Nowadays we are all used to the ubiquitous availability of powerful desktop video editing devices, in the form of either standalone workstations, or as part of networked, collaborative systems.  The capabilities of the software (and the platforms that the applications run on) are impressive - here's an example of editing that would have been extremely time-consuming in the old videotape-based editing era.

To the more, ahem... mature readers, this post will likely come across as rather familiar and obvious; however for someone starting to learn video editing today, the techniques and technology involved in the early years of videotape centric production would no doubt seem very archaic and tedious.  At the time of course, whilst there were a few experimental and/or very expensive systems that mimicked some of what we have nowadays, in a practical sense there was no real alternative to tape for all day-to-day editing tasks.

Cut!

If we go back to the very early days, then videotape editing really was just that - physical cutting and splicing of two-inch Quadruplex tape, using a magnetic indicator fluid called Edivue to 'develop' the video track and display a pre-programmed edit 'flag' indicating the point where a cut was required.  Then the splicing would typically be done using a Smith Block (pictured). 

Preroll...

Before too long a more sophisticated (and practical) editing method emerged, involving the copying of video/audio segments from one or more source VTR to a record VTR; the latter having a 'blacked' tape mounted, onto which the chosen material was painstakingly recorded, one section at a time, normally as a series of insert edits.  This technique spawned a niche industry of editing controller manufacturers, some of whom (e.g. Ampex and RCA) also made the VTRs themselves; with others being independent suppliers. 

Familiar names from the heyday of bespoke videotape editing controllers include Ampex Ace, CMX, Calaway Engineering, Cezar International, Convergence Corporation, Datatron, EECO, ISC, Mach One, Paltex, Videomedia and United Media.

The 'new' methodology introduced such concepts as a preroll and bumping (synchronising one or more source VTRs to the record VTR during preroll, using tape speed override).  Then, by using multiple, sync'd source VTRs, together with external or in-built digital time base correctors (TBCs), plus a video switcher and audio mixer, we progressed to A/B-Roll editing - with (drum roll...) effects!

Initially this was all a linear process, with the edited master being built in a piecemeal fashion, usually in final viewing order from from start to finish.  However, once EBU/SMPTE timecode (originally devised by EECO) was in widespread use, this led to the development of the Edit Decision List (EDL), list management, off-line vs. on-line editing, and auto-assembly (a.k.a. conforming), and so on. 

These latter capabilities allowed changes to be made within the already created edit sequence (e.g. add, replace or delete segments with or without ripple), albeit with the requirement to re-record the sequence afterwards - a big leap forward in terms of flexibility and efficiency.

Joysticks at the ready...

An advantage that the independent edit controller manufacturers had was that they were able to control VTRs (and later, switchers) from various different manufacturers.  This was no mean feat in the days before the advent of serial (RS-422) VTR control protocols, and necessitated the development of bespoke interface modules and chunky parallel control cables for each specific VTR (and switcher) product family.

The first ten years of my career in the broadcast industry were spent working at Convergence Corporation, one of the pioneers of videotape editing controllers.  The company's early models were quite modest (by later standards) in terms of functionality, but the amount of engineering development required in their creation was still considerable.   The ECS-1B (pictured) was built around discrete components and TTL logic ICs (i.e. no 'CPU'), and controlled Sony Model 2850 U-Matic tape decks. 

These early VTR models normally only produced pictures at Play speed; however, a substantial internal modification to the deck's internal mechanism and electronics allowed the operator to visually search for edit points at variable speed, under the control of the ECS-1B's joysticks.  Again, this is something that editors take for granted nowadays, but at the time it was a big step forward.  As a side-note, the internal VTR modifications led to disagreements with Sony over warranty invalidation; Convergence argued that it was helping Sony to sell a lot more U-Matic tape decks, but relations between the two companies remained frosty for several years afterwards.

The company then developed a series of microprocessor-controlled editing systems, including the Convergence ECS-200 range, whose operator console is pictured here.  These controllers were based around a single Motorola MC-6802 8-bit CPU, which was able to handle all the VTR / switcher controls as well as performing sophisticated EDL management (including an implementation of the 409 list cleaning program originally developed by Dave Bargen - see below). 

All the software was written in 6800 assembly language and the assembled machine code was stored on a memory card stuffed full of EPROMs.  The source code listing, when printed out, stood as tall as one of those upright 2-inch tape containers we see above, and the ECS series was once described by a Motorola representative as being a 'monument to the 6800'.  It's hard to imagine in these days of multiple-core CPU equipped home computers, just how much could be achieved with so little CPU power (see also: Apollo moon landings for an even more extreme example).

Serious computing power

A different approach was taken by CMX Editing Systems, who used a DEC PDP-11 minicomputer as the main processing unit.  And whereas Convergence Corporation started at the 'industrial' (U-Matic) end of the market and then developed interfaces for higher-end VTRs, CMX began by focussing on 2-inch, then 1-inch VTRs, followed by other formats. 

CMX systems were extremely powerful for their time, incorporating distributed control elements called Intelligent Interfaces (I²) alongside the central processor - with a big price tag to match.  One of the key development team members at CMX, Dave Bargen, left the company to initially develop his own software, Super Edit, which included enhancements such as the 409 list cleaner and Trace - clever tools that allowed EDLs to be reworked and streamlined, before being auto-assembled to create a revised edit master in the most efficient manner possible.  Super Edit was initially designed to run on standard CMX hardware but Bargen's Interactive Systems Corp. (ISC) went on to develop and sell their own complete systems. 

ISC was acquired by Grass Valley Group in 1985, who further developed the product line into the successful VPE series.  Then in 1998 Editware acquired the product rights from Grass Valley's then parent, Tektronix.  Editware went on to evolve the product into the DPE range, with a hybrid approach allowing control of video servers alongside VTRs.  More details on the evolution from CMX, through ISC to Grass Valley and Editware, are provided in this article, marking the end of the production of this venerable 'family' of editing control systems.

Changed a few bytes, added some diodes - ended up looking like a hero...

This was a great era to be working as a field engineer.  Armed with an oscilloscope, soldering iron and EPROM programmer, it was possible to not only repair defective equipment at a customer site, but fixes for design or 'interfacing' problems could also often be effected in the field, by means of a hastily developed circuit board modification, or by changing the machine code or data tables.

This was so much more rewarding than the situation often faced by engineers these days, where a broken piece of hardware usually has to replaced in its entirety under RMA (when was the last time you saw someone actually remove the top cover from a rackmount unit?).  Similarly, software problems can typically only be corrected by gathering all the relevant log files, for a faraway development team to then analyse and (eventually) come up with a fix.  This is equally frustrating for the customer and the engineer!

Postroll

The independent editing controller manufacturers were very successful throughout the 1970s and 1980s.  However, in time Sony began to produce very good edit controllers of their own - quite sensibly, they combined the best features of all the independent suppliers' products into their evolving BVE series.  Being the (then) dominant player in the market, Sony were then able to throw in an editing controller for next to nothing (or for free as a system deal clincher) as part of a full post-production system quotation. 

This new competitive landscape made it very difficult for the smaller companies to survive, resulting in a good deal of consolidation in the late 1980s.  Convergence Corporation was sold to EECO, whose combined video products division was in turn acquired by Paltex (who already owned the Datatron and Cezar product lines).  Other independents were merged into larger conglomerates (e.g. CMX was acquired by Chyron Corporation and ISC became part of Grass Valley, then Editware, as mentioned above). 

Then companies such as Avid began making practical, affordable non-linear editing systems, but that's a topic for another day...

For more information on this subject, check out the Museum of Early Video Editing Equipment and Techniques, part of the excellent BBC VT Old Boys site:

Did you work with videotape editing controllers in the past?  Please use the comment form below to share your memories and experiences.

And if you'd like to be informed about future history articles and blog posts, use the signup form to subscribe to the WHATcast email newsletter.