FS7: How should you expose this?

Doug Jensen said:
The only variable is the target level of the zebra setting. Everything else about setting exposure applies equally to S-LOG, RAW, HG, STD, REC709, with a LUT or without a LUT, etc. In other words, it doesn't matter what mode you are shooting in, what matters is if you have hit the proper exposure for that mode. So 61% might be right for one mode, and 90% for another, but that's the only variable that changes if you want to keep things consistent, very easy, fast, and reliable. This is not rocket science.
Click to expand...

This is not really correct, because linear raw (and some extent log) can be exposed right up to the right, while knee compressed codecs cannot. IMO



S
 
Last edited:
I disagree with that method and if we were in the same room I'd prove to you how it delivers inferior results. We are not shooting RAW on a DSLR. This half-baked idea that correct exposure for S-LOG is just to push the brightest element right up to the edge of clipping -- and then back off just a little bit -- is ludicrous. Really, really, bad advice to follow. Not to mention that fact that it complicates your life because now you'd have to employ different techniques for shooting RAW, LOG, HG, STD, etc. because there's no way in hell you'd expose those other gammas right to the edge of clipping even if you thought it was acceptable thing to do for LOG or RAW.
 
Of all the ways to hit correct exposure Doug's is the easiest to learn and it works perfectly in pretty much any situation.

I use that myself, as well as scopes and just how it looks in the EVF/Monitor/Odyssey/Field Monitor. Exposure is just something that becomes automatic the more you shoot, learn from your mistakes. On some cameras I just use the histogram. But the Zebras are what I have traditionally used on video cameras for over 3 decades. They used to be the only R&G tool and still work fast and accurately.

It is great to always carry a polarizer and grad NDs, even w/o a matte box. With the old cameras it was a must.

Todays cameras are so easy and forgiving compared to what we had 10 years ago, that with a bit of practice anyone can figure it out with some experimentation. But we must still use good lighting and sometimes matte box and grad filters to get the best results. It is just easier now that cameras are so much more capable.
 
On my most recent commercial shoot I left the light meter at home.

I wasn't shooting people or narrative scenes, I was shooting some staged kitchen scenes with food as the subject so I strictly used zebra's and my waveform monitor to put the highlights exactly where I wanted them. (ie. highlights on rims of utensils or bright bowls, and on the edges of some food surfaces that might have had reflective edges, etc..)

I shot two completely different segments from each other that day. One was with the F55 and the other with the Blackmagic Pocket Cinema Camera. This footage was not intercut nor connected to each other in any way, we just had a a studio booked for a day so shot various spots. I had the cameras in completely different modes from one another. Slog3 on the F55, and video mode-rec709 on BMPCC.

I haven't worked with the BMPCC all that much before, but just trusting my zebras and waveform monitor to expose my highlights like this at the right level for the mode I was shooting worked really well for me. The images are absolutely gorgeous.

I wonder if some would consider this ETTR. I don't really, even though it sounds like you are edging towards the end of the upper portion of the scale.


[Edit 1] I 100% agree with Doug that ETTR then backing off doesn't really sound like a great option. I mean it will probably work but it sort of sounds unnecessary and I think you'd get better results to just analyze the scene you need to expose and decide what's important, what should I expose for, etc..

[Edit 2] Who ever voted down my Leonardo DiCaprio image has no sense of humour. Zero.
 
Last edited:
Doug Jensen said:
I disagree with that method and if we were in the same room I'd prove to you how it delivers inferior results. We are not shooting RAW on a DSLR. This half-baked idea that correct exposure for S-LOG is just to push the brightest element right up to the edge of clipping -- and then back off just a little bit -- is ludicrous. Really, really, bad advice to follow. Not to mention that fact that it complicates your life because now you'd have to employ different techniques for shooting RAW, LOG, HG, STD, etc. because there's no way in hell you'd expose those other gammas right to the edge of clipping even if you thought it was acceptable thing to do for LOG or RAW.
Click to expand...

We are recording quality signal and avoiding noise.

Now some cameras/settings do not allocate enough data steps per stop of brightness at the high end to produce 'quality' signal at the top end.

Raw the signal is good right until the top, Slog (3) the signal seems good until the last stop, Hgammas, we know the signal is compressed (and poor quality) at the top end.

For all there is noise at the bottom end.

We therefore need to expose as far right as we can for the way the data is collected (raw, log, HG) - the way the data is collected is dependent on camera/setup of camera.

Coming back to the FS7 Id say..
Slog3 ETTR-1
Slog2 ETTR-2
HGammas, expose to keep critical image out of the knee area.

S
 
I would say
1) arrive at correct zebras for any mode
2) make the creative decision to either stay or go a bit brighter/darker

In order to know 2) you need to know what matters in the shot and how your camera and the mode (gamma, gamut etc) your shooting in works. For example on the F3 in slog I would not worry about having to lift the shadows a stop but on the FS7 in Cine-EI slog3 I would feel more confident knowing the shadows would be crushed a bit because of how that gamma curve and sensor works.
 
Doug Jensen said:
The only variable is the target level of the zebra setting. Everything else about setting exposure applies equally to S-LOG, RAW, HG, STD, REC709, with a LUT or without a LUT, etc. In other words, it doesn't matter what mode you are shooting in, what matters is if you have hit the proper exposure for that mode. So 61% might be right for one mode, and 90% for another, but that's the only variable that changes if you want to keep things consistent, very easy, fast, and reliable. This is not rocket science.
Click to expand...

Yes well thats what I mean.. exposure is exposure.. thats a given.. but where you put your exposure for log and custom is different.. you are not having to save your highlights so much with log.. one has 6 stops over grey one has 6 or 7 stops total DR..
Alot of people will purposely over expose Slog by 2 stops.. and still have detail in highlights .. because you have so much leeway .. you would do that for REC709 .. ?
My point was in regards to the landscape shots of the OP.. with bright sky.. that scene shot in Log would be very different than 709..dealing with the DR.. as you say not rocket science
 
Last edited:
Doug I think maybe there is a misunderstanding on the ETTR front.. one is the stills way.. where every frame is of course a separate entity .. to do this for moving images would not be a good idea.. with every shot being different .. and pain to grade and cut together..totally agree with you there. but in EI mode its a uniform over exposure ..to be corrected in post.. same as under or over rating ASA on film.. for a particular scene or for your whole film/shoot.. this is a valid way to shoot log.. and used by alot of DP,s.. better SNR.. and you can do it due to the massive over head of log.. very often you dont need 6 stops over.. and in that way its a different way from working "ENG" style rec709 or close variants ..
 
Last edited:
I never said there was anything wrong with a uniformly and consistently over-exposing. Nothing wrong with that at all. I do it every day when I expose RAW and S-LOG up around 75% instead of Sony's recommended 60%. It is basically the same thing as rating the camera at a lower ISO. Easy to do and an accepted way of working.

But my point is that WHATEVER level you designate as being the correct exposure it should be easy to hit consistently from shot to shot with barely any thought. But that is not the same as recording every scene right up to the edge of clipping. That is wrong. Yes, wrong. BTW, the image in the OP's original post is not that unusual and shouldn't even pose a seconds delay in knowing how to expose for it. It's amazing to me how this thread has gone to 5 pages. This is just shooting 101 and the camera's shooting mode doesn't matter once you have determined where you are going to rate the camera's sensitivity.
 
Last edited:
In the screeningroom, I posted a very short film about the pond I enjoyed participating in. It is right in front of my house. I often filmed from a bedroom window. The view from there is eastward so an important part of the day, when the animals are active, there is backlight and reflections.

Next project is placing a remote controlled camera or build a small cabin to film from. :-)
 
I'd go for a mix. Like Neill, I like the idea of a ND grads and polarizers...I always carry a few in my lens case for these occasions. But even after shooting for decades, I'll admit I sometimes don't trust myself, so I do it the old fashioned way...bracket the exposure over a few shots and use the one that looks best.

If you're sequencing the scene with wides and tights, just make sure your overall exposure stays as consistent as possible.
 
Sometimes I prefer "baking-in" something like a heavy ND grad because it's the way I want it to look. No changing it in post.
It might be "1985" to some DP's, but 9 times out of 10 the clients still go "Ohhhh, that's nice..."
 
You must log in or register to reply here.
Back
Top