Matching Color Temps

[Wes Overvold]

Trying to match the color temps in post between an Epic Dragon and a Gemini and cannot figure out why 5600k on the Dragon looks nice and warm yet 5600k on the Gemini much cooler. Both tints are 0. Any idea as to why this might be?

 

[Aaron Lochert]

This is common. Don't get too caught up in the numbers. Different sensors can have different offsets. I see this all the time when I work as a colorist -- with many different camera brands. Don't be alarmed if you have to set each camera to its own set of WB numbers. If they look right, then they're right. So just WB them both by eye or even better: match both to a grey card.

 

[Jacek Zakowicz]

We made a tool for that. Color chart in self contained backlit device.
https://optitek.org/products/optimator-pro/

 

[David Rasberry]

Matching RGB picker values between shots in the Resolve color page works well.

 

[Mike P.]

ISO probably won't match either, FYI. Gemini typically defaults to ISOCal2 which will be a stop brighter at the same ISO (and maybe why it seems cooler as the blue channel would be comparatively "boosted" by a stop and, even though it's not noisy, maybe gives it a cooler bias.)

 

[David Rasberry]

Perhaps a better approach to this issue for exposure consistency would be to adjust the various IPP2 lut settings to match middle gray values and not worry about the ends. A low contrast LUT will not fill display REC709 IRE range on your monitor scopes, if it did it wouldn't be low contrast. Same with differences in high end rolloff that may cause the maximum IRE values for whites and near whites to shift in a non-linear fashion. But in REC709 display color space IRE scope value for middle gray doesn't change. Middle gray is middle gray is middle gray.

As far as the raw exposure is concerned, with DSMC2 cameras you will have approximately a 5 or 6 stop linear DR range beyond the display range DR of 2.4 gamma REC709 which is 10 stops. Exposure for middle gray can vary by +/- 2 plus stops from the a mid point gray exposure ISO with equal highlight and shadow DR distribution without clipping raw sensor values.

This is where shooting charts to determine ISO settings for your particular noise vs highlight protection preference for optimum exposure lies. I prefer using an incident light meter for these kinds of calibration tests as it will give a true ISO value for the light illuminating a middle gray chart test chart.