Of all the special effects that lurk in the racks of Britain's recording studios, reverb is the most mysterious. It's also the most useful.

DO YOUR DEMO TAPES lack the sparkle of the professional offerings you hear on the radio? Do you often wonder why your rough mixes somehow don't "cut it" when compared with the latest 12-inch dancefloor smash?

The reason could be that your sounds are wrong for your songs. It could be that your instruments are out of tune. It could even be that your voice is totally unsuitable for anything more artistic than flogging newspapers on street corners. But if you feel your performance is good and that you have a reasonable ear for musical balance, the answer is probably a simple missing ingredient: reverberation.

Before we go on to look at how reverberation ("reverb" to you and me) is created and how it's used, let's spend a moment or two examining its role in nature.

WHAT IT IS

AS YOU MAY already know, when a sound hits a solid object, some of it bounces back at you. Check this out for yourself by standing a foot or so in front of a brick wall with your eyes closed, and talking to yourself. Place a cushion or a sheet of foam rubber between your mouth and the wall, and the change in the tone of your voice is dramatic - when the cushion is in place, the sound seems dead and lifeless. You'll also get some mighty strange looks, so be sure to conduct this test in private if you're not to blow what little cred remains after the sub-Bros cover of 'Silent Night' your band crashed through at that Christmas Party last month...

We live in a world of both natural and manmade objects that reflect sound, so everything we hear is a mixture of sound coming directly from the source and reflections of that same sound bouncing off walls, cars, pavements, and tasteless repro furniture. Take away that reflected sound, and everything suddenly becomes unnaturally dull and unpleasant to listen to.

A single sound reflection from a distant obstruction is called an echo, and can easily be recognised as such - it's just a delayed repeat of the original sound. This is the effect you create by standing at the top of a mountain and yelling down at the valley below: as your words bounce off the valley sides, they are repeated with uncanny realism. But when a sound is produced in an enclosed space, such as a concert hall, the process is far more complicated. This is where reverb comes in.

The best way to describe what happens is to imagine a single short sound - a handclap, say - occurring somewhere in the hall. Sound travels at around 1100 feet per second (comfortably faster than a Ferrari) so before long, the handclap has reached the walls, floors and ceilings, and some of it is reflected back to the listener. You might expect several short echoes of different lengths, depending on the distances between the handclap and the reflecting objects. Which is true up to a point, but it doesn't stop there. Each of these short echoes acts as a new sound source - so a short time later, you have a whole new set of far more complicated echoes, but at a lower volume level because even hard walls tend to absorb quite a lot of sound. The same thing happens again and again, with the pattern of echoes getting ever more complex, until the sound finally dies away altogether as the last of the sound energy is absorbed by the walls and furniture.

To recap, then, our imaginary handclap is followed by a series of closely spaced echoes which rapidly build up in density and complexity, blending together to form a continuous, decaying sound which we call "reverberation ".

Depending on the materials from which the room is built and the furnishings within it, not all the sound is absorbed equally. What usually happens is that the high frequencies of the sound (the cutting edge of our handclap, for instance) are absorbed quite quickly, leaving a boomy, bass-heavy reverb. The harder the walls and the less soft furnishings, carpets and curtains, the brighter the reverb.

To check this out, try clapping your hands in a variety of different interior spaces. Concrete stairwells and tiled bathrooms should produce a bright sound, while soft, cosy bedrooms and lounges produce something much duller and mellower. Again, remember to do all this in private, unless you feel "come up and hear my reflections" is an appetising invitation to members of the opposite sex.

Why is all this relevant to music? A fair question, and one that needs to be addressed before we move on. Because we live in a world of reflected sound, our ears (and brains) have evolved to use this reverberant information to give us sub-conscious "clues" as to the nature of our surroundings. Take away the reverb, and we suffer a kind of sensory deprivation which makes the experience of listening thoroughly unpleasant. Reverb also adds a "stereo" dimension to sounds, because the sound reflections arriving at each ear are slightly different in both level and the time of arrival; this is due to the distance between a pair of human ears, and the fact that the shape of the human head obscures some of the sound.

Now, modern music is usually recorded in studios that are acoustically "dead" (ie. they are designed specifically not to introduce any reverb). Very directional microphones are used, and these are set up very close to the instruments and voices being recorded. This is done to keep the sounds as separate from each other as possible. After all, you don't want to have a lot of guitar in the drum microphones, or drums in the vocal mikes. The more "spill" you get between instruments, the harder things are when you want to process the sounds separately in the mix. The other effect of this so-called "close miking" is that what little natural reverberation exists in the studio is almost totally excluded, along with the unwanted spill.

Mix your song without reverb, and no matter how good the performance and the basic sounds, the result will be uncomfortably lifeless. Much the same applies to electronic sounds like drum machines and synths, most of which need reverberation to bring them alive.

HOW IT'S FAKED

ALTHOUGH SOME STUDIOS have specially constructed echo rooms or chambers, this approach isn't practical for the smaller studio, and is right out for home demos unless you have a large tiled bathroom. The way a live chamber is used is that the sound you want to treat is fed into a speaker positioned in the chamber, and picked up again by a microphone somewhere else in the room. You can optimise the effect by experimenting with the position of the speaker and the mike - but it's a time-consuming job, and it's all ruined if someone flushes the loo during a take.

Guitar amplifiers often include a spring reverb, which is used to create a sense of space around whatever instrument is plugged in. These are somewhat crude devices, and have only one basic sound - but they are relatively cheap to build into amps. The principle on which they operate is simple enough: a coiled metal spring is suspended between a kind of miniature speaker and a pickup. I say a "kind of" speaker because there is no cone as you would expect in a conventional speaker. The mechanical energy is used to send vibrations directly into the end of the spring; these travel to the other end where they are picked up a short time later. But some of the energy bounces backwards and forwards along the spring, creating a very dense - if unnaturally metallic - reverb effect. Unfortunately, percussive sounds tend to make the spring "twang" so they aren't totally successful for use with drums. But for vocals and guitars, springs can really help make a good demo.

The studio equivalent of the spring was the so-called plate reverb, which was similar in principle but used a large suspended steel plate instead of a spring. Once again, sound energy was directed into the plate by a "transducer", and the reverberant sound picked off the plate using contact microphones. To give the illusion of stereo, it was normal to use two contact mikes located in different positions so that the two outputs would be subtly different. If these outputs were positioned respectively left and right in the mix, a fairly realistic illusion of stereo width could be created. Additionally, the decay time (the time the reverb takes to die away) could be reduced by damping the plate with a felt pad, so it was possible to vary the effect from under a second to around five seconds, often by remote control using motorised dampers.

But both plates and springs have to be isolated very carefully from external vibrations, otherwise these end up in the music as well. So, not surprisingly, the world was very glad when the affordable digital reverb came along.

Unlike a digital delay or a tape-loop echo unit, which can only produce repeating echoes with a fixed time spacing, digital reverbs provide an accurate simulation of the way reverb density builds up in real life. It can take as much as three thousand different echoes each second to create this illusion, so the computer technology used has to be damned clever. Fortunately, it is. Recent advances in microchip technology mean that you can now buy a high-quality reverb unit for under a couple of hundred quid - less than a half-way decent spring might have cost you three years ago.

Apart from the obvious fact that the sound quality of these units (like the Alesis Microverb, for instance) is high, the immunity to vibration excellent, and the size diminutive, digital technology allows you to recreate several different types of stereo reverb with the same unit. Apart from offering different lengths of reverb time, from half-a-second or so up to cavernous effects lasting tens of seconds, they allow more subtle differences which actually create the illusion of different types of room or hall. This means you can choose any reverb - long or short, bright or dull - to suit your music. You can even get so-called gated reverb from most reverb units - in effect a solid burst of reverb around half-a-second long which cuts off very abruptly. Used on percussion instruments, this gives the familiar Phil Collins/Miami Vice drum sound.

HOW IT'S USED

LIKE ANY EFFECT, reverb can make things worse as well as better - depending, of course, on how you apply it. Quite often a short reverb is sufficient to bring a sound to life, yet still leave space for other instruments. In music, the spaces are often as important as the sounds themselves, so to fill all the gaps with reverb is obviously counter-productive.

In general, reverb is used little if at all on bass instruments or bass drums, because it tends to muddy up the bottom end of the mix without contributing anything artistic (though some recent hardcore dancefloor mixes have turned this rule on its head). Far better to create a contrast by leaving some sounds fairly "dry", and only using more of the reverb effect when you need it.

A reverb time of between one-and-a-half and three seconds is usually fine for vocals, while drums seldom require longer than a second-and-a-half unless you want to sound like early Pink Floyd. If you choose a bright reverb sound for the drums, it will help them cut through the mix more incisively - this is especially true of the snare.

The final decision on how to use reverb is yours. Just be careful not to be seduced by the convincing quality of modern reverb systems. It may initially have been created to replace the natural sound reflections that had been excluded by the recording process, but with the advent of digital units and their enormous flexibility, reverb is now the most important creative effect at your disposal. No question.