Colin Cat tests the water from the great ocean of SMPTE/MIDI synchronisation, drops in the XR-300 and sees if it floats
Bear with me a moment. Mumbling on (as I will in a moment) about the Society of Motion Picture and Television Engineers, you might mistakenly think you had picked up a copy of Empire or What Film Speed incorporating My Aperture. But as I said, bear with me a moment.
You see, back in the sixties - when most people were listening to the bit in the middle of Sgt. Pepper or trying to levitate the Pentagon - the aforementioned society of engineers were wiping the audio soundtrack from their canisters of film and replacing it with a stream of binary code. Not the stuff of social revolution perhaps, but destined to become that rarest of commodities - a world standard. The binary format was named after the society - SMPTE coding.
Very simply, SMPTE codes run on the audio track for film or (more usually these days) video, counting the frames from one end of the tape to the other. Thus instead of taking the video out of the VCR to see it's about halfway through, you can consult the SMPTE track to find it is actually one hour, 20 minutes, 12 seconds and 14 frames through. This makes for far more accurate editing, soundtrack creation and so on.
The SMPTE system was so widely accepted that once music sequencing got past the dull and dismal Depeche Mode level, SMPTE was thought by many to be the obvious way to time-code musical operations in synchronisation. The emergence of MIDI gave huge versatility to a controlling system, and pretty soon equipment capable of linking MIDI with SMPTE started appearing.
Here is one such unit, the XR-300 Synchroniser from XRI Systems. It's a neatly styled 1U rack number, black with white markings. The display consists of easily read LEDs rather than those weedy angle-suffering LCD versions. The selector buttons are rather unusual cylindrical push switches, which are unfortunately just the sort of thing that might snap off when badly loaded in the back of a Transit van.
Reviewing a synchroniser such as this really comes down to sorting out a few simple questions. Does it work? How tolerant is the unit of bad signals? Is it easily operated by non-technical musicians?
Well straight off, the XR-300 passed the first Colin Cat review test - can he save time wiring it up by using the lead from the kettle? The answer is yes. (Never try that the other way round kiddies - kettles take more power than synchronisers and things may melt).
The second Colin Cat test is whether the unit is designed simply enough to operate without the instruction manual - after all, in a studio environment you never know who is about from one day to the next. Surprisingly for a unit as potentially technical as this, the basics were immediately obvious. No problems in striping or reading, no level hassles, quite delightful.
What is striping? Striping just means recording, the SMPTE society just use posh words sometimes, that's all. You use one track of a multitrack to store the SMPTE code, and then later the XR-300 reads it and pumps out MIDI messages to sequencer/drum machine et al.
There is a small problem here. The SMPTE code actually sounds like any high frequency binary signal, quite unpleasant and definitely intrusive. Using one track of a multitrack creates the danger of crosstalk onto adjacent tracks - this is a fact of life and not specific to the XR-300. In 48-track systems a blank track is left between code and music tracks but obviously in smaller systems this would be wasteful to say the least. It comes down to a compromise between crosstalk (keep the SMPTE quiet as possible) and problems with reading it back (keep the SMPTE as loud as possible).
There is also the choice between using an inside track (two channels subject to crosstalk) or an outside track (more susceptible to dropouts and damage). The XR-300 performed extremely well with low level SMPTE, even under DBX or Dolby B, C noise reduction, so that an outside track shouldn't cause any problems. (Unfortunately poor old Colin has not yet got a Dolby S system so couldn't check that.) The XR-300 picks up the SMPTE signals astoundingly quickly for someone, like myself, used to an FSK MIDI sync unit. Within five frames the unit locks on - not just at the beginning but even if you start the tape well into the code stream. Five frames is just a fifth of a second, that's pretty pronto.
You soon get used to watching the frames tick by, even though video is nowhere to be seen. In any case, the display is in hours, minutes, seconds and frames, with the frames moving too fast for casual counting. And chances are that you'll be watching your MIDI software most of the time, letting the XR-300 get on with the job unattended.
The XR-300 allows a choice of frames-per-second to be made, with four options relating to various TV and film standards. This includes the very silly 29.97 frames per second (American colour TV systems - don't you know!), while over here in Europe we prefer 25 fps. For music purposes it really doesn't matter, but 25 is probably the easiest number to calculate from - plus you'll be ready for the Channel 4 director that wants to use your last release for his documentary about Efe archers in the rain forest.
OK fine, so we can record and play SMPTE pretty effectively with this XR-300. How does it drive our sequencer and what options are there when it does?
First of all, you tell the XR-300 at what frame you want your sequencer to start, plus the tempo of the song and the number of bars that tempo runs through. The XR-300 then issues a MIDI start message at the appropriate point, followed up with clock signals at the prescribed tempo intermingled with song pointer messages to keep the bar count.
Now I know what you're thinking. Once you've programmed the whole song into the XR-300, you're tied to a rigid format because the timing code is recorded (sorry 'striped') on tape and mustn't be touched. Not the case at all. Unlike FSK synchronisers which just tape the MIDI information and do indeed have that problem, the SMPTE stripe is just the SMPTE stripe. The extra bits are software in the XR-300 and can be changed at any time (hallelujah and break open the Baileys). In fact changing time signatures by retiming individual bars on the XR-300 is much easier than doing it the proper way using some brand name sequencers I couldn't mention without being sued.
All the programming is performed using those white easy-to-snap buttons mentioned earlier. The use of increment/decrement keys slow things up slightly when you're talking hundreds of bars - a keypad would have been quicker.
A problem with many synchronisers is this. The synchroniser sends MIDI clock signals to the MIDI IN of the sequencer, which unfortunately means there is no input for recording a MIDI signal. This makes the whole thing rather pointless and usually necessitates a merge box.
The XR-300, on the other hand, has three separate MIDI OUTs and a built-in MIDI merge, adding an external input (such as your overdubbing keyboard) to the time codes. Hoorah (and another glass of Baileys).
There are always times when people get exhausted, drink too much Baileys and just lose concentration. Pressing the wrong button can be disastrous. The moment you accidentally wipe a bit of the SMPTE code ranks right along there with using the loo in your girlfriend's house then finding they're out of paper. Oh whoops, head against the wall time.
The XR-300 holds hope for such disasters. In copy mode it will listen to the SMPTE code presented at its input, clean it up and send it away as a new output. If the dropout in SMPTE code is reasonably short, when it resumes the two codes should lock back together with barely a slip. The mended code can be recorded on a separate track to the original and, if it helps, a combination bounce could be made.
Obviously there is a limit to the cleansing power of such a system - if you wipe the whole sync track you are, in a word, completely scuppered.
The only other points to note with this unit are the ability to dump each song's programming details as a systems exclusive, the back-up memory that will keep the current programme up to a month without mains, the unit's ability to generate MTC for those that want it, and the fact that playing Capital Radio into the XR-300 instead of SMPTE code produces some pretty amusing results on the display.
I looked hard for failing points on the XR-300, but apart from the aforementioned knobs there is nothing to criticise.
The price makes it attractive for both home and professional recording-type people, it beats the socks off FSK synchronisers and is stiff competition for some of the far more expensive units on the market.
Supplier: XRI Systems, (Contact Details)
Review by Colin Cat
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!