Magazine Archive

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

Art Class

ART Multiverb Alpha 2.0 & DRX 2100

Article from Sound On Sound, December 1992

ART's new Alpha 2.0 adds new features to the well-established Multiverb Alpha multi-effects unit, while the DRX2100 adds analogue processing facilities as well. Dave Lockwood finds that ART's policy of constant updates pays off.

The Multiverb Alpha 2.0 from ART (Applied Research and Technology) is in fact not the major revision that the retitling might imply, but more a series of subtle enhancements. Whether any of these alone, or indeed all of them together, might be enough to make an 'original Alpha' owner want to upgrade, I rather doubt, but taken as a whole they do represent a significant degree of added value to the package. Reviewed alongside the Alpha 2.0 this month is the DRX2100, a similar model in many respects, incorporating all of the Alpha algorithms, but with an additional range of analogue processes — a compressor, an expander, a noise gate, and an interesting new 'Digital Contour' (effect/dry-feed crossover function). The original 7-band EQ stage has been enhanced with an effective 'graphic' display of gain within each band, and incorporated within both models there is now an auto-ranging guitar/bass tuner.


Both units retain the 'alpha dial'-style 'multi-function rotary encoder', for data entry, that presumably gave the Alpha variant of the Multiverb series its name. Both the Alpha 2.0 and the DRX2100 are based on the same 24-bit ASIC processor, allowing high bandwidth, and conspicuously clean yet highly complex effects combinations. The two processes which make the heaviest number-crunching demands, pitch transposition and reverb, could not be used together in any of the earlier 20-bit Multiverb models, but could for the first time in the Alpha. In the Alpha 2.0, and naturally therefore the DRX2100, we finally have a stereo (2-channel to be strictly correct) pitch shifter — at last, simultaneous up and down micro-shifts on an ART!

With the inclusion of analogue effects, the DRX2100 allows combinations of up to nine simultaneous processes, whilst seven seemed to be the maximum I could cram into an Alpha 2.0 preset. Both, however, offer the same 20kHz audio bandwidth (44.1kHz sampling) and highly respectable nominal signal-to-noise ratio in excess of 90dB.

In common with just about all recent ARTs, the Alpha 2.0 and DRX2100 offer 200 memory locations, with the first 115 slots initially occupied by the factory programs. Presets are protected to prevent accidental overwriting, but can be 'unlocked' once you run out of 'blank' locations. Factory Presets remain in ROM, however, and can therefore be recalled individually if you find you have ditched something valuable in favour of one of your own creations. There is also a global recall function for restoring the factory programs en masse — but this, unfortunately, resets all the user-locations at the same time. A Preset Copy facility allows you to organize your patches into a logical sequence for live work, or move a factory patch into another location to use it as the basis for an edit.


Housed in the now familiar 1U, predominantly black rack case, with integral PSU (no power switch!), and dodgy paint job, all audio and external control connections are made to the rear panel. Unbalanced 1/4" jacks (1MOhm input impedance, 1kOhm output impedance), handle both input and output, with the 'dry' signal remaining in stereo through to the output, whilst the processor input is derived from the sum of the two channels. Separate MIDI In, Out and Thru connections are featured, plus two assignable footswitch jacks (momentary, ie. non-latching) — the principal use for these seems likely to be remote selection of presets, in conjunction with the user-programmable MIDI Patch Table. Finally, there is the unusual inclusion of a 9V DC output — although ideal for powering any 9V pedals you might also be using, I suspect the primary reason for its inclusion is as the power source for an X15 Ultrafoot MIDI remote control footswitch, should one be used in the system.

Two short-throw faders, mounted horizontally on the front panel, are used to control input and output levels, monitored by three LEDs: green for signal present, yellow for nominal level, and red for clipping. Centre detents indicate unity gain at the nominal (+4dBu) line level, with enough gain available within the system to cope with varying interface requirements, including direct connection of relatively low-level instruments; but definitely not microphones, nor passive guitars — despite the analogue processes, the DRX, unlike ART's SGE model, should certainly not be viewed as a self-contained guitar pre-amp.

There is no front-panel 'mix' control; rather the output mix is a programmable parameter which can be customised for every preset. This is fine for everybody except those who need to use their processors both in the studio and on-stage. A global parameter for switching en masse between the individual programmed mix values needed for use with an instrument amplifier and the 100% 'wet' set-up appropriate for use with an aux send/return loop on a desk would surely be a most welcome addition. The primary information display is the obligatory 2-line LCD window, augmented by a 3-figure LED display, showing preset numbers and memory locations.


Beginning with the analogue stages unique to the DRX2100, first up is a useful little compressor. The three principal parameters are Slope (ratio), offering 1:1, 2:1, 3:1, 4:1, 10:1, Limit, and two 'Reverse' settings, and the Attack and Release controls. With just seven settings to cover 2ms to 50ms (attack), and 200ms to 2s (release), there is always going to be some measure of compromise involved in setting up a precise compression envelope — particularly with a short-term signal like a snare drum, the attack time can be critical to within a few ms. More resolution here would certainly be welcome. The programmed values, however, appear to be starting points, for both attack and release seem, to some extent, to be program dependent. Personally, I would prefer the option of being able to defeat the 'auto response' element, for those occasions when am specifically seeking compression side-effects. Generally it is useful however, making the compressor a little easier to setup in most applications.


One thing you will notice when you switch in the compressor stage is an increase in noise level. This is to be expected; all compressors, by the very nature of the function they perform, will accentuate source noise, ie. noise present at the input. This compressor also seems to contribute a little more of its own than perhaps it should, but to combat this the DRX2100 offers a complementary Expander stage. Sharing the same envelope values as the compressor, the Expander offers merely Range (acting as Threshold) and Ratio controls, but works highly effectively nonetheless. Naturally, a low ratio (<3:1) slope will leave the dynamics largely intact, whilst hard 'gating' will be provided at higher settings. Using the Expander for this is probably fairly pointless, however, as there is a separate Gate process, triggered directly from the input signal (ie. independent of any reduction in dynamic range performed by the compressor). With just a basic Range setting to play with, the Gate can only be expected to work faultlessly on signals where a simple threshold action is appropriate, and is logically placed last in the analogue process chain.

Another process which inevitably contributes to HF noise is the Harmonic Exciter. The choice of two process frequencies, 2.5kHz or 5kHz, is actually a fair compromise — anything lower into the midrange tends to sound extremely harsh, and all frequencies above the nominal point are lifted anyway. In the context of a multi-effects patch you can certainly get what you want out of this, and I welcome its inclusion — it often seems to be the factor that makes the difference between an OK effect and a really stunning one.


This new addition is audibly very subtle, yet demonstrably effective in some situations. In essence it is a frequency dividing network, but it offers more than just a simple high/low frequency split. As well as a choice of seven 'crossover' frequencies you have the option to send just the high-band to the digital effects stage, whilst maintaining a full bandwidth signal through the analogue chain, or alternatively to send the lower band only to the analogue bus, whilst the digital stage gets a full bandwidth signal. The two signal paths are subsequently recombined at the output mixer, which allows for independent level settings for Dry signal, Analogue FX stage output, and Digital FX stage output (in addition, the DRX2100 now has an overall level parameter for each preset, allowing a series of presets to be matched in level for live performance perhaps, without needing to alter their internal balance parameters individually — a very worthwhile inclusion).

Naturally, the first application that springs to mind is high-pass filtering the digital FX feed when using reverb. The audible effect is much cleaner, particularly with a dense program, or a long decay time; bright reverbs seem to 'sparkle' a little more, and even the warmest of programs will stand out from the dry signal a little better when pre-filtered in this way. The effect is subjectively more pleasing than reducing the decay time at LF, or merely filtering the returns (whether this is done within the algorithm, as some units allow, or externally). The more complex the chain of effects, the greater the subjective benefit of using the crossover seems to be, particularly when modulation effects (chorus, phaser etc.) are combined with reverbs and delays. The chosen crossover frequency options cover the entire range that would seem to make any kind of audible sense; 110Hz, 160Hz, 230Hz, 350Hz, 560Hz, 1kHz, and 2kHz.


Both the Alpha 2.0 and the DRX2100 incorporate this one, and very useful it could prove to be, both on-stage and in the studio. The tuner can be simply recalled as a preset, whenever it is needed, and allows either silent (output muted) or audible (normal signal pass) tuning. In addition, a reference oscillator can be routed to the output. The tuner must be selected to either Bass or Guitar ranges, and you then have a choice of individually selecting each note to be tuned, or activating the Auto option. Auto mode however, requires you to be roughly within a semitone above or below the correct pitch or you will be outside the detector window. Nevertheless this should be ideal for the majority of minor corrective (ie. not tuning-up a brand new string) tuning requirements. Sensibly, the bass range incorporates a low B (31 Hz) to cater for the increasingly popular 5-string bass format. Unlike many tuners, this one seems to give the most reliable reading on open strings, rather than harmonics, but it is certainly stable and, as something of a bonus inclusion, I really could not fault it in use.

The display format makes the best of the limited space available to it, with indications for out-of-range sharp or flat, and subsequently the precise number of cents sharp or flat as you come within a semitone. Finally, correct pitch is indicated both by the cents figure and a specific 'in-tune' marker. Perhaps the only criticism I could makers simply that a 2-line LCD could not be said to be ideal for an onstage tuner, if you were designing one from scratch, being sometimes more than a little difficult to read under stage lighting conditions.


The 7-band equaliser, featured in both units, is much enhanced by its new graphic display, making it much easier to visualise the overall contour you are applying. 15dB of boost or cut is available at 'stretched octave' intervals of 40Hz, 100Hz, 250Hz, 640Hz, 1.6kHz, 4kHz and 10kHz. With only 12 steps to cover the whole plus and minus range, Gain is not in individual dB increments however, but the 2, 4, 6, 9, 12 and 15dB settings will undoubtedly prove adequate in any likely application of this EQ. There is no bandwidth, or 'Q', parameter; the compromise Q chosen is subjectively broad enough to avoid the coloured sound of high Q filters, whilst still allowing the equaliser to be reasonably selective in creative work.

There is a second EQ stage in both these units; first seen in the Alpha Mark 1, the Acoustic Environment Simulator (EAS) is again present. This provides a range of preset EQ curves, simulating varying amounts of high-frequency absorbent or reflective material in the soundfield. Despite highly specific and evocative names such as 'Wood + Tile', and 'Stone Ceiling', which lead to my cynically dismissive remarks in my review of the original Alpha, I actually appreciate this facility very much. Presumably comb-filter based, it certainly provides a level of complexity beyond just simple frequency-amplitude EQ, but I still contend that it falls some way short of the claim of 'simulating real-world acoustics', made on its behalf. What it is very good at is 'pre-voicing' a signal prior to FX treatment; warming it up, deadening it, or adding a bit of natural presence — alterations made to a signal via the EAS sound like they belong to that signal, rather than being imposed upon it. Quite an achievement really.

Completing the EQ group there is the humble Low Pass Filter. This is active only on the feed to the digital F/X stage, and its range of 30 different roll-off frequencies between 630Hz and 17.8kHz, allows the filter to applied to an exact degree, to smooth or soften the feed into an effect, rather than acting on its output (which produces an audibly different result).


Newest member of the modulation group is a Phaser, an effect which appears to be making something of a revival — you couldn't give away a phaser a few years ago! This is the classic swept notch-filter effect, with adjustments for Speed (0.04Hz to 27.3Hz) and Width (there is also a Depth parameter but that seems to be actually merely the level of this effect within the preset), plus a Regeneration facility with which to go totally over-the-top. There is a preset that sounds exactly like my classic old MXR Phase 100 (which I'm now glad I've still got, albeit only because I couldn't give it away!). All the modulation FX can be pre or post any delay/reverb effects that may be also in the chain.

The Flanger similarly offers Speed, Width and Regeneration parameters, but is based on a swept short delay time, generating more complex comb-filtering, rather than a swept preset notch characteristic. ART'S flanging programs are always strong, with the ability to get quite a decent 'flange' out of sources that you might normally expect to be too short-term, or lacking sufficient HF content. Phase inversion of one or both flanger outputs is now possible, making for a more dramatic effect when operating in stereo (although of course it simply cancels in mono).

In the Chorus algorithm Regeneration is replaced by a Delay setting for determining the actual base delay time — using a longer base delay with less width and a slower modulation produces more of a doubling effect. The single modulation source and waveform is the primary limitation to achieving sufficient richness without excessive pitch variation — if a second modulator was out of the question, then inverse modulation for the other side of the stereo might have been a good compromise.

An Autopanner is included, with just two parameters; Modulation, controlling the width of the pan, and Speed (0.04Hz to 27.3Hz). More interesting perhaps is MIDI Pan, allowing stereo position to be governed in real-time from any available source of MIDI Controller Change information. The Tremolo algorithm (amplitude modulation), offers the same pair of parameters, allowing a very realistic simulation of the old guitar amplifier effect — I can't think what else you might use it for though.

All in all, the modulation group of effects on this ART series certainly stand comparison with their counterparts on any comparable units that I've used.


Whilst retaining the original single-shift option, ART have at last added a dual-shift option, allowing either two separate harmonies, or far more likely, simultaneous up and down microshifts. Rather than define one shifted output as left and the other as right, ART have created a Position parameter — shifted notes can be placed anywhere across the stereo picture, with 63 steps to play with. This is almost overkill; personally I would happily settle for left, half-left and centre, and the same for the other side. It seems all the more like overkill on one of the less important parameters when you find that the Fine shift parameter is still dealing in six cent increments. I find this simply too coarse for some of my preferred detuning effects. The total range of the Fine parameter is actually plus or minus 400 cents (four semitones!). It would be far more logical to offer less range at greater resolution — you surely can't need more range on the Fine parameter than a single Coarse step provides.

Both dual and single Pitch Transposers offer an octave shift in either direction, with a choice of three algorithms for balancing processing speed against subjective quality: Smooth, Normal and Quick. The Quick setting is is recommended only for use with percussive sounds where almost any amount of processing delay would be audible. Smooth, whilst significantly tidier in its splicing, and therefore almost audibly glitch-free, inevitably does delay noticeably. Normal is certainly an acceptable compromise in the vast majority of applications.

External control of the shift interval by MIDI note information from a keyboard or sequencer is possible — for example, with C3 defined as the base key, entering a G3 will set the transpose interval to +7, giving a fifth above. It is just about possible therefore to overcome the lack of 'intelligent', scalar harmonising by allowing a different transposition interval to be set for every note, although in practice it requires the use of a sequencer to replay the trigger notes with sufficient accuracy to achieve a usable outcome.


The delay effects divide into three types; Stereo, Regenerated, and Multi-tapped, with further Short and Long sub-divisions. Regenerated delays offer Delay Time (with Coarse and Fine parameters), Regeneration, and Level, with maximum delay times of 1100ms (Long), and 960ms (Short). Genuine Stereo Delay is available with up to 1350ms per side (1150ms Short). HF damping, rolling-off high frequencies from successive delays to produce a more natural effect, is also included in this algorithm.

ART'S Multi-taps are always particularly versatile. A maximum of seven taps are available, with choice of three amplitude characteristics (taps getting louder, quieter, or staying the same), and variable spacing — taps can move progressively farther apart as they approach the pre-determined delay time, or get closer together. These algorithms are often very useful for adding a spatial dimension to a multi-effect program, without the softening effect of using a reverb.


Dividing once again into obvious sub-sets, the Alpha 2.0 and DRX2100 offer a choice of Natural and Gated reverb algorithms. Natural Reverbs have four basic 'characters' — Hall, the most natural and diffuse, through Room, and Vocal down to the denser and brighter Plate — with three levels of 'quality' process-complexity, selected by choosing between Type 1, 2 and 3 reverbs. Type 1 programs are recommended for combination presets where reverb is not the primary effect, whilst Type 3s offer the smoothest, densest, most natural reverb effect that the ARTs can muster.

'Natural reverbs' all offer the same basic parameters: Decay Time is variable, in increasing step sizes, up to a maximum of 25s, and HF Damping available to prevent an unnaturally bright decay. Additional subtlety is gained from the Position parameter which subjectively seems to allow you to place the listening point anywhere between the front and the back of the soundfield. Only Type 3 reverbs however have a specific Diffusion parameter which determines the extent to which discrete reflections can be heard in the decay and smoothness of the reverb field.

Both units offer some excellent non-linear or 'gated' reverb algorithms. There are two main parameters here: Decay (up to 400ms), and Diffusion, controlling density, with a further choice of Sloped, Flat or Reverse decay characteristic. Sloped allows some decay before the typically abrupt cut-off, whilst Reverse has effect of almost 'swelling' the decay. Flat is of course the familiar burst of dense ambience cutting-off into silence.


Three Sampler options are provided; Short and Long sample times, plus a Sampler and Pitch Transposer combination. Maximum sample time is just under two seconds, with a choice of Audio Level, Manual activation, or MIDI-note triggering to initiate recording. Playback options include Single (one shot), Repeat (looped), and MIDI mode (which plays the sample whenever a note on message is received) plus audio triggering. The Sampler+Transposer option enables playback of the sample at different pitches under MIDI control, rather like using the base-key facility with the transposer. The Short Sampler algorithm can also be combined with a limited range of reverbs (Types 1 and 2 only). The recorded section's playback can be truncated from both ends via the Start and Length parameters, but there is still no means of storing samples.


Despite their differences, these two units obviously still have a lot in common, and it is no surprise to find them offering a number of identical presets. All my favourites on the original Alpha seem to be still there; a number of very effective general ambience treatments, in varying sizes and characters, from 'Bright Vocal Room' to 'Carnegie Hall', and some excellent dedicated vocal and guitar treatments; compound effects, somehow contriving to add up to more than the sum of their parts.

Evaluating presets is always a rather hit-and-miss affair; if you are using an inappropriate source, the effect is unlikely to shine, and yet inevitably there is a tendency to try a series of programs all with the same source, before moving on to a different one. ART'S continued fondness for unusually descriptive names, apart from the usual quota of totally off-the-wall stuff ('Pigs in Space'?) as ever, goes a long way towards helping you make initial evaluations in a sympathetic context. Don't be fooled by preset 56, 'Red Rocks', on the Alpha 2.0, however — its a complete fraud. It's not red and there are no rocks in it.

Seriously useful dedicated drum treatments abound in both units, with personal highlights being 'Deep Gated Snare' and 'Rock Concert Kit' on the Alpha 2.0, and 'Machine Gun Snare', 'Thunder Snare 2', 'Punch Percussion', and 'Dense Drum Room' on the DRX2100. I am spoilt for choice here really; I could name half-a-dozen others just as impressive — 'Glistening Plate', featured in both models, is a bit special too, as is the evocative 'Abbey Road Rm2'. I am sure even the most jaded of listeners would have trouble auditioning either of these units without coming across something that makes them sit up and listen.


Both these units can be thoroughly recommended for their excellent sound quality, versatility, and ease of programming. The user is presented with any number of 'instant treatments' for when you need something special, as well as a strong assortment of good solid 'basics'. For many, I am sure that will be sufficient on its own. The more adventurous will find these models immensely rewarding to experiment with, both in programming, and via imaginative MIDI control assignments — the options here are very wide indeed.

The evidence seems to suggest that, as with synthesizers, effects units have to sell themselves initially on the strength of their presets alone, for that tends to be all that potential purchasers are prepared to audition in the confines of the showroom. These ART models have nothing to fear in that respect. They offer effects that are immediately appealing, with a highly individual character, and yet easy to envisage in context. ART'S policy of constant revision and upgrading, mixing and matching features from other models in their range, can be seen to pay dividends when it results in optimal units like these.

Further information

ART Multiverb Alpha 2.0 £399 inc VAT
ART DRX 2100 £499 inc VAT

Harman Audio, (Contact Details).


These units are particularly easy to operate, both at the top layer, simply using the presets, and as you delve deeper into the system. The main input device is the Encoder (a continuous, mechanically-stepped pot), operating in conjunction with its four mode switches; Preset, Mix, Parameter and Value. In Preset mode, operating the encoder scrolls the preset display for instant recall of patches. Alternatively, numbers can be entered directly from the numeric keypad allowing full random access. Full use is certainly made of the LCD, with the preset name, an abbreviated listing of the effects (and their order), and an indication of the number of real-time MIDI controllers assigned to the preset, all being displayed simultaneously.

The important Mix parameter remains always instantly accessible as o primary mode via the dedicated Mix key. Pressing Mix cycles you round separate output values for Dry level, FX level and Dry Post-EQ level, on the Alpha 2.0, and Dry, Digital FX, and Analog FX, on the DRX 2100. Where the front-end processes are performing fundamental tonal or dynamic shaping, it obviously makes more sense to use this output as the 'normal' signal to balance against the effect, rather than the totally unprocessed 'Dry' signal, hence the provision of the DRX's separate feed from the analogue stage, and the Alpha's 'Dry EQ'.


Edit mode is activated either via the dedicated Edit key, or either of the Parameter or Value selectors. The parameters and data values of the preset can then be scrolled either via individual key presses or with the rotary encoder — the most efficient method of editing undoubtedly has to be using the Value key to advance the selected parameter whilst the encoder sets the values. Thankfully, the Parameter key allows you to step backwards, in this mode, if you overshoot the item you want.

Adding or removing effects from a preset is achieved via dedicated 'Add Effect' and 'Delete Effect' switches, which really couldn't be more explicit. Whilst in Edit Mode, repeatedly pressing Add Effect, or Add Effect followed by operation of the Encoder, cycles round the available effects. When you get to the one you want. Enter completes the process — effects are always activated with sensible default values, which gives you the encouragement of instant gratification from your creation, before you get down to fine-tuning it. As each new effect is added, the options remaining are automatically reduced to reflect your use of the available processing power — for example, if the first effect you enter is the Dual Pitch Transposer algorithm, you no longer have access to any of the reverbs. Logical, simple, indeed about as efficient as you could want. I have always felt ART's operating systems to be among the best.


Sensibly, many of the presets already include useful MIDI control assignments, and a number of them are specifically dedicated to take advantage of ART's own X15 Ultrafoot MIDI footswitch control system. Up to eight parameters within an effects patch can be assigned to real-time control. MIDI assignments are added via the Add Effect routine, and displayed in Edit along with the FX data, consisting of four elements: the effect parameter to be controlled; the source controller number; plus maximum and minimum data values (inverted response can be achieved when required simply by setting minimum above maximum).

Available control sources are Controllers 0 to 120, plus MIDI note number and velocity (for both note on and note off messages), channel pressure (aftertouch), polyphonic pressure (poly aftertouch — rare) and pitch wheel. Real-time MIDI control from a sequencer obviously allows more sources than you can physically manipulate at the same time, such as assigning a controller to each of the bands of the equaliser, creating a most effective dynamic EQ — given the detailed editing possibilities afforded by using MIDI control data and a sequencer, extremely precise effects can be achieved. In practice, multiple controllers zapping up and down constantly will cause you problems, taxing MIDI bandwidth, overflowing the MIDI buffer, and sometimes audibly glitching the output, whereas 'real-world' usage tends to be a little more conservative — a few key parameters, adjusted subtly in sympathy with a mix makes a far better demonstration of the true worth of real-time MIDI FX control.

Also featuring gear in this article

Featuring related gear

Previous Article in this issue

Zoom At The Top?

Next article in this issue

Sampling Techniques

Publisher: Sound On Sound - SOS Publications Ltd.
The contents of this magazine are re-published here with the kind permission of SOS Publications Ltd.

The current copyright owner/s of this content may differ from the originally published copyright notice.
More details on copyright ownership...


Sound On Sound - Dec 1992

Review by Dave Lockwood

Previous article in this issue:

> Zoom At The Top?

Next article in this issue:

> Sampling Techniques

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 June 2024
Issues donated this month: 0

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

Funds donated this month: £0.00

All donations and support are gratefully appreciated - thank you.

Magazines Needed - Can You Help?

Do you have any of these magazine issues?

> See all issues we need

If so, and you can donate, lend or scan them to help complete our archive, please get in touch via the Contribute page - thanks!

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