The primary challenge facing director Cameron Crowe‘s editorial team when they started building a documentary on the rock band Pearl Jam for the band’s 20th anniversary this year was straightforward: deciding how to sort, synthesize and strategically utilize the reams of material available to them. The result of their efforts, the two-hour documentary Pearl Jam Twenty, debuted on Oct. 21 on PBS’ American Masters series.
Pearl Jam (L-R): Mike McCready, Matt Cameron, Eddie Vedder, Jeff Ament and Stone GossardPhoto courtesy of Steve Gullick
PBS promotes the movie as having been “carved from over 1,200 hours of rarely and never-before-seen footage,” plus more than 24 hours of more recent concert and interview footage shot by Crowe (@CameronCrowe). Co-editors Chris Perkel (@Perkel76) and Kevin Klauber clarify that the two-year process of picking through Pearl Jam concert and candid footage required the film’s research team to sift through closer to 3,000 hours of footage, much of it unlabeled or mislabeled and existing in just about every kind of film and video format imaginable.
Long before there even was a documentary, Klauber was working for Crowe and producer Morgan Neville as an assistant editor, projecting Super 8 footage on a wall to get a sense of what was available. A few months later, Perkel joined the project and cut together a two-minute video designed to sell the film, a video that later became the basis of the documentary’s trailer and the foundation of Crowe’s storytelling method. Klauber had been so instrumental in sorting the original material that he was asked to co-edit with Perkel. The two set about building a narrative based on a specific reference from Crowe.
Cameron Crowe and Pearl Jam”s Eddie Vedder, Italy, July 1993Photo courtesy of Cameron Crowe archives
“Cameron’s direction was that he wanted the documentary to be like The Kids Are Alright, the Who documentary that is basically whole performances of songs with interview bits in between that slowly reveals a portrait of the band,” Klauber explains. “That became the template—he insisted we approach it as fans who have been allowed to raid the vaults of Pearl Jam.”
Thus, the methodology was to find defining pieces of concert footage from the band’s 20 years on the road, layer that material into chronological story points, and then insert archival and modern interview pieces where appropriate.
“Our original job was to whittle down material to make it manageable enough to give Cameron a sense of what we had to work with. Then he went out and did interviews based on that,” says Perkel. “The initial performance pieces we constructed became the chief building blocks, and the interview material was supplemental. That was an unconventional process—usually you do the interviews first and build the skeleton from there. We pretty much did the opposite.”
Pearl Jam concert
There was so much material from so many places and in so many formats that Klauber describes the film as “a textured collage of media.” Indeed, footage poured in from sources ranging from MTV to boxes in garages, but particularly from Pearl Jam itself and the band’s longtime documentarian, Kevin Shuss. It came as 35mm, 16mm, Super 16mm, Super 8, DVCAM, HDCAM, HDCAM SR, Betacam, DVD, VHS dubs, Hi8 and various digital files, among others. The production sent material it was interested in to post house Lightpress of Seattle, the company where colorist Jeff Tillotson has been grading Pearl Jam videos for years.
Lightpress had transferred and re-transferred so much Pearl Jam material over the years that it ended up being a main source of footage for the documentary, and it already had originals or high-quality versions of many clips the filmmakers were looking for. The company used its Teranex Xantus to convert certain clips that existed in odd frame rates, but Lightpress relied primarily on its Quantel iQ system to clean up, assemble, conform and color grade material chosen for the documentary—the same system Lightpress would later use to create final deliverables. For editing, Lightpress provided the production with standard-def tapes or hard drives containing uncompressed native Apple Final Cut Pro files.
Co-editor Chris Perkel at the “big board.”
“We spent a good year giving them material we had transferred,” explains conform editor Shane Dillon. “The iQ let us conform everything in native resolution, allowing for a better transfer.”
Perkel, Klauber and assistant editor Adi Cabigting made up the entire editorial team, working closely with researcher Susan Ricketts and Crowe. It was Cabigting’s job to organize and track footage in such a way that the two editors, working in Apple Final Cut Pro v.7 on Mac Pro workstations, could seamlessly build individual chunks of the story.
Cabigting logged material as it returned from Lightpress using a customized database logging system to track all media sources, owners and other important metadata. That system churned out trackable labels for each piece of physical and digital media that came in—an approach that proved crucial when it came time to order masters or get clearances for third-party material.
Along the way, the production strategically limited the amount of media that could go into any single Final Cut Pro project in order to prevent the editing program from getting bogged down.
The organizational process was hampered by the fact that some of the material that came in was not labeled, or was mislabeled to begin with. Pearl Jam’s status as a cult band with an obsessive fan base played in the production’s favor, however. The fan web sites Two Feet Thick (www.twofeetthick.com) and Five Horizons (www.fivehorizons.com) feature documentation from fans regarding almost every single Pearl Jam show in detail. As a result, those sites became a handy tool for the editors. “If we discovered something useful [on those sites], we e-mailed Kevin Shuss and asked for specific footage,” says Klauber. “About 99 percent of the time he had it and would send it to us.”
The other crucial organizational tool on the project was the analog “big board.”
Co-editor Kevin Klauber
“It was just a giant board, but it was very important,” says Perkel. “We created a management system where we laid out our story beats in a line, like a timeline on index cards, and married them to a performance or song. Lots of times we found performances we liked but had to figure out how to use them narratively, so we’d shift them around the big board until they found a home that stuck. Above that line we had other cards that represented cool scenes or shots or fragments that we knew we wanted to use somewhere. We would stick them together, and eventually it became a road map.”
In Final Cut Pro, the editors built multiple project files initially to group relevant footage, such as particular concerts or interviews. As the documentary evolved, specific project files in FCP were created based on particular story themes or structure points. Besides avoiding overloading their systems with thousands of hours of media, this method was a convenient way for the editors to locate footage that matched a particular topic or theme they were working on.
Despite their caution, both editors say, there were moments when the volume of material caused temporary crashes—but they add add that the only significant glitch was an audio render problem on playback whenever the load got too heavy. Occasionally, when exporting large QuickTime files, they would notice audio dropouts or sudden level changes that were not in the actual cut.
“If we exported a QuickTime of cut sequences, sometimes certain render files appeared to be lost or would not connect during export,” Klauber explains. “They would sound okay in Final Cut, but after export there were problems. We eventually rectified it by exporting an AAF of the whole sequence, dropping it in, and then exporting the other audio. When we got to the point of doing full assemblies of the movie, we would do that for entire scenes. We would nest each scene into an assembly sequence, and then also export an audio mixdown of that scene. Before we nested the sequence, of course, we would make sure it was all rendered. Once we figured that out, it significantly cut down our export times.”
Lightpress’ Jeff Tillotson used the iQ for final color correction and other tweaks. To make it easier for Crowe to view and approve the work efficiently, the company eventually arranged to use an edit suite at FotoKem in Hollywood to do that final work, archiving all files directly from the Lightpress iQ project to FotoKem’s iQ system. Final deliverables were created in the iQ in HDCAM SR formats ranging from 23.98 fps to 30 fps to 50i, as well as Digital Betacam (29.97) and 16-bit DPX versions.