Hi, Asher,
Now we've mentioned some players, what about the rulers we might use in this discussion?
Define "neutral"
If we are speaking of a reflective target, then two (different) ideal concepts of interest are:
Spectral uniformity. This means that, over the visual band, the reflective spectral density is absolutely flat.
Chromatic neutrality. This could mean that, when illuminated with a certain illuminant (light with some particular spectral power density), the CIE chromaticity of the reflected light will be the same as the chromaticity of the same illuminance reflected from a "spectrally uniform" reflective surface.
If we have the former, we will have the latter, but not necessarily conversely.
Well, of course that means, "not precisely". We will get later into some discussion about "quantifying" chromatic discrepancies and setting "bogeys" for such.
Where are the deltas, (differences between ideal and measured) for each wavelength?
This of course ideally would be presented as continuous curve over the visible band, although of course values at a large number of discrete wavelengths can be perfectly useful.
What about effect of exposure to light on longterm stability?
What about different batches?
Important issues.
How far does one have to depart from perfect neutrality to have practical influence on perception?
That is a very complex issue, and depends on the context in which a visual comparison is made. We must contemplate a defined premise of "comparison" before it is even meaningful to give even a "broad" idea of how much discrepancy in the properties of the measurement target is consequential.
We can for example speak in terms of the minimum perceptible shift in the chromaticity of a viewed image region in an A-B comparison.
Or, at the other end of the range, we can speak of how much of a departure from the "theoretical ideal" of color correction will lead to a delivered image which (viewed under a specified illumination) will "not seem natural" to an observer (and this answer is greatly influenced by the nature of the "scene"). (Often here we find that an image with a substantial departure from "theoretically ideal" color correction will be scored highest by an observer.)
Do all wavelength deltas have the same power to alter perception of the color in an image?
We need to be careful about for what we are speaking of "different wavelength deltas". Perhaps from the context here, we are speaking of imperfection in the reflective spectral density function of a white balance measurement target.
There can be innumerable departures of that function from "flat" that will cause the same shift in chromaticity of any particular region in a corrected image (even if we are speaking of operation under essentially the same illuminant).
We can have a measurement target whose spectral density function departs considerably from flat which will nevertheless (for operation with illuminations having a certain spectral power density) will produce "theoretically-ideal" color correction.
And we can have another target with a wholly-different reflective spectral density function, for which the same will nevertheless be true.
So its all maddeningly complicated.
But, take heart. These are all concepts of color science, and this thread is about the ColorRight tool, in connection with which we been assured that scientific concepts need not be fooled around with.
Best regards,
Doug
I'm hoping that profiles created with the CC will give me a closer baseline for facial tones in mixed or complicated lighting regardless of the WB setting.
