Hi Jason

reading a previous reply of your's to a post in this forum, an offshoot question pops to mind:
you said:
"BTW, the "magenta highlights" issue is because the actual RAW data, the green channel has clipped before the red and blue channels, so there is still color information in those other two chanels, but green on the other-hand has hit the full 1.0-white-point. This typically happens most often with daylight balanced scenes, since the green-channel is the most senstitive channel on the camera, and daylight actually has quite a bit of green-light in it (much more so than tungsten)."

so does that mean that when shooting outdoor chroma key, we would be better off using blue screen instead of green screen when shooting with the SI?since "daylight actually has quite a bit of green-light in it" and will be "contaminating" the foreground subject, making the seperation/keying harder?

Thanks
Lior

Not necessarily . . . consider the fact that skin-tones already have quite a bit of green in them, etc., the fact that daylight has a lot of green-light spectrum in it doesn't greatly impact the actual ability of the camera to key a subject of a green-screen.

Also consider that more than green-spectrum light, daylight has much more blue-spectrum light as well . . . and combined that with the fact that the blue-channel is the noisiest channel in the camera, and green-screens will tend to make cleaner keys.

That being said, if you have to shoot blue-screen, do it under daylight-balanced lighting, so that you don't end up with a very noisy blue-screen like would be the case with tungsten lighting and tungsten white-balance where you end up having to kick-up the blue-gains quite a bit.

For instance, under tungsten lighting, the blue-channel can typically require 2x gain or more in order to be properly white-balanced, the green may need a 1.4x gain, and red stays at unity gain. On the other-hand under daylight balanced lighting, blue will get a 1.1x gain, green a 1.0x gain, and red a 1.2x gain or something very nominal like that.

The green channel no matter what the white-balance is, will always be the most sensitive and best-sampled color-channel on the sensor. So I would pick the screen color based on what you are trying to key out, rather than the type of lighting you will be under, although again, keep in mind that blue-screens will get very noisy under tungsten lighting. And also again, keep in mind that for color-sampling, the green-channel has the highest native sampled resolution in the Bayer array.

Thanks,

Jason

OK..i see your point, thanks for the tips about lighting with the SI

BTW, is there a place to read up a bit on the insides of the SI ? topics such as color , matrix , look files and how they work?

Thanks!
Lior

No Problem :)

We do have information in our FAQ, and the manual has some information on exposure control as well. There is also a series of instructional materials on using tools like SpeedGrade, etc.

For information on matricies, try visiting Bruce Lindbloom's site. He has a lot of great material there on color theory and matrix math.

Thanks,

Jason