Color Space and Gamut mismatching question
I have not been able to find a technical explanation to a question: what is actually happening when you set your TV's color space or color gamut to something other than auto, specifically in HDR10 mode? Is there a conversion happening or just the equivalent of a C++ reinterpret cast? What is the TV doing to the color values it's getting from the HDR signal?
From what I've been reading, all HDR10 color data uses the Rec.2020 gamut and thus a Rec.2020 container (i.e. the signal metadata marks the color data as Rec.2020 data), but is mastered such that color values do not extend beyond the limits of the DCI-P3 gamut (even though it technically could). For example, 10-bit RGB data for full green (0,940,0) would never show up because the DCI-P3 gamut within a Rec.2020 gamut would have some lower value, say 900 for arguments' sake, so "full green" in an HDR10 signal would be like (0, 900, 0). Please correct me if I'm wrong on this part, since it's kind of important to get this right as a foundation to understand my actual question.
Now, my Samsung QLED TV has essentially four modes relating to gamut when in HDR10: color space auto, color space native, color space custom DCI-P3 gamut, color space custom BT.2020 gamut. And thus it seems to me that only two of these settings make any sense: color space auto and color space custom BT.2020 gamut. The former will just "do the right thing," and the latter is correct since all HDR signals are Rec.2020 and it allows me access to the manual CMS for calibration.
However, my question is: what does color space native and color space custom DCI-P3 gamut *actually* do?
Is it just interpreting the Rec.2020 color values in a new gamut? If that were so, I would expect the colors (especially green) to be washed out since the Rec.2020 gamut is larger. Or is there something more nuanced going on with some conversion math?