A Guide to Digital Reverbs, Delays and Multi-Effects Units
Julian Colbeck takes a look at what's on offer in the world of digital reverbs, delays and multi-effects units, and explains how to choose between them.
Digital effects are everywhere — in our racks, built into our synths and samplers, and in digital recording systems. Julian Colbeck takes a look at what's on offer in the world of digital effects units, and explains how to choose between them.
Five years ago, during a visit by foreign journalists to Roland in Japan, company president Kakehashi claimed that DSP would be the "next big thing". At the time we thought he was simply talking about Roland's then forthcoming DSP2000, an undoubtedly clever box of tricks that sought to pep up recorded music through the use of delays, phase, and presence controls.
This outrider for Roland's recent RSS system hardly set the world alight (conservative old lot we are at heart), but of course Kakehashi was speaking of a wider picture, where Digital Signal Processing would one day find its way not only into dedicated units, but into synths, samplers, new concept recorders and the like. And that day has been upon us for a while.
Today, not only is there a massive number of dedicated 'effects' units to choose from — designed to suit all instrument types, applications, and budgets — but no manufacturer would dare release a synth without a comprehensive effects section, and even samplers and home studio recorders are getting in on the act. There are those who find this a most unhealthy exercise in papering over the cracks, helping to gloss over our collective lack of musical and technical ability. This is hard to argue with. All effects, in anybody's hands, are essentially used to tart things up. It is just unfortunate that the lower down the musical food chain you travel, the greater their power of enhancement. Conversely, up at the top of the musical tree effects rarely spread beyond the importance of icing As always, however, manufacturers in their infinite capacity to overestimate continue to offer effects in ever more complex and far-reaching packages. But speak to even some of the world's top engineers and they'll tell you that, yes, they'll whip out their ready reckoner delay time-to-tempi chart, fiddle about with reverb lengths, and even use some EQ if pushed. But that's about it.
Real-time MIDI control over room size? Density of room coverings? Precise room shapes? Endless tinkerings over whether the reverb should be flanged or whether the flanger should be reverbed? Interesting, fun, challenging things all, but back in the real world we have a track to mix.
Korg it was, back in the pre-M1 days when they were about as hip as Wersi — no disrespect intended — who first introduced the notion of synths with built-in effects units. The lucky fella was the DW8000 with its smattering of DDL parameters, and the year was 1985. Built-in effects came of age with the Roland D50, took off completely with the M1, and are now firmly established as an integral part of, and major selling point within, all instruments from a £200 kid's keyboard to a Kurzweil.
For the synthesist, as has been the case for so many guitarists since Hendrix ate his wah wah pedal back in the '60s, it is now almost impossible to separate sound from effect. The sound is the effect. This situation presented no real problem around the time of the D50 (apart from remembering how to switch the effects off, the request of most every engineer I encountered at the time!) because instruments were generally only capable of generating one sound at a time: program your sound, add your effects, play it.
Problems began to occur, however, when multi-timbralism reared its seemingly attractive head on instruments like the M1. You program your sound, you add your effects, you play it. Wonderful. Now you group together a whole bunch of wonderful sounds to be triggered by your sequencer, and... Goddammit... where are my effects?
The problem is soluble by providing instruments with enough DSP chips to complement their multi-timbralism. But that's expensive, and even big boys such as the JD800 don't do that. In fact the JD is, if anything, worse than normal because in multi mode you don't even have the same number of effects at all, never mind per multi part.
However, while built-in effects are undeniably seductive, they are not the answer to all the world's problems. A dedicated effects unit is still a vitally important purchase for the serious player. But which type of effect? And what model?
The effects units in this survey comprise three basic types: dedicated reverb units, in other words units which only offer reverb effects; dedicated DDLs, in other words units that only offer discernably time-based delay effects; and multi-effects units. Whereas the first two types are essentially pieces of studio equipment, now assimilated into the musician's grab bag thanks to technology's continued ability to drive prices down, the comparatively recent creature known as a multi-effects unit — offering as many effects as can reasonably be squeezed into one box — has always been aimed fairly and squarely at the musician rather than professional recording studio, and whereas dedicated reverbs and DDLs are designed to appeal to all, multi-effects units are often tailored to suit a more specific application or musician (ie. general/home studio, guitar, or bass).
Guitar and bass-inclined effects units are almost always of the multi-effect type because in addition to a range of digital effects they offer analogue pre-amp facilities like compression, EQ, filtering, or speaker simulation. Such units are not only applicable to guitarists of course. As many a seasoned guitar sound imitator will testify, all you need to produce a convincing guitar sound from a synth or sampler is a reasonably attacky, hard synth patch cross-faded with a sine wave (to simulate feedback), and a guitar effects processor to scrunch the whole thing up a bit. Plus the nouse and musicianship to be able to play lines like a guitarist, of course.
This is a hard one to justify if you are simply an instrumentalist looking for something to spice up your playing. Where a dedicated reverb might start to make sense is if you are a vocalist, and/or have, or are wishing to inaugurate, a reasonably advanced home studio.
By their very nature reverbs are seldom as dramatic as other effects. But the range of quality is vast, and the difference between a cheap and an expensive reverb should be clearly discernible by all but the clinically deaf. The difference lies in processing power: the ability to manipulate data smoothly, without undue modulation or signal degradation. And this will invariably cost money.
High quality reverb can spread across an entire recording a sheen that says 'this is expensive'. Reverb gives life. It can make otherwise clinical, 'synthesized' music sound like it is being performed by real people in a real environment. Although a dedicated 'verb unit will primarily offer reverbs, of varying length, colour, and style (gated, reverse), you can generally expect to find some other 'effects' tucked away in the background. These will normally comprise EQ (of vital importance to the feel of your reverb), plus a range of limited delays and chorus effects. What you will probably not find on this type of unit, however, is the ability to pile one effect on top of another. For this you will need a multi-effects device.
Although it is tempting to quote specifications, or get carried away by the offer of a thousand and one room size/shape/coverings parameters, numbers of controllable early reflections and the like, the most telling judge of digital reverb units are your ears: does the reverb sound natural, or electronic? Leaving aside the fact that there will be times when you actually want an electronic, synthetic effect, the most important variables in reverb are simply size and amount — ie. where do you want the listener to imagine this music is being performed? And no matter where you decide this should be — in a room, a small hall, a cathedral — the more natural the effect, the better. Harshness, modulation, tinniness, are all to be avoided. Smoothness, length without breakup, and transparency are properties to be sought after.
A Digital Delay Line, as the name implies, is an overtly time and movement based effect (so are reverbs, in fact, but they work to recreate the sound of restricted spaces where reflections don't have time to be heard so clearly), whose remit can extend from plain and simple echo effects to chorussing and pitch shifting.
Perhaps because the basic effect has less to be compared with in real life (you're more likely to encounter halls and cathedrals than canyons. And chorussing?), DDLs are less dependent upon sounding natural. Where cheap DDLs come a cropper is noise — more hiss than chorus, which I'm sure everyone has suffered at some time or another.
In recent years, and in spite of the odd piece of token rearguard action in favour of single effect units, dedicated DDLs have tended to be superceded by multi-effects processors. But the case for a dedicated unit has never been put better than by Eventide with their H3000SE Harmonizer, a name that conjures up rarely used vocal harmonising effects but that in fact describes one of the most powerful sound enhancing devices used in studios over the past few years. Here again, though, you pays your money for such quality. The H3000SE costs considerably more than most synths, so you'd expect it to sound good.
Here, specifications can paint a more telling picture, not only in terms of basic processing power (as always, the higher the number the better) but in terms of the length of delays a unit is capable of delivering. Provided the basic quality is good, ie. 16-bit D-to-A and with a good frequency response that extends up to 16kHz or beyond, you'll pay more for models that offer longer delays. And unless you'd be satisfied just using chorus effects, you will need as much flexibility as you can afford so that you can, for example, synchronise delays to the tempo of your part or song.
"Clearly the multi-effects processor offers a good deal, in both senses. But whether a single unit can satisfy your every processing need is another matter.
Long delays can effectively become samples, so if this is the type of use you have in mind, maximum delay time is the parameter to look out for. Sometimes true sampling is offered too. Although AMS are rightly famous for offering such a facility, especially with the keyboard interface on the SDMX, on less sophisticated DDLs or multi-effects you may find the restrictions and limitations limit usefulness to a significant degree. Check this out beforehand if sampling appears an eyecatching feature for you.
In terms of short delays, modulation or LFO is something to look for in order to breathe life into chorusing or flanging effects.
A final physical feature to make note of concerns inputs. Do you want, or really need, to be able to create what are called true stereo effects? Frankly a lot of nonsense is talked about this subject, with people claiming that an effect is useless it actually takes a stereo signal, processes the channels separately, and kicks the signal back out in stereo. Sure, there are times when this is precisely the effect you must have, but it is worth nothing that one of the all-time great professional delay units, TC Electronics' 2290, only uses a mono input.
Do I really need to put the case for multi-effects units? Surely the prices and features — little of one and a lot of the other — tend to make eulogies redundant. Presumably you'll know if you need a guitar or bass-orientated effects unit. Chances are you'll have those funny-shaped bits of wood with strings on lying around the house. However just because you may play funny shaped bits of wood with strings on doesn't mean that you won't be better off with a more 'general' effects unit that can suit your home recording requirements, or help out with occasional flights of keyboard playing/sequencing as well.
Clearly the multi-effects processor offers a good deal, in both senses. But whether a single unit can satisfy your every processing need, as the Quadraverb GT would seem to do for instance, is another matter.
The key to the relative merits of multi-effects units is their limitations. You read the spec sheet, and it seems you can utilise every effect known to man, simultaneously, and at full bandwidth. Back home, you discover that there are only so many 'algorithms', ie. groupings of certain effects (chorus + reverb + EQ, or delay + reverb + compression), and the particular combination of effects that you want ain't one of them. Or perhaps there is an algorithm that has the effects you want, but they're in the wrong order. You want a reverb that has chorus applied to it, and the closest you can get is chorus with reverb to follow.
Feature by Julian Colbeck
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!