By Bob Turner
By Bob Turner, from Videography magazine
I had heard rumors on the Internet. I called Pipeline Communications' Robin Hoffman to find out if they were true. It wasn't as though I hadn't been expecting the news for a long time, but it nonetheless saddened me to have it confirmed. The news was the demise of CMX.
Chyron's President, Edward Grebow, confirmed that his company's CMX subsidiary would no longer be developing new editing systems. They will, however, continue to sell from the current product line. And even though the company will continue to support those CMX systems already in the field with parts and remaining bug-fixes, they will not continue to develop the software.
This news really got me. Although I haven't touched a CMX keyboard in years, my resume still states "CMX Editor" and I proudly list my involvement on what once was known as the CMX Editors Advisory Panel. To me, CMX was not just a tool; being a CMX Editor meant you had attained a high level of accomplishment in the video editing profession. To me, "CMX" meant clean EDLs, high numbers of edits per hour, and mastery of an online editing suite with some of the most powerful tools available.
Many believe CMX offered the first video editing system. Actually it was eight years after the 1962 Ampex introduction of Editec that CBS and Memorex formed CMX to create an experimental editing system. CMX's first product was announced in 1971, and Adrian Ettlinger was listed as the inventor. That product was in two parts: the CMX 600 nonlinear offline editing system and the CMX 200 online editing system. Together the price was $500,000. The first commercial application was a made-for-TV movie, Sand Castles, for CBS.
Did you know that the CMX 600 was the first digital nonlinear offline video editing system? That system had a removable stack of 11 digital-storage platters. Each 29 MB platter held five minutes of half-resolution black & white moving images, audio, and what would later evolve into what we refer to today as SMPTE timecode. Together the six-platter system cost $30,000 and held 30 minutes of "dailies" at a mediocre image quality.
The CMX 600 output a paper tape, sometimes 30 feet or more in length, with holes similar to the old computer punch cards. This tape was fed into the CMX 200 linear online assembly system. The CMX 200 controlled quadruplex source and record VTRs, which would build the program. Besides the paper-tape reader, the only other interface was a keyboard on the ASR-33 teletype, which was reputedly quite difficult to use.
In 1972, with five units sold (three to CBS, one to Teletronics, in New York, and one to CFI, in Los Angeles), users begged for a direct human interface controller to the online-assembly editor. A direct computer keyboard terminal was added, which became the CMX 300, introduced at NAB '72. The high-cost, poor image quality, high-maintenance CMX 600 nonlinear editing system was discontinued. Perhaps it was too far ahead of its time. In any case it didn't fit in with union job-description rules.
In 1973, with the introduction of the U-matic tape format, CMX unveiled its CMX 50 edit controller, then considered the first practical VCR offline editing system. In 1974 CMX was sold to Orrox, and CMX's Manager of Product Development, Dave Bargen, left the company to start writing his own software programs (409 Clean, Trace, and Wizard, which went on to become the SuperEdit program). In 1978 the CMX 300 evolved into the CMX 340. Within the next 18 months over 90 percent of all videotape editing for broadcast was done on CMX systems.
By 1979 CMX was developing its CMX 3400 and the radically new CMX 3400 Plus. That latter product was to have voice-activation control, a radical new mappable keyboad with LEDs on each key to describe the available functions, a windowed GUI, advanced database-management capabilities, and other "space age" features. Unfortunately, CMX/Orrox had other priorities, among them DBS, which the company invested heavily in (almost 20 years too soon). The dream of an electronic nonlinear editing system, however, did not die with the CMX 600.
Other companies picked up the torch and started their own development projects. And alternative linear technologies soon began development, including Ettlinger's work on an ED-80 editing system--later to become Ediflex--and Bargen's development of ISC SuperEdit, which would soon become a Grass Valley product. CMX came out with a "laptop" (inasmuch as it was promoted as being able to sit in your lap) controller, the CMX Edge. The CMX Edge interested editors but failed for lack of power, high cost, and the tendency of its small built-in CRT to skew in such a way that the buttons along the side of the screen failed to line up with the CRT choices.
In 1984 Montage was announced, followed by Ediflex, TouchVision (later D/Vision), and others. CMX closed down its expensive DBS operations and started laying off employees (including 3400 Plus designer Rob Lay, who went to work for Lucasfilm on the EditDroid project). CMX began losing market share to ISC SuperEdit on the high end and to many companies at the low end. Convergence, Sony, and Paltex edit systems became quite popular. The planned CMX 3400 Plus never made it past prototype. A 3400 Plus did come out later on, but was just a new version of the 3400.
It was during two CMX Editors Advisory Panel meetings that I realized just how much trouble CMX was in. In a 1986 meeting during SMPTE, CMX introduced their nonlinear LaserDisc-based CMX 6000 (later evolving into a digital video system known as the CMX Cinema). The introduction surprised and angered most of the advisory panelists because none of them had been asked for design input. This loyal group of experienced professional editors also felt that CMX was off course, pursuing film editing over video editing for nonlinear technology.
It was, I recall, at the next New York CMX Editors Advisory Panel meeting that the infamous "revolt" took place. Instead of a working meeting on how to improve the CMX line, an argument erupted between those who had purchased the 3400 Plus based on the promise of certain features being developed, and CMX, which decided that those features would be incorporated on the next model, which would require an extra purchase. I attended several more Editors Advisory Panel meetings after that, but the group was no longer useful and most members stopped coming.
As the 3500 and 3600 were introduced, I had already edited on Grass Valley SuperEdit systems and appreciated that system's advanced features. Soon afterward I discovered nonlinear editing systems and was won over by them. By this time CMX had become a wholly owned division of Chyron, which I saw as a very positive development. I was excited when the CMX Omni came out, but was frustrated by its bugginess, its right-handed keyboard, and its linear orientation in a world that was increasingly going nonlinear. At Chyron's NAB press conference two years ago, journalists had a hard time believing a statement that "CMX Omni editing was alive and well." It was clear to many of us that it was only a matter of time before CMX was history. And so it is.
What's most ironic in the demise of CMX is the fact that the CMX EDL is the communication language that most of today's nonlinear offline editing systems communicate to their online counterparts.
On a personal note, I can't help but to recall certain memories of using CMX systems:
The incredible number of keystrokes it took to preview, and then perform, a single GPI event on the original 340.
My first CMX job, where I was told "Real editors do not need a Gismo!"
The system crashes and paper-tape reboots.
Greg Featherman, one of the best editors I have ever known, patiently teaching me what it meant to be a true CMX editor.
My favorite CMX offline suite at Media 1, in Boston, with a large window overlooking the Beacon Hill townhouses and the Charles River beyond.
Most of all there were the wonderful people of CMX. The enthusiastic Gary Hinderliter and his CMX bulletin board system--the first I ever accessed. The energetic Denise Gallant, with her focus on effects compositing and the wonderful CMX newsletter she produced (she replaced Christin Hardman, one of the most competent and knowledgeable product managers I've ever known). Hardware guru Gene Simon (one of editing pioneer Jack Calaway's heroes). Gary Attanasio, an excellent sales rep. And the evangelistic John Shike, faithful to the end. Shike, VP of Chyron's International Business Development, asked to add this to my story:
"CMX developed great machines, including, ultimately, the Omni product line. The market for linear editing systems, however, has been largely replaced by nonlinear systems. We will therefore not be developing new CMX linear editing systems. Chyron is dedicated to the ongoing support of its CMX customers, as well as the development of new products and technologies for our industry."
I find CMX's passing a rather sad occasion. I will always think of myself as a CMX Editor.