Proper exposure on URSA Mini 4.6K

Hi Guys,

I’m a little confused about exposure on the URSA MINI 4.6K. I’m thinking, if this camera isn’t overrated? Does it really has native ISO 800?
I did some tests using an incident light meter set to ISO 800 and the appropriate frame rate and shutter settings. I was very disappointed as I saw the results in DaVinci Resolve 12… Proper exposed face by light meter is about 370 (range from waveform). RAW footage seem to be ca. 1.5 - 2 stops under.

Am I doing something wrong?

Thanks for Your answers!
Fryderyk

Like the other BM cameras, you don't expose it, you saturate the sensor (unless you shoot in video mode). So your lightmeter is only useful for ratios.

In a nutshell: set your zebra to 95%, expose till it goes away (except stuff like headlights, chrome etc), and than stop down a little bit more to make room for a nice rolloff.

In a nutshell: set your zebra to 95%, expose till it goes away (except stuff like headlights, chrome etc), and than stop down a little bit more to make room for a nice rolloff.

Hey Frank, thanks for the tip on exposure. Has your method changed since you created you post footer which reads:

"Set your zebras to 100%, ETTR and you're golden - and NO, you can't use TB as output for an external monitor, and you can't download the footy via TB ether."

Either way, are you sure you can't download the footy via TB?

I was under the impression you didn't have to ETTR on the UM46.
Mine's not back from RMA yet so haven't done much testing yet.

I always used franks ETTR methd for exposing on the BMCC and pocket.
Since I have had the UM46k that method has not proved useful.
The dynamic range is so high that if you expose for the highlights it just doesnt work well.
I asked around on these forums and found that false colour is very much the way to go.
My footage looks amazing since switching methods.

Ok, but what with light meter? Do I understand it right - the light meter is useless?! Can we really use it only for ratios? Is there any possibility to make the proper exposure (especially face) with light meter?

How do you develop RAW footage in Camera RAW in Resolve? Especially when we talk about gamma.

Quote from manual:
„Gamma: -> BMD Film 4K: A log-encoded gamma setting that approximates Log-C encoding, but that’s optimized for the BMD 4K and URSA cameras.”

I think we can become better results with Gamma 2.6. I read another posts about big difference in developing between ACR and Camera RAW in Resolve 12. I really see this difference, but I don’t mean the highlights. The results in DaVinci seem to be underexposed.

I think we can become better results with Gamma 2.6.

Gamma 2.6 is the standard for theater projection with DCI-P3 color. Gamma 2.4 is the standard for reference monitors with REC709 color and dim environment. 2.2 Gamma is the standard for computer monitors with sRGB color and was chosen for brightly lit office/home environments.

I'm assuming you are not grading in a theater. If you are used to grading ProRes log images, then choosing BMD color and BMD film as the gamma in the raw tab will get you a log image to begin your grade. If you prefer to begin from a contrasty REC709 image, you can choose REC709 as the color option and for the gamma option choose either 2.4 or REC709.

Hey Frank, thanks for the tip on exposure. Has your method changed since you created you post footer which reads:

"Set your zebras to 100%, ETTR and you're golden - and NO, you can't use TB as output for an external monitor, and you can't download the footy via TB ether."

The footer was for the "old" BM cameras, the UM sensor/CS is a bit different, so 95% it is - at least for me.

Hey Frank, thanks for the tip on exposure. Has your method changed since you created you post footer which reads:

"Set your zebras to 100%, ETTR and you're golden - and NO, you can't use TB as output for an external monitor, and you can't download the footy via TB ether."

The footer was for the "old" BM cameras, the UM sensor/CS is a bit different, so 95% it is - at least for me.

Cool, thanks. I've been setting my zebras to 100%, so I'll try that method on my next shoot.

Respectfully disagree that 95% zebras should be a hard and fast rule. I was shooting with 90% zebras and saw no clipping, but when grading the video, a blue channel appears to clip using BMDFilm. I could pull it down though so I guess this wasn't a hard clip. I've set my URSA Mini 4.6K to 85% zebras to keep me a decent margin below full ETTR.


Sent from my iPad using Tapatalk

Respectfully disagree that 95% zebras should be a hard and fast rule. I was shooting with 90% zebras and saw no clipping, but when grading the video, a blue channel appears to clip using BMDFilm. I could pull it down though so I guess this wasn't a hard clip. I've set my URSA Mini 4.6K to 85% zebras to keep me a decent margin below full ETTR.


Sent from my iPad using Tapatalk

I've also noticed that the blue channel is easy to clip.

I set the zebras to 85% when shooting outdoors in bright light to help preserve the blue channel.

Troy, that's a good point. I am finding that is more likely to happen with blue outdoors. Haven't clipped on red or green.


Sent from my iPad using Tapatalk

Are we all talking RAW for the most part here? When you ETTR in RAW at ISO 800, isn't the image too bright to see in many cases? You can also just stop down the ISO to 200 to get something better to view while still starting with ISO 800 to set exposure based on zebras. Since everything is ISO 800 anyways right?

With the 4.6k, I've also been using false color more than waveform or any other type of monitiring system. I find that ETTR can work in certain conditions (I also set it at 95%).

Unless the scene includes sufficient dynamic range, ETTR generally requires reducing the exposure in post to get skin tones in the proper range. And, in my experience with the defective 4.6K sensor I got, dropping the exposure in post made the magenta vignetting problem worse. Has anyone else noticed this when using the ETTR method on the 4.6K?

David, the URSA Mini 4.6K raw image is not as flat as the BMCC for example. When I expose with 85% or 90% zebras, I can always see an acceptable image on the 5" monitor or the Viewfinder. Of course I'm not doing true ETTR which would have zebras at 100%. I find the 4.6K a pleasure to view through the Viewfinder. I think I'll give false colour a try this weekend.


Sent from my iPad using Tapatalk

Jamie, I've seen that too.


Sent from my iPad using Tapatalk

How about this...

Use a light meter or false color, or both. Set up your light meter, click the button, it tells you the fstop. Set the lens and ND if needed. Look at the monitor, adjust to taste or to fit the scene. Check where the skin falls with false color. Check your contrast ratios.

Look at the image in Resolve. Too much noise? "Overexpose" by 1 stop. Get it to where you can accept the noise level.

All this extreme ETTR or ETTR blindly isn't going to do you any good. It won't be consistent.

So do it this way. Be as consistent as possible.

You know... the old fashioned way... the way that works.

Jamie, I've seen that too.

I hope it hasn't been on images from the PL 4.6K you got.

Same experience from mine. Lowering exposure in post might end up in very obvious magenta vignetting.

Sent from my Nexus 5X using Tapatalk

I believe ETTR is wrong. I would never ever use ETTR or "Back off from clipping"
I'll outline my reasoning here and would invite anyone who has used these techniques with better results to defend their use.
So. every sensor has a response curve. What the OP is talking about is where the midtones on the UM4.6 fall on the curve which is not linear and he's seeing them fall under where he would like for skin.
The best response would be to shoot tests to find where he's comftorable with skin landing plus or minus what his meter reads for ie. shoot a stop over and under etc and then correct for his look. Another option would be to use a Sekonic Exposure Profile and map the sensor for the same effect.
There are two things that need to happen on set. The scene needs to be lit so that skin works both for exposure and ratio, and that exposure and ratio needs to match across cuts in the scene. The chances of these very important and I would say primary goals go down very dramatically when we start talking about saturating a sensor or exposing to 95 and backing off. Why? Well first off skin is not a easy thing to work with on any camera when it's over or under exposed, you want to be within half a stop of your target exposure that you found during pr-production with the camera tests I mentioned earlier. Second as we move around on set the way to keep shots consistent it to meter skin and to hit the same exposure in different setups. Again this is not going to work with backing off of 95 as the angle to light or the background changes, now we are in a position where we have to match varying levels of light on skin and that's going to hurt the image and the whole scene.
The reason these methods are out there is that they are a way to keep the image out of the noise floor. It's easier to say to a DP that they didn't saturate the sensor than it is to acknowledge that the noise floor on different camera's may be higher or lower and this has led to a lot of DP's losing sight of the primary responsibility of a DP mainly: to expose the actors so that they look best and to deliver scenes that don't need color correction just to look coherent. False color and meters are the only valid tools here and a DP must have a target value for a camera (or film stock) that he or she has come to by testing in pre-production.
It's really that simple.
Now as for the noise issue. If after testing a DP decides to let skin take a hit and to expose a stop over so as to land shadows higher vs where the noise floor of a particular sensor is that's completely fine. DP's have been over and under rating stocks and sensors forever and some have preferences one way or the other and those preferences may change given the look of the film.
ETTR insists that we feed the sensor light until it's saturated as a way to make up for a shortcoming, in this case the higher noise floor and really is one of the worst ideas in my opinion, I mean are we are trying to convince the OP here not to meter and to just back off from zebras? Thats wrong. That's totally confusing and leads to terrible results. Again I would really like to here from anyone who is pushing this ETTR idea show some work examples and outline the methods they used.

I believe ETTR is wrong. I would never ever use ETTR or "Back off from clipping"
I'll outline my reasoning here and would invite anyone who has used these techniques with better results to defend their use.
So. every sensor has a response curve. What the OP is talking about is where the midtones on the UM4.6 fall on the curve which is not linear and he's seeing them fall under where he would like for skin.
The best response would be to shoot tests to find where he's comftorable with skin landing plus or minus what his meter reads for ie. shoot a stop over and under etc and then correct for his look. Another option would be to use a Sekonic Exposure Profile and map the sensor for the same effect.
There are two things that need to happen on set. The scene needs to be lit so that skin works both for exposure and ratio, and that exposure and ratio needs to match across cuts in the scene. The chances of these very important and I would say primary goals go down very dramatically when we start talking about saturating a sensor or exposing to 95 and backing off. Why? Well first off skin is not a easy thing to work with on any camera when it's over or under exposed, you want to be within half a stop of your target exposure that you found during pr-production with the camera tests I mentioned earlier. Second as we move around on set the way to keep shots consistent it to meter skin and to hit the same exposure in different setups. Again this is not going to work with backing off of 95 as the angle to light or the background changes, now we are in a position where we have to match varying levels of light on skin and that's going to hurt the image and the whole scene.
The reason these methods are out there is that they are a way to keep the image out of the noise floor. It's easier to say to a DP that they didn't saturate the sensor than it is to acknowledge that the noise floor on different camera's may be higher or lower and this has led to a lot of DP's losing sight of the primary responsibility of a DP mainly: to expose the actors so that they look best and to deliver scenes that don't need color correction just to look coherent. False color and meters are the only valid tools here and a DP must have a target value for a camera (or film stock) that he or she has come to by testing in pre-production.
It's really that simple.
Now as for the noise issue. If after testing a DP decides to let skin take a hit and to expose a stop over so as to land shadows higher vs where the noise floor of a particular sensor is that's completely fine. DP's have been over and under rating stocks and sensors forever and some have preferences one way or the other and those preferences may change given the look of the film.
ETTR insists that we feed the sensor light until it's saturated as a way to make up for a shortcoming, in this case the higher noise floor and really is one of the worst ideas in my opinion, I mean are we are trying to convince the OP here not to meter and to just back off from zebras? Thats wrong. That's totally confusing and leads to terrible results. Again I would really like to here from anyone who is pushing this ETTR idea show some work examples and outline the methods they used.

Thank you. Worded much more elegantly than I did in my post. THIS is exactly it. There are no shortcuts.

Aharon, good summation of exposing for skin as the key common criteria across scenes and using lighting where you can to control the ratios within each scene. I haven't shot anything yet with false colour but will this weekend.


Sent from my iPad using Tapatalk

I mean are we are trying to convince the OP here not to meter and to just back off from zebras? Thats wrong. That's totally confusing and leads to terrible results. Again I would really like to here from anyone who is pushing this ETTR idea show some work examples and outline the methods they used.

Interesting theory.

Can you tell me than, why all my shots look great (especially skin) and consistent using my method (besides the fact, that I do very precise lighting and rations on my sets)?

Frank, I would really prefer to talk about craft and not your work. Sure you footage looks great but you might get better results with skilled metering . I want to be very careful not to offend as I do feel very strongly about the craft, and I do feel your technique is wrong. Again, not in any way offending you or your footage, I have a problem with the technique. Perhaps as we clarify our positions we can both learn from this conversation.
Let's look at overexposure on skin for a minute. Yes it's less of a problem than underexposure but the UM4.6 and every other manufacturer is playing highlights vs shadows vs midtones on the sensor, it's a balance between how much range on the curve we give each at base ISO. The toe and shoulder on that curve are designed to roll off highlights etc but the center of the curve is setup to maximise information in skin ie. Midtones. So when you land skin according to the meter readout you are placing it where the camera manufacturer designed it to land on the sensor exposure curve.
Now. It's most certainly a good idea to run a over and under test on skin with a new camera and see what works to preference. You may walk away from that saying ok I like my skin a stop over, that's fine because you don't really need to get very technical with sensor curves or whatever. You know what you like and you found that in the test. Thats what the OP was doing. Simple.
Very different story to start saying that the camera needs to be saturated and we should always expose to maximum and then back off. Why? Because that focuses on one area just one that of noise and forgets about skin. Suddenly you are actually messing with what the camera sensor is set up for. You don't really know if skin would have looked better if it wasn't three stops over and then pulled back. There's a reason Alexa skin looks so good and it has a lot to do with the sensor response curve as it relates to skin.
Now. Again about noise. This is not a this or that proposition. This can also be a this and that idea.
If you are seeing noise at 10 ire or the last stop and you want to get away from that you can make that trade-off by metering on set and adding a stop on the lens, that's a valid way to work but again it's a known push that still working from the meter, still working from skin. In darker shots it's the shadows that take a hit and are noisy not skin. The ETTR approach will put skin all over the place in order to keep shadows consistently high and noise free. If you are shooting a scene for a few hours or even different days when we get to color we are going to have to match to the worst looking shot. Much better to just focus on the thing that's most important, the midtones and that way we can work the highlights down in some shots or bring a window up or whatever but not need to start making our actors look worse to match that one shot with the bright skye behind the guys face to the shot where he's walking back to the car through tree shadows. Again not a this or that approach. We can still watch our zebras so we don't blow out but we now have consistent actors... You get the idea...
Hope this helps!

I think we can become better results with Gamma 2.6.

Gamma 2.6 is the standard for theater projection with DCI-P3 color. Gamma 2.4 is the standard for reference monitors with REC709 color and dim environment. 2.2 Gamma is the standard for computer monitors with sRGB color and was chosen for brightly lit office/home environments.

I'm assuming you are not grading in a theater. If you are used to grading ProRes log images, then choosing BMD color and BMD film as the gamma in the raw tab will get you a log image to begin your grade. If you prefer to begin from a contrasty REC709 image, you can choose REC709 as the color option and for the gamma option choose either 2.4 or REC709.

Sorry for butting in, Davinci Resolve just removed gamma 2.6 from it's workflow since V12. It's only 2.2 and 2.4. How and why is that if gamma 2.6 is the standard for theatre projection?

Yes, this works best for situations with basically uncontrolled lighting, like outdoor shooting. But Frank's approach is valid, in that he lights all of his scenes with carefully metered lighting ratios, so the mid tones,/skin fall into a consistent exposure. When I set up a shot, I set exposure mid level for the subject so it is correctly exposed based on desired IRE level, then light highlights and shadows accordingly for the camera being used. Same technique used for film, only now the camera sensor replaces the film type. Then you use Zebras to insure you are not clipping any highlights in the process, or setting the roll off where you want it. With experience and using constant lighting ratios, like Frank does, you can the Zebras to check the overall scene exposure.

Back in my ENG days, we would set Zebras to indicate the skin/mid value and expose for that, especially out of doors in uncontrolled lighting situations. Both systems work, it is a matter of how you are used to working, and when using a new sensor/cameras doing the required tests to check the new system out against your work methodology. There is no "right" or "wrong" way to meter, IR use other exposure tools. I use meters, waveform monitor and zebras when setting up a shot, others like false colors, we all have our own unique way of working. And that's my two cents worth...
Cheers

Yes, this works best for situations with basically uncontrolled lighting, like outdoor shooting. But Frank's approach is valid, in that he lights all of his scenes with carefully metered lighting ratios, so the mid tones,/skin fall into a consistent exposure. When I set up a shot, I set exposure mid level for the subject so it is correctly exposed based on desired IRE level, then light highlights and shadows accordingly for the camera being used. Same technique used for film, only now the camera sensor replaces the film type. Then you use Zebras to insure you are not clipping any highlights in the process,,or setting the roll off where you want it. With experience and using constant lighting rations like Frank does, you can the Zebras to check the overall scene exposure.

Back in my ENG days, we would set Zebras to indicate the skin/mid value and expose for that, especially out of doors in uncontrolled lighting situations. Both systems work, it is a matter of how you are used to working, and when using a new sensor/cameras doing the required tests to check the new system out against your work methodology. There is no "right" or "wrong" way to meter, IR use other exposure tools. I use meters, waveform monitor and zebras when setting up a shot, others like false colors, we all have our own unique way of working. And that's my two cents worth...
Cheers

Denny I respectfully disagree.
As outlined earlier there are two fundamentally different approaches to exposure. Please explain the benefit of ETTR and backing off of clipping vs using a meter after camera tests.
Thanks

Oh I'm enjoying this thread.

I believe ETTR is wrong. False color and meters are the only valid tools here and a DP must have a target value for a camera (or film stock) that he or she has come to by testing in pre-production.
It's really that simple.

Hi Aharon, I'm new to using false color but wish to understand it better. So if skin tones are best with a light grey or pink false color (which color depends I suppose on darker or lighter skin tones?), does this mean that wether I'm shooting a bright beach scene or a dark cave scene with torches, I want to expose faces for that same light grey or pink false color consistently?

I shot some interesting (well of interest to me) tests today with exposing by zebras and false colour. Hope to upload them later. But I did one test exposing with zebras set at 100%, 95%, 90%, 85%, 80%, and 75%. CaptainHook is correct that the change in exposure is not linear. The greatest difference in light levels was from 100% to 95% as determined by looking at the waveform in Resolve.

I shot raw of an interior scene with a window in the background. Outside was very bright but most of what was in view was just my backyard fence, a weather-greyed red cedar fence. Exposing for zebras is not reliable as we all know because if you are clipping in one colour channel, zebras do not show. But what I saw looking at the parade surprised me. At 100% zebras, I was strongly clipping red as I expected. At 95% still strongly clipping red. Even down to 80%, red was clipping. At 75%, red was almost all within bounds! From 100% to 75% is a difference of one-half stop exposure level.


Sent from my iPad using Tapatalk

I did another test that taught me something that could be useful in exposing for EXT Night scenes. I haven't read anything detailed about exposing day-for-night. So this might be old news. I thought to do day-for-night, you simply reduced the aperture or added so many stops of ND. So I compared a bunch of exposures with the aperture going from T22 to T2.1. Sure T22 is a lot less light. But T22 or T16 just looked darker of course, but didn't look right.

Then in one clip, I ramped the day scene from T22 to T2.1. Still boring, right. Same as the individual clips at various T-stops. In Resolve, I positioned at a frame (in the clip with ramped aperture) that was at T2.1 and hit Auto Colour. It was boring at T2.1, but at T22 and T16 in that ramped clip, it was a lovely day-for-night radiant deep blue. Just needed to add some star light! Much better than Auto on the clip that only had T22 frames. Hope that was clear, but when I can add the test to Vimeo, you'll see what I mean.

If Ken Burns can attach his name to a simple technique of making a movie by panning and zooming around a still frame, then I'll call this technique the Lang Night shot.


Sent from my iPad using Tapatalk

Lang Night - love the ring of it

All Night Lang?

Joerg and Frank, thanks! My Scottish great-grandfather (or was it his father?), changed his last name from Laing to Lang just because he thought it was simpler apparently. I understand Laing means 'long' so it is a bit of a play of words that Joerg picked up on exactly.


Sent from my iPad using Tapatalk

I recently tried to ETTR the UM46k sensor and I had to work extra hard to fix the skin in post (I'm not a colourist).

My process is to use a light meter to get it in the ball park and use false colour to dial the skin and I get great results. Without false colour I feel I'm shooting blind.


I


Sent from my iPhone using Tapatalk

Joerg and Frank, thanks! My Scottish great-grandfather (or was it his father?), changed his last name from Laing to Lang just because he thought it was simpler apparently. I understand Laing means 'long' so it is a bit of a play of words that Joerg picked up on exactly.


Sent from my iPad using Tapatalk

Haha, a perfect match then. Great!

Olan, from the tests I did yesterday, shooting with false colour on skin was giving me a brighter exposure in camera than shooting with my usual 90%-85% ETTR and that surprised me. So feeding the sensor may actually sometimes be helped by relying on false colour in exposing a scene rather than using a 'safe' ETTR.

So reviewing the history briefly, but in general terms:
1) with the BMCC we were told to feed the sensor and use ETTR with 100% zebras,
2) when we found out a single channel could clip in ETTR, we were advised to shoot ETTR with 95% zebras,
3) the concept of exposing for the scene average when shooting on the URSA Mini 4.6K sensor gained traction since the wider latitude could help take care of the highlights and shadows when mid-tones are exposed correctly, presumably using an exposure based on the 18% grey card or light metering and controlling the zones like Ansel Adams advocated knowing what the media could handle,
4) Frank reminded us that ETTR with 95% zebras can still be effective for skin when using well-controlled lighting and many of us have seen and admired his capture of skin in many different and challenging situations,
5) the use of false colour to match exposure of skin tones across shots and scenes is now recognized as a very useful technique in camera to help speed up the grading process
6) I and some others thought even 90%-85% might be safer but yesterday I realized if you really want to ensure no channel clipped 75% zebras may be required in camera (but we don't have the tools in camera to know), and
7) yesterday I realized sometimes exposing for pink false colour skin can actually be a higher exposure than ETTR with zebras protecting the highlights which means some clipping may be risked unless the brightest highlights are controlled by gels or other means.

All in all, it seems good to know what works best for you and how to use that technique effectively. Zebras, false colour, zones, average, 18% grey are all tools for the DP. When I turned on false colour, I always had 'safe' zebras on anyway just to give me some additional information.


Sent from my iPad using Tapatalk

What I would really like to put forward here is a movement away from the DIT or color or some technical aspect of using a digital camera taking responsibility for the image and towards the DP taking responsibility for the image.
Case in point the Zaccuto camera tests from a while back. Different DPs with different cameras using the same room, actors and lights and delivering very good results across the board.
This is happening because they aren't backing off of highlights you know saying there's this technique to always use etc which is sort of throwing responsibility on to the camera and instead working that dynamic range into the camera by taking responsibility as a DP to understand the camera by shooting lighting tests with a meter ahead of time.
Case in point: Blue Ruin which is fantastic and shot on a C300.
The question can be framed: do you know what you want and how to get it with this camera?
Very easy to say "Yes I like a denser, darker negative, yes I like skin a bit muddy, I like a smoother lower contrast vibe, I like shadows and blacks, and I'm going to shoot everything a stop and a half under the meter because on this camera that will put me in that place and I know I've got 6 stops under midtones from my camera test but those last two are really noisy so I'm looking at zebras to keep my sky from blowing out but I'll never let skin go below 4 stops under because it will fall apart in post. So yes sometimes I'll take the call to let the sky clip.
This is a DP in control of the image.
I mean you show up with the ETTR thing and backing off of 90 percent or whatever and all you really know is that you like to have a saturated sensor. You've got a technical approach that's not really technical at all. It says I did my job by feeding this camera a lot of light and now someone else will have to grade it to match and really grade it to look like anything really.
Again nothing against zebras, but use them to know when you are clipping not as a weird all in one exposure tool

Hmmm...

Here is one ETTR (Zebras 95% and back up a hair) vs. a shot exposed with a light meter

All shot in uncompressed raw, CC in Resolve, just a bit curve tweaking plus LUT and lowered exposure on the ETTR.

ETTR

ETTR.jpg (852.09 KiB) Viewed 18506 times

Let's look at overexposure on skin for a minute. Yes it's less of a problem than underexposure but the UM4.6 and every other manufacturer is playing highlights vs shadows vs midtones on the sensor, it's a balance between how much range on the curve we give each at base ISO. The toe and shoulder on that curve are designed to roll off highlights etc but the center of the curve is setup to maximise information in skin ie. Midtones.

You are mixing up film and tone mapped delivery curves (i.e. rec709 with a soft clip, or anything else meant for delivery) on one side, with linear and log signal (meant for post). There is no shoulder with any linear signal, or any real log curve (except Cineon and its derivatives, which are supposed to encode film scans and do replicate the shoulder of the negative): there is no shoulder over log-exposure on Arri's LogC, or Sony's S-log, or any other log curve with a true log upper part.

The consequence is that with most digital cameras shooting raw or log you will get the cleanest signal using ETTR (and yes, this includes the best and densest skin). Whether this is the best approach for your colorist, who will then need to match exposures shot-to-shot, is another story.

How do you guys know that the 2 methods don't produce similar results? I've tried both that are mentioned and 9 times out of 10 they're very similar. Only shots that may be an exception are very bright backgrounds.

Would love to see your tests, Rick.

Okay, Scott, I'll post some still images hopefully today. Can't upload video to Vimeo yet. The stills are useful. And they show that the hint of magenta I had on the right side of the frame may have disappeared. Perhaps the comment about the cause of the problem being the sensor coating that had not yet evaporated is more correct than we thought! You'll be the judge.


Sent from my iPad using Tapatalk

Like a few of you have already noted, I'm also finding this thread to be interesting. Coming from stills I was weaned on Uncle Ansel's methods of exposure and development. For years plotting film curves was standard operating procedure. There's no doubt in my mind that sensors work the same way as film, each having individual characteristics in how they deal with saturation as well as more basic exposure information as they roll off in both the highlight and the shadow direction.

Ansel's rule of thumb was if in doubt shoot for the shadows and print it down, which meant shoot a little more open and burn the highlights. That was a little easier when the choices were between Royal-X and Tri-X, though even then if you really dug into these curves there was much more than meets the eye. In any case his exposure guide is what I've presumed ETTR is similar to.

Like most of you I've tried both. With field monitors having improved so much in recent years in addition to using a good meter, most of the time I have access to a tremendous amount of exposure information as I shoot. After looking at 3-4 exposure tools and the meter I've found it's pretty hard to be way out. Super contrast daylight is the most difficult scene for me as there are so many choices as to what to do. Once you're controlling the light and the ratios, it gets easier.

Here is a look at an interior exposed in raw using false colour to get the palm of the hand well into the pink. Note the considerable clipping of the fence outside on the right as shown by the parade and waveform.

False colour for a pink palm, no correction

Screen Shot 2016-06-07 at 9.56.01 AM.png (994.85 KiB) Viewed 18350 times

Here is the same shot exposed with zebras at 75%. Note the slight clip of the red channel! The scopes illustrate how much lower the levels are compared to false colour on light skin.

Zebras at 75%, no correction

Screen Shot 2016-06-07 at 10.09.32 AM.png (996.27 KiB) Viewed 18350 times

Now with zebras at 80%, green about to clip, red slight clipping. Scopes show modest increase in levels.

Zebras at 80%, no correction

Screen Shot 2016-06-07 at 10.17.56 AM.png (996.95 KiB) Viewed 18350 times

If you don’t see the scopes under the picture, scroll the image.

Here are zebras at 85%. Slight increase in levels.

Zebras 85%, no correction

Screen Shot 2016-06-07 at 10.26.11 AM.png (998.6 KiB) Viewed 18345 times

Zebras at 90%. Slight increase in levels.

Zebras at 90%, no correction

Screen Shot 2016-06-07 at 10.36.49 AM.png (1000.64 KiB) Viewed 18345 times

Now for zebras at 95%. Shows more clipping. Modest increase in levels.

Zebras 95%, no correction

Screen Shot 2016-06-07 at 10.41.09 AM.png (1000.03 KiB) Viewed 18342 times

And finally zebras are at 100% red clipping strongly. And showing a larger increase in levels from 95% zebras.

Zebras at 100%, no correction

Screen Shot 2016-06-07 at 10.47.05 AM.png (1002.88 KiB) Viewed 18342 times

Here is the false colour exposure once again to show a brighter image than exposing ETTR at 100% zebras based on the light coming in the background window!

False colour pink palm, no correction

Screen Shot 2016-06-07 at 9.56.01 AM.png (994.85 KiB) Viewed 18340 times

Very interesting to see the changes in light levels. Zebras at 75% are one-half stop below zebras at 100%.

Screenshots illustrating the Lang Night effect to expose day-for-night in DaVinci Resolve. It also illustrates any magenta that may be apparent as the aperture of the lens changes from minimum to maximum on the SLR Magic APO 50mm T2.1 prime. So i ramped from T22 through T2.1 in the same shot and then applied Auto colour correction in Resolve for a T2.1 frame to the whole whole shot. The results are very interesting and an easy way to get a nighttime backdrop shooting in the early afternoon on a sunny day.

Scroll the image if needed to see the scopes.
T2.1 used to determine Auto correction for the ramped shot.

Frame exposed at T2.1 Resolve Auto correction

Screen Shot 2016-06-07 at 11.54.32 AM.png (941.41 KiB) Viewed 18326 times

Best viewed in a darkened room with 7 stops less light:

Frame exposed at T22 using T2.1 Auto correction

Screen Shot 2016-06-07 at 12.02.34 PM.png (472.24 KiB) Viewed 18326 times

6 stops less light.

Frame exposed at T16 using T2.1 Auto correction

Screen Shot 2016-06-07 at 12.05.12 PM.png (615.65 KiB) Viewed 18326 times

5 stops less light. T11

Frame exposed at T11 using T2.1 Auto correction

Screen Shot 2016-06-07 at 11.33.24 AM.png (741.06 KiB) Viewed 18345 times

4 stops less light. T8

Frame exposed at T8 using T2.1 Auto correction

Screen Shot 2016-06-07 at 11.36.55 AM.png (855.74 KiB) Viewed 18345 times

3 stops less light. T5.6

Frame exposed at T5.6 using T2.1 Auto correction

Screen Shot 2016-06-07 at 11.40.05 AM.png (986.97 KiB) Viewed 18345 times