you have the software for the larger LUT calibration, but your meters may not cut it for this type of calibration at-least in your time frame that you are willing to spend doing your standard calibrations.

I am trying to get a good LUT 4096 calibration using the latest Beta for Calman pro, but am having sync problems when doing that big of a cube Lut. Do you have any idea what could be causing this, and I am asking from a stand point of your hands on experience using Calman's latest Beta for a 4096 LUT calibration.??

btw, when you do a standard 10 point RGB/gamma/grayscale and a 6 color point (RGBCMY) do you use your Jeti for the full and complete calibration or do you use a lessor meter some of the time?? .

I am glad to see we basically agree on Jims question.

ss
buzzard767's Avatar buzzard767 06:53 PM 04-07-2013
Quote:
Originally Posted by JimP View Post

Hi SS,


Maybe I should clarify.

In one tutorial about adjusting CMS, (not display specific) it's recommended that you first get the brightness of your primaries done first. Then it recommends getting the hue of the secondaries adjusted. Then go back to the primaries and adjust hue and saturation and correct any shifts in brightness. Then do brightness and saturation of the secondaries and recorrecting for changes in hue.

Another source just shows you going through each of the primaries and secondaries and getting brightness, saturation and hue done at each color point at the same time..

Just thinking out loud, it would seem that due to the interaction between color points, you might want to start with the ones that's out the most. Correcting those first may get the ones that are out the least to properly fall in place.....but again, this is just thinking out loud.

I swear I saw pretty much this same post a couple months ago. Was it you?

Quote:
Originally Posted by Chad B View Post

I respectfully disagree. As you can see from the ISF Day and ISF Night modes calibration reports in this post, which shows saturation sweeps and gamut luminance sweeps, taken without the help of a LUT type calibration, the blue result is textbook perfect (ISF Day is a little more perfect in the blue than ISF Night, but even Night leaves little room for complaint).

As to Jim's question, personally I prefer to run through each color in order, doing sat, hue, and lum for each color before moving on to the next. At least 2 passes should be taken, particularly to nail down the red and blue.

Absolutely. What possible benefit to doing it any other way?
Chad B's Avatar Chad B 07:05 PM 04-07-2013
Quote:
Originally Posted by sillysally View Post

Well then I must respectfully disagree with you.

Have you ever seen a good 771 color point (or higher) LUT calibration ???

You may want to to offer the advanced LUT cube calibration to your clients. I know you have the software for the larger LUT calibration, but your meters may not cut it for this type of calibration at-least in your time frame that you are willing to spend doing your standard calibrations.

I am trying to get a good LUT 4096 calibration using the latest Beta for Calman pro, but am having sync problems when doing that big of a cube Lut. Do you have any idea what could be causing this, and I am asking from a stand point of your hands on experience using Calman's latest Beta for a 4096 LUT calibration.??

btw, when you do a standard 10 point RGB/gamma/grayscale and a 6 color point (RGBCMY) do you use your Jeti for the full and complete calibration or do you use a lessor meter some of the time?? .

I am glad to see we basically agree on Jims question.

ss

I've done 125 pt with the Lumagen; I haven't yet done or seen the ee color box or similar except for seeing it demoed at the CEDIA trade show, but that's not saying much. However, that isn't relevant to our disagreement, which stemmed from you saying you didn't think blue could be tamed without the help of that equipment and procedure.

I have nothing against it. I have no time limits on my normal calibration, only on my cheaper "Geek Buster" calibration service.

I use my Jeti for everything unless either there is too much ambient light or I need to take an accurate black reading on a high contrast display. In those situations the profiled C6 does nicely.
Chad B's Avatar Chad B 07:12 PM 04-07-2013
Quote:
Originally Posted by buzzard767 View Post

Absolutely. What possible benefit to doing it any other way?
None that I know of. I don't see an advantage to the alternate method mentioned in post #6.
sillysally's Avatar sillysally 07:35 PM 04-07-2013
Quote:
Originally Posted by Chad B View Post

I've done 125 pt with the Lumagen; I haven't yet done or seen the ee color box or similar except for seeing it demoed at the CEDIA trade show, but that's not saying much. However, that isn't relevant to our disagreement, which stemmed from you saying you didn't think blue could be tamed without the help of that equipment and procedure.

I have nothing against it. I have no time limits on my normal calibration, only on my cheaper "Geek Buster" calibration service.

I use my Jeti for everything unless either there is too much ambient light or I need to take an accurate black reading on a high contrast display. In those situations the profiled C6 does nicely.

Well then you don't know what you are talking about. Doing a 125 point Lut cube calibration using a Lumagen Radiance 3D is just a start and very easy to do.
btw just because you use a Lumagen doesn't mean it will store and processes a 3D LUT cube calibration, even if its just a small LUT (125 point).

I use my Radiance Mini 3D for generating triplet patterns not single patterns that you use . However before Calman started to support eecolor I used my Radiance Mini 3D for the small LUT cubes.

Yes I have a C6 but imo it just will not compare to my K10-A in any of what you are using your C6 for.

Yes all this is relevant to my statement.
1. I said "really taming Blue" ( I am sure your standard calibration also helps tame blue but not to the point I am talking about). That leads to my second point.
2. At-least a 771 color point LUT, that's what "really taming Blue"

You seem to be a decent man, and try and help others, if you want I can help you if you want to stop by my house I will be happy to show you how all of this works. wink.gif

Also understand there is one differences between you and me, I don't do calibrating for money so I am void of most suspicion ( not saying you have any ulterior motive) smile.gif

ss
Chad B's Avatar Chad B 07:53 PM 04-07-2013
The fact that I haven't done a 771 point LUT does not mean I don't know what I'm talking about, when I just proved in my ISF Day mode calibration that blue can be really tamed on a VT50 without it. That was my point. So you're saying that a maximum dE2000 of under 1 of any blue measurement whatsoever isn't really tamed?
sillysally's Avatar sillysally 08:50 PM 04-07-2013
Quote:
Originally Posted by Chad B View Post

The fact that I haven't done a 771 point LUT does not mean I don't know what I'm talking about, when I just proved in my ISF Day mode calibration that blue can be really tamed on a VT50 without it. That was my point. So you're saying that a maximum dE2000 of under 1 of any blue measurement whatsoever isn't really tamed?

Don't get huffy, this is just a friendly point counter point. My god man I have even offered to help you.

However you proved nothing when you are just talking about dE's you should know that. I am surprised you are hanging your hat on some ISF Day calibration dE's that you did on a 55VT50. btw Jim and I both have 65VT50, and as you so stated there is a difference between the 55 and 65VT50 in the total outcome of the calibration. .
You being a "Pro" calibrator should know you can have a calibration that has very low dE's as is the case with the ones in my sig, but still there is no guarantees that the PQ is that great or in this case the Blue is tamed as much as I am talking about or probably not even that close..

So until you actually understand (and it seems you don't) this advanced type of calibration and see it in your home display viewing many well mastered Bly Ray movies, you shouldn't get involved in this type of a friendly point counter point discussion. .

Do you honestly believe a 6 color point as (pictured below) can achieve what a 4096 color point can achieve (pictured below). Please take a look at these to examples of color points and tell me.wink.gif



btw, are you saying you will (once you learn how to do this type of calibration) do a ISF Night and Day 4096 LUT's plus a 3D 4096 LUT calibration for the same money as your non ""Geek Buster calibration service" whatever that is. (no need to explain to me).


ss
buzzard767's Avatar buzzard767 06:11 AM 04-08-2013
Quote:
Originally Posted by sillysally View Post

a 4096 color point can achieve (pictured below). Please take a look at these to examples of color points


I looked, and what I see is a 4913 point cube, not 4096. biggrin.gif
airscapes's Avatar airscapes 06:24 AM 04-08-2013
Hey SS, in case you had not noticed, the OP, you know the one who started this thread with a question.. he don't have a LUT box, so why hijack his thread with this crap!

OP asked
I've come across information for adjusting CMS that is somewhat in conflict with each other and would like to know if there is a best practices in adjusting the CMS on a Panasonic 65VT50.

Notice no LUT box or question pertaining to LUT box..
markrubin's Avatar markrubin 06:29 AM 04-08-2013
move on please
Doug Blackburn's Avatar Doug Blackburn 12:14 PM 04-08-2013
JimP...

One thing you have to understand about written calibration "procedures" (like "do color luminance first, then do ___________, then do _______________) are not EVER requirements. And they are unlikely to apply for EVERY calibration... in fact they often don't even apply to MOST calibrations. And in some cases, for some video displays the any given written procedure could be WRONG in that it won't work for one reason or another.

Your job as a calibrator is to START SOMEWHERE, pay attention to how the TV responds. If it does something you don't expect, figure out what it is doing that is different than your expectation and WORK AROUND it by modifying the procedure you were going to try to follow in any way you think might be more appropriate for THAT TV/projector. A Pro calibrator with a lot of experience is likely going to make those decisions quicker and with more certainty that someone who has taught himself calibration and has calibrated 1 or 2 different video displays. There is an advantage to just having to know 1 or 2 video displays... you can spend a LOT of time on them figuring out their response and how to best calibrate those 2 displays. And it will take a lot of time. More time than a Pro can spend on a calibration in a single work day.

LUT calibrations are, undoubtedly, the bomb when they work right. And more points are always better than fewer points. 125 points is a HUGE advantage over 26 points or 16 or 8 points you calibrate while doing a conventional calibration. Thousands of points is better than 125 points (obviously, duh - does not require an argument or telling someone they don't know what they are talking about). But there are only two devices that currently support video calibration LUTs for home theater... Lumagen Radiance processors, and the eeColor box, neither is particularly cheap. If you were using an HTPC with the right sort of video system, you might be able to have a software app for LUT calibration with thousands of points, but I'm not aware of such an app for home computers existing today. HTPCs have their own set of problems though, and a lot of people simply don't want to deal with a home computer being the only source component for their home theater system. Consumer TVs today don't directly support LUT calibrations so if you don't have a Radiance or eeColor box and software capable of creating the appropriate LUT to load into the box you are using, all the discussion about LUTs is un-helpful in the extreme.

Back to the original topic - It boils down to nothing about calibration ever being "this way" and "this way" only. You have to find your way. Remember there was one Harry Potter movie (and book) that described how the corridors and stairs would move constantly so you always had to discover a new way to get to where you were going? Same thing with calibration... you may find doing grayscale first works most of the time, but are you paying enough attention to what the TV is doing to realize that there may be a time when you have to do CMS before grayscale in order to get optimum results? This is all part of the art of calibration. Calibration is a process that is constantly fluid and variable. That's why there has never been a single place you could go to find written calibration instructions that will lead you to the best possible calibration every time. With the current state of things, that's simply impossible. Every calibration procedure you ever see written down is guaranteed to NOT WORK just about as often as it DOES work (some variation there, of course... some written things may only work 25% of the time, while others might work 75% of the time).

The next issue calibration has is how do you define whether a process worked or not? Take good calibration software like CalMAN... up until version 5, you verified your calibration by looking at your results for 2 grayscale adjustments, 10 (or 11) grayscale adjustments or 21 grayscale adjustments (depending on the display and video processor being used) and by looking at 6 colors. CalMAN 5 comes along and makes it easy to automate checking calibration results in FAR more detail... with so much more detail, you can find problems you didn't even know existed if you had an earlier version of CalMAN (unless you did extensive manual measurements of points you never look at with the 2, 11, or 21 step grayscale and 6 point CMS the previous versions used. With version 5 you now have automated confirmation options to check MANY more points. Now you may see problems you never saw with previous versions. That might lead you to find out (eventually, after much trial and error) that 75/75 color patterns work better for your TV than more conventional 100/75 patterns.. Problem is, until someone has TRIED this and found it works on multiple samples of the same video display (with measurements to show that it works), then documented it somewhere (probably specific to the video display in question as this may be a WORSE procedure for other models or brands)... you are out there on your own having to try different options to see how they affect your final (and much more complete) post-calibration measurements. And you still have to visually evaluate the images to make sure something bad didn't pop up during the calibration process (example, I've seen Toshiba video displays that would have HORRIBLE images loaded with block artifacts and posterization if any of the color calibration controls were moved more than +/- 3 (control range +/- 10) and just a year or 2 earlier, Toshiba TVs had CMS controls with ranges of +/- 30 and you couldn't use more adjustment than +/- 10 before the images would turn to total crap. You won't find that written down anywhere unless you read my reviews and kept a copy or made notes). Every video display is a puzzle and there's no one solution that is necessarily the answer for any given model because there might be 2 or 3 or 6 firmware versions released during a production run and they might all require different techniques.
Up
Mobile  Desktop