But isnt the Spyder about characterizing the device, ie generating an ICC profile?
Two steps involved :
- first, setting the device in a convenient and reproductible state, ie calibrating,
- second, characterizing the device in this state with measurements, to build an ICC profile.
If you haven't already, read
this book : it's quite easy, and yet comprehensive on the subject.
When I do white-balance by eye, I am essentially changing the color temperature, am I not? How can I trust my eyes to do that task if I am to "normalize away" color temperature "errors" of my display?
From what I try to understand, your eye assumes a neutral reference mixing the white or grey parts of the screen, the mean colors of it and some surrounding light.
The thing about the mean color is that when you change too much the neutrality of the screen, its color response is so biased that the eye can't take it as neutral anymore : the correction is more efficiently done by the eye itself.
And if you wear red glasses long enough, your vision will adapt and you may see colors not very differently from without glasses after a while.
But generally, the main rule for calibrating a display to a convenient state is to try and see what works.
Ideally, I would want an image pipeline that tried to recreate perceptually what I would have seen in the original scene, on display and on paper.
No you don't!
The original scene has to be rendered, ie compressed tone-wise and color-wise, to be decently represented on a display, and even more on paper.
But I'm playing with words here, and I agree that :
AFAIK, Lightroom should be able to do this as long as:
1. My display ICC is correct
2. The camera "ICC" is correct
3. The printer ICC is correct (wherever it came from).
The problem may reside in characterizing the display so that it matches visually your viewing environment, and mostly characterizing the printer driver so that it doesn't apply any correction behind your back (as LR cares all about it).