Calibration, Camera ICC, Color System

[lumines]

I notice Pantone has their own color matching, numbering system. What about the color spectrum gamut in Lightjet or Inkjets? (eg. prophoto or rgb)
Based on an exact rgb number within a specific color space, how close will a color match or differ?

Can a lightjet 430 take and print a ProPhoto rgb file? (with a wider gamut than rgb?; capable of 16 billion colors?) Though ideal is a custom, just fitting the gamut.

I have two different cameras, each shoots raw from the company's own algorithm. I'd like to calibrate (ICC) both cameras, so they'll print with a more standard color base.
Is the 24 colorchecker passport too limiting? Or should I collect paint sample swatches as well as different color products to make a test target; profile?

Will this be similar; better than a standard color chart? (colorchecker sg chart? has good with negative reviews) Silverfast has a similar product for digital cameras, (studio target, pocket target). Reviews on this?
Is the surface on these charts more accurate in construction in determining rgb values?

Should each lens photograph the color target charts, or just the camera body with whichever lens?

Upon photographing the targets, what software or process is best to match then find discrepancies in the printer's test profile;target;chart and a person's own camera profile target charts ? (achievable in all printer types? eg. inkjet's, lightjets, mimaki)

[digitaldog]

The Pantone system (and there are a number of variants within the system) is really for a CMYK process. That isn't to say you can't reproduce some of the colors on such a device but the trick will be getting the proper RGB values that produce these CMYK color appearances.

IF you profile the Lightjet then you have the decoder ring that could, with the proper software (something like ColorThink Pro) tell you numerically how close another converted color (in this case CMYK or process Pantone color) can be on that Lightjet using deltaE values. Some products can simulate visually on-screen the differences but that takes the display into the loop and the hard numbers are a better way to know if a spot color is close or way off from your goal.

Forget number of colors. That's encoding and has nothing to do with color gamut (ProPhoto versus the gamut of say the Lightjet itself).

Just what are you trying to figure out in terms of this RGB output device?

[Ellis Vener]

I have two different cameras, each shoots raw from the company's own algorithm. I'd like to calibrate (ICC) both cameras, so they'll print with a more standard color base.
Is the 24 colorchecker passport too limiting? Or should I collect paint sample swatches as well as different color products to make a test target; profile?


Forget ICC calibration for cameras.

There are currently three different "camera color calibration" systems.

1) Adobe DNG Profile editor (free @ https://www.adobe.com/cfusion/entitlement/index.cfm?e=labs_dngprofileeditor)

2) X-rite Color Checker Passport. http://xritephoto.com/ph_product_overview.aspx?id=1257 The software alone  is still free (go to http://xritephoto.com/ph_product_overview.aspx?id=1257  >Support> Downloads) but you can also buy the ColorChecker passport target set.

Options one and two require an X-rite 24 patch Color Checker target. A small version of the 24 patch  is in the ColorChecker Passport target set which also includes a biasing target for both landscape and portrait work and a large white balance target, all in a pocketable hard plastic shell.
Both require you to shoot the target in raw mode, open it in either Adobe Camera Raw or Lightroom, export the photo as a DNG and use the respective software to build a "Camera Calibration" profile which you name and install. You then quit ACR or Lightroom and when you restart either the profile should appear in the Camera Calibration menu. In Lightroom you can create a preset for applying the profile during import.

An option with both the Adobe and X-Rite CCPP software is the ability to build a dual illuminant (flash/daylight and 3200˚K)profile which works well when working in mixed light conditions. This requires shoot the target under both types of illumination.

In Lightroom (at least you can apply the profile to proprietary raw formats, DNG raw format (if you use that) or JPEGs,

3) The Datacolor Spyder ColorCHECKR - target and software. Instead of using the Camera Calibration settings you build an H/S/L profile for use in Lightroom or Adobe Camera Raw.

I have used all three. All three do a fine job, but it is like ordering vanilla cream from three different creameries - there are variations between the two. When tuning a new camera these days I will build profiles at least two ways (usually Xrite CCPP and Datacolor ColorCHECKR) and see which works best. The important thing is for you to decide which calibration system you want to use for the cameras you want to co-ordinate color with.

Keep in mind that different camera makers have their own built in bias and I do not think it wise to expect o get completely identical color out of a mixed brand set of cameras. a close to very close match but not an exact match, and that  may even be true of different generation cameras of the same brand (examples : a Canon EOS 1D mark III and a 1D X or a Nikon D3x and D800)

What exactly are you photographing? What type of lighting and how varied is the lighting from session to session?

[Vladimirovich]

Forget ICC calibration for cameras.

if you can't profile that does not mean that others can't.

There are currently three different "camera color calibration" systems.
1) Adobe DNG Profile editor
2) X-rite Color Checker Passport.
3) The Datacolor Spyder ColorCHECKR

for an "expert" you for example totally missed QPCard 

[Ellis Vener]

if you can't profile that does not mean that others can't.

for an "expert" you for example totally missed QPCard  

I didn't say it couldn't be done but my experience is that unless I was  constantly shooting under the same lighting conditions it wasn't worth the hassle,.
I was writing only of the systems I know and have used. I haven't tried the QPCard system and frankly prefer not to be in constant measurebation mode testing new calibration systems all of the time. I'd far rather be making photographs of things other than targets. Also as I have some very exacting and color critical clients and they continue to be extremely happy with what I am delivering - far happier than with other photographers they have worked with -  I don't feel any pressure to change.

[digitaldog]

if you can't profile that does not mean that others can't.

It can be done, it appears to be an inconsistent process (hit or miss). I've seen many cases where a supplied 'canned' camera profile works much better than a custom one. There are a number of variable that could account for this. But with display and output profiles, even scanner profiles, I've never seen this unless bad measurement data or a bug in the profile software surfaces. Cameras are not scanners so I suspect treating them like one is a large part of the issue. That we have to supply a rendered image (output referred) is another possible issue. There's a lot of baked processing going on which while true for a printer, a camera isn't a very consistent process and that may account for the issues with ICC camera profiling.