Dragon Sensor, Tungsten or HMI (Native WB?)

[Stacey Spears]

Does a CTB or CTO filter alter the CRI rating? I'm assuming it does...

If you are using LED as a light source, I would not use CRI as a measurement of quality. TLCI is a better alternative.

 

[Tom Gleeson]

I don't want to start a war but go with Mike P's advice. Maybe on a M or MX sensor there was some value in CTB filters but on the Dragon you don't need them. If you got so much tungsten light you are using ND filters then sure instead use an 80C CTB as this will lose you a stop or two stops if you go 80A. Assuming its not a super high dynamic range scene then you could just drop your ASA to give the same blue channel boost. These CTB filters won't hurt if you got the lighting budget but be wary of advice that states you NEED these filters to shoot in tungsten.

 

[Ulf Krentz]

Does a CTB or CTO filter alter the CRI rating? I'm assuming it does...

To my understanding it just alters the CT balance, not the CRI. Regarding filtering I like to keep part of the original colors and don´t completely filter against it, also you might split up between a lighter gel and a camera filter (or leave that subtle adjustment to CT setting in post). While their is no absolute NEED to filter tungsten it definitely helps to get a cleaner result in certain situations. I´m a big fan of not leaving everything to post.

 

[Hrvoje Simic]

Does a CTB or CTO filter alter the CRI rating?

No.

a) Filter filters out a part of the spectrum, it doesn't add the missing part.
b) CRI defines light source properties, not light setup properties.

 

[Ron Michaels]

What's better, gels or dichroic filters? I read dichroic filters have better light transmission.

 

[Chris Sharman]

Everywhere I look it seems people recommend shooting tungsten with a 80c CTB filter on the lens to get the best color, like film...

I tried this for a while on the MX sensor, which I found a little noisy at ISO 800. It improved things, but I found results just as good and far less hassle to rate the sensor at ISO 500 instead of using the filter, in the end.