Systems dedicated to the Digital Intermediate process have evolved over time and have only really become commercially viable in the last five years for smaller more independent production houses. Although broken down in to three components; scanning, recording and the Digital Intermediate itself, it is not as cut and dry as it sounds. There are a myriad of vendors that have grown from other products and now claim to be reaching into the DI market. Others have purposefully driven and marketed their product, designing and refining to fit the needs of the DI posthouse. These offer turn key solutions direct from the vendor, which straight from the box can command a profit - however they are not cheap and the infrastructures surrounding a Digital Intermediate can also be costly and prohibitive, however more and more facilities are starting to invest in the technology and stretching the profit margins within the DI market, it may simply not be as lucrative or attractive as it was even three years ago.
There were a few post houses who invested and adopted technology early within development (early adopters) and helped shape the interfaces and workflow ideals. However a lot of these efforts were bastardised constructs, cobbled together utilising many pieces of equipment to function as a whole. Not necessarily at the highest optimum or quality either, previous incarnations of hardware may have a bottle neck with rendering through 8 bit processors which is not ideal for film or high definition work. A lot of in house proprietary software has gone on to become fully operational software, hardware or a combination of both. The problems with a lot of these efforts was being able to play back your film in real time, 2K logarithmic material takes a huge amount of bandwidth to play back. Proxies or smaller versions of the 2K frames were used and these in themselves could give problems. Also ways of splitting the image in to smaller easier chunks of data have been used but any form of processing on a single chunk as a whole can lead to image quality problems if hardware errors corrupt a small part of the image.
There are several ways of configuring a DI facility. I suppose the obvious would be to build one from the ground up but this is costly and an expensive procedure, although admirable, I have done this. Most though have built themselves around existing post houses and infrastructures; however this in itself can pose problems. Although fundamentally it offers a better starting point as many things will already be in place which will help the post house know where to start from. Although an existing structure is harder to change because networks exist and a working etiquette is already at play. A DI suite will suddenly draw huge amounts of resources - IT, personnel and bandwidth, if the setup is out of date to start - this can cause extreme bottle necks when you are passing large chunks of data around a network or set of production disks.
Central storage or SAN disks are the most relevant and talked about DI structures. Do you have one central storage area where everything is held and work from that, or do you copy the material to a local work station so that everything is backed up - which makes sense, especially if there is a hardware problem - such as a disk dying.
Myself in action in Dragon DI's Grading Theatre
The grading suite itself is also something which will need careful consideration - you have to decide how it should look, whether to grade from a monitor or use a projector. Space is usually the deciding factor if a post house is building a DI suite. Facilities who have the luxury of plenty can make dramatic theatres, which sits pride of place within the facility and recreates the viewing of a full preview theatre. At least with this the client believes they can sign off with confidence and what they are seeing is what they will get on the final print. This environment, I believe can possibly give the client a false sense of security (see Calibration). Ultimately this environment allows the client to view their project with all senses, such as sound and a large screen - small problems may be picked up easily where as a small screen may hide them.
Budgeting to cater for this market is an interesting position to be in, how do you cater for the demands of an ever increasing market. One that is demanding more bang for its buck and a heavier and more intricate/individual look associated to their project. Clients that have grown up through TV and the Telecine suite are already very familiar with what a colourist can do to the look of a project, so assume that it will become even more sophisticated for film. Usually they are unaware of more technical issues surrounding the output, in honesty they often don't care, they just want to see their project in their intended vision - sooner rather than later. Real time is the name of the game and the intended ideal. For 2K it is now becoming a reality albeit with a cost to the quality. 4k is now the new 2k and to get that things are still slower, more time consuming and ultimately more expensive.
From a scanning point of view, as has been mentioned, the choice of scanner is paramount to the operation. Do you look at speed over quality, cost over specification, or simply purchase the newest on the market, hoping it will attract the client just because it is the new kid on the block? Or do you stick with tried and tested methods which might not be able to be easily upgraded, as and when the industry moves on?
The DI equipment is all a matter of personal choice, as most of the equipment available all do much the same job (in this writer's point of view). A colour corrector does exactly what it should and the user/fan base will all tell you which is better and the reasoning(s) behind their decision. The tool sets are being developed with the Colourist in mind - however that said, a lot of the user base is coming from an effects/editors background borrowing a lot of the tool set from their individual background(s) as well, such as compositing. A lot of the newer systems are now an amalgamation of the best tools for the job, bundled hopefully in an intuitive and user friendly way. Star Trek interface user panels are being developed, which not only make colour more intuitive to those who manipulate it but also to differentiate a grading suite from a standard computer that anyone can use.
This approach is certainly making the Colourist feel at home because they can see the tools match the ways in which they have worked for years, jumping across platforms is daunting and manufacturers are learning that they need to help this transition.
For the filmout certain recorders are paramount to the operation such as the Arri Recorder - in fact some clients will not even work with you unless you can output via Arri, it seems to have become the industry standard. I have even known particular projects to disappear from a facility when another output method has been mentioned - Arri seems to have such a reassurance to the client that it helps fully with the operation itself. I have often wondered whether it was its legacy within the film industry for such amazing cameras and lenses that has given it such a stronghold?
The One Stop Shop (A Myth?)
As a facility grows, unless it is developed as a purely dedicated DI facility, certain parts of the operation may not be available to start with (scanning for example). It is only when you can offer all the services for a Digital Intermediate that is when you can truly be known as a one stop shop. Many facilities have gradually offered different parts of the service, until they can offer scanning, through to the final output negative you may be dealing with several production facilities to undertake your one individual job. Some would even say this is not enough of the process to truly be a one stop shop.
Nearly there: True one stop facilities as far as vision is concerned have started to emerge as the big laboratories are also offering the DI experience as opposed to an optical grade. Companies such as Technicolor are in the unique position to be able to offer everything because they can even offer the final print run, which would give them an advantage with pricing over their rivals because they can also offer incentives to use them.
The Ultimate Boutique One Stop Shop: For me would be to also have the audio facilities alongside the DI/grading process as the two run in tandem towards the end of the production. In my opinion it would make the final process easier because the production would be able to see their film finalised with the incarnations of new sound and grading treatment together, gradually updating as the project nears completion. Traditionally Audio and DI post houses are entirely separate, the film is watched under entirely different resolutions and formats, the audio often using a low resolution offline version, while the DI may use the rough cut audio alongside the final grades? Personally I believe if this happened together more seamlessly, it would also allow the director/DOP to concentrate their efforts in one place, thus not flitting back and forth between facilities in the last, often vital moments before the deadline when the two are married together.
Finally: The final piece(s) of the DI puzzle would be able to offer film finals without having to go to a traditional lab. This ultimately would keep all of the work under a one roof which will control everything within a fully calibrated loop. One piece of equipment that might be useful for limited print runs and sound prints is the real time recorder coming from Cinevation, by piql. An amazing piece of kit. This innovation could allow subtitling to be added, another cost that can be kept in house and not out sourced.