What is the best non-linear editing system?

Throughout my work in non-linear editing, I have often encountered the same question: what is the best system for editing? My answer has always been the same: the system is as good or competent as the publisher who uses it. This answer has to do exclusively with the very personal opinion that the craftsman makes the tool.

This is also the case when talking about editing, both in film and television, especially at a time of rapid and drastic technological advances. Since the beginning of my career in cinema, first as an assistant and then as an editor, there have been great changes in the reproduction of image and sound, and in the way of manipulating them during editing.

I consider myself lucky to have started working in film and not in video. At the beginning, in 16 mm film, then moving to 35 mm format. The only direct contact with video editing occurred for a short time, three years after I started working in the film industry in Britain, and that as a response to my need to expand professional knowledge. And it wasn't out of love that I sought out of video, but out of a selfish desire to promote my availability in the publisher market. From the outset I was attracted by its immediacy, the work is there: it is shot, then the material is looked at, not as it happens in cinema where after shooting it is necessary to take the rolls to the laboratory, wait for the development and printing, and later look for a projector that allows to advise the work on the screen. I hated the inconvenience that this created for me as an editor, when it came time to play with the shot.

In video, a matrix tape is copied to a secondary tape, either as a copy in order to prepare an EDL (cutting decision list) cut list, a process called off-line and that makes the final mastering process faster and more efficient. During the off-line you can experiment with the structure and shape of the final program. Once finished, the cut-off list is taken to an online room where the final copy is prepared, and music, narration, effects, etc. are added. If during assembly the editor wants to make any changes that involve cropping shots or adding a new longer or shorter shot, the assembly process must be started again, from the point where the alteration is executed, because everything that is extracted must be replaced exactly.–otherwise it is a clean slate. This is what is commonly known as the final process.

In cinema, on the other hand, this is not the case. With the film there is a freer and more agile process to develop the assembly of scenes. If you want to move a scene to the end of the program, to put it at the beginning, it is only necessary to peel the splice tape, extract the required segment and put it in its new place. Once the assembly is finished, the copy is delivered (which in video would be the EDL) to the negative cutter, who will shape it so that the laboratory is responsible for producing the final copy: the exhibition master, once the sound is mixed.

As we can see, the "non-linearity" of cinema is sacrificed for the immediacy of video However, these two working methods are laborious by the amount of time they consume in their execution, and above all they reduce the opportunity that creators have to experiment with language and audiovisual form. This inconvenience has been alleviated with the system of zeros and ones, because it converts analog information, which constitute image and sound, into digital information.

Precisely, digital technology (not so new, because for some years it began to be used for recording and reproduction of texts and sound, exclusively) converts the analog information contained in the magnetic tape and, or, into cinematographic film, in allogarithmic combinations. This information can thus be manipulated immediately – that is, in real time – in online systems , or secondarily once the material has been compressed – the amount of information in each frame goes through a reduction process, affecting its quality, but allowing its storage on hard drives to facilitate its manipulation. The non-linearity of cinema and the immediacy of video are combined in digital technology, thus allowing quickly and efficiently access and handling of audiovisual material.

Let's return to the question proposed at the beginning of this article, what then is the best system for non-linear non-random assembly? I would answer with another question, what does the editor want to achieve with the chosen system?

If the primary consideration is to keep the cost of equipment low, and what matters is the off-line assembly of the material – in order to produce a cut list – there are more than five systems on the market that allow great compression of the image; it breaks down into a greater number of pixels (the points that constitute the pictorial representation), thus minimizing storage requirements, which today translate into very high costs – a 9 Giga byte disk, for example, costs around US $ 5,000.

It should be noted here that compression does not reduce the operational capacity of the systems, as long as you have a good amount of RAM, but simplifies the complexity of the signal that is digitized. There are systems such as MATROX, D-VISION, PREMIER or VEMOS (a system developed in Spain by an Argentine engineer recently launched on the market), which allow the digitization of VHS tape material as a timecode printed on the image. With the information of these systems, lists are generated to make up the master material. These off-line devices cost around $15,000 a unit.

For this price what the user acquires is a computer, either Mac or PC, image compression cards, sound, and communication between computer, storage disks, and above all software, the program that allows the typing, assembly and preparation of lists.

If, on the other hand, what an editor is looking for is greater control of the assembly and finishing process, there are non-linear equipment that not only allows off-line assembly, but even without producing the EDLs, they serve to execute the mastering or process of on-line.

The off-line only uses one of the two fields that make up the video image. The video frame is composed of lines, so one field contains the odd lines, the other field the even. To illustrate the above it is necessary to add that a Giga only allows the storage of three minutes of high-resolution image, with a compression of 3x1.

Until a few months ago with video images, although using two visual fields, the data was compressed, but minimally, thus making maximum use of the capacity of hard drives However, currently mounting equipment that offers images without compression is coming to the market, maintaining the dynamics of assembly that non-linearity brought to the industry. Such is the case of systems such as Avi, Lightworks and Heavyworks, Video Cube, Media 100 by Data Translation, Sony, Postbox by Panasonic, D-Vision, Night Suite by Amtel and Fast.

These systems, although they differ considerably in price and configuration of software and hardware, allow to carry out the same operations. The most notable differences lie in the way in which the systems develop the interface, that is, the way the user uses to carry out work operations. The computer is the platform on which software and hardware are related.

I wanted to make a distinction between on-line and off-line because in our Latin American environment (where economic resources are smaller compared to industrialized countries), the advantages of off-line assembly have not been fully exploited. In Latin America almost exclusively, assembly is done on-line. By the above I mean that a lot of time is wasted in making assembly decisions in a room, time that could be dedicated to mastering. In this way, cost efficiency would not only be attractive to those who pay for the finished work, but to those who offer the service.

A rationalization of the combined use of off-line and on-line equipment would offer greater flexibility when post-producing programs, both for film and television.

When choosing between non-linear assembly systems, the fundamental aspects to consider are, in my opinion, the following:

1) Image quality: determined with compression and storage.

2) Final objective of the program: corporate use or commercial transmission. This determines the quality of the final product.

3) Format compatibility: the state of current technology does not allow to invest all resources in a single absolute system, but in one that provides flexibility of communication with others that offer different facilities. Not all the resources of a total post-production system are always used, so it is not worth investing in facilities that will not be used regularly.

4) Management of the database: this has to do exclusively with the way of organizing the material with which it works, in a way that allows its capture and handling quickly and efficiently.

5) Hardware and software platform of recognized stability for users: systems proliferate that are not yet sufficiently tested in real work situations, and that allow to have a reference point of their reliability.

Finally, it would be necessary to insist on the adequate preparation of the operators in charge of carrying out the assembly. At present there is an abysmal lack of information material in a language other than English in all systems, and manuals are generally whole volumes of complex information that users do not consult, because they do not understand them. That is why many editors still do not know in depth the way in which the systems operate and as a result their work ends up being complicated and full of moments of frustration, since what was expected as a day of creative work, can become a bleak battle with the computer. In this extreme case, it would be worth asking what is digital technology for?