Cinegamma 1 Thoughts/Questions

[Philipp Eierund]

I have a few thoughts/questions on Cinegamma 1:

I'd like to use Cinegamma 1 for an exterior shoot to protect as much of the highlights as possible. However, I am not sure how to use Cinegamma 1 in collaboration with a light meter.

With REC709 (regular) gamma, my light meter reading and a test grey card show the same exposure reading. Sekonic says F6.3 and the grey card reads 50% with the lens stopped down around a 6.3. Easy.

Now, with gamma set to Cine 1, however, I need to open up my lens to 4.6(ish) to get my grey card reading back up to around 50%. As demonstrated by Any Shipsides (AbelCine Tests the Sony F3 Dynamic Range | CineTechnica) cine gamma 1 shifts the grey point downwards to allow for more data to be captured in the highlights.

My question is: should I compensate for this shift or is the hole point of cinegamma 1 to keep exposing at 6.3 and then simply bringing up the mid-values by a stop or two?

Any thoughts are highly appreciated.

Thanks,

Philipp

[Dave Sperling]

Typically the Cinegammas, particularly Cine1 and Cine2, will work fine with more exposure. I wouldn't recommend trying to use a light meter with them, since setting the 'level' will also have an effect. However using a waveform as your light meter should work quite nicely.
With the Cinegams, you will also want to be aware of your post workflow -- are you planning to do a full color correct on the footage, or are you planning to use it without further grading? If doing a full correct later, you may want to 'protect' your highlights a little more, whereas if no correct has been budgeted, you will want to get as close as possible to your finished look.
Remember, light meters were designed originally for use with film, where virtually everything received color correction (and I used to routinely overexpose by 2/3 of a stop in 16mm to optimize my grain structure). Also remember that when doing a film shoot, you would never walk on set without a contrast viewing filter around your neck. Now what you need is a good monitor with an accurate waveform.

[Alister Chapman]

Because the Cinegammas have a non-linear response a light meter is slightly less useful. I was interested to see the way Andy carried out his evaluations. I'm not saying he did it incorrectly, but if you were measuring latitude with film you would expose your 50% grey according to the light meter and then measure the number of stops of under/overexposure. This would translate to setting mid grey at 50% for each gamma and then measuring under/over as opposed to allowing the difference in gain at 50% to skew the mid grey level up and down.

In the real world, IMHO you would tend to expose your mid tones accurately using a WFM, Zebras or just the VF and then evaluate what your highlights are doing, only stopping down if necessary to protect the highlights. I'd have liked to have seen Andy's test done keeping the 50% grey at 50% for each gamma to show the changes in highlight and low light response relative to that. Again this is not a criticism of Andy's tests, they looked to be accurate and true. If I had a decent latitude chart, I would have done it a little differently.

[Philipp Eierund]

Interesting thoughts! So where would you guys place caucasian skin on a WFM using CG1?

[Dave Sperling]

Typically I'll keep the skin highlights around 68, assuming that I want the face brightness looking 'normal'. It's obviously going to be different in a dramatically lit scene, or if I'm keeping a face in the shadows.

[Philipp Eierund]

Interesting. Don't you have to watch out that the skin tones don't creep into the compressed area of CG1, however? Where does that start kicking in?

[Alister Chapman]

Yes, that's one of the things you have to watch out for with the Cinegammas, compression of skin tones. Compression starts at about 60IRE but becomes more aggressive above 70IRE.

[Leonard Levy]

That's always kept me away from cinegammas, but I've never spent a lot of time with them. Seems if you keep the skin tones safe you need a lower exposure which neccessitates grading. That's fine as long as you are posting yourself or you trust your client which I generally don't with mine unfortunately.

[Philipp Eierund]

Right. Hence I was wondering about using my light meter since it will 'automatically' underexpose the cg 1 a bit and then bring the mid tones up in post. And by using a light meter I'd at least make sure that I keep exposing consistently.

[Dennis Dillon]

Originally Posted by Alister Chapman

Yes, that's one of the things you have to watch out for with the Cinegammas, compression of skin tones. Compression starts at about 60IRE but becomes more aggressive above 70IRE.

That is why I set skin tones to 60 IRE.

[Alister Chapman]

Frankly with the F3 I'd use a monitor rather than a light meter to expose. A monitor will show you not just your single optimum exposure (as determined by a light meter) but also how your over and under exposure is behaving. A light meter will not tell you this.

[Leonard Levy]

I've always found light meters to be dangerous in video as a way to set exposure because video simply is not as linear as film and the your gamma, highlight and grey scale range are not predictable. Its still essential for lighting though and as a guide to exposure issues. As Alister said, the monitor and waveform are the best ways to reliably set exposure in video. That's my 2 cents anyway.

[Doug Jensen]

Leonard, I agree with you completely and will even go one step further. I refuse to use ISO numbers when discussing the sensitivity of a video camera. ISO for what settings?? Out of the box?? Nobody I know shoots with the default settings, and as soon as you start changing paint menus (especially gamma) you have changed the sensitivity of the camera. So to say a camera is "ISO 400" is totally meaningless unless you also define exactly what paint settings you are using at the time. My camera might be ISO 400, but your camera might be ISO 800 because I have set it up differently. How do we know? Therefore, the use of light meters to set exposure, without calibrating the meter for the actual sensitivity of a camera as it is currently set up, will certainly lead to exposure errors.

ISO isn't arbitrary. It is an industry-standard, definable, measure of sensitivity that actually means something specific. Every camera, film stock, etc. that is rated at a given ISO should have exactly the same sensitivity to light. But just because your F3 may say "ISO 400" in the viewfinder doesn't mean it really is. This is a topic I demonstrate in my F3 training DVD, but after talking to some other shooters at NAB, I wonder if it will be understood.

[Brian Tori]

Doug is right, video cameras will have different ISO speeds depending on what picture settings are dialed in. The only way to accurately find the ISO for a particular camera is to rate the camera after choosing picture settings. I actually like using a meter for exposure purposes. I find it to be more objective when determining exposure than say using zebras on skintones for example. The reason being that zebras are based on a reflected reading. The reflected reading is subjective, and every skintone reflects a different percentage of light. I use one in combination with my Canon DSLR which gives me good exposures.

[Leonard Levy]

Brian,
Certainly I would not argue with your preference for using a meter, but i would disagree with your saying that zebras are a reflected reading.

Zebras are a measurement of the actual exposure on the chip, and yes the angle of light and the skin tone will affect where the zebra lies , so not all skin should be placed at exactly the same zebra reading. Caucasian and say dark African American skin tones should not both get 70% for example and in that case an incident reading could be more reliable. But as explained earlier ISO equivilents are dicy in video as well. Also my experience is that sometimes video is simply not linear.

It all comes down to knowing how to interpret whatever your tools are and what you're comfortable with though.