Organic nonlinear editing

In addition to covering new products, trends and developing technology, Video Innovations is about how people are using the new tools at hand.

Producers working in various sectors of the industry across the country are at different stages of delving into the nuts and bolts of the digital toolbox, newer formats, or fiber-optically linked production networks. In order to explore some of the learning curves and pass along the experiences of fellow techno travelers, vi invites readers to share their technology odysseys (call the editor for details).

The following piece by editor David Ransley, describing a post-production solution for a budget/time-challenged project, is the first in a series of articles wherein the people on the front lines pass along the rewards and the pitfalls encountered along new production paths.

Return to Regent Park was a monster of a job. Using D/Vision (dos program on ibm platform), it took me 16 weeks to cut this documentary, which included a 56-minute version for the National Film Board and a 45-minute version for CBC Newsworld.

Summer 1993: Regent Park is Canada’s first and oldest Government-subsidized, low-income housing project. Over 10,000 people live there. Producer/director Peter (Bay) Weyman of Close Up Films set about capturing the mood of its people as they confronted the problems of poverty, crack addiction and crime in this city planning-begat ghetto on its 40th birthday.

December 1993: Weyman ended up shooting 74 hours of original Hi-8 footage. He felt that Hi-8 cameras were less intimidating than a Betacam or film cameras, and he also enjoyed the freedom that the hour-long Hi-8 tapes gave him.

All of the footage was dubbed over to Betacam sp tapes and they became our masters. vhs viewing tapes were made at the same time; they included a time code window and a user bit window (to identify the tape numbers). There were also approximately 10 hours of archival footage that had been taken from various sources – the CTV Television Network, cbc and the nfb.

Weyman was concerned about the cost of renting a Betacam player to digitize the shots into my system. I made a few phone calls and discovered that a Betacam player would probably cost $8,000 over the planned eight-week editing schedule, an expense Weyman wanted to avoid.

I suggested we use the vhs copy of the Betacam masters instead. Weyman was concerned about the image quality and I was thinking about the time code we would need. I knew that I could read vitc (Vertical Interval Time Code) off a vhs tape, and I suggested we do a test to resolve Weyman’s concern about image quality and my technical interest.

I’m glad that we did the test because we discovered a problem. I captured about five minutes of footage and Weyman and I agreed that the image quality was fine for the off-line, but the cbc, which did the first set of dubs, could not put vitc on the vhs tapes, a procedure that any post-production or duplication house could do. As a result, Weyman had all of his dubs done at the nfb in Montreal.

The first scene we cut was titled ‘Fight Night.’ Weyman was doing an interview at Regent Park when they heard a commotion outside. They looked out the apartment window and saw a fight brewing across the street. The welfare cheques had just been issued and some of the locals had gone out drinking and, sure enough, fisticuffs ensued. A couple of minutes later the police clashed with the rowdies in an effort to control the situation. It made for quite an exciting scene.

This was Weyman’s first time cutting on a nonlinear system, and once he got comfortable with the process he was more willing to experiment. For the next six weeks we would spend each day cutting a new scene.

This is how a typical 10-hour day would unfold: Weyman would sit at his desk making notes on his transcriptions while viewing his vhs tapes. He would highlight the sections of a scene that he wanted me to capture the next day. In the meantime, I was capturing the footage from the previous day’s notes. I would spend two hours capturing the shots and approximately three hours cutting a scene, and over the last four hours Weyman and I would refine the edit.

After a few days, Weyman told me he really enjoyed the nonlinear process, describing it as a ‘very organic way to edit.’

I was working at medium resolution and only had space for eight hours of footage. Because we were inputting an hour’s worth of material each day, I had to find a way to manage the space available on my hard drives.

The solution: cut five scenes and have five hours worth of material in my system. Each scene was approximately five to 10 minutes in length. I would make a new edit that included the five cut scenes. This new select scenes edl would then be copied and converted to a new edl called a bumper edl. The bumper edl would have three seconds of extra material for every cut, so I could add up to three seconds to any shot or sound cut. The bumper edl would be perhaps 45 minutes in length and I would use it to make a new 45-minute file which would take the computer 70 minutes to create. To make better use of my time, this process was done overnight, and the next day I would come in and delete the original five hours of material.

After six weeks I was capturing maybe two or three hours and saving perhaps 20 minutes.

February 1994: When we started our fine cut I had six hours of material that had been copied as new files, which left two hours of free space on the drives for additional footage.

One day, without warning, my computer gave me the ominous message, ‘cannot write to drive g.’ It was a new problem for me and an unusual one. ‘g’ hard drive, a 2.1 GIG Seagate, could play my files but it would not allow me to write new files to it. I called my dealer, Digital Renaissance, and they suggested the problem might be a faulty cable, so I changed it and it worked.

The next day the problem returned. My drive was under warranty so the dealer said he would replace it. I took my computer in on a Friday and the following Tuesday morning it was ready. Luckily, the dealer was able to copy the files from my damaged drive to the replacement, which saved me several hours of recapture time.

March 1994: Everything worked well for the first few weeks and then I encountered another problem. A file was somehow corrupted and we were unable to get into my directory (job) ‘Regent.’ I sent a fax to Chicago. An hour later the software folks at TouchVision called back and I spent an hour on the phone, checking files and trying to isolate where the problem may have occurred.

We didn’t find the problem but we found a solution. I normally started the D/Vision program by typing dv at the c: prompt, which would take me to the main D/Vision screen. I would then call up the job I wanted, in this case, ‘Regent.’ If I followed this procedure, the computer would hang and I would have to reset the computer. I discovered that if I called up the job directory ‘C:Regent’ first then typed dv, everything worked fine.

A few weeks later I installed a BETA 2.2 version of the D/Vision software and the problem corrected itself, although working with Beta software can sometimes bring a few surprises. When I first tried to install Beta 2.2 it was a disaster. The new program would lock up the computer. I called Chicago and spent time with a junior tech rep trying to find the problem, but after two or three hours I reinstalled my old version.

A week later, I arranged with the people in Chicago to reinstall the Beta 2.2 version again, and this time I had the senior tech rep on-line. We had to remove all evidence of version 2.1 from my files and then everything worked out just fine.

April 1994: When it came time to complete the project I let the D/Vision do its traceback, which it did flawlessly, and it gave me an edit decision list for the hour-long show of approximately 780 edits. We mastered from Betacam to Betacam and it took two full days to do the edit, which is about 50 edits per hour. Normally I would have expected to do closer to 100 edits per hour, but the large amount of material we had to go through made the on-line go longer than usual.

The end result: Weyman is very happy with the outcome of the nonlinear process and can’t see how he could have finished the project in the same time frame if he had cut it on tape. As an editor who has a nonlinear system, I’m very pleased that I could use my hard drive space so effectively, although in the future I suspect hard drive prices will drop dramatically and that I’ll chuckle about the way we used to manage our footage.

David Ransley is an independent editor of 12 years standing, working out of Motion Picture Video in Toronto.