|
|||||||||
|
Thread Tools | Search this Thread |
July 29th, 2008, 11:59 PM | #1 |
Major Player
Join Date: Mar 2004
Location: Toronto, ON, Canada
Posts: 479
|
Gels and mixed colour temperatures
I've been lighting interviews for the past year and a half and have a good grasp of colour temperatures and light sources and a basic understanding of gels.
I'm currently out of town with a 6-bulb Barger Baglite (6 x 650w), a roll of half-blue gel and two Lowel kits (2 x 500w omnis, 1 750w tota, 2 full-blue gels, 2 diffusion and 2 ND in both kits). I plan on trying some new things (new to me, that is) with mixing colour temperatures and will likely white balance to 5600k in a windowed room. Playing around with different full- and half-blue gel combinations has shown me that what I thought I knew about gels isn't necessarily so. What I know for sure is this... A Barger or Lowel light is 3200k A Barger with half-blue is ~5000k A Lowel with full-blue is ~5600 This seems to suggest that a half-blue cools the light by ~1800k and a full-blue cools by ~2400k. If I were to combine a half- and full-blue on one 3200k light, it seems logical (to me!) that the 1800k and 2400k would combine to make 4200k and the resulting light would be ~7400k. I tried this, white balanced and the camera told me 12,600k (9400k cooler than without gel, 5200k cooler than I expected). Am I right in now thinking that stacking gels multiplies their cooling or warming effect, rather than adding? Why is this and is there a way to calculate it or is all just trial and error? Any insight would be appreciated. Thanks!
__________________
Mark Utley |
July 30th, 2008, 12:32 AM | #2 |
Inner Circle
Join Date: Nov 2006
Location: Tallahassee, FL
Posts: 4,100
|
The answers you seek are called Mired Shift Calculation. and there is a formula to do it. I am attaching a PDF chart which will help, and I'll give you the formula too...
The Conversion Poster will give you the basic idea. Conversion formula is: (1,000,000/Filtered source in Kelvin) - (1,000,000/Original Source in Kelvin) = Mired Shift Value Example (converting 5500k to 3200k): (1,000,000/3200k) - (1,000,000/5500k) = 313 - 182 = 131 (Equivalent to a 85 filter), and why you see this filter on movie camera lenses shooting outside or with HMIs when they have tungsten balanced film in the camera like Kodak's 5218 that is nearly a Hollywood standard. Each gel has an associated Mired Shift value. And by looking at them, you should be able to calculate what you need. In terms of combining filters, the math should work fine. BUT, you have to remember that as you give the camera a lot of light in one part of the spectrum, and starve it of light in another, your camera will probably not be able to judge white correctly, because it will no longer be getting "full color spectrum light" to make judgements with.
__________________
DVX100, PMW-EX1, Canon 550D, FigRig, Dell Octocore, Avid MC4/5, MB Looks, RedCineX, Matrox MX02 mini, GTech RAID, Edirol R-4, Senn. G2 Evo, Countryman, Moles and Lowels. |
July 30th, 2008, 03:43 PM | #3 |
Major Player
Join Date: Mar 2004
Location: Toronto, ON, Canada
Posts: 479
|
This is very helpful. Thanks so much, Perrone!
I did a lot of reading about colour temperature, spectral power distribution, etc and am understanding the concepts. Do know of any DVDs that go over colour temperature in detail? I think it would be easier to understand visually.
__________________
Mark Utley |
August 1st, 2008, 10:29 PM | #4 |
Trustee
Join Date: Apr 2006
Location: United States
Posts: 1,158
|
generally speaking 2 1/2 CTBs should give you about 1 full CTB. I use 2 or 3 layers of CTB to shift tungsten as needed, and sometimes HMI's if they are too warm - as HMI bulbs age they shift warm! 1/2 CTB/CTO is pretty flexible to work with rather then just having full as you can tune the temp better. having a few peices of 1/4 is nice too, but if i can only have one thing, its 1/2.
|
August 2nd, 2008, 10:50 AM | #5 | |
Regular Crew
Join Date: Dec 2006
Location: Tampa-Orlando, FL
Posts: 124
|
Quote:
There are two kinds of gels, one is actually better referred to as filter, that's the color conversion filter because it doesn't add colors, it subtracts or filter-out any color that are not belonged there, this is why you lose so much brightness when you make a conversion. This is when you should color balance after placing the gels on your lights. This is why I can't understand why double gelling, you are no longer correcting one light to match the rest of existing lights, you are now trying to obtain a color effect. Then there are color enhancement gels and those are for color effects. Correction filters don't make good color effect gels. If you white balance after placing color enhancement gels on the light the camera will pretty much wipe-out everything that you are trying to accomplish and the result is usually a mess. The camera always sees neutral gray and will try to take away everything that is not neutral gray thus killing the gel's effect. Your best friend is always a GOOD monitor that can be well calibrated. |
|
August 3rd, 2008, 08:08 AM | #6 | |
Trustee
Join Date: Dec 2006
Location: Little Rock
Posts: 1,383
|
Quote:
One other tool that I still use to this day is a color temperature meter. They are portable, give instant color temp readings, and tell you which gel to use for perfect color temp matching. Something along the lines of a used Minolta Color Meter II can be purchased on ebay. |
|
| ||||||
|
|