Название | Timeline Analog 3 |
---|---|
Автор произведения | John Buck |
Жанр | Историческая литература |
Серия | Timeline Analog |
Издательство | Историческая литература |
Год выпуска | 0 |
isbn | 9781925108682 |
For actual editing tasks. the IVES had a rotary knob with a centre detent like that found on a Sony VO-5850. The knob was mounted on a horizontal slide which allowed it to move it left or right and in turn select the Source or Record deck for control. In addition, the knob could be pushed down to initiate an Edit Preview. Other features included a built in black burst generator, programmable A/V facer and mic/line mixer.
But most importantly what IVES didn't have was or need any adjustments to tweak in order to repeatedly perform frame accurate editing. It just worked time after time after time.
Hester had only just returned to California after the launch of IVES at NAB when he took a call from a headhunter acting for Convergence Corporation.
I was flattered that Convergence wanted to hire me and I ended up at Convergence's corporate offices getting a big sales pitch from Julian Hanson and George Bates. But I think what Convergence was really up to was to simply get me to leave EECO, because if I left EECO, so too would have most of the ideas and enthusiasm for the IVES and that could have ended or stalled its continued development.
FRIENDLY’S DINER
Ron Barker quit his temporary office at Adams-Smith and moved home to research editing. He needed to raise money to go any further with the project and arranged to meet Chet Schuler who he knew had managed to get the Masscomp Computer project bankrolled.
The two met at Friendly's Diner in Sudbury, Massachusetts. Barker recalls:
Chet seemed like a very competent engineer and he had raised money. He convinced me he could help, so I asked him to put a financial plan together and manage the engineering team as well.
Schuler remembers the first discussions.
Ron had become an expert at his hobby of building and flying model helicopters and he had recognised that video editing was terribly cumbersome and counter intuitive and that if one tried to fly a helicopter by keyboard (for instance) it would be a complete disaster. He surmised that this created a unique opportunity for us to revolutionize the video editing process by inventing a more user-friendly product.
I came from an aerospace and computer technical/engineering background and had very little awareness of the practices of the video industry (which ultimately may have been to my advantage) causing me to express some reservations about this opportunity.
He didn’t know exactly what we were going to do, but he had an itch so bad he had to scratch it. It would not have happened if he hadn’t been involved. He was so bent on fixing what he saw as a hopelessly frustrating process, the linear online editing system.
With a home PC and dot matrix printer they formulated a business plan. Barker recalls:
Wang Computer had been successful with the Wang 1200 Word Processing System that was marketed as replacing typewriters with computer word processors, so I pitched an editing system that was a visual equivalent - a picture processor.
Barker and Schuler created a partnership that they wanted to call Picture Processor (PIP) but a quick search discovered a printing company Postal Instant Press already had the same initials.
Instead their new venture was christened Composition Systems Corporation (CSC) and included in its assets register Schuler's personal frame grabber, Barker’s Sony Betamax and his Sears monitor. Barker doggedly tested the theory.
I spent several weeks at the dining room table trying hundreds of picture setting combinations to finally determine that 4kb images were sufficient to "see where the star falls off his horse".
He tried unsuccessfully to interest West Coast based investors or business partners, so Barker changed his focus to the other hub of video editing in America, New York City.
John Storyk and Alex Major were the new owners of Metropolis Studios and I knew that they were keen to build the best studio production facility in the US and that they would benefit from being able to promote a state of the art editing system. So I wanted to solicit seed money in exchange for first delivery. Storyk recalls the meeting, vividly.
I can remember first meeting Ron, almost like it was yesterday. A meeting was set up for him to visit our offices on Union Square in Manhattan. Alex and I listened as Ron presented a well thought out presentation of this concept that Ron called 'picture processing'. We continued to look at each other asking the other (silently) if either of us had ever heard of anything quite like this.
When he was finished, we asked Ron if he could actually do this project. We believed him and agreed to inject $10,000 into this project, even though we barely had this money ourselves, but considered it a good investment. I remember having to actually lend Ron money at the end of the meeting for him to get back to Boston.
With money to continue working, Barker reconnected with Chet Schuler and the two men decided to set out on a research trip. Barker knew the solution to the creative impasse of electronic editing was to a visual interaction between editor and machine.
The user interface should be similar to the way he flew his helicopter, using eyes and ears but not looking down at the controls. Barker had also tested the concept of using the 'pause' function on his home Betamax machines to create in and out points, visually. He believed that there must be an existing method to combine both concepts for editing. Schuler recalls:
Ron decided that I should experience video editing from start to finish in order to convince me what a great opportunity this could be. First he explained the long editing process beginning with the editor reviewing his takes offline with burned-in time code while taking voluminous notes concerning their sequence, his preferences and all corresponding reel and time code locations in preparation for an expensive on-line session.
Anybody who watched an online session back then without any previous prejudices, like I was, just couldn’t believe how laborious it was and how expensive it was!
I was told before the session that it was costing the client over $500.00 per hour which (even though it wasn’t my money) made me feel very fidgety all though the long drawn out discussions concerning the time code locations of the edit entry and exit points for each edit point of the ad spot.
I guess because I was new to this entire field. I had no prior experience, no prejudices and so I looked at electronic editing and said to Ron, “This is crazy!” At one particular scene showing a skier making a wide showy turn in powder snow, there was a discussion of whether the exit point at location time code should be moved 50 or 75 frames forward or back causing me to whisper to Ron “
Why are they chatting in eight digit numbers to describe the particular image location they want to designate?”.
He started to explain rudimentary time code and I interrupted him. 'That's simple, I get it." What seems to be missing is that timecode number relates to a specific image, a specific frame, so why not have the computer use images of the frame in question to let the editor select his location of choice and have the computer keep track of the numbers, in the background.
Take actual timecode out of it and replace it with the images. There was a long pregnant pause after which Ron asked “Can you do that? To which I replied "of course, just digitize and store the images in a data base that cross references the time code locations and let the computer keep track of the editor’s time code choices”.
Schuler’s lack of familiarity