RE: Approximate (or exact) relationship between illuminant color temperatures?
RE: Approximate (or exact) relationship between illuminant color temperatures?
- Subject: RE: Approximate (or exact) relationship between illuminant color temperatures?
- From: Wayne Bretl <email@hidden>
- Date: Wed, 30 Mar 2016 20:49:11 -0700
The concept you need is an old one in photography, called micro reciprocal degrees or mireds:
https://en.wikipedia.org/wiki/Mired
Express both direct sun color temp and blue sky color temp in mireds, subtract them, and that is the difference in mireds you need from any base color temperature to get the same relative color difference.
Example: suppose direct sun is 5000K = 200 mired; and blue sky is 16000K = 63 mired . This is a difference of 137 mired
If you start with 6500 to represent blue sky, that is 154 mired, and adding 137 gives 291 mired, or 3436 K to simulate direct sun against 6500 as blue sky.
-----Original Message-----
From: colorsync-users-bounces+waynebretl=email@hidden [mailto:colorsync-users-bounces+waynebretl=email@hidden] On Behalf Of Ben Goren
Sent: Wednesday, March 30, 2016 7:29 PM
To: ColorSync
Subject: Approximate (or exact) relationship between illuminant color temperatures?
So, I have a photographic project in mind that involves a bit of remapping between illuminants, where the relationship between two illuminants is going to be more important, I think, than the actual illuminants themselves.
Specifically, I'm thinking of using a pair of strobes, one in a softbox and another hard and directional to simultaneously simulate the overhead sky and the Sun. Gelling the softbox is likely to be impractical, so the thought is to leave it at its ~6500 native temperature and heavily gel the "Sun" correspondingly.
I'd white balance the final image using the same sort of approach one would for a scene in direct Sunlight with shadows -- either neutral for the primary subject, or something that leaves the subject warmish and the shadows bluish, or some other variation along those lines. The actual temperature targeted would be much visually warmer / thermally cooler than outdoors, but I'm hoping that it'd still look as if it were shot outdoors.
My obvious question: how red should the "Sun" get gelled to, and how do I know that without trial and error? (I'm expecting trial and error, of course, but I want to understand what's going on.)
The sky is often north of 15,000 K, and horizon sunlight can be below 5,000 K. The most naive of math would suggest that, to pair with my 6500 K "sky" I need a "Sun" with a color temperature thousands of degrees below absolute zero -- obviously absurd and obviously nowhere near the correct approach.
I'm guessing instead I'd want to compare distances on the Planckian locus and figure out x, y coordinates to target. Is that reasonable?
And, yes, of course -- I realize that I'm seriously oversimplifying all sorts of things and that there probably isn't an actual answer. The SPD of the strobes isn't a good match for D series illuminants, for an obvious starter. Instead, I'm hoping for some sort of "close enough" approximation.
If there happens to be an actual answer, fantastic, and it'd be a good exercise for me to work through the math. But some sort of a pointer to a general vague heuristic would be welcome, too.
(And, yes, I can think of some other alternatives, such as covering the ceiling with an appropriate bluish color and bouncing a strobe off of it. Other alternate suggestions are fine and welcome, but the geometry of the studio setup is likely to work least miserably with a softbox and a directional light, and only the directional light is relatively easily gelled.)
Thanks,
b&
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden