I'm trying to figure out the flaw in my logic when thinking about color spaces and conversions between them. If you could follow along with me, and correct me when needed, that would be very helpful.
I'm just going to focus on non-print situations at the moment, for that is what I am mostly confronted with. I'm going to pour the things I've been thinking about for quite some time, so I apologize in advance for the length of this text:) and I would also like to thank you in advance.
Here's where I am at at the moment:
A digital image its colors are made up out of values which represent three channels: Red, Green and Blue (RGB). These values are just numbers (let's assume we have an 8 bit image, so 256 values) and the software program with which you open an image tells the monitor what numbers are in the photo and what color profile is incapsulated in the photo (so how the monitor should show that color).
Question 1: When I am looking at an image on a monitor in a software-program, like Windows Photo Viewer (or any other simple software program for that matter), and there is no color profile attached to the image, how does my monitor show the colors? Does all software have some sort of default-color space with which it interprets colors? Or does some software have no default-color space and is the numerical information simply transferred to the monitor to interpret according to its factory settings?
If I open a photo in Photoshop that has no color space encapsulated, the interpretation of this photo will be done according to Photoshop's working space (you need to choose a working color space, because otherwise Photoshop can't do anything with your photo. It wouldn't be able to do anything the numbers which are the photo in essence).
Question 2: Here my knowledge falls a bit short because of practical experience with calibration. I have an Imac at home, and there are a lot of color profiles to choose from in the Preferences-window. There is the standard color profile which I assume is the profile which tells my monitor that, for example 255, 0, 0 is the most saturated red it can produce. In other words, this is the profile that uses all the color capabilities of my Imac. Then there are the device independent ICC-color spaces such as sRGB, Adobe RGB,... I assume that when I bought this computer, Apple made sure that these profiles would counter the abnormalities in my monitor. What I mean is, they actually calibrated my computer for me and made profiles of all these calibrations for the different color spaces?
I understand that my monitor could have shifted in the passing of years, so perhaps I need to calibrate my monitor again. What I do not understand is how the color settings in Photoshop work with the color profiles in the Preferences window? When I work in sRGB in Photoshop, does Photoshop actually tell my OS to use its sRGB-color profile to show the photo on the monitor?
And if I see in the Photoshop color settings that I can choose to emulate a Fuji-filmstock, how does Photoshop communicate with my OS to make sure that what I see is correct? There is no Color Profile in my preferences-settings of my Imac for Fuji-filmstock.
To conclude a question about Photoshop's Profile- vs. Convert To-settings: When I open an image with no color profile incapsulated, the photo is shown in Adobe RGB because that is the working color space in my case. Nothing happened to the actual numerical values of the image. The numbers are just interpreted and show the color as if these numbers were always meant to be Adobe RGB. This is also what happens when you Profile a photo.
If you convert a photo in Photoshop, say from Adobe RGB to sRGB, the numerical values will change in an attempt to stay visually as close to the original as possible.
However, when I convert an image without color management but viewed with an Adobe RGB-profile, to Adobe RGB, the colors change drastically. I don't understand why that happens. I mean, the numerical values should stay the same, right? So the colors should stay the same as well, since I was viewing the photo in the same color space as I now want to convert it, no? Photoshop could not possibly know how the original image was intended to look because there was no color space encapsulated? So what is Photoshop's logic in this case? Just in case you're wondering, this is a question to better understand how Photoshop thinks. I don't think this is actually a useful action to do, right?
If some questions aren't clear, please let me know. I'm quite anxious to finally get an answer to these questions, so I'll be more than glad to make it all clearer.