As a broadcast consultant and college media instructor, I'm often asked about NLEs. Probably no other product technology has had such a huge effect on the amount and type of video content that's available. Being able to easily assemble video segments to form longer-form content is key to today's vast array of programs.
There are basically three types of professional editing systems. The simplest is a newsroom editing system. These systems allow journalists to complete simple pieces or rough cut their items and send the script, cut list and notes to a craft editor for final conforming. They typically require minimal computer resources, allowing them to be easily deployed at journalists' desktops. In many cases, they are tightly coupled with a particular manufacturer's storage system and have direct interfaces to the other newsroom components.
At the other end of the editing spectrum is the finishing system. Finishing systems are used in high-end post for such tasks as color correction, motion graphics and compositing. These systems can be expensive, but they provide all the bells and whistles needed to create cinema-quality programming.
In between these two extremes is what's called a craft editing system. This is what most people think of as NLEs.
The craft editor
Let's first examine the task of craft editing. In my classes, I tell students that it takes talent to get the audience to buy into the story you are telling. Give the same project with the same footage, script, etc., to 10 different editors, and each of the resulting videos will have a different feel. Craft editing systems are designed to allow the editor to tell the story the way he or she “feels” it. So, craft editing goes beyond simple cuts, fades, titling and even some effects. It requires skill to merge video, audio and effects to creatively tell a story.
Most craft editors rely on one of two operator interfaces: a timeline or a storyboard. While other systems have come and gone, these two remain the most common. And, although some manufacturers may claim to do both, you will usually find that they excel at one or the other.
The storyboard interface displays the first frame of each available clip, sometimes called picons, in a storyboard format. (See Figure 1.)
A timeline NLE interface focuses more on tying clips to a visual time reference. (See Figure 2.) Clips are placed (drag and drop) onto the timeline. The NLE connects the clips into a sequence, and the timeline automatically adjusts.
The above is a simple description of the two major interfaces. I'll make no attempt to venture into each system's benefits or drawbacks. Such discussion is best left to those whose jobs are dependent upon them.
What the buyer needs to understand here is that storyboard or timeline interfaces have strengths and weaknesses. One may be a preferable interface for the type of material you are producing. Ask your editors what type of interface they prefer and why. However, know that human editors are as passionate about their particular choice as PC and Mac users about their computer religions. You will need to filter out opinion from fact.
Features
With hundreds of NLE features and options, comparing each across multiple vendors is impractical. Start by narrowing the list of desired features to something manageable. Here are important aspects to consider.
GUI, ease of use
An NLE's GUI provides the interface to all the sophisticated processing that is going on in the background. How responsive the GUI is given certain tasks, i.e. trim to fit, link/unlink sync, multicamera cutting, change fade duration, type and timing, etc., is not only dependent upon the hardware, but also on the system's design. That is one reason NLE prices range from hundreds to tens of thousands of dollars. Like other technologies, you often get what you pay for.
Sequencing forms the basis of all storytelling. A storyteller needs the ability to quickly and easily move clips around, and trim the start and end points to create the tension and release required to hold viewers' attention. The ability to arrange clips to control time, not in the technical sense of time remapping, but rather to move back and forth in time and stretch or shrink time by using the arrangement and length of the available video material is at the heart of editing. Such basic tasks should be easily learned and quickly executed by the NLE.
Both the speed of the underlying hardware and the interface design affect the overall responsiveness of the system. If the cutter cannot respond to the requests of the director and achieve the required results without delay, then he or she is lacking in skill, or the system is not up to the assigned task.
Metadata is king
In today's connected production environment, the handling of metadata is often the key to improving performance. Whether the metadata is in XML or EDL, MXF or AAF format, the way the system handles the data in the containers' headers can be critical to a smooth workflow.
This means that NLE performance is highly dependent on accurate and inclusive metadata. Comprehensive content metadata may not be required for every craft editing task, as the shots have already been selected and may already be sequenced. But for long-form content, having good metadata hooks can mean the difference between a stressful and stress-free operator environment.
The first place to look at how an NLE handles metadata is within the bins. What data fields can be selected for display? Is it possible to automatically fill the bins (or storyboard) with data from the asset management system?
Is versioning information stored in the resulting file? In other words, can we come back later and easily change logos, voice-overs and subtitles?
Consider audio
Close your eyes and watch television. Now open your eyes, but plug your ears. In which case was it easier to follow the plot? I've asked nearly all of my media production students to perform this simple test to help them understand the importance of audio.
Be sure you understand how the NLE handles panning, multichannel sound, EQ and various bit rates. Does it have the ability to record voice-over? Can it cut on the beat? Will video-centric editors be able to easily handle similar audio tasks?
When a fade to black is made, the outpoint audio should be adjustable without having to unsync the sources. Resampling should automatically happen in the background as required. When a clip is stretched or shrunk slightly, the system should know if repitching the audio will suffice. Is it easy to add a live voice-over into the timeline?
While there are high-end editors that focus on audio sweetening, the required audio toolset for craft editing need not be as comprehensive, but it does need to be a lot easier to use.
Mixed timeline
Being able to implement a mixed timeline is the ability to drag and drop a file, regardless of source, into the timeline and simply edit. Editors should not need an engineering degree to edit together divergent video formats.
Examine how long it has taken the manufacturer historically to implement new formats. How flexible is the frame server running in the background? Obviously there are certain minimum performance requirements for each format such as EX, D10, ProRes, AVC Intra, DNxHD, etc. The question is: How are these implemented? Is there a plug-in to run or a delay while some conversion takes place, or is it transparent, with no operator intervention?
Look at how the system handles various wrappers such as MXF, QuickTime, AVI, etc. Can the NLE still use the essence even if the wrapper is not 100 percent conformed? Ask about any intermediate codec used when rendering effects. Does the use of mixed codecs increase render time? How good is the frame rate conversion and scaling?
Effects
Probably the most used effect is the crossfade. But how is a crossfade inserted into the edit? Can the editor specify a group of cuts and apply crossfades to all of them at once with or without a black or white frame?
No editing system can compete with the variety of effects available through plug-ins. If effects are an important element in your production, consider how many third-party suppliers are out there creating products for the system being considered.
Calculate TCO
Over the past 20 years, engineers have had to add an increasing number of business skills to their technical toolboxes. In addition to price, CFOs and GMs are asking about total cost of ownership (TCO) and return on investment (ROI).
Determining TCO for editing systems is difficult because they are configured and supported differently. In addition, performance (throughput) is highly operator dependent.
Some vendors offer support contracts, which may include software upgrades and loaner hardware. Determine exactly what you are paying for with each supplier.
In some cases, the NLE software package is a download, and support comes through user groups. In a top-market TV station, this option may not be viable. In other situations, this approach may provide the high-quality video you desire at the rock-bottom price you can afford.
Warning: Be careful about relying exclusively on your own in-house engineering resources for installation and maintenance. While this may allow you to reduce the initial investment, the TCO will probably be the same.
Involve the users
Finally, involve your editors in the investigation and decision process. You must have their early buy-in to ensure long-term success. Be certain that your decision is firmly based upon the organizational goals and needs, not just price. Initial cost is but one factor in these highly operator-intensive systems.
Start with a prioritized list of features and performance. Add to that your own operators' input. Follow up by talking with both vendors and their customers. There is no one best editing system and plenty of good products from which to choose.
Christopher Walker is a broadcast consultant and college media instructor.
