Home -> Magazines -> Issues -> Articles in this issue -> View
SMPTE Uncovered | |
Article from Home & Studio Recording, March 1986 |
A brief look at the basics of this important synchronisation system.
Almost everyone involved in recording will have heard of SMPTE code but how did it evolve and just what can it do?
In the early days of motion pictures it was discovered that films and soundtrack tapes could be made to run in sync by using the now familiar sprocket hole system by means of which the film and the magnetic tape could be mechanically advanced one frame at a time. The time code approach involves recording a signal onto one track of a multitrack tape so that any position on the tape can be accurately identified by reading the code.
The time code now accepted as the standard in the film and video industry is the SMPTE code. This conforms to a standard laid down by the Society of Motion Picture and Television Engineers (hence the name). This has also been adopted by the European Broadcast Union so if you see the letters EBU, you'll know what it means. Embedded in the code is information pertaining to hours, minutes, seconds and film frames as well as other less obvious functions to do with video work, and as the format may demand up to 30 frames a second (as in the American NTSC television standard), and as a block of code is recorded for each frame, the timing accuracy of the system is very good. Furthermore, if greater accuracy of resolution is required, it is possible to use conventional counter circuits in the reader system to register fractions of frames.
Before proceeding further it's interesting to look at the code structure, but for those readers who don't have a head for figures, you could safely give this next section a miss.
The code is digital and as such comprises a series of digital ones and noughts, though the implementation is not quite as you might imagine. Rather than a logical one being a pulse and a zero being a space, a one is recognised by a level transition occurring within a certain time window or bit cell. If no transition occurs, the status is read as zero and Figure 1 should help to clarify this where A is the time window or bit-cell stream and B is the time code itself. This system goes by the inauspicious title of Manchester Bi-phase Modulation. For a frame rate of 30fps, the bit cell rate is 1200Hz though the code would be twice this frequency if all the bits were ones.
Each word of SMPTE code comprises 80 bits and the function of these is made clear if you refer to Figure 2. In addition to the time and frame information previously mentioned, there are other pieces of information embedded in the code and the first is made necessary by the fact that not all video standards use the same frame rate; for example, the British system operates at 25 frames per second. To match up two different systems, it's necessary to drop or discard a frame every so often and the code contains what is known as a drop-frame flag (bit 10) which is enabled when set to a one.
Another bit is made necessary by the fact that colour video is transmitted in frame pairs and this is the colour-frame-bit which ensures that edits take place between frame pairs rather than splitting them, and this is bit 11. Admittedly, it's not much use for audio, but you never know when the video bug will bite.
So how does the SMPTE reader recognise where one chunk of code starts and another one stops? It's a good question. At the end of each block of code is a unique 16 bit word which cannot be duplicated by any other part of the code either in its standard or reversed form (as may be encountered when the tape is being wound backwards). As this appears at the end of the block of code, the time data is always displayed one frame late and so SMPTE readers generally have a plus-one function which automatically corrects for this.
In Figure 2, which shows a single SMPTE word, the bit groups one to eight are reserved for user-assignable functions and the four bits marked U are unassigned, though they may be used in future updates of the system.
"...it's only logical to combine both MIDI and SMPTE in a piece of equipment that can sync tape position to a particular place in a musical score and keep sequencers running in sync with the recording."
To recap then, SMPTE code is a stream of code arranged in separate blocks or words, each containing 26 time code address bits, two assigned bits (drop frame and colour frame) 16 sync word bits, four unused bits and 32 user-assignable bits. These may carry information that help to identify the work or anything else the user dreams up, but in practice they are seldom used. Up to thirty of these SMPTE words may be recorded onto tape each second and each word carries all the information required to identify that particular section of tape to the nearest frame. The time part of the code may be set to start from zero at the start of work or to local time, the latter being preferred by the film industry.
Of course the code doesn't appear by magic; it must be generated and it must be read by something. The generator creates the code which is recorded onto one track of the tape machine ready for use. This may be a separate unit or part of a combined generator/reader system and is of course built around a micro-processor. Only now are specially designed ICs making cheap SMPTE systems possible and because of this, SMPTE is now a practical proposition for use in music only systems where its features have a distinct advantage over simple click track type codes. Apart from the operational advantages, the fact that something will actually be standard is more than welcome and, as music and video are now converging fields, this has got to be good news.
The reader on the other hand interprets the off-tape signal and displays the code and maybe passes on a buffered and cleaned up version of the code onto another machine. When copying tapes where the SMPTE code needs to be preserved, it's unreliable to make a straight tape to tape copy because of the deterioration in the code caused by the copying process. The code is processed in the reader so that a regenerated version may be recorded onto the second machine so the copy is always as good as the original. As the codes on the two pieces of tape being synchronised may not start at the same time, the reader has an offset facility so that the desired synchronisation can be achieved regardless.
"What started out as a highly sophisticated and expensive system for use in the high budget movie industry looks set to become the norm for even modest audio and video recording set-ups..."
Even if you're not involved in video production work, SMPTE can be a valuable tool and the most common application is to sync two or more recorders together, an example being the established Q-Lock system which can control up to three tape machines simultaneously. To accomplish this, the code must be recorded onto a spare track of each machine, starting about ten seconds or more before the point on the tape where recording is due to start so that the units have time to sync up properly. One machine then acts as master with the others being slaves. The slave machines are servo locked to the master and if the codes start to drift out of sync, a control signal is generated which speeds up or slows down the offending machine. In most professional applications, the slaves will follow the master in play, record and fast wind modes and it's possible to program in an offset or delay between the machines if required to correct for any timing problems that might exist between one tape and the next.
As well as locking machines together, it is possible to make very flexible auto-locaters using SMPTE reading circuitry and a sophisticated model may memorise and execute several drop-ins automatically. Because the code is a physical part of the tape, the drop-in point is always accurate and there are no tape slip problems to worry about.
With the MIDI system being well established as the standard for controlling synthesisers and drum machines, it's only logical to combine both MIDI and SMPTE in a piece of equipment that can sync tape position to a particular place in a musical score and keep sequencers running in sync with the recording. In fact Roland and Fostex have already implemented such systems and there are sure to be many more in the pipeline. What started out as a highly sophisticated and expensive system for use in the high budget movie industry looks set to become the norm for even modest audio and video recording set-ups in the very near future. We'll be looking at the new Fostex system as soon as we can get our hands on it so watch this space!
Technically Speaking |
Time For SMPTE (Part 1) |
Using Timecodes - An Introduction To Timecode Synchronisation (Part 1) |
Short Circuit - Time Machine Revisited |
MIDI Matters - Song Position Pointers (Part 1) |
Rock Around The Clocks - Synchronisation Explained |
Adrift On An MTC - MIDI Time Code |
That Syncing Feeling (Part 1) |
15 Fab Software Tips |
Getting into Video (Part 1) |
When Is Sync |
The Sync-ing Feeling |
Browse by Topic:
Feature by Paul White
Previous article in this issue:
Next article in this issue:
mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.
If you value this resource, you can support this project - it really helps!
New issues that have been donated or scanned for us this month.
All donations and support are gratefully appreciated - thank you.
Do you have any of these magazine issues?
If so, and you can donate, lend or scan them to help complete our archive, please get in touch via the Contribute page - thanks!