The Agfa Kine Handbuch
Historic tinting and toning guides are a major resource for the historic and scientific study of applied color in film preservation. The best known among these is Kodak’s , of which numerous copies and editions survive in archives, as well as in the library of the Haghefilm Foundation. The Foundation also holds a copy of Pathé Frères’ Manuel de Developpement et de Tirage from 1926. For that reason, it was all the more exciting to recently complete the collection of the major silent era tinting guides by the acquisition of two copies of the German version of the Agfa Kine Handbuch. Upon receiving these rare books, which like the Pathé and Kodak equivalent contain charts with actual, tinted and toned nitrate frames mounted in cardboard, we were surprised by the intense colors and pristine condition of the frames, surpassing anything we have seen in extant copies of the other books. Why are the Agfa samples in such good conditions, and does this explain their vibrant colors also? Chemical differences in the fixing, washing and tinting of the samples compared to the French and German samples might offer an explanation, but the quality of the mounting board can be a factor, too – acidic paper will certainly trigger and accelerate nitrate decomposition. However, an alternate, aesthetic explanation is offered by Elfriede Ledig’s seminal essay on silent movie color in her book Der Stummfilm, Konstruktion und Rekonstruktion. In footnote 59 (p.105) she dates the Agfa Kine book to 1921/22, as its toning recipes are replicated in another, 1922/23 source. Earlier in her essay, Ledig also elaborates on a shift towards lighter colors in the 1920s, substantiated by a 1921 quote from Urban Gad (p. 100). Does the Agfa book represent a snapshot of more vibrant tinting preferences prior to the mid-1920s, or is the reason for the colors a question of condition and thus rather chemical in nature? The question remains open for now, but is a strong reminder of the archeological importance of original film samples, and how color film preservation and study need to be informed as much by technical and scientific as by historic and aesthetic judgment. - Ulrich Ruedel (submitted February 25, 2010)
Faded Prussian Blue Tone
Stability questions are an important consideration when making color decisions in film preservation. We had previously first heard about potential fading of Prussian Blue tones from Film Preservation Officer Anthony L’Abbate. In that archive, we had even investigated a suspected case for this in a major film from the collections using X Ray Fluorescence. However this method only detected silver in the nitrate film and no iron or any other toning metal (such as vanadium or uranium). This confirmed that the sample at hand was a mere tinted black and white film, rather than also, as had originally been suspected based on the look off a fairly recent safety print of the same film, a faded blue tone underlying the tint.
Courtesy of
Working on a nitrate original of an entirely different film, from the Nederlands Filmuseum (now known as eye), Film Preservation Specialist Daniela Currò recently encountered an example of Prussian Blue fading, powerful anecdotal evidence demonstrating that this effect can indeed be observed in historic nitrate film. Working on the original print of The Apache (UK, 1925), she noticed a section in which the image looked completely neutral. However, the Kodak edge code is clearly still of the blue toning color typical of the iron compound. Both fading by projector light and exclusion of air are known as factors in Prussian Blue fading and offer a potential explanation why the edge code was not affected by the change. - Ulrich Ruedel (submitted February 26, 2010)
SENSITOMETRY OF HERITAGE SOUNDTRACKS
A lot of attention in film restoration is dedicated to parameters that assure the quality of photographic images in film preservation. And interns at the Haghefilm Foundation have so far often been surprised how some exposure (no pun intended) to sensitometry, the science of the behavior of photographic emulsions, helps them more deeply understand the processes and parameters in photochemical preservation.
But how about soundtracks? Essentially producing photographic images of sound and turning them back into acoustic waves was a new challenge mastered with the introduction of the sound-on-film systems in the early talkie days, replacing the earlier, separate soundtrack on discs. Diving into the sensitometry of soundtracks turns out surprisingly interesting, and challenging, as we learned when recently asked to summarize lab aspects of sound for a sound restoration panel at the in Oslo, Norway, this past May.
Like with color, a plethora of analog soundtrack systems have been invented throughout Hollywood history. Some even were promoted as catchy trademarks: Fantasound, Perspecta, Sensurround … However, in terms of their photographic properties, they all fall conveniently into the two distinct groups most preservationists and technicians are aware of, that is, the variable density and variable area soundtracks.
Perhaps not surprisingly, many of the key photographic issues were explored, and succinctly summarized, as early as 1930 by Kodak’s Loyd Jones, known as the father of modern motion picture sensitometry.
In vol. 14 of the Journal of the Society of Motion Picture Engineers ( available online at the ), Jones and his co-author Savik describe how for variable density film,
“ … exposure .. is directly proportional to the instantaneous sound pressure on the microphone. (…) This problem is fundamentally identical, in principle at least, with the usual photographic problems involved in the production of motion pictures, portraits, landscapes (…) where the correct reproduction of a series of variable brightnesses is of prime importance.”
Consequently, for a direct reproduction of the sound in the theater, the characteristic sensitometric number, gamma that is, needs to equal unity (or 1.1-1.2, to compensate for the inevitable flare in optical systems).
But for its then-experimental successor, the variable area track (or variable width, as called back then) the situation was and is quite another one. Again, Jones and Savik:
“ The general requirements (…) for variable width are radically different (…) One portion of the sound track area should be (...) opaque, while the other should (…) remain almost completely transparent.”
In other words, gamma (contrast) needs to be very high.
Variable area tracks require careful optimization in soundtrack printing, though, due to a unique problem related to their stark contrast. In recording or printing soundtracks, so called image spread invariably occurs, that is, the effect of light scattering within the emulsion into areas where it is not supposed to be. In the sound signal, this manifests itself as distortions, especially of sibilants. Fortunately, given careful and correct exposure of the sound positive, the two image spread phenomena that occur separately in negative recording and positive printing can effectively cancel each other out. This means that there are unique optimum printing lights for every sound track negative. This effect is generally known as cross-modulation (XMOD) cancellation, and tested to this day periodically in any lab engaged in soundtrack recording and printing
Of course, for heritage elements the exposure required for distortion-free printing is effectively an unknown number for each and every case. Different soundtrack emulsions throughout the years do not make the problem any less complex. Consider, for instance, Kellog’s seminal review of the History of Sound Motion Pictures in Journal of the Society of Motion Picture and Television Engineers, Vol. 64 (1955), available on the website of the
This review gives details on ten different materials for the 1930-45 period, including the especially ironic case of Eastman1360, an emulsion that was actually too good for its time in terms of XMOD cancellation.
Nor is it easy to locate any technical information on heritage film stocks. Corporate archives may typically have discarded such “obsolete” material, and only recently have FIAF archives become more proactive in saving such technical materials as photographic spec sheets before they vanish.
Working with historic soundtrack negatives thus requires trial and error until the best quality sound positive is achieved. Such testing and printing from a historic sound track negative is traditionally even necessary when its mere purpose is to subsequently digitize the sound.
Fortunately, digital sound track scanners that employ either image capture or line scanning of soundtrack can now facilitate a direct readout and electronic or digital XMOD cancellation. A couple such instruments are indeed available in facilities or for purchase by labs or studios both in the US and Europe. Since the autumn 2010, such a device - the SoundDirect Laser Soundtrack Scanner from is also installed at Haghefilm BV, and thus not only allows state-of-the-art soundtrack restorations in the lab, but also be available for learning and study to the interns of the Haghefilm Foundation. ( submitted September 3, 2010 )
A lot of attention in film restoration is dedicated to parameters that assure the quality of photographic images in film preservation. And interns at the Haghefilm Foundation have so far often been surprised how some exposure (no pun intended) to sensitometry, the science of the behavior of photographic emulsions, helps them more deeply understand the processes and parameters in photochemical preservation.
But how about soundtracks? Essentially producing photographic images of sound and turning them back into acoustic waves was a new challenge mastered with the introduction of the sound-on-film systems in the early talkie days, replacing the earlier, separate soundtrack on discs. Diving into the sensitometry of soundtracks turns out surprisingly interesting, and challenging, as we learned when recently asked to summarize lab aspects of sound for a sound restoration panel at the in Oslo, Norway, this past May.
Like with color, a plethora of analog soundtrack systems have been invented throughout Hollywood history. Some even were promoted as catchy trademarks: Fantasound, Perspecta, Sensurround … However, in terms of their photographic properties, they all fall conveniently into the two distinct groups most preservationists and technicians are aware of, that is, the variable density and variable area soundtracks.
Perhaps not surprisingly, many of the key photographic issues were explored, and succinctly summarized, as early as 1930 by Kodak’s Loyd Jones, known as the father of modern motion picture sensitometry.
In vol. 14 of the Journal of the Society of Motion Picture Engineers ( available online at the ), Jones and his co-author Savik describe how for variable density film,
“ … exposure .. is directly proportional to the instantaneous sound pressure on the microphone. (…) This problem is fundamentally identical, in principle at least, with the usual photographic problems involved in the production of motion pictures, portraits, landscapes (…) where the correct reproduction of a series of variable brightnesses is of prime importance.”
Consequently, for a direct reproduction of the sound in the theater, the characteristic sensitometric number, gamma that is, needs to equal unity (or 1.1-1.2, to compensate for the inevitable flare in optical systems).
But for its then-experimental successor, the variable area track (or variable width, as called back then) the situation was and is quite another one. Again, Jones and Savik:
“ The general requirements (…) for variable width are radically different (…) One portion of the sound track area should be (...) opaque, while the other should (…) remain almost completely transparent.”
In other words, gamma (contrast) needs to be very high.
Variable area tracks require careful optimization in soundtrack printing, though, due to a unique problem related to their stark contrast. In recording or printing soundtracks, so called image spread invariably occurs, that is, the effect of light scattering within the emulsion into areas where it is not supposed to be. In the sound signal, this manifests itself as distortions, especially of sibilants. Fortunately, given careful and correct exposure of the sound positive, the two image spread phenomena that occur separately in negative recording and positive printing can effectively cancel each other out. This means that there are unique optimum printing lights for every sound track negative. This effect is generally known as cross-modulation (XMOD) cancellation, and tested to this day periodically in any lab engaged in soundtrack recording and printing
Of course, for heritage elements the exposure required for distortion-free printing is effectively an unknown number for each and every case. Different soundtrack emulsions throughout the years do not make the problem any less complex. Consider, for instance, Kellog’s seminal review of the History of Sound Motion Pictures in Journal of the Society of Motion Picture and Television Engineers, Vol. 64 (1955), available on the website of the
This review gives details on ten different materials for the 1930-45 period, including the especially ironic case of Eastman1360, an emulsion that was actually too good for its time in terms of XMOD cancellation.
Nor is it easy to locate any technical information on heritage film stocks. Corporate archives may typically have discarded such “obsolete” material, and only recently have FIAF archives become more proactive in saving such technical materials as photographic spec sheets before they vanish.
Working with historic soundtrack negatives thus requires trial and error until the best quality sound positive is achieved. Such testing and printing from a historic sound track negative is traditionally even necessary when its mere purpose is to subsequently digitize the sound.
Fortunately, digital sound track scanners that employ either image capture or line scanning of soundtrack can now facilitate a direct readout and electronic or digital XMOD cancellation. A couple such instruments are indeed available in facilities or for purchase by labs or studios both in the US and Europe. Since the autumn 2010, such a device - the SoundDirect Laser Soundtrack Scanner from is also installed at Haghefilm BV, and thus not only allows state-of-the-art soundtrack restorations in the lab, but also be available for learning and study to the interns of the Haghefilm Foundation. ( submitted September 3, 2010 )
