Zprávy památkové péče 2016, 76(1):76-90

Digital Restoration of Cinematic Heritage. The DRA Method

Marek Jícha, Jaromír Šofr

Just as there exists "belles lettres," there is also "belle cinématographie": a precious set of artistic cinematographic works that cannot escape worldwide attention for their unquestionable quality and contribution to culture. Cinematography of this quality is art that, due to its superior parameters, should be considered part of a nation's cultural heritage and recognized as cultural heritage.
With the advent of new digital technologies, how to preserve this cultural heritage in an unchanged form and quality for future generations becomes more and more urgent as new methodological problems appear. Digital restoration can significantly expand work opportunities of restorers in their quest for a permanent preservation of this heritage. The digitally restored film copies become a source of creating further digital copies of films necessary for their further presentation - projection in digital cinemas, digital TV broadcasting and dissemination by the means of data media.
The new profession of the digital restorer brings with it increased ethical responsibility for the results of work and requires high professional expertise. Digital restoration techniques clearly exceed conventional analogue restoration techniques previously used by archivists working only with archived film copies. Work with the data needs to be carried out in the interest of the desired result, which is the appearance of the work indistinguishable from its appearance during its first-night presentation; otherwise, there is a risk of creating new, damaged versions of films. Poor result of digital processing cannot be considered a work identical with the author's original work, sometimes a case of mere plagiarism occurs, which is to the detriment of the authors of the original work. It is the more serious offence because the spectators, to whom the low-quality output is presented as the "right one", are misled and drastically deprived of the true appearance of the original work and of the level of artistic qualities of its authors, since the spectators usually do not have any possibility of making relevant comparison.
In 2013, under the Applied Research and Development of National and Cultural Identity Programme (NAKI), a research project called "Methodology of Digitisation of the National Film Fund: Methodology for evaluating the quality of film image from the perspective of the viewer's visual perception with the aim of creating an equivalent restored digital copy when compared with the original archive sources" was launched. This project defined a clear mission to set rules for providing new digital access to cinematographic works in the Czech Republic. The researchers under the project, AMU professors and experts from the National Film Archive (NFA), embarked on a collaboration which, unfortunately, did not last long. The collectively developed Methodology of Digitisation of the National Film Fund, based on the DRA (Digitally Restored Autorizate) method, became a subject of a professional dispute. The reason is that digitisation is bringing new views of a cinematographic work as such, of the related copyright aspect as well as new contexts, including restoration and a clarification of terminology. The impulse to the formation of the DRA method (where the film digitisation workflow involves a professional digital restorer working with an expert group to yield an outcome certified as a new original source of the cinematographic work) was the original request of the NFA to invite the authors of the film's image, namely cinematographers, to work on the digital restoration of the first important works of the Golden Fund of Czech Cinematography: Marketa Lazarova, The Firemen's Ball and All My Compatriots. Although these first digitally restored films in the Czech Republic can be described as the result of a successful harmonious cooperation of archivists, authors and post-production workers in the area of film image and sound, after a year and a half of the work on the project, NFA specialists left the research team because the new director of the NFA disagreed with the DRA method. His main reason was a disagreement with the participation of professional associations in supervising the restoration of the film fund by which the NFA met the directives of the European archival association.
In the approach of archivists to the digitisation of a cinematographic work, it is, unfortunately, sometimes possible to trace a rather misleading way to finding "authenticity" in the preservation of marginalia. In these matters, however, the essence of things should be very carefully considered. For example, maintaining the image format of the final work is an essential quality of the film image since it concerns the possibilities for creative work with the linearity of the image, including the so-called linear composition. In contrast, an irrelevant characteristic of the film image, preserved by the negative, are marginal marks made for specific purposes that are unrelated to the will of the authors.
Researchers from AMU including cinematographers, sound masters and cooperating external experts on digital and film technologies (from CTU) thus had to proceed to complete the methodology without film archivists from the NFA. The DRA method is now complete and ready for certification. It was this method which was used in 2015 to restore the very first film - The Stone Bridge by director Tomáš Vorel. The world première was already on 26 September 2015 at the International Film Festival Manaki Brothers in Macedonia, where the NAKI project representatives, prof. Marek Jícha and prof. Jaromír Šofr, introduced the DRA method to the world public, also in the form of the project presentation.
It is meaningless to talk about improving or failure to improve in the process of digital restoration because image information contained in the original negative transferred by being scanned into digital files brings a completely new qualitative benefit even without intentional interventions of restorers and their expert groups or of authors, if these are available. The inevitable amount of image-related damage due to ageing is compensated by digitisation in the sense of returning the image to its original appearance which cannot be considered improvement. No film author, whether a director, cinematographer, sound engineer, script writer or composer, did not envisage, when creating a film, that the film would be faded, its colours shifted and its sound interrupted. Films are created in the faith that they could be screened in their unaltered original form. Rigid insistence on digital fixing of the poor quality of a copy under pretence of so-called "historical authenticity" renders an ill service to cinematography as a performing art and represents misunderstanding of what the film medium actually is. The current digital era obliges both filmmakers and archivists to revise such notions of authenticity and to condemn them as worthless from the point of view of preserving cinematographic works for future generations.
The age of digital technologies has given origin to a cardinal problem that affects the current practice of film restoration and fails to be addressed by any of the codes of film archiving of the past. An evident conflict arose between the opinion held by the personalities representing the archive in terms of what the appearance of the digitised film copies should be like, and the interest concerning the appearance of the digitally restored works on the part of their authors. Recent experience from presentations of digitised film copies demonstrates that the interest of the authors of the works does not differ from that of the viewing public. Although this viewing public is not professionally trained in the area of technologies, it is not correct to provide it with substitutes (versions) of the original films. The archivists' interest is satisfied when they digitally document the current condition of the works in their analogue form, irreversibly impacted by devastating physical and chemical changes. Authors and, clearly, spectators themselves have no reason to accept this age-degraded form, provided that it can be avoided in today's digital age. Both authors and spectators are interested in digital restoring which restores the original appearance of the work. It also becomes an ethical question whether the above mentioned devastated form should be forced upon authors and spectators if we have available means and procedures - the DRA (Digitally Restored Authorizate) method - to guarantee the preservation of the original visual and acoustic quality of the work.
The DRA method strives to create new digital copies using the original negative. Just as an analogue 35 mm film copy once used to be made from the original negative, today, an additional new digital copy may originate. A restorer is responsible for ensuring that a new version of the work does not emerge and is the guarantor of this process, even in the case when authors are no longer alive.
Within the research of the Methodology of Digitisation of the National Film Fund, the DRA method has been further elaborated and is ready for certification. The method accurately identifies and characterizes individual assumptions and workflow steps of digital restorers who strive for the emergence of the original source of the cinematographic work in a digital form. The application of the method does not give rise to a new version of the work but to its Digitally Restored Autorizate (DRA). It can be considered the original source of a cinematographic work if it meets the following criteria:
1. The image is digitised exclusively from the original negative (if it has not been preserved, then from other duplicated generations, in the worst case, from release prints) at a resolution corresponding to the original cinematographic material, in the original frame rate, with the aspect ratio and image size corresponding to the signed copy used at the time of the première screening, with sufficient range of brightness and colour depth of the image preserving full fidelity to the tonal range of the analogue original.
2. A professional workplace of officially recognized expert film and digital restorers or such restorers that are university educated in the fields of cinematography and sound engineering participated in the restoration of the film.
3. The authors of the film participated in the restoration of the film - cinematographers, sound engineers and directors (if available) and representatives of their supervisory professional associations of authors which are professional authorities representing the highest expert corrective supervision.
4. The restored film is approved by an expert group consisting of restorers and authors (officially recognized film restorers, the aforementioned authors of the cinematographic work, if they are available, and representatives of professional associations of authors, film historians, technologists and other required experts, hereinafter referred to only as the "expert group"), whose members should sign, after mutual consent, an official certification document on the DRA or a restoration report documenting that the DRA method was used.
5. Differences in quality between a reference copy selected by the digital restorer or a digital facsimile of the reference copy and DRA must be, as far as the appearance is concerned, in order to preserve the author's concept of the cinematographic work - within the meaning of the Copyright Act - approved based on a competent analytic opinion of the expert group.
What is used as the single source for archiving the DRA is the so-called Master Archive Package (MAP) and the Intermediate Access Package (IAP) from which, subsequently, all the copies of any distribution formats are made, namely without any intervention into the appearance of the work as per the above-defined criteria (except for the changes in the overall size and resolution of the image and the different levels of compression depending on the respective distribution format).

Keywords: cinematic arts; film restoration; DRA method; preservation of cinematic heritage

Published: March 1, 2016  Show citation