Dubner joined Computer Applications Inc., then the largest software company in the East, and became a Vice President in charge of all special hardware projects.
All my projects included a computer, programming the computer, and designing and building peripheral equipment for the computer. As an example, we designed and built Ticketron, a ticket reservation system (1966) that could service 1200 ticket selling devices in real time.
Dubner's Ticketron work, done with Joseph Abate, defined a generational change in using microcomputers (CDC 1700) to solve problems previously assigned to mainframes. Dubner explained his frugal 5000 lines for the entire TIcketron system as a system where sophistication and generality must be traded for economic and reliability.
Their work was seminal: "Ticketron - A Successfully Onerating System Without an Operating System."
Dubner continues:
Unfortunately, CAI was forced into bankruptcy in November, 1970. Without planning it I suddenly found myself starting a new company because several of the companies for which I was doing projects, wanted to continue with me.
This was the start of Dubner Computer Systems (DCS).
2. Right now
The CMX 600 System could now record and play video. Its capacity was about five minutes of black and white video per Memorex disk pack and the system usually ran four to six drives per system thereby getting 20 to 30 minutes of storage. VP Martin Fletcher was readying the marketing campaign. Yves Faroudja was impressed.
Our marketing guy, Martin was a very sharp guy and he understood what the market wanted with regards the image resolution for editing.
Strobele recalls:
It was amazing to see it working as a unit for the first time even though it was strung out across a number of work benches. We had all of the audio and video coming off a disk pack in sync and sure it was a black and white image on a standard definition monitor but we were shouting and cheering.
Jerry Youngstrom recalls:
The skip field disk drive was a marvel but it was a real challenge for David Bargen and Jim Adams when edits had to be “assembled” into a broadcast quality tape. Probably more time was spent on that technical issue than any other in the CMX development.
There was more work to do because put simply, the original specifications weren’t a true list of what was required. For various reasons many things had been missed. We just didn’t know any better but despite that I started to think we might have a real success on our hands. As an engineer that’s what you strive for.
Bill Butler recalls how the team used resources like the Stanford Research Institute which was newly independent of the university:
In those first days we spent a lot of time at SRI, which was the seminal source for input/output devices concepts. We had to find a system that was appropriate for creative editors who were mostly terrified of computers, more used to pencil and paper for records and being able to see each individual frame.
Lon Priest wrote of early interface development.
Originally, the console was designed to be alive with push buttons marked 'splice,' 'dissolve,' 'forward,' 'reverse,' 'transport’. A push button for all the myriad of electronic orders needed to edit videotape. When it was determined this would be incredibly cumbersome, an idea was taken from the computer industry itself.
CMX used the CC light pen originally chosen by Adrian Ettlinger on the CBS’ Autocue lighting system and CBS RAVE system.
Joe Flaherty later explained:
At the tip of the light pen is a photocell that senses the control words and characters on the monitor when the light pen is pointed at them and pressed. The identification of the specific control word or character is determined by the timing of the light detected by the photocell relative to the sweep timing of the picture monitor display.
The pen was used to select editing tools from a superimposed Menu displayed on the console’s right hand monitor. The interface presented four function options:
Splices - Edit - Play - Scenes.
The CMX interface acted as an electronic equivalent of a film editing system. Single letters on screen could be tapped with the light pen to set the rushes in motion.
The letter F was used for fast, N for normal speed, S for slow and J for jog, one frame at a time was. A still-frame was selected by touching any of the hyphens/rectangles between the letters. These variable speed capabilities were breakthrough achievements by themselves.
The CMX digital team had expanded on the work of Ampex engineer Anthony Poulett who had experimented with recording and reproducing television signals with an altered time base effect.
The video rushes signals had been recorded to the Memorex discs at a predetermined head-to-medium writing speed, and could be played back at the same head-to-medium writing speed, single frame access or half head-to-medium speed.
When the editor selected the ‘Scenes’ option from the right-hand monitor, he was able to call up the various rushes material. He could then play a shot and chose the first frame he required. Then he pressed the light pen to the word ‘Splice’ and the chosen frame was instantaneously transferred to the left-hand monitor and spliced to the preceding material.
By selecting ‘Play’ on the right hand monitor material, the editor then played through the scene to select a desired exit point. The exit frame of the previous one scene was always present on the left-hand monitor and the entrance frame of the next scene on the right-hand monitor.
As an alternative to a direct cut, the editor could call up the Edit option which gave him the ability to add a transition such as a (fade,dissolve,wipe,matte) or a special effect of any length from one frame on.
The 600 had the option of splicing video only or audio only from the "Edit" option.
Jim Adams continues:
From the software side, the 600 was pretty straight forward. Disk drive positioning was well known, all of the video was 30 fps and the PDP-11 real time clock (RTC) provided timing interrupts at 60 Hz.
Dave Bargen worked out the timing to switch between the primary playback disks and the auxiliary at the various playback rates and I devised a satisfactory for the Edit Decision List (EDL) as we named it. This was nothing more than an ASCII formatted list of source material, in-point and out-point frame codes and the computed record points.
The operator could also specify whether the recording was to be audio only, video only or both, and if the video was to be switched or fed through an effects generator. The audio and/or video transition was provided within the 600, but no effects were simulated.
Some of the original material was recorded in color and was frame coded in what was known as 'Drop Frame Code'. This code periodically omitted two frames from the frame count. For the 600, this was a minor issue and I solved it easily with special add and subtract routines, not unlike leap year day counting.
Interestingly the CMX team realized that the editing system could be controlled by other devices. From their 1971 patent:
It will be appreciated that while the light pen character generator described herein is particularly suited to convenient operation of the disclosed editing system, other interface terminals could be utilized if desired, such as keyboard, push buttons, joystick, etc