What is your favorite workflow for keeping all DR available on MX?

[jake blackstone]

That makes sense...

It does? How so?

 

[Gunleik Groven]

It does? How so?

If you've been looking at this example:
http://reduser.net/forum/showthread.php?t=51308&page=7

Post#3 downwards, the point is that redGamma doesn't hold the highlights @ 800, and that's kinda fine if you know it and bring them back in post.

One of the points with RedlogFilm is to hold the highlights, even at different ISO settings, while redlog is (I guess, because it seems to be) calibrated for 320.

You'll have to compensate by lifting the bottom. If there's one thing redlog does, it is lifting the bottom...

 

[Fredrik Harreschou]

But it is a bit low in the lowlights, so I usually end up using a curve in the bottoml, lifting the lowlights into a range where the colorist can easilly see that he/she actually doesn't have to clamp it all down. It is important for the signal that this happens when still in RAW. When you're in RGB/YUV (which also applies when grading RAW: post the "RED development stage") the info is gone...

And this is a favourite subject of mine... So far only Clipfinder and RCX have a curve to use on the raw file prior to development. Resolve, Color and AFAIK even Scratch doesn't have this, and that leads me to frequently suggesting DPXs/10-bit AJAlog as online format even for RAW-capable applications. And after some A/B tests, I am usually listened to...

SCRATCH has curves in the RED (RAW) settings. To me, there are so many benefits of working with the full 4K image in a 1080/2K workflow (retiming, animated repositioning, stabilizing etc) that you to some extent give up if going the DPX route. But for grading in Color, doing a good DPX conversion before grading is probably a good thing.

 

[Gunleik Groven]

BTW...

I am advocating redlogFilm as an option for the preview on the RED to be capable of seeing more of the available latitude on-set. But to be clear: I want this as an option, not to replace the current redGamma. I actually happen to like redGamma a lot.

Simplifying a lot here... There's two main ways for looking at an on-set preview...

One is to go for "finnished look".
In that scenario, it is good to have some headroom in the top and in the bottom to take care of those occational peaks in the highs and lows - as long as you have someone in your signal-chain who knows how to reclaim that data in post.

As a DIT/DP you can calibrate the headrooms to your needs, by hightening the ISO in the camera and hightening or lowering the contrast on your monitor. The higher the contrast on your preview and the higher the ISO, the more protection you have. (Yup - the combination of those two is the key to that sentence...)

On the other hand, you have the philosphy of NOT seeing a sexy image, but what is available for processing in post. I am a big fan of that because it gives ME more controll - but at the expense of a crappy on-set image... -

Sorta...

because if the on-set image was based on a cineon curve - like redlogFilm - I could load a LUT on he directors monitor, to get the punchy "finnished" look - while still see "what is in there" on my own.

At the moment I am in situations where - depending on the DP - both options would have been great.

"View RAW" doesn't really cover the bases, as it is calibrated for rec 709 @ 320...

 

[Gunleik Groven]

SCRATCH has curves in the RED (RAW) settings. To me, there are so many benefits of working with the full 4K image in a 1080/2K workflow (retiming, animated repositioning, stabilizing etc) that you to some extent give up if going the DPX route. But for grading in Color, doing a good DPX conversion before grading is probably a good thing.

Cool.

I deliberately tried to be carefull with what applications I mentioned, as it is a long time since I tried Scratch and I know of other RAW graders that will be implementing the curves shortly.

Actually Color was not the only app I had in mind, even though it is the app here used here... I have had a fun thing going on with EQ lately and also other applications. I think it might get fixed shortly, though for the ones I have tested...

I def agree with the benefits of working from a 4k image - if you can get the whole DR in without transcoding.

My point is that that is not allways accessible. -

 

[jake blackstone]

redlog is (I guess, because it seems to be) calibrated for 320.

Now you're talking crazy
You keep mixing up ISO, EI and Gamma, as they somehow interchangeable.
Correctly exposing sensor is THE most important aspect of the process and it has nothing to do with Gamma choice in post. As I already said, if you miss the exposure, any choice in post of different Gamma, Color space, FLUT, ASA, Exposure etc. will not do you any good.

 

[Gunleik Groven]

Now you're talking crazy
You keep mixing up ISO, EI and Gamma, as they somehow interchangeable.
Correctly exposing sensor is THE most important aspect of the process and it has nothing to do with Gamma choice in post. As I already said, if you miss the exposure, any choice in post of different Gamma, Color space, FLUT, ASA, Exposure etc. will not do you any good.

LOL

Agreed that getting the "right" exposure is the most important! - (Not that I am mixing up the terms, though...)

But there are different ways of getting there depending on the DP' preference - as to the "correct exposure" part of your reply.

BUT
The gammas etc affect a lot on how people use the camera on-set and in post. Thus it is not irrelevant.

Just take a Redgamma wide-latitude exposed image and play with the ISO/gamma options in post, and you'll see what I mean.

Basically you have already done that, as you go redlog 320 for posting a redGamma 800 image...

What you see matters too, not only what is "in the file" -

 

[jake blackstone]

LOL

Agreed that getting the "right" exposure is the most important! - (Not that I am mixing up the terms, though...)

But there are different ways of getting there depending on the DP' preference - as to the "correct exposure" part of your reply.

BUT
The gammas etc affect a lot on how people use the camera on-set and in post. Thus it is not irrelevant.

Just take a Redgamma wide-latitude exposed image and play with the ISO/gamma options in post, and you'll see what I mean.

Basically you have already done that, as you go redlog 320 for posting a redGamma 800 image...

What you see matters too, not only what is "in the file" -

Are you saying, that depending on the Gamma curve used in post, you recommend changing your camera exposure? Wow!
What I would've done, if I was a shooter, is I would try to come up with some kind of rule of thumb or a table for adjusting the effective camera EI for over and under the base exposure, depending on scene contrast ratio. In other words, if I'm shooting with MX camera and I'm using ASA 800 as a base, what would be my ASA increase or decrease for a scene with different contrast ratios? And, yes, you should use the FilmLog to get the maximum dynamic range in post.

 

[Pablo Villegas S.]

Now you're talking crazy
You keep mixing up ISO, EI and Gamma, as they somehow interchangeable.
Correctly exposing sensor is THE most important aspect of the process and it has nothing to do with Gamma choice in post. As I already said, if you miss the exposure, any choice in post of different Gamma, Color space, FLUT, ASA, Exposure etc. will not do you any good.

First...
Coming from a film background, I couldn't believe it myself, before actually doing it. But yes, starting grades from redlog @320 iso gives me a higher range of correction than grading from redgamma @iso 800, which was used to shoot it.

Second...
What is a correct exposure? In film it used to be placing your middle gray exactly at the middle of the curve, but then you usually overexpose in a dark environment, to get more shadow detail, and underexposed in a light environment to protect the highlights from blowing out. And then you used to take that middle gray back at the middle in the printing process.

Right now, at least to me, the correct exposure is getting all the DR you can out of the sensor, and all detail you can as well, so then you can bring the DR to where you want it in post.

And to me that means shooting and exposing @800 ISO and then grading using Relog @320 ISO. That's what I found it gives me the best DR in the MX sensor, in my limited experience with the sensor, I still need to try some other different things, but I don't have a chance of doing tests, I'm doing them with broadcast footage...

Regards.

 

[jake blackstone]

First...
Coming from a film background, I couldn't believe it myself, before actually doing it. But yes, starting grades from redlog @320 iso gives me a higher range of correction than grading from redgamma @iso 800, which was used to shoot it.

Second...
What is a correct exposure? In film it used to be placing your middle gray exactly at the middle of the curve, but then you usually overexpose in a dark environment, to get more shadow detail, and underexposed in a light environment to protect the highlights from blowing out. And then you used to take that middle gray back at the middle in the printing process.

Right now, at least to me, the correct exposure is getting all the DR you can out of the sensor, and all detail you can as well, so then you can bring the DR to where you want it in post.

And to me that means shooting and exposing @800 ISO and then grading using Relog @320 ISO. That's what I found it gives me the best DR in the MX sensor, in my limited experience with the sensor, I still need to try some other different things, but I don't have a chance of doing tests, I'm doing them with broadcast footage...

Regards.

You can't compare Redlog@320 with Redgamma@800. It's insane. They are two different animals. Shooting with Redgamma camera setting has nothing to do with your recorded material. It's just a matter of monitoring the signal, before it gets recorded. But once it's recorded, it's just a metadata. You should really monitor your camera in RAW mode. It is much better way of doing it and will alleviate the confusion, which you seem still have, coming from film background.
And, if you don't know what is a proper Red exposure, may be it is better to do camera testing for yourself, rather than going by what someone else said, that may be correct, but may be not Are you willing to trust someone else's woo-do? Not being able to test, doesn't mean you just give up...

 

[Pablo Villegas S.]

I do not shoot with redgamma, I use Redcolor, and set the ISO to 800, that way You can properly expose using that ISO, If You use RAW, you're exposing @320.

I'm not comparing the two, all I'm saying is that having shot and exposed using Redcolor @ 800 ISO, grading from redgamma @800 gives you a lower correction level than choosing Redlog @320 ISO.

I'm not exposing from what someone else has told me. I don't know where you got that idea.

 

[Gunleik Groven]

No offenses here... -

But if you are using the 800/redcolor setup, the in-camera gamma is redgamma.

You can change that in post to whatever you want.

We really don't disagree at all, actually we agree totally.

LOL

 

[Bill Anderson]

Good. Looking foreward to testing it -
I am not all that certain, but I am easily persuaded if I see that things work "right" -

Key question is how the signal comes out of the metadata-stage, what you can do with metadata before you start grading and what kind of signal you'rer grading from.

So far, the RAW graders I've been looking at have had issues in this first step, but I haven't done them all...

@ D Fuller

If for nothing else:
Digital ISO works the exact opposite way of how you would think with ISO on a filmstock.

Heighten your ISO and get MORE headroom in the top, lower it and get less. (I am not sure that saying the oposite about "protecting your lowlights" is entirely practically right, , though. Theoretically: Yes. But still in dayly use...)


Generally you would have chosen a high-iso stock when you have little light - shooting film, and low one for outdoor shooting at daytime.

I don't have a good alternative term comming up, though
It would lessen the confusion if there was some other easilly explainable term, though.

Gunliek, does "Digital ISO" work in the "exact opposite way" from film?
If I'm shooting an ISO 200 stock and find that the highlights have a tendency to blow out too quickly then I might be tempted, in some situations, to set the (camera meter) to ISO 400, this would give one stop less exposure. ISO 400 would help protect my highlights from all but the shoddiest work habits.

And, again, if you are in a low light situation with red, you'd up your ISO (stock) (just like film) to pull out as much info .... So I can't see how it works the "exact opposite" ?????

However, saying "expose to the right" to mean protect the highlights IS exactly the opposite of film practice. "Expose to the left" means to protect the highlights. Exposing to the right (of the film curve) is to push the highlights up and off the curve. Perhaps we mean "expose FOR the right".?????

 

[Gunleik Groven]

Are you saying, that depending on the Gamma curve used in post, you recommend changing your camera exposure? Wow!

LOL!

Not at all.

I am saying that if things are exposed well to the right @ 800 (which would be redgamma) I need to step back to reclaim the highlights in post.

That leads to a darker image than the one desired, so I have to compensate in the low/mids with a curve. That way I can have all the highlights and lowlights actially recorded (but quite often not seen on the monitor) available in post.

 

[Pablo Villegas S.]

Gunliek, does "Digital ISO" work in the "exact opposite way" from film?
If I'm shooting an ISO 200 stock and find that the highlights have a tendency to blow out too quickly then I might be tempted, in some situations, to set the (camera meter) to ISO 400, this would give one stop less exposure. ISO 400 would help protect my highlights from all but the shoddiest work habits.

And, again, if you are in a low light situation with red, you'd up your ISO (stock) (just like film) to pull out as much info .... So I can't see how it works the "exact opposite" ?????

However, saying "expose to the right" to mean protect the highlights IS exactly the opposite of film practice. "Expose to the left" means to protect the highlights. Exposing to the right (of the film curve) is to push the highlights up and off the curve. Perhaps we mean "expose FOR the right".?????

I don't know why he says is the opposite way...

The thing is ISO is just metadata, the recorded image is RAW, which doesn't take ISO into consideration, the "native" RAW ISO is @320, or at least that's the closest thing to redcolor, monitoring.

So, if you are shooting @ 800 ISO, you are already underexposing 1.5 stops the recorded RAW, so you are already protecting the highlights in the RAW image. So, if you expose to the right in redcolor, making sure your highlights won't blowout, there is no way they will be blown out in the RAW image, since it's underexposed, that way you get the most detail in the shadows, protecting the highlights.

If you shoot lower than ISO 320, you're overexposing the RAW image, so, you need to monitor both the redcolor, and the RAW data so you wont blowout the highlights. I did this once, but you have a smaller DR at the top, so you need to control your highlights real well.

Right now, I just shoot @800 ISO, since I need the extra speed.

 

[Gunleik Groven]

Gunliek, does "Digital ISO" work in the "exact opposite way" from film?
If I'm shooting an ISO 200 stock and find that the highlights have a tendency to blow out too quickly then I might be tempted, in some situations, to set the (camera meter) to ISO 400, this would give one stop less exposure. ISO 400 would help protect my highlights from all but the shoddiest work habits.

And, again, if you are in a low light situation with red, you'd up your ISO (stock) (just like film) to pull out as much info .... So I can't see how it works the "exact opposite" ?????

I see your point and you are of course right as long as you are exposing with the SAME stock all the time - which is comparable to shooting digital.

My point, though - is still valid - (stubborn ass I am!)

I have seen a lot of photographers here and in real life treating the ISO as "changing stock".

Thus:
You'd like to have a slow stock outdoors in daylight, and a fast stock indoors under tungsten.

Both these stocks could (theoretically) have the same latitude on both sides of middle grey.

So people tend to want to lower their ISO when there is high light levels, while they should rise it to protect their highlights (as long as they reclaim any lost highlights they've been protecting on set by lowering the ISO in post...)

However, saying "expose to the right" to mean protect the highlights IS exactly the opposite of film practice. "Expose to the left" means to protect the highlights. Exposing to the right (of the film curve) is to push the highlights up and off the curve. Perhaps we mean "expose FOR the right".?????

Again you are right, and I really don't like the term and it has become a lot more irrelevant lately as the latitude of the cameras increase...

ETTR is a theory that rightfully stems from the fact that the precission is better the higher you expose (as long as you don't clip)
Thus ETTR only has relevance for low-contrast scenes where you could choose to set the exposure "too bright" to get higher precission for post.

REDs advice now is actually the opposite (for the MX), by advicing to shoot 800 on a (for all practical matters) 320 sensor.

It pushes the exposure down, which is good, since all the extra latitude in the R1 MX is in the lower part of the images due to extremely low noise generation. The image is more usefull lower down than the M, but to get to use that extra latitude, we'll have to expose more usefull image-info lower than was possible with the M.

Thus:
ETTR is just becoming a confusing term these days.

It is still possible, though for low-latitude images.

Photographers tend NOT to be so happy for the idea, and I think it is mostly nothing to worry too much about these days.

Cheers!
G

 

[Bill Anderson]

Ah, I see what you're saying, Gunleik, and in that context you are correct.

And just in case I've given the impression that I'm only involved/interested in the stills' end of this technology: I also have a (very modest) background in short film (festivals) and TV, and have a passion (and aspirations) for all things cinema. So most of my queries/observations are definitely cinema oriented.

 

[jake blackstone]

Both these stocks could (theoretically) have the same latitude on both sides of middle grey.

But that's the most important point, Red ASA is NOT the same, as in film. For higher dynamic range scenes you need to increase the ASA. But how much for a given scene? Well, that's a real question. I can't understand, why nobody has come up with a table, that for say, a 14 stops of dynamic range, the ASA should be increased by a 1 stops from the base exposure, i.e ASA 1600. That's just an example...

 

[Bill Anderson]

But that's the most important point, Red ASA is NOT the same, as in film. For higher dynamic range scenes you need to increase the ASA. But how much for a given scene? Well, that's a real question. I can't understand, why nobody has come up with a table, that for say, a 14 stops of dynamic range, the ASA should be increased by a 1 stops from the base exposure, i.e ASA 1600. That's just an example...

Jake, when you say "higher dynamic range scenes", don't you mean "brighter scenes" (higher values)? Granted, brighter scenes are often of a greater dynamic range than avarage/low value (darker) scenes, but far from always being so. Higher DR scenes can exist at the extremes of both high and low light scenarios.

If you do mean "higher DR scenes", what would simply increasing the ISO achieve if the optimum (DR) ISO is bypassed?

 

[Blair S. Paulsen]

Perhaps this makes Gunleik's desire for a formula even more difficult to define, but I am more interested in the range of discrete values in an specific scene that can be recorded. Further, in any scene with more than 10 stops of inherent DR I advocate intelligently biasing exposure to the more critical portion of the scale rather than just targeting 18% gray by rote.

In other words, if there is lots going on in the mids/shadows that needs to read well (or even "pop") then I ETTR to hold greater precision and better chroma knowing I may have to work harder to get the highlight rendition to look pleasing. If the shadows are less important and/or the skin tones are mostly well lit then I am happy to stay well out of the clip and finesse the shoulder to taste.

While it may facilitate discussion to divide the shooting/exposure, virtual telecine and post grading into separate steps it is my contention that in practice they are co-dependent.

FWIW I typically start with 800 in camera and pull back to 320 or 250 as the starting point for "developing" the image. I concur that using curves to manage shadow reproduction is effective and strongly recommend using the best monitoring you can get while dialing in that curve. By using 800 on camera and being conservative in exposure I trust that I will be able to shape the shoulder to suit the look with minimal/no clip.

If you prefer to think in terms of latitude over and under 18% gray then I reiterate my philosophy of making a conscious choice based on the specific exposure environment and desired end result.

Cheers - #19