If you think MIDI is a four letter word, you may never have discovered the sequencer. But in 1989 sequencers are arguably the most important part of any home studio. Paul White explains why.

FOR ONE REASON or another, sequencers have had a lot of bad press. This is partly because they often get used to make such god-awful records, and partly because they still suffer from the "mechanical" stigma earned by their less sophisticated ancestors. In their infancy, sequencers could only store a short sequence of notes, (hence the name), and these could then be played back with robotic monotony using whatever synths were then available. Early models were often restricted to working with monophonic synths (that is, only one note at a time, so no elaborate six-note chords) and often had a maximum capacity of 16 notes — you could get more feeling out of a musical doorbell!

Since that time — MIDI happened. Nowadays, even drummers have heard of MIDI, and next month PHAZE 1 will be going into more detail on what it can actually do for you. In advance of that, and for those who've still been untouched by progress, a few words of explanation are in order.

You probably know that the letters MIDI stand for Musical Instrument Digital Interface — but what exactly does that mean? Most recent synths and drum machines are digital, they are really dedicated microcomputers that interpret our musical ideas as electronically generated sounds. The keyboard synth, for example, has a keyboard that looks much like a piano, but in reality, the keys just control a bunch of switches and sensors to let the computer inside know what we are doing with our fingers. The computer reads the switch movements and turns them into a language that it can understand.

MIDI simply allows this electronic language to be transmited along wires from one instrument to the next. The simplest application is to connect two synths together so that you only have to play one keyboard to be able to use both lots of sounds. Just about anything you can press or twiddle on your synth (bar the mains switch) can be translated into MIDI language and sent from one instrument to another, including things like pitch bend, program change, keyboard velocity and so on. It all sounds very complicated, and so it is if you are the designer of a MIDI system, but fortunately, the user doesn't need to know how it works any more than he needs to know how a telephone exchange works in order to make a phone call.

So where does the sequencer come in? Imagine not sending MIDI information from the synth you are playing to another synth, but to a computer that can store it for later use. A computer that can store not just a few notes but a few thousand, or even tens of thousands — not necessarily with cold mechanical precision but with your own timing faithfully preserved. Sounds great doesn't it, a far cry from the old 16-note sequencer, but what does it offer that a simple tape recorder doesn't?

When you record MIDI information, you are not recording sounds, you are simply recording what your fingers did on the keyboard of your synth. When the sequencer is set to play back the recording, it actually plays your synth for you via the MIDI cable — you are hearing a live performance every time you play back your piece of music. This means that you can select a different voice on your synth so that, for example, the brass part you just recorded could be played back with a string sound or whatever sound you have at your disposal. You don't even have to play the tune back through the same synth — so long as it's got MIDI, you can use any synth or sampler you can lay your hands on.

The other major difference between MIDI recording and tape is that you can slow the song down or speed it up without the your system, manufacturers are making keyboardless synths available, better known as MIDI expanders. These are physically small and are considerably less expensive than a complete instrument.

This brings to light the first problem with MIDI — the more parts you write into your song, the more expanders, synths and samplers you'll need to play back the result.

HARDWARE

SO FAR WE'VE seen that the sequencer is some form of computer designed to store and regurgitate MIDI information. That would be of little use if it wasn't easy to operate or if it didn't allow the user to do the things he needs to do. For example, if you play a duff note, it would be nice to be able to change just that one note rather than play the whole part again. This is known as editing, and the sophistication of the system depends very largely on the sequencer in question.

If you flick through the pages of this or any other modern music magazine, you'll soon realise that there are two basic kinds of sequencer: the dedicated machine, like the Alesis MMT8, Yamaha QX series or the new Cheetah MQ8, and the software package that runs on the home computer — the C-Labs, Steinbergs and Intelligent Musics of this world. The self-contained, dedicated systems have appeal because everything is in one box and they are compact enough to use live. They score highly in the convenience stakes but they have their weaknesses too. For a start, the cheaper models may not have a built-in disk drive so you have to store your music on cassette rather like the old computer games. This can take a long time and isn't always reliable. Even if they do have a built-in floppy disk drive, you may find the lack of a computer screen makes them harder to operate as you can pitch changing — if you want to hear your song faster, you increase the tempo at the sequencer end and all that happens is that the MIDI data gets sent to your synth faster.

The best news is that, apart from being able to record polyphonic MIDI data, MIDI sequencers can handle several different sets of data at the same time, much like the separate tracks on a multitrack tape recorder. MIDI has 16 channels which means that up to 16 different synths can be controlled at the same time. Some sequencers, like Hybrid Arts' 'Edit Track' even allow multiple streams of MIDI data, so you can have up to 64 independent channels. There aren't many people with 64 MIDI synths lying around and most music uses fewer parts than this, but some modern synths are designed to play several sounds at the same time, using different MIDI channels. These are known as multi-timbral instruments and, because you only need one keyboard to run only see a small amount of information at any one time in the display window. Even so, there are some very friendly, dedicated sequencers on the market. The only real disadvantage left to whine about, is the fact that they only run their own sequencing software so you can't use them to play space invaders, do your accounts or edit your synth sounds.

The most serious computer based systems in the UK make use of the Atari 1040ST computer because it is a relatively affordable yet powerful computer that also has a built-in MIDI port so you don't need any external hardware. It also has a built-in disk drive, but the monitor screen has to be purchased separately which means finding around £500 for the computer/monitor system and anything from £30 to £500 for the software, depending on your requirements.

This may seem like a lot of money, but with the addition of a few synth modules and a good drum machine, you have the basis of a really serious MIDI studio. If you're into recording, the great thing about this kind of system is that the sound quality is limited only by the machine you use for your final mix, as all the synths are playing in real time. In effect, you're recording a live performance by an infallible musician, the computer. The same computer will also run voice editors for all the popular synths, games, business and word processing packages — the only disadvantage is that it is less portable than the dedicated sequencer. As a general rule, the computer based systems are more sophisticated, especially when it comes to editing, and the large screen area means that it's easier to keep track of what's going on.

FEEL AND FEATURES

COMING BACK TO the business of feel, sequencers do have to rationalise the timing of a performance to some extent in order to process the data, but on a good modern system, this is so slight as to be negligible. However, there are occasions when you want the timing to be rigidly accurate in which case you get the option to quantise your music. Quantising a bar of 4/4 music to 16ths ensures that all the notes end up accurate to the nearest quarter of a beat which is fine unless there's any really fast playing that needs a finer resolution. For most pop music 16ths is fine.

On the system I use, (C-Lab's Creator) you can really go to town in the cheating department (for cheating read editing). Once you've corrected any duff notes simply by finding the mistakes and changing the pitch or timing position, you can quantise the part to make it sound perfectly in time. If this is too perfect, you can then use a special randomising function to introduce tiny random timing errors to make the playing sound more human — but not as sloppy as you first played it.

I think by now you've got the general idea — these machines shouldn't really be called sequencers any more, they're really MIDI recorders with really useful editing facilities, although that doesn't roll off the tongue so well. And now you've got used to the idea that they can record and play back your performances, what else can they do?

For a start, if you have a drum machine, you could program it as normal, then connect it to your sequencer via MIDI to synchronize its tempo with that of your sequence. Alternatively, you could do as I do which is to use the drum machine as a box of MIDI drum sounds which can be played from the MIDI synth keyboard just as if they were notes on a piano. This makes it easy to tap out rhythms on the synth keyboard which are stored by the sequencer just like any other song parts. I find it easier to work in this way because you aren't limited to working with drum patterns of a fixed length, you can play along with your tune and do a whole verse at a time or a whole chorus which gives you a better, more natural feel.

Of course it would take a pretty good drummer to play the whole drum part perfectly all the way through from a keyboard, so the usual approach is to make the recording in layers. I tend to put down a hi-hat pattern first, then the bass and snare drums, then finally the toms. You can keep these on separate tracks on your sequencer or you can merge the data together onto one track to save space. You can even plug MIDI drum pads into the sequencer instead of a keyboard and let your drummer play the part more or less naturally.

You may remember that I said earlier that MIDI can cope with all sorts of control information like pitch bend or program changes. You can insert these simple commands into your compositions to change sounds at the appropriate times or you can even change the settings on your MIDI effects units — real automation for the price of a MIDI lead.

COMPOSING

MOST SEQUENCERS ALLOW you to compose on what is known as a pattern basis. In other words, you make up a series of musical sections and then join them together to form a song. Unlike tape, you only need to record each section once so if you have four identical verses, you could use the verse section four times. If you want to add subtle variations, you can do this too by recording additional parts and then using one or the other of them at different points in the song. An example might be a verse that has bass, drums, string pad and brass parts. You could also record a piano part which would be muted first time round, but second time through, you could mute the brass instead and let the piano take over.

Even a fairly complex song may only need a handful of patterns such as an intro, a verse, a chorus, a bridge section and an instrumental solo. And if you want to change the arrangement when the song is nearly finished, you don't have to go back to the start, you simply insert the extra verse or whatever it is you want and continue.

As well as being able to rearrange songs very quickly, sequencers allow you to transpose tracks or even whole patterns so you can easily do your Song for Europe complete with obligatory key change after the middle eight.

AT THE END OF THE DAY

AN ARTICLE LIKE this can only touch upon the versatility of the modern sequencer, but I hope that it's dispelled any doubts that the sequencer is a mechanical monster with no capacity for feel. Think of it instead as a MIDI multitrack recorder with magical editing capabilities. You can store several songs on one inexpensive floppy disk, and you have the option to hear your work played back on any MIDI synth you can beg borrow or steal — at any tempo and in any key you choose. If you are a virtuoso musician, you can preserve your original timing, or if, like me, you're a little more human, you can quickly quantise the lot before anyone else gets to hear it.

You can automate your MIDI effects and your synth's program changes, and with the addition of an inexpensive MIDI sync unit, you can run your sequence together with a multitrack tape recorder to give you a really powerful recording system. Even a simple four-track Portastudio leaves you three tracks for vocals and guitars after you've recorded the sync track and your sequencer can provide anything from eight to 64 tracks of MIDI instrumentation, depending on the model.

If you're a guitar player, sequencers give you a way to record and edit the music from an inexpensive MIDI guitar. These things are notoriously finicky and tend to miss the odd note or play an accidental you hadn't planned, but with a MIDI sequencer, you can clean up the performance in the edit mode until you're left with a perfect take.

MIDI sequencing is already an irreversible part of modern music production and now we've seen just what the sequencer can do, there's no going back. Technology is enabling us to do more things more easily all the time, and so long as we bear in mind that these machines are here to serve us and not the other way round, the future looks very bright indeed.