Having been forced to match s-log and non s-log footage from the same scene (due to pre v1.3 firmware that didn't allow for 60 fps capture with s-log), I would say the differences between the two are far from "slight". I like the images from the F3 with s-log. I don't particularly care for them without. For me it's not about satisfying a client, it's about capturing a look that I have in mind and having flexibility in post to be able to manipulate values as I choose.

The point I make is , are we being over concerned about a slight loss of detail in the highlights?


Not myself. I understand where you're coming from, I've been on plenty of shoots where somebody will latch onto one thing and try to make it unnaturally perfect for no good reason...

But badly burned highlights are a distraction in my book...and the F3 is not terribly good with them in the standard/Cinegamma profiles.

I myself am not overly concerned with detail and latitude for the sake of either, I just want my highlight burnouts to please the eye. Like on 35mm. Like on an Alexa. Like on a Canon DSLR. S-Log is the only way to get that I've found on the F3. Having the latitude to go with it benefits the image in other ways.

I have just had my first test using S-LOG filming my son relaxing after school. What do you think, I am sort of happy with the results though I may have been half a stop to open. It seems really easy to over expose slightly. Focussing seems a little more tricky as well as you are focussing on something with less light /exposure - if this makes any sense! Here is my test:

First F3 S-LOG trial on Vimeo

Mark,
Looks nice for a first test. Curious as to what you used to grade your work.
Color, Colorista, Avid, Prermier or FCP internal color correction?
Care to share?
Thanks

Thanks Chuck. I used Sony Vegas (have been using Vegas for 9 years now!). I just altered the levels (brought them up) added some saturation, then a little colour correction. These three effects, in this order, seem to work with my S-LOG footage. I don't know how many other S-LOG users use Vegas, but I haven't heard from many. I didn't play with the color curves with these clips but I think they look pretty good. I just think they may be half a stop over exposed. I am looking forward to trying out some shots in bright conditions. From what I gathered in the forum, you should only let the brightest part of any shot reach 80% on a waveform monitor and keep skin tones (on average) around 45. Would this be about right? Or this limited the dynamic range? Thanks for feedback, it's a real learning process for me. Sparky

There is one very obvious problem with exposing for white. White is not always 100% white depending on the lighting conditions, take for example a white Vortex card in sunlight, this will be a different shade of white in the evening. I am not referring to colour balance here, just exposure.

Hi Vincent,

I respect that you have a different approach, but my method of nailing the exposure via zebras has worked well for me for about 30 years. So, even if you think it wouldn't work and would destroy the ambience of the scene, I have to disagree. Underexposing is rarely the right approach no matter what the ambience is of the scene.

An exception would be if I was shooting a night time dramatic scene where I really did want my video to be very dark, but that is not what I shoot.

I can see the point of using Zebras, and do so myself. My concern is that we can become too dependant on instruments that measure exposure or colour temperature etc. Sometimes our own eyes and judgement skills are the best instruments.

I am sure that after 30 years experience with video you don't need your Grandmother teaching you how to suck an egg.

Quote; "Sometimes our own eyes and judgement skills are the best instruments."

That is true if you are talking about naked vision into the eye. But with video, the light is going through the camera first and then being represented on a display that isn't your eyeball. I have learned that you often cannot trust what the viewfinder or flip-out LCD is telling you about color or exposure. No matter how well they are setup, the ambient lighting conditions where you are shooting will affect your judgement of what those displays are telling you.

I don't feel like I am too dependent on instruments. And I hope the pilot of my next flight doesn't decide to ignore his instruments and fly the plane by instinct because he knows better. And I hope my audio tech doesn't decide to ignore what his level meters are showing just because he thinks it sounds fine in his headphones.

There are good reasons why video professionals have used scopes, zebras, histograms, etc. for decades. And photographers and cinematographers have used light meters for more than a hundread years.
Scopes are better than your eyes and will not lie to you when used properly.

Making decisions based on what the LCD or viewfinder is telling you, will put you on the road to disaster. Not everytime, but often enough that it is a bad habit.

I think one problem with this discussion is that we are talking about living and dying by rules in a an area that has room for rules to be broken, depending on the circumstances and what style of shooting one does.

There is a logic in assuming that the goal is to fit skin tones into a specific area of the exposure curve, with whites falling into their own place. The result of that is going to be a consistent and respectable image which is exactly what is required of many types of jobs (and will deliver the most range for post adjustment). It's what I would define as a "video" approach to image acquisition.

Conversely, with a "film" approach, the choice can be made to peg these values elsewhere. Rather than attempt to preserve the highlights from a window, one may opt to let them blow simply for the look (one example would be having to paper and light windows for day in a night or studio environment--you don't want to preserve detail of any kind for it to look real, the windows NEED to approach or enter clip to sell as daylight). And there may be times that one may want skintones underexposed as a dramatic choice. There is a big difference between exposing skin two stops under vs exposing "properly" and then attempting to darken them in post--at a certain point it just looks artificially suppressed in the latter circumstance.

This is by no means a indictment or endorsement of either style of shooting. I'm not a fan of the phrase "horses for courses" but it is applicable here. Video has always required a more delicate and precise hand with exposure than negative film due to the limited dynamic range, which resulted in adherence to rules to produce consistently good results. I personally rebelled against this and pushed and pulled it six ways to Sunday; sometimes it worked and sometimes it didn't (certainly with analog video it was far tougher to mess around with underexposure and avoid the whole thing sinking into murkiness). But if I was hired to deliver traditional, clean video, then by gum I toed the line and observed the zebras and fanatically avoided overexposure as Doug and Alistair have outlined here.

We have now the blessing of log and raw video which allows for a much more filmic approach if that is desired. And this gives us the opportunity to make creative choices, to break the "rules". Properly exposed video is no longer an absolute; it's a choice. Clipping is no longer a dirty word--if you want your highlights to blow, you can do so without them looking "electronic" (well, there's still a little way to go on this--it's still possible to get rogue color tints in overexposure, but so far I've been able to manage these in color correction).

Another way to consider this is whether one is acquiring the image or creating it. If the former, the goal is to find the single exposure that will deliver the greatest range of tones in the image and that, again, is what Doug and Alistair are describing (I hope--gents, please don't let me put words in your mouth if this is not accurate). In the latter circumstance, one has control over the individual exposure of everything in the frame and thus all are a creative choice.

I would submit that those seeking a specific answer to how to expose in log read as many opinions and watch as many "test videos" as they want, then turn off the internet and proceed to shoot a lot of tests for themselves. Take the footage into color correction and experiment. What is the difference between underexposing skin tone in camera vs knocking it down in post? Which delivers the look you are going for? Experiment, make mistakes. This is how we all learned to shoot before there were blogs and message boards.

Could not agree with you more Charles.

Exposure is subjective at the end of the day. With S-Log the way it over exposes is not unpleasant, so in an extreme lighting situation, or where that is the look you desire, then you over exposing is not the terrible sin that it was in the past.

However, if you are going to over expose or allow your highlights to blow out, make sure you have discussed this with the director, editor, colourist and however else is might be paying your fee, as once you do it, there is no going back. Of course sometimes you may find yourself in a lighting situation where it's unavoidable and it's then that your own skill and knowledge of how you over exposure is going to behave that becomes critical as you balance blown highlights with under exposure and try to find that optimum balance that the colourist can get the most from.

One issue with zebras and S-Log on the F3 is that Zebras don't go down low enough to be useful for anything other than your whites which if "going by the book" will be at 69IRE. Mid grey and skin tones are below the minimum F3 zebra levels.

Sony Australia seem to be matching the US deal for S-Log now. Videocraft are offering the same huge discount.

My concern is that we can become too dependant on instruments that measure exposure or colour temperature etc. Sometimes our own eyes and judgement skills are the best instruments.

Not to sound too snarky but try telling the DP or Gaffer with a light meter around his neck all day long that he's 'too dependent on instruments' . . .

and yes, our eyes and judgement skills are important, and more accurate when used in conjunction with those instruments.

The reality is that our eyes/brain combo is one of the least reliable instruments for measuring light there is on the planet. Great for seeing under varying conditions that cameras can't handle but horrible for an absolute measurement. We see relative and a camera sees absolute.

Real light measuring instruments (either in camera or not) are mandatory tools that our eyes just can't replace.

Yeah, but a hand held light meter and measuring tape are both 'old school' film techniques when you couldn't see what you were shooting in the camera.

It's struck me more than once that the light meter and measuring tape in their hand isn't going to tell me more than the spot meter in the camera, the wave form, histogram, zebras, peaking and focus assist in the monitor. Plus they usually take longer.

Often I have the aperture and focus set when those types are still futzing around.

I'm not saying it doesn't work IF your lenses are marked accurately and IF the spot meter is used correctly. There's just no need to do it that way anymore because we can see the results of any adjustment and double check it.

I've filmed plenty of musicians with well tuned guitars that still made crappy sound. I believe it will always be a combination of the available tools and your eyes that make it work.

Oh yeah... and your experience.

Duke, that's all assuming that you are able to use the camera itself as the meter--doesn't help much if you have to light one set while shooting on another.

I'll admit that I use my meters far less often than I used to, but they still come out when needed.

As far as focus, that is a changing art but I have yet to see a focus puller be as deadly accurate just by working off a monitor as the best of the guys who came up the old-school way (who are all now incorporating monitors into their workflow, but they still will generally pull a tape from time to time). And I have yet to see an operator consistently pull their own focus on a large sensor camera with enough speed and accuracy.

Technology has changed a lot of how we used to shoot in the film days, but there's a lot of relevance left.

I think in this case it's important to separate focus from exposure.

I once watched an "old school" AC pull focus by eye on a gymnast running directly towards lens and doing flips along the way - not once in many takes did he miss even a frame of focus & all done without a monitor. Now trying to do such a thing with exposure would be suicidal not to mention stupid. Horses for courses, use your instruments, be they in-camera or external meters in whatever way is most efficient and double check it by your subjective and experienced photographic eye.

And I have yet to see an operator consistently pull their own focus on a large sensor camera with enough speed and accuracy.

That's strange. I do it every day and have a number of friends who manage to be able to do it too. I have never used a focus puller in my life. I can see the advantage of that luxury, but I have never found it necessary.

adding another .02 cents on focus -
When you see the actor in the monitor starting to go soft it's too late - the take is blown.
With large format sensors and long, fast lenses your depth of field may often be far less than an unexpected lean or head turn -- the types of actions that are often easier to sense and adjust for when looking at the actor and not concentrating on a monitor. Combine this and actor improvisations with 'let's shoot the run-through' and an AC with magic fingers is worth his or her weight in gold.

I'd love to have a focus puller on all my s35 shoots. Doing it yourself is hard, not impossible, but when your also trying to judge framing, exposure and checking for all those other things that can spoil a nice shot, it's not really surprising that when doing it yourself you don't nail focus on a moving subject all the way through the shot. But focus is different to exposure. Focus is either right or not. Exposure is much more subjective, some people will prefer a brighter or darker image to others. When shooting a night scene you will want an overall darker look to a sunny beach scene. Faces in the night scene might want to be darker than in the day scene, then again you might need the night time face to pop against a dark background. A meter won't tell you how the exposure "feels" all it will tell you is the level.

If it was really as simple as exposing using a meter or scope, setting point "a" for level "b", then there would be no great cinematographers, as it would be something that anyone and everyone, including an automatic camcorder could do, just follow the rules and you'll have the perfect shot. But that's not how it works in reality. It's the personal touch, the DoP's "eye" that makes the difference. That's part of the reason why people like Roger Deakins and Conrad Hall get paid the big bucks that they do.

Doug, I stand by my statement. I simply haven't witnessed anyone pull their own focus with moving subjects consistently on a S35 or greater sized sensor camera to the degree I consider as accurate as a good focus puller can. I think that as usual, we are talking about different worlds, different expectations, different kinds of shooting.

I've spent the past few years watching 5D and 7D videos online and seeing shots that are lightly to horrifically buzzed, and I fully expect to see the same with C300 and F3 and whatever else down the road. I'm talking about bringing someone from head to toe into a closeup on, say, a 50mm at T2.8 and focus being solid all the way through the walk, for multiple takes. That's for starters. Then try it at a 75mm, or even 100mm. Granted, few focus pullers would be able to nail that last one. There actually is a threshold where it does become "easier" for the operator to ride the knob themselves, at very long lenses.

Pulling focus on a S35 camera is no different than it is on a 35mm film camera. In fact, when we started working with S35 HD six years, we all felt it was slightly shallower and thus more focus-critical than 35mm film (something about the depth of the emulsion vs the sensor). I never saw anyone pull their own focus on 35mm film, except perhaps in a documentary situation. AC's were never considered a luxury, they were a necessity. I can imagine that there are those out there who are learning the ability to pull their own focus out of their own necessity but whether they have achieved what I would consider critical success at it; there's the rub.

I spend my days buried in a 24" monitor, switching between two cameras and whispering focus notes to the AC's over the walkie. Two years ago I was the one with my eye in the finder, hearing the DP doing the same and me and AC looking at each other and shrugging because we weren't able to see what the DP could see in his monitor. Critical, dead-on focus is very hard to see in today's viewfinders and onboard monitors, focus assist notwithstanding.

Charles, yes two different worlds, but not with different expectations. As Alister says, there's really no debate about what is in focus and what is not.

It is my theory, and you can correct me if I am wrong, that focus pullers were needed back in the days of geared heads because the operator did not have a free hand to control focus. And often the cameras were too big and bulky to physically make it possible to reach the focus barrel and look through the viewfinder at the same time. But with a nice fluid O'Connor head, and an F3, I can pan, tilt, and focus simultaneously.

So, whether they are really needed or not with today's equipment, focus-pullers are grandfathered in and expected to be part of a high-end shoot. I totally understand that. And if I was used to that way of working, I'd be making your argument. But that doesn't mean they are absolutely necessary for focusing.

Hey, I'm not trying to change anyone's way of working, I'm just saying that I DO know operators that can pull their own focus on large sensor cameras. Perfectly every time? Nope. But, then again I I see plenty of soft-focus shots in big budget films, so nobody is perfect . . . even when they have only one thing to do.

I was going to mention something about soft focus shots in studio features. Usually that's a function of a bad take that was noted and discussed and then rectified, but in the edit, the choice is made to go with performance even though there are sharp takes to choose from.

Obviously focus pullers are fallible. And it is possible for an operator to pull their own focus for certain kinds of shots. But I do maintain that there are many, many instances when it is simply not practical to attempt to do both. Has nothing to do with union requirements or grandfathered-in traditions. Here's an example from one of my last operating jobs, on the short-lived series "No Ordinary Family"; we are pushing in on young boy walking down a hall towards us; camera is low to play the adjacent railing as foreground. As he moves to a doorway, camera pans around and booms up to his eye level and as he opens the door, the shot becomes over his shoulder to reveal a girl sitting inside the room.

In the film days, where one had to keep an eye on the viewfinder at all times, this would have required me to start the shot hunched/leaned over to the left, viewing a sideways image, then contorting through the 120 degree pan and three foot boom up. In the digital era I became more inclined to use the monitor for a shot like this, in fact I believe I removed the onboard and mounted it directly to the dolly so that I could pan the camera without having to twist my body around. Still, it took 100% of my attention to operating to make the shot perfect, with all of that happening in a matter of a few seconds. In that time, the actor was moving towards the camera that was converging on him; by the time the frame had landed, focus had shifted from foreground to background at a specific point and speed. The chances of any operator getting the converging focus part during all of that let alone the rack to the background, while performing a difficult move, would be extremely slim. That's just a reality.

Obviously not all shots are that hard. But on any given project, you never know when something tough might be thrown in there. Add to the complication that a job that doesn't include an AC is probably going to require the DP to operate himself, and that's a pretty heavy workload for one person.

If anyone has links to shots that they have operated and pulled focus on that are demonstrably difficult, please post.

That does sound like a tough shot, and I have to admit that I have had no reason to attempt anything at quite that level. All I'm saying is that for the vast majority of what I shoot, and what I see in films and broadcast TV, there is no reason why a focus puller would have been needed to obtain the shot. But, they do earn their living in a few isolated instances.

Can't speak to what you shoot, but as far as what you see in prime-time and on the big screen; we are going to have to agree to disagree on that end. Things are simply not what they seem. Any given piece of coverage may be just part of a much longer shot that runs the length of the scene, with multiple camera moves and other challenges. Plus, these days we get far less time to finesse anything. Schedules are getting absurdly tight. Rehearsals have been replaced by "take 1" and it's not uncommon to move on with a single take in the can.

I'm not going to say that one type of production is harder than another. I spent ten years shooting largely as a two-man band, so I understand that world as well. All I'm saying is that there's a lot of factors under the hood that make this much harder than it may seem and having an operator pull their own focus is not appropriate in many cases. There's a lot of interest on the manufacturing end to simplify the process of pulling focus and I'm sure we will see technological advancements that will make this a much more likely scenario in the future.

I recently shot a project that was spoofing reality TV on F800's, think it came up in a previous thread. I had two operators; one who had shot plenty of times with that form factor of cameras and had done some network reality, and the other came from an episodic/feature operating background, and no focus pullers of course. Without question the first guy was more attuned to pulling focus on the fly. But I still needed to constantly micromanage focus to get it absolutely tack-sharp.; mmany times it was off just a tiny bit. And that's on 2/3" cameras.

While closeups are tough to maintain focus when the person moves slightly, the harder stuff tends to be medium shots where the subject's eyes are quite small in the frame and it's hard to see in a viewfinder or onboard if you are 100% of the way there.

Anyway--we should probably get back to the subject of the thread!