ColorChecker PassPort and CS5-ACR questions.

[buggz]

Hello,
I am trying to understand the PROPER usage of the ColorChecker PassPort.
I've watched some videos of it's use, and lightly went through the manual.
Seems simple enough.

My questions:

 - Do you keep ALL sliders in ALL tabs of ACR to the default values?
   What if some of the color squares are overexposed?
 - Then do you then adjust values in ACR?
   I would guessing changing anything in ACR to generate the DNG kind of defeats the purpose of the target?
Especially changing values in the Camera Calibration tab.
Yes, I should take the time and perform a proper exposure each time, instead of hurrying along.
That is probably my biggest fault, trying to hurry through everything.

Another question:
 - Are you take another target photo for different lens changes during the shooting session?

[digitaldog]

1. Yes*.
2. If its way off, the software should pop a warning you are over exposed.
3. You just leave everything alone and let LR’s Export module for Passport do the work (or you can use the standalone software which is necessary to build dual illuminant profiles anyway*)

* I suspect but don’t know for sure that the export module in LR kind of ignores all the presets anyway, it just sucks up the DNG like the standalone and builds the profile. We’d need someone from X-Rite or better, Eric of Adobe to conform this.

[wolfnowl]

In the last Colour Checker Passport webinar I watched (with Crash Taylor), he recommended making a profile for each sensor (camera body) for each ISO used in each light.  He only works with a few specific ISO values for that purpose.  Now you get to decide if daylight at sea level is the same as daylight at 10,000 feet, and if the daylight in say, LA is the same as the daylight in Rome!

Mike.

[digitaldog]

In the last Colour Checker Passport webinar I watched (with Crash Taylor), he recommended making a profile for each sensor (camera body) for each ISO used in each light. 

Each sensor, no question, each ISO (and lighting) seems overkill, at least for many users. I guess if you doing commercial work where you have the time and control to shoot a Macbeth for each setup, great. Otherwise I find one for daylight, one for tungsten and individual profiles for “weird-ass” illuminants is fine. Keeping track of dozens upon dozens of DNG profiles could get messy too.

[buggz]

Thanks for the replies and information!

I found one answer in the manual, going thriugh it now.
Exposure .
The ColorChecker Classic should be properly exposed in the image.
If color channels are clipped you will not be able to make a profile with the image.
 
When creating a profile, correct exposure in the camera is necessary.
Using your software to correct bad exposure will not make it usable.


I would be pretty certain that you aren't supposed to change the sliders in ACR, even though I haven't found the manual to show that as yet.

[PhilipCummins]

Each sensor, no question, each ISO (and lighting) seems overkill, at least for many users.

I would have thought doing a single pass of profiling through the ISO's on the known settings would be able to determine how the sensor reacts with varying levels of sensitivity to light that you could apply later on once you know what ISO was used.

Now you get to decide if daylight at sea level is the same as daylight at 10,000 feet, and if the daylight in say, LA is the same as the daylight in Rome!

It almost certainly won't be... just take a really bad smoggy day in LA; that would throw the quality control right out of the window!

[bjanes]

1. Yes*.
2. If its way off, the software should pop a warning you are over exposed.
3. You just leave everything alone and let LR’s Export module for Passport do the work (or you can use the standalone software which is necessary to build dual illuminant profiles anyway*)

* I suspect but don’t know for sure that the export module in LR kind of ignores all the presets anyway, it just sucks up the DNG like the standalone and builds the profile. We’d need someone from X-Rite or better, Eric of Adobe to conform this.

Any clipping of color channels in the exposure could easily affect color balance. I would not advise use of recovery, since it is nonlinear. A small amount of positive exposure in ACR might be permissible, but there are reports of hue twists with the exposure control and this could affect color balance. Negative exposure can not accurately correct for clipping in the color channels, but small amounts of positive exposure should be permissible if the exposure control is linear.

I would be interested in the opinions of Eric Chan, Thomas Knoll, and Jeff Schewe regarding this matter. In my work with Imatest Color check, exposure is important since delta E (the standard measure of color accuracy) involves differences in luminance, chroma, and hue. Increases in chroma are often desirable as evidenced by the success of Fuji Velvia back in the film days, but changes in hue are less welcome. Slight changes in luminosity are not that perceptually evident.

Regards,

Bill

[joofa]

Oh Bill (BJanes), your link made my day. I love it when suspected bugs in software are identified as intended by design  .

Joofa

[Electromen]

For each job, I shoot the card in difficult lighting and do it again as lighting changes.  In LR I name them for that shoot.  After awhile I have too many profiles.  After photos are exported I delete old profiles.
Others that were created in controlled conditions like strobes, I keep.

[digitaldog]

IF you use DNG, the profiles you select can be embedded into that container, after which, the profile itself doesn’t need to be accessible to LR (unless you want to use it for other images of course).