Jump to content


Photo

The Thin Red line


  • Please log in to reply
15 replies to this topic

#1 anthony le grand

anthony le grand
  • Basic Members
  • PipPip
  • 80 posts
  • Director
  • france

Posted 25 October 2009 - 07:35 AM

Hello everyone,

I just searched for it on the forum but couldn't find anything about it. The Thin Red line remains a mystery to me, not only the film which is one of the best I have ever seen, but also the cinematography of John Toll.
This gorgeous film has very rich and strong colors, but keeping everything natural. It's the same with the contrast I guess. Even if they're higher than in a lot of movies, Toll managed to keep them natural and quite soft I think because the light in Queensland is VERY hard. And they didn't shoot everything at dusk or dawn as Malick loves to do.
I'm thinking for instance of the long scene on the hills, during the attack.
Also, Toll used the '79 film stock I think which is more saturated and contrasted than the new stocks (well, except the '80) and used only natural light with some negative fill.

So my question is: do someone has any clue concerning how Toll and Malick got this look? Did they use any particular processing or anything unusual, a particular exposure of the film stock??

Thanks for your help!
  • 0

#2 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 25 October 2009 - 09:42 AM

There was no special processing, it's just good photography: well-exposed 35mm negative using anamorphic lenses. It wasn't just shot on 5279, but also 5248 (EXR 100T) I believe.

It was printed on Kodak Vision Premier stock (2393), which is slightly higher in contrast and color saturation than Kodak Vision (2383). No D.I., just contact printing.

Originally the idea was to release it using Technicolor dye transfer prints, but the post took so long and the release date was set, so there was no time. Technicolor did make one dye transfer print of the movie later as a gift to Malick -- it gets shown now and then at places like the American Cinematheque. I saw a screening of it and the dye transfer process gives the movie a somewhat period Kodachrome look, which is appropriate.
  • 0

#3 Dominic Case

Dominic Case
  • Basic Members
  • PipPipPipPip
  • 1357 posts
  • Other
  • Sydney Australia

Posted 25 October 2009 - 05:02 PM

So my question is: do someone has any clue concerning how Toll and Malick got this look? Did they use any particular processing or anything unusual, a particular exposure of the film stock??

I can answer this one!

David is right - processing was normal. But Toll loves a thick negative - and much of his work on Thin Red Line printed in the high forties: that is, between two and three stops overexposed.

The rushes (dailies) were all printed on film. (It was before HD transfers etc). In Australia it was rare to grade (time) rushes, as a fixed light was always held to give the DoP better feedback - and it was quicker and cheaper. But US DoPs wouldn't work that way, so Atlab's leading grader of the time, Arthur Cambridge, was pressed into an early morning shift for the rushes grade. Arthur was a fabulous grader, but a little conventional. Every day he grew more desparate as the grading lights climbed closer and closer towards the limit of 50-50-50! The light is indeed very hard and very bright in Far North Queensland (then known as "f'n Q"), and Arthur was convinced that Toll simply wasn't believing his meter.

The final grade was also a much more prolonged exercise than we were used to, but I know that John Toll was eventually very very happy with the result.
  • 0

#4 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 25 October 2009 - 05:16 PM

Sometimes I notice a tendency for younger filmmakers to assume that if a movie looked especially good, some special "trick" was involved, when generally the good results were just due to working very hard using traditional tools in the best manner -- good filmstock, good exposure, good lenses, the right light in the right location, etc. Plus solid post work through the chain all the way to exhibition.
  • 0

#5 anthony le grand

anthony le grand
  • Basic Members
  • PipPip
  • 80 posts
  • Director
  • france

Posted 25 October 2009 - 06:18 PM

I can answer this one!

David is right - processing was normal. But Toll loves a thick negative - and much of his work on Thin Red Line printed in the high forties: that is, between two and three stops overexposed.

The rushes (dailies) were all printed on film. (It was before HD transfers etc). In Australia it was rare to grade (time) rushes, as a fixed light was always held to give the DoP better feedback - and it was quicker and cheaper. But US DoPs wouldn't work that way, so Atlab's leading grader of the time, Arthur Cambridge, was pressed into an early morning shift for the rushes grade. Arthur was a fabulous grader, but a little conventional. Every day he grew more desparate as the grading lights climbed closer and closer towards the limit of 50-50-50! The light is indeed very hard and very bright in Far North Queensland (then known as "f'n Q"), and Arthur was convinced that Toll simply wasn't believing his meter.

The final grade was also a much more prolonged exercise than we were used to, but I know that John Toll was eventually very very happy with the result.


Thanks very much! That explains why the blacks are not so hard in the film and why the colors are so rich. The look of the film is quite unusual, vibrant but natural. At first, I thought that Toll overexposed the film a lot but then I didn't understand how you could get those blacks and not having too much contrast with the Queensland light.
Do you know Dominic if they exposed the film stock at 100 for the '48 and 500 for the '79 then? I just remembered that Toll said that he exposed for the shadows in the Daintree rainforest but don't know for the rest...

You're right David, we often tend to think that there are tricks when it looks so good but I believe hard work is sometimes not enough. I guess you say that because you're a hard worker and all your work shows it BUT you also have a vision, a great talent and fantastic ideas for light. I guess experience and work can give it to us, but I think that the best work comes from something else, a kind of sensibility.
I've never seen a film shot in natural light looking as good as The Thin Red Line except some part of The New World even it's a bit softer and difficult to compare.

I tend to think that when you push the limits of the film, you have unusual result and a particular texture which can give a lot of emotion and help the storytelling. Maybe it's just my taste but The Thin Red Line is just one of those examples.
I also think of Kubrick pushing all of the film 2 stops for Eyes Wide Shut or Harris Savides having very very thin negatives, sometimes underexposed 3 or 4 stops. For me, those extreme methods create something on the texture of the film, and not in the light itself and make the film almost alive.
But you're right, hard work is necessary as well. What would be The Yards without the beautiful lights (and shadows) of Savides and those perfect frames...?
  • 0

#6 Dominic Case

Dominic Case
  • Basic Members
  • PipPipPipPip
  • 1357 posts
  • Other
  • Sydney Australia

Posted 25 October 2009 - 07:09 PM

Do you know Dominic if they exposed the film stock at 100 for the '48 and 500 for the '79 then? I just remembered that Toll said that he exposed for the shadows in the Daintree rainforest but don't know for the rest...

I honestly don't know exactly what John Toll's procedure was. But if a stock is rated at 500 by the manufacturer, and the DoP rates it at 500 too, then in normal terms, the negative has to print back in the 25 to 30-odd range of printer lights. Not 48-50. So, in normal terms, he either rated the 500 stock at around 100 and exposed for that, OR he rated at 500 and overexposed by 2 1/2 stops, OR . . .

OR. . . you say these aren't normal terms, Toll isn't a normal cinematographer, the lighting wasn't normal, and the look he wanted wasn't normal. As you say, the light there is harsh - even in Sydney, a building casts a 4+ stop shadow on an average day: in the Queensland rainforest, it's a lot more. So if he exposes for the shadows, the sunlit highlights are going to blow away in a normal print, whereas grading down will bring the highlights back. But there is detail in the shadows, and the Premier print stock goes on into the blacks a lot further than normal to reproduce that detail (in a dark-enough cinema!).

In other words, there is a lot more to exposure than simply how you rate the stock. Which goes back to David's comment about just using the tools you have in a competent (or more than competent) way.
  • 0

#7 anthony le grand

anthony le grand
  • Basic Members
  • PipPip
  • 80 posts
  • Director
  • france

Posted 26 October 2009 - 06:04 PM

I honestly don't know exactly what John Toll's procedure was. But if a stock is rated at 500 by the manufacturer, and the DoP rates it at 500 too, then in normal terms, the negative has to print back in the 25 to 30-odd range of printer lights. Not 48-50. So, in normal terms, he either rated the 500 stock at around 100 and exposed for that, OR he rated at 500 and overexposed by 2 1/2 stops, OR . . .

OR. . . you say these aren't normal terms, Toll isn't a normal cinematographer, the lighting wasn't normal, and the look he wanted wasn't normal. As you say, the light there is harsh - even in Sydney, a building casts a 4+ stop shadow on an average day: in the Queensland rainforest, it's a lot more. So if he exposes for the shadows, the sunlit highlights are going to blow away in a normal print, whereas grading down will bring the highlights back. But there is detail in the shadows, and the Premier print stock goes on into the blacks a lot further than normal to reproduce that detail (in a dark-enough cinema!).

In other words, there is a lot more to exposure than simply how you rate the stock. Which goes back to David's comment about just using the tools you have in a competent (or more than competent) way.



Thanks very much for those words and for your time.
As you say, Toll's work here is not "normal" and the final result looks gorgeous. It's always incredible to watch those kind of movies where the work of the best DoPs like Toll or Lubezki meet the vision of a great artist like Malick. Not to forget Hans Zimmer's beautiful score and the work of all the people who worked for and with them...

Thanks again!
  • 0

#8 Keith Walters

Keith Walters
  • Sustaining Members
  • 2219 posts
  • Other
  • Sydney Australia

Posted 26 October 2009 - 08:04 PM

Sometimes I notice a tendency for younger filmmakers to assume that if a movie looked especially good, some special "trick" was involved, when generally the good results were just due to working very hard using traditional tools in the best manner -- good filmstock, good exposure, good lenses, the right light in the right location, etc. Plus solid post work through the chain all the way to exhibition.

"Sometimes"...? :rolleyes:

I think you can safely apply that scenario to just about any sort of creative activity, whether it's film-making, writing, cooking, acting, songwriting, painting, even ... gardening. The notion of effortless-bolt-on-competence-at-sensible-prices, sadly permeates our entire culture.

Sorry, my years of working for a famous film equipment rental company have imbued me with an enduring sense of wonder: You sometimes wonder how some projects ever made it all the way to the screen, large or small. :lol:
  • 0

#9 Keith Walters

Keith Walters
  • Sustaining Members
  • 2219 posts
  • Other
  • Sydney Australia

Posted 26 October 2009 - 09:38 PM

This gorgeous film has very rich and strong colors, but keeping everything natural. It's the same with the contrast I guess. Even if they're higher than in a lot of movies, Toll managed to keep them natural and quite soft I think because the light in Queensland is VERY hard.


The reason the light in Queensland is so harsh is simply that the Tropic of Capricorn runs through the middle of the state. In particular around the Summer Solstice (near the end of December), the sun shines directly overhead at midday, which means it gets minimal attenuation by the Earth's atmosphere.

In other parts of the world, even at midday, the sun strikes the ground at considerably greater angles, which means it takes a longer path through the atmosphere. Apart from increasing the simple intensity attenuation, airborne dust particles also tend to remove some of the blue light, giving the sunlight more of the characteristic reddish "sunrise/sunset" cast of late in the afternoon shooting.

So generally, in most other parts of the world where shooting is likely to be taking place, you would most likely be shooting under a combination of less-intense, reddish-coloured direct sunlight, and blue-tinted "fill" light provided by the sky.

Around the Tropic of Cancer (and Tropic of Capricorn in the Northern Hemisphere) you have a stronger, more bluish-tinted sunlight, but with much the same level of sky-scattered "fill". Hence, shadows tend to be deeper.

Most of the US and Europe is a long way North of the Tropic of Cancer, which means cinematographers there are more used to less harsh natural lighting conditions. (Actually Southern Florida is quite close to the Tropic of Cancer, so they should experience similar conditions there to North Queensland).

Regarding over-exposure for obtaining a specific "look" the pictures below might help you get some insight into this, although in a back-to-front manner.

These photos were taken in Sydney around 3PM in December. They were originally designed to demonstrate the unrivalled ability of film emulsion to at least get some sort of recognizable image even over extremes of tonal range.

The picture on the left was simply taken following the directions of the camera's inbuilt light meter; the one on the right is the result of increasing the exposure time in one-stop-equivalent increments to six stops. (I've only shown the 1/500th and 1/8th second shots)

What you see in the first pair of pictures is simply what came out of an ordinary supermarket minilab/processor/film scanner.
Posted Image

The second pair of pictures is what I was able to obtain by using the "curves" function in Photoshop.
You'll note that even though there were 64 times as many photons hitting the emulsion of the second image, (that is, it 6 stops over-exposed), the pictures don't look all that different. Remember also that this is an 8-bit JPEG from a supermarket minilab, not a 10 bit+ scan from a professional slide scanner. You will be hard-pressed to find anything in the correctly exposed image that is missing from the one over-exposed by 6 stops!

Posted Image

Anyway, as regards the current discussion, (North Queensland and Overexposure) you will note the one thing I was not able to do was restore the correct skin colour. Anything that is meant to be orange or red looking comes out looking pale or bluish.

And the reason for that is that, because the emulsion was getting heavily over-exposed by reddish-tinted sunlight, the red-sensitive layer tends to reach saturation before the green and blue layers. While this can be corrected to certain extent by Photoshop, because the red layer has less contrast range left than the green and blue layers, there is not enough information to accurately reproduce flesh tones.

So, consider what would happen if the same shot has been taken a few thousand miles North of here. Then, the sunlight would have a distinctly bluish cast. If we over-expose film now with the intention of correcting in Post the blue layer would tend to saturate before the red and green layers. This would result in tonal-range supression of the blue channel, and produce the much-desired "warmer, richer" texture.

Anyway, the other message here is that you can probably acquire a lot of the necessary photographic skills with an ordinary stills camera, particularly an old-fashioned non-automatic model that you can pick up for next to nothing these days. OK the film is not exactly the same, but it's not all that different either.

This also explains the current popularity of 8mm negative, since it gives aspiring film makers access to a real "working model" as far as exposure and filtration goes.
  • 0

#10 Paul Bruening

Paul Bruening

    (deceased)

  • Sustaining Members
  • 2858 posts
  • Producer
  • Oxford, Mississippi

Posted 26 October 2009 - 10:05 PM

Stunning ensemble, there, Keith.
  • 0

#11 Keith Walters

Keith Walters
  • Sustaining Members
  • 2219 posts
  • Other
  • Sydney Australia

Posted 26 October 2009 - 10:56 PM

Stunning ensemble, there, Keith.

It was about 40 Degrees C (about 105 F) here at the time...
(As a rule, it does NOT usually snow down here come Christmas)
Also, the rayon Hawaiian shirt has a very high reflectivity.
  • 0

#12 Chris Millar

Chris Millar
  • Basic Members
  • PipPipPipPip
  • 1642 posts
  • Other

Posted 27 October 2009 - 02:18 AM

gawd I love these forums :lol:
  • 0

#13 Tom Lowe

Tom Lowe
  • Basic Members
  • PipPipPipPip
  • 1211 posts
  • Director
  • somewhere worshipping Terrence Malick

Posted 27 October 2009 - 11:15 AM

Thank you for those posts Dominic. Very interesting.
  • 0

#14 John Holland

John Holland
  • Basic Members
  • PipPipPipPip
  • 2248 posts
  • Cinematographer
  • London England

Posted 27 October 2009 - 12:17 PM

Main thing about the "look" anamorphic and no DI . Why is it so many people put up with the bad finish you get going through a 2k DI ??
  • 0

#15 K Borowski

K Borowski
  • Basic Members
  • PipPipPipPip
  • 3905 posts
  • Camera Operator
  • I.A.T.S.E. Local # 600 Eastern Region

Posted 27 October 2009 - 03:30 PM

John, you and I are going to get banned from these forums if we keep getting caught making personal attacks on the DI process!

If I were king of the world, I'd make banning 2Ks from scope neg's my first mandate :-D
  • 0

#16 Keith Walters

Keith Walters
  • Sustaining Members
  • 2219 posts
  • Other
  • Sydney Australia

Posted 27 October 2009 - 08:02 PM

Main thing about the "look" anamorphic and no DI . Why is it so many people put up with the bad finish you get going through a 2k DI ??

The biggest problem with a 2K DI of 35mm neg is that it really needs to be an 8K DI, that is 16 times as many pixels in the scan.

In the normal photochemical process you'd start with a pristine camera negative that arguably needs an 8K scan to derive most (but by no means all) of the information it contains, and then starting the inevitable downhill-all-the-way run to release print, leaving you with something approximating 2K resolution.

But with a 2K DI you’re effectively starting with 2K “negative”, and then people wonder why what comes out the other end is so ordinary looking.

It’s like trying to do some multiple backflip Winter Olympic ski-jump routine, but only being able to start a quarter of the way up the ramp because they couldn’t afford enough snow for the top 75%.

This is one reason 2K digital projection tends to look better, because you’re then effectively projecting a 2K scan of the original negative, albeit with the crap most likely “photoshopped” out of it in post.
  • 0


Gamma Ray Digital Inc

Paralinx LLC

Ritter Battery

Tai Audio

Rig Wheels Passport

Opal

The Slider

Wooden Camera

Abel Cine

CineTape

Glidecam

Broadcast Solutions Inc

Visual Products

Aerial Filmworks

Technodolly

Metropolis Post

FJS International, LLC

Willys Widgets

Media Blackout - Custom Cables and AKS

CineLab

rebotnix Technologies

Metropolis Post

CineLab

Opal

Visual Products

Willys Widgets

Aerial Filmworks

FJS International, LLC

Paralinx LLC

Wooden Camera

CineTape

Gamma Ray Digital Inc

Rig Wheels Passport

Ritter Battery

Glidecam

Tai Audio

rebotnix Technologies

Abel Cine

Broadcast Solutions Inc

Technodolly

The Slider

Media Blackout - Custom Cables and AKS