Magazine Archive

Home -> Magazines -> Issues -> Articles in this issue -> View

The Performing Art (Part 3)

A synth is only as good as its sounds - and that goes for any synth controller too - sax, violin, guitar... Ollie Crooke And Simon Thomas look at sound programming for MIDI controllers.


THE MOST POPULAR guitar synth of the moment is the Roland GR50 - if shop sales are anything to go by. The reason would seem to be the immediacy of the system and the quality of its internal sounds. Keyboards, too, sell on the quality of their presets, and Yamaha's WT11 wind sound module seems designed to cash in on wind players with an ear for a useful preset. It's hardly surprising that most people just want to be able to plug in and go.

But you discriminating MT readers are more interested in the potential for greater control and the ability to program unique and personal sounds. Aren't you? So what's the difference between a "playable" keyboard sound, a "playable" wind synth sound and a "playable" guitar synth sound? Anyone who's tried just plugging a wind or guitar controller into an expander will know inappropriate sounds can glitch.

While there's basically only one way of triggering a sound on a keyboard - pressing a key - guitar controllers can produce sounds from left or right hand techniques. The right hand plucks, while hammer-ons and slides can be independently produced with the right hand. On wind controllers a note can be triggered by a new outlet of breath or by a new fingering with the same stream of breath. This is also true for bowed instruments but not for percussion instruments.

These two types of triggering involve two different types of attack: normal and legato. With legato playing the normal attack portion of a note is not sounded and the second note is a continuation of the first at a different pitch.

This is where MIDI Mode 4 (Mono Mode) comes into the frame. It not only means that the sound source becomes monophonic, but that when two notes overlap, the attack portion of the second note will not be played. This allows legato playing on wind controllers and, with a separate sound source for each string in Mono Mode on a different MIDI channel, it allows hammer-ons and slides on stringed instruments (bowed or plucked). It also copes with slightly uncoordinated playing techniques. For example, a series of rapidly tongued notes on a wind controller will produce extra Note Ons if the Keys are pressed fractionally before or after the note is tongued. Mono Mode will make these events sound like a single note (as they do on a real saxophone).


APART FROM KEYBOARDS there are five basic types of MIDI controller available: wind controllers such as Yamaha's WX7 & WX11, Akai's EWI & EVI, the Synthophone and Casio's MIDI Horn; guitar controllers such as the Stepp, Synthaxe, Roland GR50, Yamaha G10, Casio PG380 & MG510, Wal's MB4 Bass and many others; pitch-to-MIDI converters such as Digigram's Midimic and various systems that blur into the guitar controller category; percussion instruments like the Simmons Portakit, SDX, SDS range and Silicon Mallet, Roland Octapad and Ddrums; hybrids such as the Zeta Systems MIDI violin, MIDI Chapman Stick and things such as MIDI noise gates which can be used to trigger MIDI from live performances.

Of all these controllers only the percussion category wouldn't benefit from the use of Mono Mode. Of course some cry out for it more than others - the Chapman Stick uses a two-handed tapping triggering technique which is more akin to keyboard playing than guitar playing, but you can slide up and down strings and Mono Mode makes that a more natural sounding legato process.

There are four main considerations that make programming for controllers different from programming for keyboards. The first of these is Mono Mode. With a wind controller the question is simple - does your sound source subport Mono mode? If it does then use it. (If it doesn't then ring the manufacturers and give them a hard time.) With a polyphonic instrument such as a guitar, each string will need a separate voice on a separate MIDI channel in Mono Mode. This usually demands a multitimbral instrument, but certain synths (such as the Matrix 1000) allow several voices to play on different channels in Mono Mode (but all with the same timbre). You should not, however, assume that just because a machine is multitimbral that it will go into Mono Mode. Korg's M1R and M3 don't, for instance, and very few samplers do.

Once you've got your synth or expander set up in Mono mode, the next thing to consider is the envelope shape. This applies only to sound-producing envelopes and not to sound-altering ones - on an FM synth it applies to carriers but not modulators. The aim here is to make the sound respond like a guitar string, a saxophone or whatever. That's not to say that we want it to sound like a guitar or sax but it must react so that the musician's playing style will produce the same sort of effects as on the 'real' instrument. You can overcome most problems in this field except for effects like guitar harmonics and dampened strings, which often trigger as open strings or as nothing at all.

As far as amplitude envelopes go, for both stringed and wind controllers you want a fast attack and instant release. Before all you wind controller players rise up in a frenzy of outraged letter writing, you then achieve a slow attack through breath control, with the strength of your breath determining the (MIDI) volume. The fast attack on the envelope is so that you can play staccato as well. The instant release may seem at first to be the wrong approach for guitar strings, but sounds have to be "dampable". The decay on a plucked string should only occur as long as the note is sustained - the volume of the sound should start to decay after a Note On and should continue to decay unless a Note Off is received.

The assumption here is that a sustained guitar string decays over about eight seconds at loud volume and about five or six seconds at low volume. Unfortunately, MIDI note lengths from guitar controllers will be about the same as for unamplified guitars, and so long sustained overdrive sounds will need to be extended with a MIDI sustain pedal.

An ADSR envelope should be almost "square" - A = 0, D = high, S = 0 and R = 0 for a guitar. Level/Rate and Level/Time envelopes can be set in a number of ways, but the initial and final rates are at their fastest, and the level before the final rate should be zero. With DX7 and TX802 sounds, Rate 2 should be in the 90s and Rate 3 should be between 18-30.


With wind controllers, A = 0, D = anything, S = max and R = 0. L/R and L/T envelopes need not necessarily be completely square, but this shape does mean that the envelope is entirely under the breath control of the player through MIDI volume.

The MIDI violin should have sounds with amplitude envelopes like those for a wind controller - high sustain, fast attack and release - when being bowed, but like a guitar with shortened decay when played pizzicato. The Silicon Mallet sends a short MIDI Note when struck (like a drum pad) but also has a sustain pedal for increasing the length of a note. There are therefore two approaches to creating envelopes for this instrument. The first is to use a long decay time so that, as soon as the Note On and Off have been sent, the note takes a long time to die away. This is the least satisfactory of the two because you have no way of varying the length of the sustain whilst playing. The other way, which Luis Borenius, who plays a Silicon Mallet with MIDI Chapman Stick player Jim Lampi, recommends, is the use of the sustain pedal that Simmons provide. The sounds can therefore be programmed with any desired sustain level and any release rate. Chapman stick envelopes should be like guitar envelopes, and Midimic envelopes should be like wind controllers.

Our sound source is in Mono Mode and has a volume envelope like the "real thing". The next thing to consider is velocity response.

With wind controllers there's not a lot of considering to do, really. The things you normally control with velocity on a keyboard, should all be taken care of with MIDI volume and breath control. The overall volume is controlled by MIDI volume, while breath control takes care of any timbral responses from simple brightening of the sound when blown harder to the introduction of complex new waves and even sample crossfades.

'Should' because, unfortunately, not all synthesisers allow dynamic MIDI control of parameters, and of those that do, a large proportion only allow the 'amount' of LFO to be varied. More on this later.

Suffice it to say that if you can't get the effect you want with breath control, you may have to resort to velocity response. This is a shame because it restricts you to using the initial strength of your attack to shape the timbral quality of the sound. If you can use breath control then it's a good idea to turn off any velocity response, because high initial velocities involve high Breath Control levels at the start of a note, and you may find yourself doubling the effect you wanted to create.

With guitars the question is more complicated. What you want from the velocity response of your sound will depend on what range of velocities your guitar controller generates. For instance, the G10 has a very wide range that it delivers in a pretty smooth curve - in fact you can define your own. With the G10 then, set the velocity response at maximum sensitivity and make adjustments to the guitar or its brain to suit the player and style - plectrum, strummed, finger-picked and so on. Maximum velocity response on the sound gives the added advantage of allowing expressive dynamics to be played and also of minimising the volume of damping glitches.

With the Wal MB4 there are a few more problems. The velocity sensing on the MB4 is more like a series of five or six discrete steps than a smooth curve. This means that a smooth playing curve on the natural sound of the instrument is represented as a coarse set of velocity responses. There are two discrete ways of dealing with this; the first, and simplest, is to switch off the velocity response of the instrument and play the line at a fixed volume (and therefore timbre as well) with possible recourse to using a foot controller both for MIDI volume and timbral control. The other way is to use the real-time filters on your sequencer to compress the MIDI signal so that the steps are close enough together for it not to sound odd with a very sensitive sound. This is usually a more satisfactory way around the problem than just making the sound less responsive. Obviously the way to find out the best velocity response for any sound on any controller is to experiment, but in general, volume should be more responsive than any timbral changes. The sound should become brighter as it's played louder, but consider the options of using a foot controller or overdubbing a modulation wheel part to effect any dramatic changes in the sound, as you can get a much more natural feel out of a pedal than you can from using velocity in an unfamiliar way.

This brings us neatly to our fourth and final consideration: the use of such realtime effects as breath control and foot controllers. Most keyboard players already use mod wheels and aftertouch; the difference between keyboard and sax players (and, to a lesser extent guitarists) is that they come to MIDI with the attitude that it's in the nature of their instruments to control the timbre of a note whilst it is being sounded. With keyboard players expression controls are a sort of added bonus that MIDI has provided, and are a much less integral part of playing technique.

Until recently it was unusual to find a synthesiser (and pretty nigh impossible to find a sampler) with any parameter other than an LFO (modulating either pitch or amplitude) that could be assigned to breath control or any other continuous controller. One way around this, if you have suitable sequencing software, is to use real-time MIDI mapping to translate breath control (or any other) information into System Exclusive messages. This means that you can edit one or more of the parameters on your synthesiser as you play. Thus if you're using Notator, you could map MIDI Controller 2's value to SysEx 67 16 0 19 plus 'Value' and you'd be editing the fine frequency on Operator 6 of a DX7 with its device number set to 1. Controller 1's value Mapped to SysEx 240 66 48 36 65 22 plus Value gives your mod wheel the power to edit the Intensity of a Korg M3R's Cutoff MG.

Luckily, more and more companies are implementing that old analogue synth capacity to edit various parameters in real time whilst you're playing, but instead of sticking a knob for each parameter on the front of the box they allow you to assign them a MIDI controller number which you can then play. Thus on the Yamaha TX16W you can not only go into Mono Mode but you can also set up a crossfade point on a continuous controller and slide from your Sweet Tenor sample into Raspy Sax halfway through a note and back out again, and be controlling it all from a footpedal. On a Korg M3R, Controllers 1 and 2 and aftertouch can be set to affect Pitch Modulation, VDF Cutoff and VDF modulation. On Yamaha's TX & DX range of FM synths you can use breath control (or any other control for that matter) to directly vary the volume of one of the operators - set the EG Bias of Breath Control to 99 and then adjust the "Ams" parameter on the Sensitivity page for the operator you want to affect. This works wonders on a modulator that is introducing a lot of brightness, edge or fuzziness to your sound - find out what each operator does to a sound by turning it off and seeing what it sounds like without it.


So there you have it - now there's no excuse for all you guitarists and sax players not to start programming your own sounds. At the very least you should modify all your favourite patches so that they react in a more natural way to your playing style. One of the most persistently annoying things about the pop charts in recent times is that everything sounds as if it's been programmed by a keyboard player. Whilst there are a lot of great keyboard players around, maybe it's time for bass players to play basslines again. And why not get a real horn player to play your synth brass lines or a guitarist to strum your Pan Flute pad? Until recently technology has been one excuse - MIDI controllers are an opportunity for technology to bring musicians back again.


IT'S TIME TO dish the dirt; which instruments support Mono Mode? The first thing to say is that things are getting worse rather than better. Mono Mode 4 seems to be something that manufacturers have decided isn't important any more. This is sad because there's huge potential for guitar and wind controllers, and the only way they're ever going to communicate in a glitch-free way over MIDI is by using Mono Mode 4. As controllers become more user-friendly and glitch-free, it seems short sighted not to make the most of them by including Mono Mode.

What's also quite depressing is the fact that it's incredibly difficult to find people, even in the technical departments of some pretty heavyweight manufacturers, who actually know what Mono Mode 4 is. Most people seem to think that multitimbrality is the same thing - if you could have six or more sounds on different MIDI channels. Oh, no.

Akai's VX600, which is a Matrix Analogue synth, goes into Mono Mode and has six voices so it's fine for MIDI guitars as well as wind controllers. Unfortunately, Akai haven't yet invested the time and effort needed to convert any of the software for their samplers so that they can run in Mono. They assure us that it's something that they're looking into, but it doesn't seem too high on the list of priorities, so as yet the S900, S950 and S1000 may not be top of your shopping list if you're after a sound module for your MIDI guitar.

Casio's VZ range of Phase Distortion synthesisers are both multitimbral and use Mono Mode, but their FZ range of samplers don't. One encouraging factor is that, being makers of guitar controllers and a MIDI horn, Casio have looked into the MIDI requirements of these instruments and, as they mention it in their sales literature, you can only hope that there's a section in all their manuals about the use of the VZ synths with MIDI controllers.

Unfortunately E-mu's man in England wasn't in England when we tried to check the official E-mu line on Mode 4. Ron Lebar at the Synthesiser Service Centre reckons the Emulator II, the Emax, the E-mu III and the Proteus all 'might' support Mono mode, and that the Proteus had the most advanced MIDI spec and so was the most likely. Still, we thought a lot of machines "might" go Mono...

Ensoniq are one of the few companies that seem to have a genuine commitment to Mono Mode 4 - their EPS and EPS-M samplers support Mono Mode, as do the highly praised VFX, SQ80 and ESQ1 synths. They were even eager to lend us a VFX so that we could try it out with various controllers but we didn't get it together in time for our deadline. Still, they deserve a pat on the back considering that they don't make any guitar or wind controllers.

Although Korg synths used to support Mono Mode, or at least the DW6000 that we used to have did, the M- and T-series do not. Using multi mode and some dynamically allocated sounds makes the M3R more friendly to non-keyboard controllers, but it's not a substitute for the real thing.

Oberheim's Xpander and Matrix 12 both support Mono Mode. The Oberheim sample player is a more complex matter - it will definitely play samples recorded for the Prophet 2000 and 2002 in Mono Mode if the sounds were in Mono on the Prophet. As to whether it's possible to put samples from other machines into Mono mode on the Oberheim we haven't been able to find out.

Roland's MKS series of analogue synths support Mono mode, and both the MKS70 and MKS50 allow for six independent MIDI channels. Unfortunately the S50, D50, U110 and all the other little S-, D- and U-series do not. The MR50 guitar controller plays its own sounds with a personalised set of extra messages that allow slurs and hammer-ons, but obviously this information is unique to this system and is not sent via MIDI.

The Sequential Prophet 2000 and 2002 samplers both support Mono mode and also allow "real time" crossfades controlled via MIDI Controller 1. With the right mapping capabilities on your sequencer or MIDI routing facilities, it is possible then to translate Breath Control or Foot Controller information into Controller 1. We couldn't find out about the Prophet VS but in an old review we discovered a couple of hints that Mono mode was supported, and as Sequential Circuits were such MIDI pioneers it would seem likely.

Until very recently there was nothing but nice things to be said about Yamaha - they produced the excellent TX802 with its capacity for eight Mono DX7 voices and several features specially included for non-keyboard MIDI devices. They also produced the TX81Z - a great budget module that also offered Mono Mode. The DX range supports Mono mode as does the TX16W which, with its v2 software, seems to be the only sampler in the universe that goes into Mono Mode. Nothing but praise then for Yamaha until the SY77 and TG55 came along. The boffins at Yamaha admit their new synths only partially support Mode 4.

So there it is - Mono mode is supported by many, though not all. MIDI instruments and it's an invaluable aid to getting not just playable, but exciting sounds out of controllers other than keyboards. It's up to those of us with a vested interest in retaining the playing techniques of non-keyboard instruments in the expanding world of MIDI.

Thanks are due to Ron Lebar (and Anton) at the Synthesiser Service Centre. If anyone feels they have any wisdom they would like to pool with the authors, we would be very happy to hear from you. Also, if you'd like any more information about anything you have read about in this series of articles, contact us.

Series - "The Performing Art"

This is the last part in this series. The first article in this series is:

All parts in this series:

Part 1 | Part 2 | Part 3 (Viewing)

More from these topics

Browse by Topic:



Previous Article in this issue

Voodoo Chile

Next article in this issue

Roland R8M

Music Technology - Copyright: Music Maker Publications (UK), Future Publishing.


Music Technology - Apr 1990





The Performing Art

Part 1 | Part 2 | Part 3 (Viewing)

Previous article in this issue:

> Voodoo Chile

Next article in this issue:

> Roland R8M

Help Support The Things You Love

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!

Donations for December 2021
Issues donated this month: 0

New issues that have been donated or scanned for us this month.

Funds donated this month: £4.00

All donations and support are gratefully appreciated - thank you.

Please Contribute to mu:zines by supplying magazines, scanning or donating funds. Thanks!

Monetary donations go towards site running costs, and the occasional coffee for me if there's anything left over!

Small Print

Terms of usePrivacy