The Digital Eye: Customized Applications

In this months Digital Eye column, Autodesks Tom Ohanian asks if there is such a thing as one-size-fits-all in todays accelerated production environment?

Image courtesy of Deron Yamada. © 2004 DYA367.

Applications in the media and entertainment industry are largely designed to tackle specific tasks: digital non-linear editing systems (DNLEs), digital audio workstations (DAWs), 3D modeling software, etc. However, most users are not running off-the-shelf applications but instead are modifying and augmenting what the manufacturers version of a product does. Custom scripting and plug-ins are frequently employed in order to achieve the desired level of functionality. There is no one-size-fits-all application in todays accelerated production environment.

Over the last five years, weve seen an ever-decreasing timeframe from the content creation to final distribution stage of a project. We are being asked to create even more in less time while the sophistication level of the content has dramatically improved. In years past, a signature effect that first appeared in a big budget feature film would not appear at the same level on a television program for at least a year. Today, that window has shrunk to weeks or, at most, a few months. At the same time, toolsets and workflows are all changing. Most feature films are edited digitally, almost all sound work is done digitally and 3D animation software is rapidly replacing the traditional animation process.

Over the last 10 years, the desktop revolution that so influenced and redefined desktop publishing has had a similar effect on the media and entertainment industry. Software packages for video editing are now on many desktops and may even come pre-loaded with new computer purchases. All the while, of course, the feature sets of DNLE software has continued to grow and, in many cases, we are approaching 10th and higher versions of these products.

The Bundling Opportunity and the Bundling Dilemma

Even as individual software applications add more features, missing features or functionalities are often addressed through bundling multiple applications. Separate modules for video editing, audio editing, 3D animation, title generation, etc., are combined in an attempt to provide more functionality. Some of these attempts are more successful than others from a users point of view. One of the major challenges to the bundle concept is the level of integration or interaction between and among the different applications. Is it a straightforward procedure to begin working in one module and move seamlessly to another module?

Filling the Gaps

As software packages are developed and undergo iterations, there is always the risk of not being able to include all the desired features in the delivery timeframe. Features may take longer to code and test and often manufacturers will tend to concentrate on developing and enhancing their core functionality. Entire companies have been started around the notion of filling the gaps that exist in extending functionality that is missing in core applications. Ultimately, if a company continues to develop such functionality, it may become an indispensable tools manufacturer or mature beyond that to a full-fledged software manufacturer.

The Rise of the Plug-Ins

Software that has traditionally been developed to extend the functionality of a core application can sometimes become as or even more important than the main application. In many cases, artists and facilities are investing hundreds and thousands of dollars in obtaining plug-ins or in modifying off-the-shelf software. The use of plug-ins, combined with software modifications is also occurring at a time that the work methodology is also changing.

From Workstation to Pipeline

Not only are traditional workstation toolsets being augmented by plug-ins, filters, shaders, renderers and an assortment of other task-focused applications, but also the actual manner in which work is being done is undergoing a shift.

Workstations are increasingly being connected to network attached storage (NAS) or storage-attached networks (SAN). Facilities are evolving from video signal centric workflows to data-centric models. Metadata and essence (content) exchange are improving and slow standardization is occurring.

These shifts from core applications to core plus plug-in plus customization, and from disconnected workstations to workstations on a shared network using shared storage are being joined by the need of artists and facilities to actively manage the process that makes up their workflow.

There may be slight or more extensive customization of applications through scripting languages such as python or perl. When one combines plug-ins that may have been added to an application, customized scripts, proprietary changes to the application via its software development kit (SDK), the result is less a piece of off-the-shelf software and more a facility-specific proprietary workflow component. The number and nature of these customized components in turn become a highly tuned and highly modified system.

How Pipelines Affect Software and Solutions Development

One need only look at the future confluence of editing systems, the Web and the de-emphasis on hardware in lieu of software functionality to realize that how we work today is going to change rapidly. Editorial and collaboration systems operating over the LAN and WAN will continue to be refined; videotape usage is being rapidly replaced by disk storage systems; metadata and essence interchange is improving and plug-in architectures are becoming almost mandatory.

Tom Ohanian.

The Promise and the Cost of Customization

Large visual effects films such as King Kong and Superman Returns will certainly continue to create more requirements for eye-popping effects creation. Is further customization driving up R&D costs as filmmakers push the boundaries of what can be achieved in visual effects? Will pipelines become more complex? Will deadlines be pushed to the very last minute?

It is inevitable that both artists and facilities will continue to drive toward and demand that both the software and hardware that they invest in can be modified and support open structures. The past approach of setting out to purchase a video editor or a 3D animation package will be replaced with a mindset of purchasing products with the strict intent that that they can be significantly altered to fit the pipeline requirements of a facility.

Is off-the-shelf, out-of-the-box software a thing of the past? For the professional artist and facility, the answer may soon prove to be yes.

Tom Ohanian is senior director, Autodesk Consulting for Media and Ent. Former vp of product development at Digital Media on Demand and chief editor and corporate fellow at Avid Technology, Ohanian is the recipient of a 1994 Emmy for his pioneering co-invention of the Avid Media Composer, a 1995 Academy Award for Scientific and Engineering Achievement for the Avid Film Composer and a 1998 Emmy for the Avid Multicamera System. Ohanian, an accomplished systems and user interface designer, inventor and author, is an industry expert whose work has led to the widespread adoption of digital nonlinear video and film editing systems, as well as innovative digital media workflow strategies, and LAN/WAN digital media collaboration. He is the author of Digital Nonlinear Editing and co-author of Digital Filmmaking, which have been translated in more than 18 languages.

Tags