Jump to content




Photo

Technicolor Process IV


  • Please log in to reply
11 replies to this topic

#1 cole t parzenn

cole t parzenn
  • Basic Members
  • PipPipPipPip
  • 287 posts
  • Other

Posted 15 January 2015 - 05:29 PM

What of the look of Process IV came from the three strips and what came from the dye transfer printing? Was it a challenge to match orthochromatic and panchromatic emulsions? What Blu-Rays come closest to what Process IV films should look like?


  • 0




#2 John E Clark

John E Clark
  • Basic Members
  • PipPipPipPip
  • 780 posts
  • Other
  • San Diego

Posted 15 January 2015 - 08:37 PM

When the "Aviator"(2004) came out, and some amount of to do was made of the 2 and 3 stripe Technicolor process that Scorsese wanted to emulate, I looked into 'how would one know what these looked like...', unless of course one had a pristine print and projector in one's private home theater...

 

I was referred to a film museum in LA, I think it may have been the Academy Museum, and ask if they had any dye transfer stills from the two processes. Unfortunately I don't have the time, and have forgotten the exact reference to follow up on such things...


Edited by John E Clark, 15 January 2015 - 08:37 PM.

  • 0

#3 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 18789 posts
  • Cinematographer
  • Los Angeles

Posted 15 January 2015 - 08:50 PM

Occasionally dye transfer prints are shown at places like UCLA, American Cinematheque, AMPAS, etc.  Every two years, UCLA runs a Festival of Preservation to show what they've been working on -- one summer, the festival was devoted to Technicolor and I saw a number of dye transfer prints, plus new Kodak prints of 3-strip Technicolor movies.  I also saw a collection of dye transfer trailers at the American Cinematheque before a screening of a dye transfer print of "Jungle Book".

 

Also, the last major release of "The Wizard of Oz" back in the late 1990's included a certain number of new dye transfer prints made at Technicolor when they had their prototype printer working, before it was dismantled in the early 2000's.  So I'm sure those prints are in the vault somewhere.

 

Occasionally CineCon, the film festival for early cinema held on Labor Day weekend at the Egyptian Theater in Hollywood, has the occasional dye transfer print, though not very often.

 

The schedules to these festivals will often list whether the source is an original dye transfer print.


  • 0

#4 cole t parzenn

cole t parzenn
  • Basic Members
  • PipPipPipPip
  • 287 posts
  • Other

Posted 16 January 2015 - 01:10 AM

I may move to L.A., just for the film screenings. Do you know the answers to any of my questions?


  • 0

#5 Dennis Couzin

Dennis Couzin
  • Basic Members
  • PipPipPip
  • 149 posts
  • Other
  • Berlin

Posted 16 January 2015 - 02:54 PM

What of the look of Process IV came from the three strips and what came from the dye transfer printing? Was it a challenge to match orthochromatic and panchromatic emulsions? What Blu-Rays come closest to what Process IV films should look like?

 

I may move to L.A., just for the film screenings. Do you know the answers to any of my questions?

 

One person in L.A. who can help you plumb the question of the "look" of Technicolor movies is Ross Lippman at the UCLA Film & Television Archive.

 

Your question about matching orthochromatic and panchrochromatic emulsions is unclear.  Almost every trilayer color film includes non-color sensitized and color-sensitized emulsions.  Color sensitization is independent of tonality, which needs to be matched, and of grain, which perhaps doesn't.

 

The Technicolor process produced (YMC) subtractive color prints from three emulsions capturing BGR records, respectively.  As such it's "look" was not necessarily different from that of other, integral, subtractive color films.  It was different for technical reasons.  There are two aspects to a film's look.  One is the gamut of colors in the print.  You can recognize a gamut without recognizing the things pictured. Gamut is determined by the YMC dyes' spectrums and their maximum densities.  Technicolor had the advantage of being able to choose among a greater range of YMC dyes than those achievable by chemical coupling in integral color films.  The spectral absorptances of Technicolor's YMC dye set should be published somewhere.  If not they wouldn't be hard to measure.  (They should be known for restoration purposes.)

 

The other aspect of "look" is how the film reproduces, and especially misreproduces, colors.  This is determined by the spectral sensitivities of the BRG recording emulsions (as filtered in the taking), the characteristic functions of the emulsions, and the YMC dyes' spectrums.  The whole kit and caboodle.  It will be difficult to figure out Technicolor's effective BRG sensitivities today.  Technicolor had no advantage in this vs. integral color films (but it is difficult to figure out the effective BRG sensitivities of any color films for which this data wasn't published).  Technicolor could adjust its emulsions' characteristic functions during processing, but any well-designed integral color film has already adjusted these.  As Frank Brackett of Technicolor memorably said:

...the three curves must fit like the poor family in one bed if the color film is to be pleasing.

 

The gamut part of the "look" is easy to determine, but the color reproduction part is difficult.


Edited by Dennis Couzin, 16 January 2015 - 02:55 PM.

  • 1

#6 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 18789 posts
  • Cinematographer
  • Los Angeles

Posted 16 January 2015 - 03:38 PM

It's hard to quantify how much of the look of Technicolor was from the 3-strip camera origination versus the dye transfer printing -- probably a mix of both.  Certainly new color masters and digital transfers of 3-strip movies retain a unique look due to the way the image was captured, but some of that is optical rather than merely the color reproduction.

 

For example, the red record was grainier and softer than the green and blue record because the image had to pass through the film capturing the blue record, and since flesh tones have a lot of red in them, this probably had a mild diffusing effect for skin.  Also, the blue record (in front of the red record, which used panchromatic b&w stock) used a blue sensitive b&w stock with no anti-halation backing and was dyed red in order to filter out the blue wavelengths.  This had the effect of creating halation artifacts that were magenta.  You see this around anything that sparkles.

 

But also keep in mind that Technicolor movies were carefully designed for color, down to the make-up used, so some of this color effect was happening in front of the lens.

 

Also, the low effective ISO meant that light levels were very high and generally lenses were shot near wide-open for interiors, which also affected the look.  And contrast was fairly high in the process, so how fill light was used or not used would have a strong effect on how colors would pop.

 

Of course, the dyes used in the printing process also contributed to the look, particularly resulting reds, which were almost three-dimensional.  Some people like to say that you could get a perfect saturated yellow in Technicolor compared to Eastmancolor, not sure if that was true.

 

But even the old carbon arcs used for projection compared to later xenon, which is slightly colder, would have some mild effect on color.


  • 1

#7 Dennis Couzin

Dennis Couzin
  • Basic Members
  • PipPipPip
  • 149 posts
  • Other
  • Berlin

Posted 18 January 2015 - 12:57 PM

I must disagree with David Mullen's view that optical details contributed significantly to the Technicolor look.  Of course it was a complicated system running three films through two gates in a camera -- see nice diagram at http://www.digital-intermediate.co.uk/examples/3strip/technicolor.htm -- and then it took a small army of engineers to maintain registration through the dye transfer.  But the little artifacts, the worst being mis-registration, which make Technicolor prints detectable to experts, are quite apart from the look of Technicolor prints, which can be recognized by all, the moment they see the picture. That look is from its color.

 

Dye transfer allows a wider choice of dyes, but dye transfer per se does not affect color.  It makes no difference that three dyes are intermingled in the Technicolor print rather than layered as in the Eastmancolor print.  (OK, it makes a difference to the colors of emulsion scratches.)

 

How unsharp was Technicolor's red record?  The blue sensitive and red sensitive emulsions were in contact (ignoring the thickness of the red coating).  The red exposing light did have to pass through the full thickness of the crystal-packed blue sensitive emulsion before reaching the red sensitive emulsion, but this is comparable to the red exposing light having to pass through both the blue sensitive and the green sensitive emulsions before reaching the red sensitive emulsion in a tripack color film.  Red layer MTF is typically much lower than blue or green layer MTF for all our camera films.  Does this contribute to "look"?

 

The way cole t parzenn formulated the question: "What of the look of Process IV came from the three strips and what came from the dye transfer printing?" -- in terms of machinery -- obscures the color question and make it less amenable to modern, electronic cinema, considerations. (It also omits the processings of the color separation records which figured in the looks, I think, of the Russian embodiments.)  Today's two questions are: 1. what does it take to restore, by scanning and digitizing, a Technicolor print (including the question of display) and 2. How to emulate the Technicolor look?

 

David makes an important point reminding us that Technicolor was from the era of carbon arc lamp projectors.  The different light source can affect projected color far more than just the shift in correlated color temperature -- e.g., colder -- of the white point.  For film projectors, it's the spectral power distribution of the light source, not its chromaticity, that determines what colors appear on the screen.  This is completely different from video projection where it's just the chromaticities of the projector's primaries that determine color.  (The spectral reflectance of the screen figures too.)  Xenon arcs have a nasty spike at around 470 nm.  Who's got the carbon arc SPD?   Xenon arc and carbon arc probably project films differently, in different ways for different films (depending on their dyes), and not just warmer-colder differences to which the eye can adapt. 


Edited by Dennis Couzin, 18 January 2015 - 01:00 PM.

  • 0

#8 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 18789 posts
  • Cinematographer
  • Los Angeles

Posted 18 January 2015 - 02:25 PM

Keep in mind that many of us watch these older color movies on a video monitor, so the optical characteristics of 3-strip Technicolor are the main clues (other than the date of the production) that we are watching Technicolor photography because the viewed colors are all electronic representations of the original. I would never claim something I was watching on a monitor was Technicolor because of the shade of red I was seeing.

 

But like I said, there was only about a five-year period (1950-55) of overlap between 3-strip and Eastmancolor where sometimes one is trying to figure out which was used for production.


  • 0

#9 Mark Dunn

Mark Dunn
  • Basic Members
  • PipPipPipPip
  • 2182 posts
  • Other
  • London

Posted 18 January 2015 - 03:35 PM

Specular highlights are usually a dead giveaway- a big fat reflection with a white-to-blue gradation. I'd assumed it was beamsplitter flare, if that's the right term


  • 0

#10 Dennis Couzin

Dennis Couzin
  • Basic Members
  • PipPipPip
  • 149 posts
  • Other
  • Berlin

Posted 18 January 2015 - 05:34 PM

@David and Mark, you two are reducing the original question about the look of Technicolor to an experts' game of detecting Technicolor originated material.  Cole's question was about the look:

 

What of the look of Process IV came from the three strips and what came from the dye transfer printing? Was it a challenge to match orthochromatic and panchromatic emulsions? What Blu-Rays come closest to what Process IV films should look like?

 

We rarely get to see Technicolor projected but I'm sure that Technicolor prints can be properly scanned and digitized and hopeful that these can eventually be projected or viewed on monitors. Today are gamut issues for digital display of colorful motion picture prints.  See how the Kodak 2393 gamut dwarfs even the DCI P3 gamut in this diagram. Monitors and digital projectors must improve by adding at least a fourth primary. 


  • 0

#11 John E Clark

John E Clark
  • Basic Members
  • PipPipPipPip
  • 780 posts
  • Other
  • San Diego

Posted 19 January 2015 - 07:16 PM

@David and Mark, you two are reducing the original question about the look of Technicolor to an experts' game of detecting Technicolor originated material.  Cole's question was about the look:

 

 

We rarely get to see Technicolor projected but I'm sure that Technicolor prints can be properly scanned and digitized and hopeful that these can eventually be projected or viewed on monitors. Today are gamut issues for digital display of colorful motion picture prints.  See how the Kodak 2393 gamut dwarfs even the DCI P3 gamut in this diagram. Monitors and digital projectors must improve by adding at least a fourth primary. 

 

There is a similar sort of problem in the reproduction of Art... I've been to a number of galleries, in the US and Europe, and in most cases the 'reproductions' found in books range from 'good' to mostly 'fair' to 'abysmal'... then there's Internet.. The Wife and I were permitted to take pictures of Klimt's "The Kiss" (actually we had been given permission to only take images of another Klimt... the morning that we were to arrive at the Belvedere, that painting shipped out to a exhibition in Korea... and our laison offered 'The Kiss' as a consolation...).

 

But even those images don't really convey the original... and we were only allowed to shoot without tripod 'in situ'... So we do have some details of the painting technique that aren't particularlly available in most books, brush strokes, etc. but still 'color' is elusive...


  • 0

#12 Dennis Couzin

Dennis Couzin
  • Basic Members
  • PipPipPip
  • 149 posts
  • Other
  • Berlin

Posted 20 January 2015 - 09:07 AM

 

There is a similar sort of problem in the reproduction of Art... 

... but still 'color' is elusive...

 

So true.  I cut my teeth in color science while working on photographing paintings.  It was whacky how one slide film could handle the reds and oranges in a painting and another slide film handle the greens and browns in the painting.  Try putting the two slides together!  But technology is better today, and the reproduction problem for Technicolor film is fundamentally simpler.

 

Today there are cameras that come close to meeting the Luther Condition -- equivalent to matching the human eye's three spectral sensitivities.  This is necessary for photographing paintings, because they're made from a large variety of pigments (and in your case of Klimt, even metal foils).  But Technicolor film is comprised of just three dyes.  Therefore almost any camera having known spectral sensitivities can, with the aid of some mathematics, photograph/scan Technicolor film perfectly, provided the spectral absorptances of the three unit Technicolor dyes are also known -- they can be.

 

We need a more scientific approach to film restoration -- not colorists spinning knobs.  The old film prints are fugitive and fragile, and the digital restorations will soon be all that's left.  The scanning and digitization can be done correctly, now, even before there are adequate display technologies.


  • 0


Ritter Battery

Rig Wheels Passport

Paralinx LLC

Zylight

rebotnix Technologies

CineLab

CineTape

Tai Audio

Visual Products

Glidecam

Technodolly

Willys Widgets

The Slider

Broadcast Solutions Inc

Aerial Filmworks

Abel Cine

Pro 8mm

CineLab

Zylight

Abel Cine

Rig Wheels Passport

Visual Products

Tai Audio

Technodolly

Paralinx LLC

Glidecam

rebotnix Technologies

Pro 8mm

The Slider

Willys Widgets

CineTape

Ritter Battery

Aerial Filmworks

Broadcast Solutions Inc