Is DCI color space mapped to regular day time reflective color range.of N Europe?

[Wayne Morellini]

I have a theory, and I wonder if anybody that does color grading would know the answer?

I am suspicious that the DCI color space is modeled on the likely reflective color range of Northen European light in summer. That is the color range is limited by the available light spectrum reflecting off of colors.

Thanks.

 

[Marc Wielage]

Here are the current specs for the Digital Cinema Initiative:

http://www.dcimovies.com/specification/index.html

Color space is a bigger topic, and I would refer you to David Stump's book on Digital Cinematography as well as Glenn Kennel's book Color & Mastering for Digital Cinema. Both have quite a bit of math, but it gets into how white points are determined and how the standards were created over a period of time. SMPTE and the EBU also have quite a few technical papers on the subject, and the EBU's are generally free.

 

[Wayne Morellini]

Thanks Marc. I know that the whitepiont is based on northern European light, but wonder if it is based on the likely reflective spectrum of that light, to cover it. If it is so, you likely would have heard of it Marc. So it might not be.

I came here to trust the knowledge of colourists who have already seen the technical spec and back story. So if you have not heard it, it is probably unlikely. Just seems limited increase over NTSC color for some reason.

 

[Wayne Morellini]

Hmm, the guy at the next table is wearing a bright fluorescent hawian shirt that is probably close to the max of NTSC in some directions, but it is night with a mix if halo lighting.

 

[Marc Wielage]

Thanks Marc. I know that the white point is based on northern European light, but wonder if it is based on the likely reflective spectrum of that light, to cover it. If it is so, you likely would have heard of it Marc. So it might not be.

I don't think it matters. I think this is one of those nebulous issues like what constitutes a meter or the value of an ounce of gold: it's just kind of an arbitrary standard that all the industry tech organizations have agreed to. So it's basically like, how they picked the standard isn't so important as that there is a standard people have to adhere to. I would also point to the American Cinematgrapher Manual as another reference that goes into color temperature in detail. I seem to recall reading a section in there talking about 6500° whites on sunlit days vs. cloudy days vs. various times of day. My general explanation to clients has been that the white temp would be measured on a cloudless day at high noon, but no question there is a difference in the quality of light on the East coast vs. West coast, and in Europe vs. Asia vs. North America. How that could be quantified, I dunno. I don't think it matters in the grand scheme of things.

I believe the 5400° K standard for theatrical projection goes back to the use of carbon arc lamps in film projectors going back 40 years or more. I think they also rationalized the 14fL brightness standard for theater screens at the same time, but the interesting thing is that I have seen people misinterpret whether this should count for "open gate" (no film in the gate) vs. "clear film in the gate." No piece of emulsion is 100% clear, so it winds up that an open gate is at least 1 or 2fL brighter.

The standard gets more nebulous when you consider that different screens have different colorimetry, and curved, reflective screens (like those optimized for 3D) are generally not as accurate as flat, lower-gain screens. I think THX will approve a gain of up to 1.30X, but that's all. Some of those curved reflective screens are bad enough that you can see a brightness variation of maybe 10-15% if you sit at a bad angle in the theater. I would bet this also has an adverse effect on white point as well.

 

[Wayne Morellini]

Thanks Mark.

I probably can see brightness variations sitting head on locally.

I think if it was set to expected reflective color range (except in some cases, such as plants greens etc) then it is a reasonable good standard to use, except in the exceptions such as plants, different colored light at night and with emmisive out of range color sources. It would mean quantifiably it is a scientifically reasonable compromise to use. I would not mind a OLED, but see them as potentially compromised going into the distant future in a lot of areas except true black (but my local cinema uses grey black which wrecks the experience for me, but they have shrunk the screens except the premium vmax, so I wonder if that is blacker. As I dont remember it being so dissapiontedly gray at Cairns Central. It really does distract me).

 

[Brett Harrison]

P3, like REC709, is a display-referred colour space, i.e. it is meant to represent a range of colours that a type of screen can reproduce reliably. The screen might be able to form more colours, but it covers 709. You have to map your device-agnostic colour space (ARRI wide gamut, DragonColor2) to 709 or P3 as some of that space doesn't 'play' on a screen or projector.

The white point is defined as D65 which as Marc says, is just an anchoring definition to reference and hold the standard which bears some relevance to a given application.

The gamut of colours produced by the quality of light on a type of day in a part of the world reflecting off of arbitrary materials is far and away wider than any display-referred space, defined by CIEL*a*b*, the human perceptual colour space.

Colour spaces don't define a quality of colour, only a range from one edge condition to another, like the visible spectrum (Lab) or CMYK (the printed colour medium). A 709 colour space can produce some of those colours such that the 2deg Standard Observer will report that they match (metamerism).

That's my understanding anyway, happy to be corrected as I am a neophyte on this stuff.

I have a theory, and I wonder if anybody that does color grading would know the answer?

I am suspicious that the DCI color space is modeled on the likely reflective color range of Northen European light in summer. That is the color range is limited by the available light spectrum reflecting off of colors.

Thanks.

 

[Wayne Morellini]

No, that's good Brett. I thought that with P3 they might have extended it beyond displays, as ultimately it was projected. The point of the color spectrum that the white point is based on is that it is limited and its interaction with most materials would limit it even more. So it was an interesting proposition that it might have been done that way anyway. However, that's good enough. Evennif it was done that way it would be ill defined for many other lighting types.

To suggest an reason why the light might ge different in different parts of the world is that, apart from local environmental factors and light from terrain etc, the time zones are probably not lined up properly from location to location. The color scheme f light changes with the direction the sun hits the atmosphere, so 12 pm at the same latitude between of fferent locations that are hour accurate is similar, but different half an hour away in the same time. The other is latitude, as the light angles in a different detect depending on the time of year and latitude. That European Summer light you can probably find here in the tropics in autum or winter after 3pm going to 6pm. Definitely in the sub tropics down in Brisbane at a certain time of the year and afternoon. interesting stuff, and why the sky is blue