Exposing + Native ISO

[Mark L. Pederson]

Though this is true in respect to exposing a film negative, there is no "knee" or "shoulder" in digital cinematography. Film responds logarithmically, digital CMOS sensors like all of the RED cameras respond linearly.

Sure Adrian - but the logic and example still applies. If you want to exploit more of the dynamic range in the shadows or want to light for a higher or lower contrast ratio, etc. etc. etc.

I don't know about you - but creatively - it's a compromise to me to shoot all subject matter - in all situations - at the same fixed ISO just because there's and equal number of stops above and below 18% grey.

 

[Adrian Jebef]

But it does. You push or pull the development of the r3d just as we did with film. And it changes the image's response, just as it did with film.

Actually it doesn't. With film negative you are in essence "pushing" and "pulling" the logarithmic response curve when you push or pull process at the film lab. The "knee" and "shoulder" of the film negative's response curve is elastic. It bends and stretches. This isn't true with digital cinema cameras like the DRAGON. Though you can change ISO later in RedCineX in the development phase, and the gamma curve will shift, and this will seem to be the same response that one would get with film, the underlying data values present in the original R3D are still linear code values. They are not elastic. The big difference between the two is that when you push or pull film negative you are actively affecting your scene referenced exposure. When doing "the same thing" with a DRAGON via RCX scene referenced exposure has already been set in stone. The top, bottom, and mid-point of your DR will not change.

With the advance of modern digital cinema cameras like the DRAGON this difference becomes much more academic than real-world. Though if one goes back in time and shoots with the RED ONE this difference becomes all the more apparent.

 

[Adrian Jebef]

Exactly the same! 100% agree, iso is a creative desicion that serves the story.

No. ISO is a technical value defined as the middle point where dynamic range is evenly distributed above and below said value.

Your chosen exposure for the scene is the creative decision that serves the story.

 

[Adrian Jebef]

I don't know about you - but creatively - it's a compromise to me to shoot all subject matter - in all situations - at the same fixed ISO just because there's and equal number of stops above and below 18% grey.

ISO is a static, reference value (or it should be... hint, hint RED). ISO doesn't dictate your exposure choices. You do.



Say you're shooting a high-key day exterior shot. Maybe something in the desert. 21mm lens. Sun clearly visible in your frame. Sand dunes in the distance. And your main character front and center. Maybe something like this but with the sun in the shot as well:

Your incident meter tells you that proper exposure for your leading actor's face is T8. OK. You take a meter reading of the sun. Unsurprisingly it's 10 stops over "key" or your nominal scene referenced exposure based on subject matter (your actor) and ISO (800 via DRAGON). The bummer is you've only got 8 Stops over Key. What to do? Well you can easily close the lens down 2 stops to a T16. Problem is that now your Key or actor or scene referenced middle-grey/18% Grey is now 2 Stops underexposed. To get your Key back to proper exposure (creatively defined by you, the DP, not technically defined by ISO) you'll need to bring out a couple of 18K HMI's and punch in some fill light into your actor's face. Problem solved. You expose at T16. Your shot looks amazing.

Same scene, different approach: It's the desert. It's hot. Really hot. You're sweating. Your actor is sweating. You've just climbed up this stupid sand dune to get your incredible shot. It's gonna look amazing. But you'd really like a bottle of water. You take a meter reading of your actor. T8. OK. But you don't want to shoot with such a deep stop. You prefer shallow depth of field. You want to shoot at a T2. So you call to your trusty AC, "Throw in a N12!" Ooops. He left the filters back on the camera truck. It's gonna be 30min to get them. "Forget about it", you decide. My DRAGON is a 200-2000 ISO shooting machine! You'll just compensate for the exposure change in camera. Who wants to shoot day exteriors at 800 or 2000 ISO anyway? So you set your DRAGON to 200 ISO. That's only 2 Stops. Now according to your meter you should expose at T4. But you wanted to expose at a T2! Well, DRAGON has awesome dynamic range and it really has super nice highlight fall-off. You think to yourself, "I bet even if I overexpose by 2 whole stops it'll still be OK. After all DRAGON's got like 16+ Stops of DR!" If you're only left with 6 Stops over Key instead of 8 you think you'll still be fine. Maybe the sun does blow out a bit but the surrounding clouds should hold. You'll be fine. Problem is you've already shifted your DR via changing your camera and meter ISO. You don't have 6 Stops over Key, you probably have less than 3 or 4.

When you get the footage back home and pump it into RCX, approach 1 is perfect. You haven't clipped any highlight detail. You've got some range in the sun and surrounding clouds. And your actor's face is perfectly exposed. You can shift the developmental ISO from 200-2000 and it still looks great. It's all about how you want the scene to play: bright and hot or a bit more silhouette-y. It's another creative decision.

Approach 2 doesn't turn out so well. You realize you totally forgot to pump in 2 more stops of light into your actor's face. But it's cool cause you can just change your development ISO in RCX. Go from 200 back to 800. Though this helps your actor's exposure you also notice that the sun is completely blown out. As are most of the surrounding clouds. Totally clipped. Completely white. Zero detail. You take development ISO back down to 200. It's a compromise. Your actor is a bit too dark. You can probably help that out a bit later with a proper grade in Resolve. But surprisingly, the sun and surrounding clouds are still blown out. Still clipped. You see a bit more detail there but nowhere near enough to call your shot properly exposed. Worst of all is when your actor turns back into the sun to walk off into the distance. When his face hits the sun it goes completely nuclear. No hope.


This is just one example where ISO ranges like 200-2000 can be more confusing than helpful when one is learning how to expose properly in digital cinema.

 

[Bill Anderson]

A lot of seasoned guys set up using an incident meter and then get out the spot to check and make sure everything is in play.

And good post, Adrian

 

[Adrian Jebef]

A lot of seasoned guys set up using an incident meter and then get out the spot to check and make sure everything is in play.

Yup, and they do this based on a reference: ISO and Middle-Grey. The two are the same.


Thanks, Bill.

 

[Bill Anderson]

Yep, good old "Zone V". Adams knew what the heck he was talking about.

 

[Jacobo Martinez]

You can shift the developmental ISO from 200-2000 and it still looks great. It's all about how you want the scene to play: bright and hot or a bit more silhouette-y. It's another creative decision. .

That is exactly what i am trying to say, exposure has nothing to do with iso, to change exposure in red camera it can only be made with shutter and changing the amount of light passing through the lens to the sensor either stopping up or down or using filters. Iso is a creative choice.

 

[D Fuller]

Actually it doesn't. With film negative you are in essence "pushing" and "pulling" the logarithmic response curve when you push or pull process at the film lab. The "knee" and "shoulder" of the film negative's response curve is elastic. It's bends and stretches. This isn't true with digital cinema cameras like the DRAGON. Though you can change ISO later in RedCineX in the development phase, and the gamma curve will shift, and this will seem to be the same response that one would get with film, the underlying data values present in the original R3D are still linear code values. They are not elastic. The big different between the two is that when you push or pull film negative you are actively affecting your scene referenced exposure. When doing "the same thing" with a DRAGON via RCX scene referenced exposure has already been set in stone. The top, bottom, and mid-point of your DR will not change.

With the advance of modern digital cinema cameras like the DRAGON this difference becomes much more academic than real-world. Though if one goes back in time and shoots with the RED ONE this difference becomes all the more apparent.

I spent a lot of time thinking exactly this. And while it is not wrong, it misses the point. It doesn't matter where the tone curve gets applied. It does get applied. And once you accept that development decisions are part of exposure decisions (as they were in film: when you decided to push a stop, or decided to let the negative go thin, knowing that it would be printed down to keep the shadows dark and moody) then the craft of exposure becomes almost exactly like film. And in fact, by the time you push-process film, the exposure has already been made (just as when developing an r3d) and the push-process is part of the decision you made at the time of exposure.

I actually disagree just a bit with Mark's description of Dragon as 3 different film stocks. I see it as one "film stock" that can be exposed and processed in three different ways. In that way, it is the way I used to thick about Tri-X. Over exposed and undeveloped, it produced a very different look than normally exposed. Under-exposed and push-processed, it produced still another look. But before you did any exposure, I had to know how I was planning to develop.

 

[Adrian Jebef]

Yep, good old "Zone V". Adams knew what the heck he was talking about.

The MAIN MAN he was. Adams defined how to expose creatively for any aspiring photographer or cinematographer.

A lot of the confusion surrounding proper exposure techniques for digital cinema in relation to ISO is that most think ISO is a singular expression when, in fact, it has become two. ISO in relation to a film stock or digital cinema camera is a rated value that when exposed at via 18% Grey will yield an image with an equal distribution of dynamic range above and below 18% Grey or stated ISO value. This is how ISO has always been defined. This is how light meters work, both incident and reflective. This is how Adams' Zone System is calibrated.

But today, in the ever changing digital landscape, ISO has made it's way into Post. Here it is no longer an exposure tool. It is now a development tool. In RCX changing the ISO does not change the captured dynamic range at all. Scene referenced exposure DR is static. It is a fixed range. When you change ISO in RCX you are shifting your captured DR in relation to a targeted gamma curve. You are actually not recovering any highlight or shadow detail. You would need to go back in time to your scene and change your lighting or exposure to do that. You are simply shifting where in the gamma curve your extreme highs and lows fall. Depending on what choices are made at this development stage you may or may not push out some part of usable DR. Again, it depends on the targeted gamma/color space and how much of a deviation you make from the camera's Base ISO.

If you haven't already, I highly recommend taking the time to read Ansel Adam's "The Negative" and "The Print". These two companion books are in essence the BIBLE on creative exposure choices across any and all photographic medium, whether digital or analogue. One can acquaint "The Negative" with a digital camera's Base ISO and "The Print" with the development ISO.

BTW: "The Camera" is also highly recommended.

 

[Adrian Jebef]

That is exactly what i am trying to say, exposure has nothing to do with iso, to change exposure in red camera it can only be made with shutter and changing the amount of light passing through the lens to the sensor either stopping up or down or using filters. Iso is a creative choice.

No. ISO is a technical, fixed value. Exposure is a creative choice. If you change your development ISO in Post it would be an intentional creative choice. One that you hopefully thought of in advance when you were on-set setting your scene referenced exposure. ISO tells you how to expose for an equal distribution of dynamic range both above and below middle-grey. Where you actually set your exposure defines what you find important in the scene.

Exposure has everything to do with ISO. How can you make a confident creative choice if you don't know the middle and limits of your exposure range (DR)?


Again, I highly recommend Ansel Adams' books on this subject. I have yet to come across a more elegant explanation of the technical and creative process than that which is found within the pages of his books.

 

[Adrian Jebef]

I spent a lot of time thinking exactly this. And while it is not wrong, it misses the point. It doesn't matter where the tone curve gets applied. It does get applied. And once you accept that development decisions are part of exposure decisions (as they were in film: when you decided to push a stop, or decided to let the negative go thin, knowing that it would be printed down to keep the shadows dark and moody) then the craft of exposure becomes almost exactly like film. And in fact, by the time you push-process film, the exposure has already been made (just as when developing an r3d) and the push-process is part of the decision you made at the time of exposure.

I actually disagree just a bit with Mark's description of Dragon as 3 different film stocks. I see it as one "film stock" that can be exposed and processed in three different ways. In that way, it is the way I used to thick about Tri-X. Over exposed and undeveloped, it produced a very different look than normally exposed. Under-exposed and push-processed, it produced still another look. But before you did any exposure, I had to know how I was planning to develop.

Personally, I think both of us "get it". We've probably both spent enough time with film and digital formats that we're now accustomed to exposing them. The bigger issue is those with less of a background or simply with less experience. Explaining the technical and creative concepts of proper or intentional exposure is very important. Saying that so and so digital camera is amazing and has so much DR and shoots RAW and has a usable ISO range of 200-2000 misses the point. Describing a camera this way makes it seem that there are no compromises when one shoots RAW at 200 or 2000 ISO, when in fact there are many, many compromises.

You said it yourself, "once you accept that development decisions are part of exposure decisions then the craft of exposure becomes almost exactly like film". Exactly! This is why Adams' books have such relevance in the digital age. The difference between you or me and the newbie who has just purchased his first RED camera is that we know this very important distinction. Someone less experienced may take these ISO specs (200-2000 for DRAGON) at face value. They may change camera ISO to suit their scene referenced exposure without realizing that they have now shifted their entire dynamic range to a point where highlights clip or shadows get noisy. And then later, in RCX, they shift development ISO again because they think as long as the numbers are within 200-2000 their shot should look great. Sometimes it works. And sometimes it's a complete mess.

A cinematographer's creative exposure choices can only be made when he/she knows for certain what his camera's Base ISO rating and range are and what additional creative compromises he/she will make in the development phase (which is intrinsically tied to the scene referenced expose / Base ISO).

 

[David Battistella]

I actually disagree just a bit with Mark's description of Dragon as 3 different film stocks. I see it as one "film stock" that can be exposed and processed in three different ways. In that way, it is the way I used to thick about Tri-X. Over exposed and undeveloped, it produced a very different look than normally exposed. Under-exposed and push-processed, it produced still another look. But before you did any exposure, I had to know how I was planning to develop.

and this was my point earlier in the thread.

I never said you could not rate the dragon at various ISO's but I do feel it's the manufacturers responsibility to "put a sticker on the film can" as it were. The concepts being discussed here are all correct. Toe and shoulder, rating higher outdoors than indoors.

I think the real problem is that most "shooters" are actually doing the opposite. They crank down the ISO outdoors and crank it up for dark interiors, not realizing that they are increasing grain and limiting DR by doing so. (Where is the craft going!)

The zone system is still one of the most solid exposure practices to go by.

So, in my poor english, or language choices, I'm not disagreeing with what is being discussed here and everyone has valid points, I'm just suggesting that the "dragon sensor film stock" be given a baseline ISO rating by the manufacturer as a suggested starting point. When I get my dragon, I would probably do exactly the same tests as mark did with different ratings and I might end up with a floating range as well (probably two) but I think the manufacturer should have one.

Oh, and the 800 ISO rating on the EPIC, I think the Alexa came out at about the same time rated at ISO 800, red was toying with rating MX at 500 (which would have been more accurate) but landed on 800 probably not to loose competitive advantage. I've always found 800 just a tad grainy on MX, and tend to process ass lower ISO's

Ok, I turn it back to back to the mad squad now.

battistella

 

[Will Keir]

Adrian,

I like this graph, much better and more intuntive than Phil's version. Sorry Phil, your graphs are great but that one I didn't visualize as well as this one. The bars being vertical does something special.

BTW: I like Phil's graph but I believe it's a bit misleading in the way it shows "expanded" or "compressed" highlight/shadow with respect to the -2 and +2 stops of exposure. Looking at it, the graph seems to suggest that by opening or closing your exposure +/-2 Stops or changing your ISO you keep the same Dynamic Range end points but squeeze in or squeeze out more or less highlight/shadow detail. This is not how digital CMOS-based sensors work. They have a linear sensitivity to light unlike a film stock whose sensitivity is logarithmic. Scene exposure values become discretely coded into a digital camera's DR (ie: 0-1023 in a 10-bit scale), they do not bend or curve like film does. Technically, there is no knee or shoulder with a digital cinema camera. The exposure curve (gamma) is interpreted directly from the captured linear code values. Though the curve can be interpreted numerous ways via post transforms it has no bearing on the top and bottom end of scene exposure.

I prefer this graph on the relationship of a sensor's native Dynamic Range and ISO changes:

 

[Will Keir]

haha, your going to hate me Mark but this guy is using ISO to adjust exposure. 2:57 haha, thanks for the video. still watching.

You are a wise man Will.

watch this video -

 

[David Battistella]

I have a hunch David B and I might have chosen a different spot. And I think that is good...

Or we could discover it is the exact same spot and that is also good.

But it is pretty futile for me to be in this discussion without having run my own tests, as you Mark and Phil (phil with the absolute latest) have hands on time with the sensor.

One thing I can say is that protecting highlights (WHILE STILL IMPORTANT) looks like it is not going to have to be as carefully monitored as the sensor clip/rolloff (even on the pre new OLF stuff) is much more gentle and organic than it is on EPIC MX.

Battistella

 

[Gunleik Groven]

Or we could discover it is the exact same spot and that is also good.

But it is pretty futile for me to be in this discussion without having run my own tests, as you Mark and Phil (phil with the absolute latest) have hands on time with the sensor.

One thing I can say is that protecting highlights (WHILE STILL IMPORTANT) looks like it is not going to have to be as carefully monitored as the sensor clip/rolloff (even on the pre new OLF stuff) is much more gentle and organic than it is on EPIC MX.

Battistella

Agree...

From what I have read of our previous discussions, I think we maybe expose similarly, but with different methodology.

 

[Evin Grant]

When I was at Brooks Institute in the 90s and taking "The Zone system" from Chris Rainer and Bob Smith (Chris was one of Ansel's last assistants) one of the first things we did was test the EI (Exposure Index) of our film. The reason was that despite the number on the box almost all films have a slightly different sensitivity depending on the developer and temperature used. I bought 100 rolls of the same lot of medium format Tmax 100 and I spent two weeks nailing down the exact EI and DR (which turned out to be EI 125 once processed). I have an entire 50 page book of various exposure and dynamic range tests that I had to produce before we were allowed to go out and shoot any real subjects.

here is one of the images I made in that class...

The jpeg really doesn't do the print justice unfortunately.

This is not any different than the approach you should take finding the EI you feel most appropriately matches your vision for the project you have on hand, regardless of the camera. I have tested the MX and the Dragon and happily expose those cameras at EIs from 320-1280 on the MX and 400-4000 on the Dragon depending on the look and needs of the scene. The rated EI/ISO on the camera are dead on accurate to 18% grey when checked with both incident and spot meters (which need occasional calibration too BTW).

However you look at it you have considerably more DR and control over your exposure than any previous group of filmmakers up to this point.
Deal with it!

 

[David Battistella]

Agree...

From what I have read of our previous discussions, I think we maybe expose similarly, but with different methodology.

And the key will always be having the tool that makes the image we(the collective we) want. Pretty, gritty, sharp, grainy, over, under, what the artist is going for. The digital tool is more flexible now than ever.

battistella

 

[Will Keir]

I see your point. It's crazy to think your adding more light on this dune to get the proper exposure. Even crazier, after your long hike up the dune that pushes you back one step for every two taken, that you will have 18k HMI's readily available.

So setting your Base ISO fixes you to a specific grain pattern in the footage? So pick the grain you like best for your movie and roll at that ISO? Cutting a scene at 250 ISO with 3200 ISO material will look distinctively different and jarring in the final edit?

So the method both Phil and Mark are suggest could be a possible total nightmare in post? Never thought of it like that before.

Variable Film Stock has met a challenging obstacle. I'd like to shoot a test scene for grain, shooting a shot reverse shot at the very different ISOs. I won't have a chance for another week or two, if anyone beats me to the punch, that would be cool.

ISO is a static, reference value (or it should be... hint, hint RED). ISO doesn't dictate your exposure choices. You do.



Say you're shooting a high-key day exterior shot. Maybe something in the desert. 21mm lens. Sun clearly visible in your frame. Sand dunes in the distance. And your main character front and center. Maybe something like this but with the sun in the shot as well:

Your incident meter tells you that proper exposure for your leading actor's face is T8. OK. You take a meter reading of the sun. Unsurprisingly it's 10 stops over "key" or your nominal scene referenced exposure based on subject matter (your actor) and ISO (800 via DRAGON). The bummer is you've only got 8 Stops over Key. What to do? Well you can easily close the lens down 2 stops to a T16. Problem is that now your Key or actor or scene referenced middle-grey/18% Grey is now 2 Stops underexposed. To get your Key back to proper exposure (creatively defined by you, the DP, not technically defined by ISO) you'll need to bring out a couple of 18K HMI's and punch in some fill light into your actor's face. Problem solved. You expose at T16. Your shot looks amazing.

Same scene, different approach: It's the desert. It's hot. Really hot. You're sweating. Your actor is sweating. You've just climbed up this stupid sand dune to get your incredible shot. It's gonna look amazing. But you'd really like a bottle of water. You take a meter reading of your actor. T8. OK. But you don't want to shoot with such a deep stop. You prefer shallow depth of field. You want to shoot at a T2. So you call to your trusty AC, "Throw in a N12!" Ooops. He left the filters back on the camera truck. It's gonna be 30min to get them. "Forget about it", you decide. My DRAGON is a 200-2000 ISO shooting machine! You'll just compensate for the exposure change in camera. Who wants to shoot day exteriors at 800 or 2000 ISO anyway? So you set your DRAGON to 200 ISO. That's only 2 Stops. Now according to your meter you should expose at T4. But you wanted to expose at a T2! Well, DRAGON has awesome dynamic range and it really has super nice highlight fall-off. You think to yourself, "I bet even if I overexpose by 2 whole stops it'll still be OK. After all DRAGON's got like 16+ Stops of DR!" If you're only left with 6 Stops over Key instead of 8 you think you'll still be fine. Maybe the sun does blow out a bit but the surrounding clouds should hold. You'll be fine. Problem is you've already shifted your DR via changing your camera and meter ISO. You don't have 6 Stops over Key, you probably have less than 3 or 4.

When you get the footage back home and pump it into RCX, approach 1 is perfect. You haven't clipped any highlight detail. You've got some range in the sun and surrounding clouds. And your actor's face is perfectly exposed. You can shift the developmental ISO from 200-2000 and it still looks great. It's all about how you want the scene to play: bright and hot or a bit more silhouette-y. It's another creative decision.

Approach 2 doesn't turn out so well. You realize you totally forgot to pump in 2 more stops of light into your actor's face. But it's cool cause you can just change your development ISO in RCX. Go from 200 back to 800. Though this helps your actor's exposure you also notice that the sun is completely blown out. As are most of the surrounding clouds. Totally clipped. Completely white. Zero detail. You take development ISO back down to 200. It's a compromise. Your actor is a bit too dark. You can probably help that out a bit later with a proper grade in Resolve. But surprisingly, the sun and surrounding clouds are still blown out. Still clipped. You see a bit more detail there but nowhere near enough to call your shot properly exposed. Worst of all is when your actor turns back into the sun to walk off into the distance. When his face hits the sun it goes completely nuclear. No hope.


This is just one example where ISO ranges like 200-2000 can be more confusing than helpful when one is learning how to expose properly in digital cinema.