All this needs much reading - there are no simple explanations.
Color appearance modells
Best regards --Gernot Hoffmann
Red is just an English word some use to describe a sensation. In terms of numbers, read the wiki. R255/G0/B0 is the most saturated red you can define numerically in RGB in Photoshop but those numbers are different colors if the color space, the scale of the numbers is sRGB versus ProPhoto RGB!
Thanks Andrew Rodney and Gernot Hoffmann.
I realise how deep this CMS thing is, while I tried to grasp some workable concepts to apply and use in daily work.
As for the main question here, will the hue-saturation change or not?
I am just looking for an answer for this, so that I will paint that part of the model with a bit less saturation next time, if so
Please let me reply phrase by phrase, though I don't like such a style.
This question has always bothered me, especially when using color in some areas of architectural models, Isn't that car too red in the model? Hey, but I used the same deco paint from the shop they use in the garage! So, is there a factor of "scale" to color?
The paint on a car is a surface color, which has a reflectance spectrum. The reflected light depends on the light source spectrum
and the reflectance spectrum. Thus the same material paint can cause different spectra for the reflected light and therefore different
Let me put it another way, if I print an A4 of some fixed value red. The same value I use in a huge
tiled poster, where the same real A4 size appear some 15 meters above, rest area other colors/content.
Will that A4 appear to be same red?
Probably not. Mounted at different places, the sample will be illuminated by different light. See above.
In the neighbourhood of other colors or mounted on other color surfaces the appearance will be different
(see simultaneous contrast below).
The sample is mounted 15 meters above: this will normally not change the color appearance by haze,
but the adaptation of eyes & brain will probably be very different, for instance if adapted to the sky.
Note: this sample shall have the same measurable color, therefore the colorimetry is of no importance
Still putting it another way, will the purple flower in close range be same purple I see in that bush
when I was kilometers away?
For objects far away, the color looks different. Haze and even pure air may filter the light. Mountains
far away look different - more blue-ish than nearby.
Perceptually I may think they are supposed to be same, but in reality they may not be, not just because
of haze etc, but simply because the distance fades away the color. So, does distance fade the intensity
(luminosity?) or the color(hue)? Or none! - as light, energy is spherically diminishing anyway.
Yes, distance fades away the intensity (luminance), but this wouldn't change the appearance very much.
The mountains are looking blue-ish though we know that they aren't. This is an remarkably paradox
situation. In our rooms we have chair with red textile. These seem to have the same color in daylight and
in lamp light, but the reflected light is of course very different. This effect is called color constancy.
The answer may need me to modfiy or not modify the hue intensity when using it for small sizes,
vs when using it for museum posters or hoardings.
In my humble opinion it's necessary to test the effects (for instance talking about architecural modells
or exhibitions) at least under some different illuminants.
Talking about colorimetry and color management is somewhat different. Nowadays we can optimize an
image on a calibrated monitor in a darkened room, print it on a calibrated printer on paper and view the
result under special proofing light in a D50 lightbox.
The effects that a print looks different under different illuminants can be described ("this prints is neutrally
gray under D50 but under sunlight it looks blue-ish and under fluorescent tube it looks greenish") is called
metamerism. This effect could be measured, but it`s difficult to counteract, besides choosing different
inks and better balanced illuminants.
A nice optical illusion:
http://www.google.de/imgres?imgurl=http://web.mit.edu/persci/people/adelson/images/checker shadow/checkershadow_illusion4med.jpg&imgrefurl=http://web.mit.edu/persci/people/adelson/c heckershadow_illusion.html&h=420&w=540&sz=77&tbnid=pBiAPm1chSrNUM:&tbnh=82&tbnw=105&prev=/ search%3Fq%3Dchecker%2Bboard%2B%2522optical%2Billusion%2522%26tbm%3Disch%26tbo%3Du&zoom=1& q=checker+board+%22optical+illusion%22&usg=__wFVg2QDPk3dRCsY_b15hCCFhIUQ=&docid=gpS7f5jw-7 lboM&hl=de&sa=X&ei=10w_UbDVAsLk4QTU84GYBQ&ved=0CEIQ9QEwAg&dur=513
More about simultaneous contrast:
Hopefully this post will help more than quoting the book by Mark Fairchild (Color Appearance...)
in my previous message.
A great scientist, who tries to introduce color appearance corrections into the color management
workflow. But of course we don't want the mountains losing their blue and we don't want to
see the chairs in the evening pinkish-darkgray instead of red.
Best regards -- Gernot Hoffmann
Thank you very much G. Hoffmann. Your detailed reply does help.
I had forgotton the optical illusion, thanks for sharing. As for the rest, yes I will have to try and check.
I am now so inclined to "perception" than "measurement" that with my struggles wiht getting thigns right with my Spyder4, I am creating combinations of monitor settings (sRGB, Blue, Red OSD settings and one of the older several profiles Spyder created) and "seeing" which comes near to my expectations.