FS7: blotchy footage

I've evolved. :-)
I no longer agree with Sony's numbers, but that isn't to say you still can't get good results using their recommendations. I just think I can do better grading in Resolve with a higher expsoure. So should I go back and update my FS7 video to reflect my evolution? Maybe, but I don't have the time and 61% isn't necessarily wrong anyway. As always -- I highly recommend people to their own testing with their own workflow to find what works best for them.

BTW, LUTs in post that are designed for the exposure to fall around 61% need an exposure adjustment node placed just prior to the LUT node in Resolve if 75% is the target exposure.

BTW again, those who have seen my FS5 training already know I now recommend 75% for S-LOG.
 
Last edited:
Yeah ive found that 61% has given me underexposed pictures from time to time as well. User error ofcourse, but it goes to show the margin for error is too small. White is not really always white. Theres 50 shades of white out there, so wich one is the correct for your zebras for that particular scene.

Ill test your 75% on reflected white and see if it gives me different results.

I have seen others post 1 stop overexposed s-log that looks better than "properly" exposed s-log, so it doesnt sound like a dumb idea in theory at least :-)
 
I actually find the highlight/lowlight warning to be a good indicator of footage going out of bounds. I used it on my F3 also with very good results. But it can't be trusted blindly and common sense is needed as with everything else.

Not that I'm ignoring my exposure tools but the warnings are great when working fast.
 
If there's one thing that small waveform on the FS7 is good for, imo, it's being able to see when you are clipping your highlights... you may not know where any % actually is... but you can easily tell the flat line of clipping
 
Martin.G said:
White is not really always white.
Click to expand...

Technically that is true, but when you put them on the scopes the differences between whites is much less than you would think. 35 years of judging exposure by zebras on bright reflected whites has not let me down and provides great consistency from shot to shot whether shooting LOG or WYSIWYG.
 
markfpv said:
If there's one thing that small waveform on the FS7 is good for, imo, it's being able to see when you are clipping your highlights... you may not know where any % actually is... but you can easily tell the flat line of clipping
Click to expand...

But if you have anything anywhere near clipping in S-LOG or RAW, you are way over-exposing. It's better to set a proper target with zebras and aim for it, a few IRE one way or the other if you miss won't make much difference. But putting the brightest stuff just at the edge of clipping is not something I would ever recommend.
 
But it's not over-exposing is it, if the scene has very bright specular whites as well as dark shades? Then I think I would need to put brightest (like sun reflections on water) near the edge or even at the edge of clipping.
I mostly use zebra's for my exposure... just saying it's easy to tell when you are clipping...
 
Sun reflections on water are specular highlights and should not be used for judging exposure because they are not reflected whites. In fact, in most instances, if you keep specular highlights from clipping -- even right at the edge -- you are likely underexposing. Same thing goes for a lot of really strong backlights on hair, shoulders, etc. Even with 14 stops of DR sometimes highlights must be allowed to blow out and if you don't let them you will be underexposing the more important subject.
 
starcentral said:
sorry guys if my post caused any confusion.

I did go back and find the clip I shot where I saw the gain applied. It was shot at 60p and sure enough meta data says 4.0db gain but I definitely didn't play with any gain settings. I agree this isn't normal however, because I have other 48fps clips shot the same day and they say 0db gain.

Sorry can't explain it and too long ago to try and remember the details.

Screen Shot 2016-03-03 at 3.23.23 PM_zpscfqf6z0n.png
Click to expand...

Hey guys,

I'm a bit confused that it says the exposure index (another word for ISO?) is 2000 and the gain is 4.0 dbs. I thought the gain was just another name for ISO on the fs7? If the exposure index is at the native ISO of 2000 wouldn't the gain be at 0 dbs? Does the gain switch itself, or is that something that you need to do manually.
Sorry if this is a silly question I'm just trying to soak in as much info on the fs7. I Just ordered one and it should be here in a week. I've rented Doug Jensen's classes on it and it's been super fun learning about it.

Thanks in advance!
 
Doug Jensen said:
Technically that is true, but when you put them on the scopes the differences between whites is much less than you would think. 35 years of judging exposure by zebras on bright reflected whites has not let me down and provides great consistency from shot to shot whether shooting LOG or WYSIWYG.
Click to expand...

Yeah, you may have 35 years of experience exposing for white but i have 3,5 months ;-) I do not shoot much indoors, so the shades of white that has me miss my target the most is clouds and snow with the range of passing shadows and highlights from the sun through the clouds. Its all a learning experience that keeps getting better, i find the reflective white exposure to be very consistent and workflow friendly. I know that ive exposed for whatever white is in the shot, and then im quickly able to balance and adjust the image based on that, before judging my corrections further. Its really neat.
 
Wow, I just gained a hell of a lot more information than I bargained for when I originally made this post. Thanks for all your replies.

Doug, I absolutely love shooting Cine EI and wish I could use it exclusively, but unfortunately I constantly find myself in situations where it is simply not possible to overexpose. I actually find this happening a lot more frequently than I would have expected - especially when shooting indoors on very time-constrained shoots with house lights only.

I was wondering if you could perhaps shed some input while you're here...!

My instinct when in these situations was to shoot in Custom Mode with Gamma set to Slog2 and Matrix set to Off as I thought this gave the closest look to Cine EI's Slog2 but with the flexibility of bumping up the ISO higher than 2000 for when it was really needed. I've always refrained from using HyperGammas or different Matrix Presets because it felt like they weren't as flat and so there'd be less leeway for grading them - but then perhaps I'm completely wrong there...?!

My other gripe with Custom Mode is the difference in how it white balances compared to Cine EI. I absolutely hate having to use the WB SET button with a white card as often my subject could be in very different lighting conditions to where the camera is (if I'm on a telephoto lens), in some instances I have to crank the ISO to 12500 just so that I don't get the NG: Low Light message when white balancing and I also never really think it sets the WB quite where I would like it - I've found the camera setting the WB as high as 10,000K when indoors before which just seems like complete madness. What's your approach when you don't have the luxury of being able to constantly get a white card out when running between drastically different lighting conditions?

Thanks!
 
Really interesting thoughts on the gain being applied... I've definitely noticed when jumping between 24fps and higher frame rates that I don't need to adjust my aperture/ND as much as I'd expect. Wouldn't going from 24fps to 180fps be about 3 stops? I haven't done any scientific testing, but I don't think I've had to open up three stops when jumping to 180fps, so there must be something else going on behind the scenes.
 
Question for Doug: If you've adjusted your zebras percentage for white to 75% (SLOG) where do you find skin tones fall best? Skin tones are typically what I'm the most concerned with in the majority of the stuff I produce.

I tend to agree that zebras are great, quick tool for judging exposure and have recently come in line with Doug's conclusion that 61% can lead to underexposed images.
 
Oh man, don't even get me started on this. Who knows where skin tones will fall? I have absolutely no idea, nor do I care because I would never, ever, expose by skin tone. If I expose for whites -- skin tones will be exactly where they need to be whether the subject is a Nigerian, a Swede, or an albino. And by the way, whatever the correct exposure is -- it is exactly the same for all three of those people in the same setup. Unless you're going to try to tell me that everyone in the world has the same shade of skin, then setting exposure based on an unknown, variable value really just comes down to guessing. But I'd rather be precise about exposure and that can only be achieved with white, 18% gray, a chip chart, or some other reference of known reflectance. In fact, using zebras on white is exactly the equivalent to taking an incident light meter reading, except its a whole lot easier.
 
Last edited:
jshbckr said:
Really interesting thoughts on the gain being applied... I've definitely noticed when jumping between 24fps and higher frame rates that I don't need to adjust my aperture/ND as much as I'd expect. Wouldn't going from 24fps to 180fps be about 3 stops? I haven't done any scientific testing, but I don't think I've had to open up three stops when jumping to 180fps, so there must be something else going on behind the scenes.
Click to expand...

Well, when you get around to doing your scientific testing you'll find that the camera is not doing anything sneaky with the gain or anything else. It is completely straightforward and predictable what happens to the exposure when the shutter speed is changed to accomodate a faster frame rate.
 
Doug Jensen said:
Oh man, don't even get me started on this. Who knows where skin tones will fall? I have absolutely no idea, nor do I care because I would never, ever, expose by skin tone. If I expose for whites -- skin tones will be exactly where they need to be whether the subject is a Nigerian, a Swede, or an albino. And by the way, whatever the correct exposure is -- it is exactly the same for all three of those people in the same setup. Unless you're going to try to tell me that everyone in the world has the same shade of skin, then setting exposure based on an unknown, variable value really just comes down to guessing. But I'd rather be precise about exposure and that can only be achieved with white, 18% gray, a chip chart, or some other reference of known reflectance. In fact, using zebras on white is exactly the equivalent to taking an incident light meter reading, except its a whole lot easier.
Click to expand...


Fair enough :) I'm not about to "tell" anyone anything, just simply curious about different people's methodology and I can't say I disagree with your logic.


I'm always interested in learning / advancing my skill set. I'm not a believer in the "there's only one way to build a house" mindset and there's quite a lot of that going around these days. I read / listen to how different people approach different challenges - if it fits in my workflow or gets me a better result, I'll adopt it - if it doesn't, I disregard it. I just appreciate other professionals in our field taking the time to contribute to forums like this.
 
digitalinnovations said:
I'm not a believer in the "there's only one way to build a house" mindset and there's quite a lot of that going around these days. I
Click to expand...

I totally agree that there may be several ways to successfully build a house, but that doesn't change the fact that there are even more wrong ways to build a house. :-)
 
I rarely jump in on the exposure conversations anymore since I now know how to nail it every time. It started with a ride back from Boston with my friend Doug Jensen years ago. Remember Doug?

Doug was too the point in the conversation which I always appreciate. Doug told me as he has here, how easy it really is, just trust your zebras. Learn the right zebra setting and use it. Done!
 
You must log in or register to reply here.
Back
Top