A Scenaria for Video

by Dean Winkler

It's that time of year again, and I've just completed my annual spring ritual: off to Las Vegas to attend NAB, a stop in Utah afterwards for a few precious turns on my monoski, and back to New York. While Vegas is always somewhat surreal, NAB was particularly odd this year; a type of collective insanity seemed to overrun the convention. If you will permit me a little humorous exaggeration, numerous manufacturers were claiming that "in six months they will have full-resolution component digital video, real-time compositing and a Jumbotron running on an Apple Newton." What I found bizarre was that so many people believed it.

What was driving this? The revolution in offline editing made practical by Avid and other manufacturers of computer/disk based nonlinear offline systems does not yet have an equivalent for online editing. That so many attendees were salivating over "vaporware" demonstrates just how strong the demand for this is. Over the past fifteen years the changes in traditional online editing systems have been evolutionary, not revolutionary. Although the signal quality of online editing systems has improved dramatically and the image and sound quality of a component digital online system is truly astonishing, the basic configuration of these systems has remained the same. Online editing systems are still assembled from individual, hardware based machines. Typically, this includes a video switcher, a few channels of digital video effects devices, an audio mixer, a character generator, an editing control computer, a signal router, and several video tape recorders and digital disk recorders. There are several disadvantages to building an editing system this way: The system does not have true random access capability. Each machine runs under its own individual "operating system," with the edit control computer functioning as little more than a master synchronizer (as opposed to having one integrated control structure for all of the system's functions). There is a high degree of operational complexity and the system is expensive to purchase and maintain.

Given these disadvantages, why build an online editing system this way? Beyond maximizing signal quality, a conventional online editing system has one capability that outweighs all of the disadvantages listed above: the ability to perform real-time previews. By this I mean that the system can play back several full resolution sources simultaneously, and while these sources are rolling, multiple parameters can be adjusted with the results viewed in real time. For example, repositioning, edge softness, transparency and color saturation of an image can all be adjusted in real time, while viewing the relationship between that image and other images in a scene. This real-time preview capability is particularly essential when compositing images for visual effects sequences.

Clearly, it would be wonderful to have a new type of online editing system that combined the signal quality and real-time preview capability of existing systems, with true random access and a totally integrated control architecture. Ideally, such a system should also be simpler to operate and less expensive to build and maintain than current online editing systems.

Unfortunately, none of the computer/disk based systems currently being touted by manufacturers as replacements for the existing online editing systems will be able to meet these criteria. Even if, for the sake of argument, these computer/disk based systems could offer signal quality equal to the quality of existing component digital online editing systems (a big assumption, particularly with regard to digital video effects), and these systems were capable of performing all of their "rendering" operations in real time, they still would not be able to offer a sufficient level of real-time preview capability. There are three reasons for this:

• First, the bandwidth required to simultaneously play back several sources of uncompressed, real-time component digital video and audio is too large for any single computer/disk system that is readily available.
• Second, the computing power required to execute all of the real-time functions of an online editing system in software is far beyond the capabilities of even the fastest existing supercomputer.
• And third, pull-down menus or similar CRT based controls will never be ergonomically fast enough. Some type of physical controls (e.g., switches, shaft encoders, lever arms, etc.) will continue to be required.

"Ah," I imagine you're saying, "if this is true, why has the film community thoroughly embraced computer/disk based systems for compositing visual effects sequences for movies?" Bear in mind that their point of comparison is not online video editing systems, as these systems are not capable of working with the high resolution images required for motion picture scenes. The film community has adopted computer/disk based systems as an alternative to optical printers. By comparison to optical printers, computer/disk based systems are a godsend. If, however, a high-resolution equivalent of current online video editing systems were readily available, I'm willing to bet that California would be littered with them. Lest you think I'm merely whining about the direction I see the manufacturers in our industry heading in, allow me to propose a solution.

An excellent model for a new type of online video editing system already exists in the audio world: The Scenaria, manufactured by Solid State Logic (SSL). We have three of these remarkable machines at Mixed Nuts (Post Perfect's audio division), and I have spent a considerable amount of time using them. The Scenaria is a complete, "online" post-production system for audio. It integrates all of the functions traditionally executed by individual pieces of hardware in audio post-production environment. For the purpose of it as a model for a new online video editing system, the following capabilities of Scenaria should be emulated:

• The signal quality of the system is excellent.
• It is a true random access system, with multiple disks that can be assigned to different projects.
• It provides complete real-time preview capabilities (in this case, up to 32 channels of full-resolution digital audio and one channel of reference quality video can be played and modified simultaneously). But the idea is that the results of adjusting any of the system's relevant signal parameters can be monitored in real time, as the controls are adjusted.
• It uses a combination of CRT-based menus and assignable, physical controls. All of the functions of the machine are accessible from the CRT menus, and in addition, functions which need to be adjusted in real time can be assigned to physical shaft encoders, switches and faders.
• It is a completely integrated system, with one master "timeline". All of the functions of the machine are controlled from this one timeline and are fully automated.
• Signal routing between the Scenaria and the external world is fully integrated into the control architecture. Even its patchbays are under system control. SSL accomplished this by building a "hybrid" system. They used standard computer components (e.g., microprocessors, disk drives, etc.) augmented with dedicated hardware to execute specific functions (e.g., signal routing, A/D D/A converters, etc.). The software they developed integrated all of these separate components together into a unified system, so that the Scenaria operates as if it were one machine.

This same approach could be used to develop a revolutionary online video editing system. I believe that such a system should sell for $650,000 or less, including everything except the video tape recorders. In the hope of getting this developed, Post Perfect is offering the following:

First, I will personally give an extended Scenaria demonstration to any interested video manufacturer (engineers from Avid have already taken me up on this). Second, hanging on the wall of my office, directly below one of my ceramic "Elvis" plates, is Post Perfect purchase order # 18042, for an online video editing system that meets all of the above specifications. Hopefully, there is a manufacturer willing to take it off of my wall.

© Copyright 1995, Dean Winkler. Used by permission. All rights reserved.

Dean Winkler is president of Post Perfect, a high-end post-production facility based in New York City.

The purchase order in question is still hanging on his wall.