Moving from tape to disk?

Although the advantages of working with non-linear editing systems are still not very clear, there is no doubt that over the course of the next five years, editing, production and post-production work will have to be done on disc.

Linear editing, the one in which you can return or advance the tape to find the sound and images that need to be worked, still has many followers and defenders in the field of professional television and video production, thanks to its large storage capacity. However, the medium surrounding teletransmission is transforming, while the needs and tastes of consumers are multiplying and more different from each other. For example, new and different types of services and operations emerge day after day – such as pay TV, video-on-demand, quasi-video-to-demand and multi-channel distribution – that cannot be provided through linear tape systems but through disc.

To achieve, then, competitive advantages, professionals in the television and video industry must use non-linear editing and the disc format, with which they can achieve randómico access, in real time – a possibility that video recorders do not offer, VTRs. However, the choice of a disc or tape depends on the specific editing needs. If what you want is to work with a high storage capacity then tape is shown as the best option; but if the priorities are randómico access and real time, the alternative is the disk format.

The fact of abandoning the tape to fully adopt the disk has so far generated discussions that have not reached a clear conclusion: both systems offer advantages and disadvantages. Tape will always be cheaper than disk and is a removable medium, but randómico access is a competitive advantage that cannot be compared. To clarify this dilemma, M. Horton, of the Sony Broadcast International of Great Britain, proposed during the International Broadcasting Convention, held last September in Amsterdam, a pragmatic strategy that he called "hybrid edition", which consists of using both tape and disc in a mixed hardware and software environment. A system that mixes current technology with future technology.

According to Horton, proponents of nonlinear editing argue that working in traditional studios means limiting creativity and uncompromisingly committing to tape; while proponents of linear work argue that the system they employ goes far beyond disk-based technology. This is where your hybrid edition proposal becomes valid.

If compared to editing rooms where you work only with tape, the great advantage of exclusive disc systems is that once the material is digitized, the editing work is done very quickly thanks to the randómico access. However, the time spent on digitization and rendering can increase the time spent developing a program.

Similarly, as it is unlikely to handle external real-time devices (such as digital effects) and slow motion needs rendering, you can not then edit "on the fly". To this we must add the lack of fineness devices, such as manual real-time control for audio or video dissolution. Now, if you want to achieve a quality/cost/storage disk ratio between 5:1 and 20:1, then you must resort to compression. This can achieve reasonable image quality but it is very likely that a large number of artifacts will have to be introduced in special circumstances. In this case, VTRs are relegated to simple loading and unloading functions in material transfer.

Even if it were possible to end these operational limitations in a disk-only environment, the question we should ask ourselves is not "will it be possible to do this work exclusively with disk?" but "does it make sense to do this work exclusively with disk?".

Another bottleneck that occurs in an exclusive disc environment is the difficulty of visualizing complex effects, which makes the editing process too slow. For example, if the file size of an image is based on 5:1 compression, systems can take a long time to manipulate frames – one that normally takes a second to present can take up to four seconds or more. As long as there are more "virtual channels" of effects in the system, the problem can be much greater.

No one can deny the great benefits of the randómico access characteristic of exclusive disk systems. However, the big question is that we can create a systems architecture that offers this benefit and also has the advantages offered by an editing room?

The answer to the above question is hybrid structures, in which the benefits of both disk and tape are combined; that is, take advantage of having real time and storage devices. In a hybrid architecture, therefore, it is not necessary to digitize all the material to the disk before editing and the same system can be used for editing on disk, tape or the two combined (i.e. off-line and on-line in the same system). It is important to remember that editing begins as soon as the tapes are in the VTRs.

In a hybrid system all inputs and outputs are given in real time by conventional signals, by means of NBC connectors, instead of using the movement of files in a central processor. Signals also pass in real time through dedicated external hardware for image and sound processing while the computer works the gui-Graphic user interface (GUI-G) and control system.

Because no rendering is given in this environment, the system has all the benefits of a conventional editing room, such as real-time effects (including manual control of video and audio dissolvences), real-time slow motion, on-the-fly editing etc. Similarly, the computer does not have to expand too much when processing heavy images (such as DME movements) since this work is done by hardware specially designed for it.

According to The same Horton, of the Sony Broadcasting International, with the hybrid system that he proposes, the signals are changed by means of a crosspoint configuration, and this allows to make cuts or transitions "live" or connect external feeders, such as title generators or cameras. It is also possible to process the image (before storing it in a disc-compressed form) directly in the VTR – something useful for chroma-key work or other kinds of photo manipulation where signals subjected to high compression have to be processed. When working with high levels of compression in exclusive disc systems, the Betacam has a YUV signal that by itself corrects the color and elaborates the chroma-key.

A disk-only workstation needs to display multiple motion sources in its VDU. In the hybrid system proposed by Horton this alternative is optional, although problems may occur with the saturation of images on the screen. Standard TV monitors can also be used, connected to the outputs of the sources. This option is especially useful for the tasks of evaluating the quality of the photograph, since it can be observed directly at the output of the device.

In a hybrid system you can edit directly from draft recordings, submasters, previously edited material, etc., but these resources do not necessarily have to be compressed on disc. This form of editing can be called "multiplane editing" because you have to work with different planes of access time and required duration. With this editing system, unnecessary artifacts generated by the problem of "compression on compression" found in systems that use heavy levels of compression are avoided, at the time of reimporting previously edited images.

As a hybrid system is modular, the possibilities of using different storage media are greater. For example, systems can be built that can work online or offline using a similar architecture, which facilitates an exchange of EDL between off-line and on-line and even the possibility of running off-line and on-line in a single system.

As the density of disks, which is increasing, is on par with the density of tapes, which is also increasing, disk and tape combinations will be possible in the near future. But can these combinations occur on a general-purpose computer or will they require dedicated hardware? In a hybrid system, such as the one proposed by Horton, both can be used: a computer to generate pictures or images of text and then perform special effects supported by dedicated hardware. Of course, computer systems are expected to become faster and faster, but as this is achieved, dedicated hardware will also become faster and cheaper. In a hybrid system, both the fast computer and the improved hardware can be used.

The goal of hybrid editing is not to stop any post-production technique that is currently working with existing editing technology, while adding benefits of randómico access and the ability to use external software. It is rather to appropriate the benefits and advantages of the tape's own storage to combine them with the real time and randómico access provided by the disk.

In the meantime, the suggestion is to first solve the problem of digitizing before editing through the direct acquisition of randómico access media. It remains to be seen whether this technology is economical and practical. There are many unresolved questions in this regard, including operating costs, image quality, and the fragility of a disc filmer.

It seems reasonable to conclude that the hybrid system, which allows direct-to-disk editing without digitizing, is a different way of looking at the problem and is based on existing technology that has been approved.