Personal Music Studio
The term workstation is officially in demise - to prove it, Ensoniq's SQ1 is dubbed a Personal Music Studio. Simon Trask investigates the best-specified workstation to date.
With some keyboard manufacturers moving away from the workstation concept, is Ensoniq's latest synth behind the times in wholeheartedly endorsing it?
SO HOW DOES the SQ1 fare in relation to the VFX SD? The good news is that it retains the SD's 21-note polyphony. However, compromises have to be made somewhere, and so the cheaper workstation omits the SD's onboard 3.5" disk drive, poly-aftertouch keyboard, large fluorescent display and ability to act as a generic SysEx recorder. Its 61-note keyboard is sensitive to attack velocity only (though its synthesis section can respond to both channel and poly aftertouch via MIDI) and has a light, fairly shallow action.
The SQ1 has also undergone a fair amount of front-panel reorganisation in conjunction with its substitution of a 2 x 16-character backlit LCD for the SD's fluorescent display. To my mind, neither the reorganisation nor the new display work to the detriment of the SQ. Ensoniq have made the SQ's LCD work extremely well for them (or rather, for you). Aided by clear parameter organisation and a readily comprehensible layout of buttons on the synth's front panel, the LCD allows you to focus in quickly on just the right amount of information, with no extraneous parameters to distract you.
The SQ1 omits the Patch Select buttons of the VFX and the VFX SD, allows you to combine up to three Voices within a Sound compared to the six of the other two instruments, allows you to layer Sounds only in Sequence/Preset mode, omits the user-definable pitch table of, and offers less digital effects than, its two relatives, and omits the VFX SD's second, dry stereo output pair. However, the SQ has more waveforms than the VFX (121 as opposed to 109) though a few less than the VFX SD, and the waveforms differ in some respects (the SQ1 has less Transwaves, for instance, but more drum and percussion sounds).
On the sequencing front, the SQ1 has 16 tracks as opposed to the VFX SD's 24, and comes with a 9000-note memory upgradable to 58,000 notes (as opposed to the SD's 25,000 notes upgradable to 75,000). Also, where the SD has 20 Presets and 60 Sequences/Songs, the SQ1 has 30 Songs and 70 Sequences/Presets.
A TOTAL OF 340 Sounds are accessible on the SQ1 at any one time. These are divided as 100 ROM Sounds and 80 RAM Sounds internally and a further 160 RAM or ROM sounds on a plug-in card. Sounds are accessed by selecting Int, ROM, Cart A or Cart B (as in card banks A and B, not two card slots) and then pressing the ten Bank and ten Screen buttons; you can also scroll through the Sounds with the Up/Down buttons located beneath the LCD. Quick and easy.
A standard SQ1 Sound consists of up to three Voices, each of which is a self-contained osc-filter-amp configuration and therefore effectively an independent sound. Depending on the keyboard range which you assign each Voice to, you can create split and layer textures within a Sound. A good example of this is 'Jazz Izz' (preset Sound 16), which uses its three Voices to create a double bass/acoustic piano split in which the bass is layered with an open hi-hat sample. This combination of bass and cymbal works extremely well in a jazzy context, not least because flexibility has been worked into the hi-hat by giving its amplitude envelope a short decay and a longer release - with the result that legato bass notes produce a closed hi-hat while staccato bass notes produce an open hi-hat.
The SQ1's raw sound material consists of 121 Waves, which are actually a mixture of looped multisamples, one-shot samples, waveforms, inharmonic loops and wavetables. These put "real" instrumental sounds such as piano, strings, flute, trumpet and guitar alongside more off-the-wall samples like 'Pot Lid Hit', 'Doorbell', 'Toy Hammer' and 'Spray Can', a range of synthesised and sampled waveforms and various weird metallic sounds. The Waves are organised into 11 categories: Strings (13), Brass (6), Bass (6), Breath (4), Tuned Percussion (10), Percussion (22), Drums (23), Transwave (9), Waveform (22), Inharmonic (5) and Multiwave (1).
Each Voice within a Sound can be assigned one of the Waves, which can be pitched up and down over a wide range. A useful feature called Start Index allows the start point of sample playback to be altered, so that, for instance, you can bypass the sound's attack (this can be particularly effective when dynamically controlled from velocity).
The SQ1 provides two filters per Voice, connected in series. Filter one is always low-pass and can be either 2-pole or 3-pole, but filter two can be either low-pass or high-pass and either 1-pole or 2-pole. The total of both filters cannot exceed four poles; if you set both filters to low-pass and 2-pole, you've got the classic analogue low-pass filter with 24dB/octave roll-off. As on the VFX and VFX SD, the SQ1 doesn't provide filter resonance, which I find a bit disappointing as it can be such a versatile feature.
The next stage is, of course, the amplifier. In addition to allowing you to set the Voice's volume, this also allows you to switch in a 12dB boost, a useful feature as long as you don't start switching it in on everything. At this stage you can also route the Voice to one of three stereo busses (Dry, FX1 or FX2), pan it within the stereo image and give it a low/medium/high note priority and a velocity window.
But what about enveloping? Well, each Voice has three five-stage envelopes which are automatically routed to pitch, filter and amplitude - though the first two can also be assigned to any modulatable parameter. You can set both times and levels, with envelope times ranging from 0.01 to a lengthy 49 seconds!
Additionally each Voice has an LFO which is freely assignable to any modulation destination. The LFO offers a choice of seven waveforms (triangle, sine, sine/tri, pos/sine, pos/tri, sawtooth and square) together with level, delay and speed parameters - the latter itself a modulatable parameter.
"The Song tracks allow you to record extended parts while the Sequence tracks handle repetitive accompaniment parts in familiar fashion."
Each Voice has a total of 15 modulators, which in addition to the envelopes and the LFO include velocity, keyboard position, a noise generator, pitch and mod wheels, an external MIDI controller, MIDI aftertouch and Timbre. Assigning the latter as a modulator allows you to make real-time adjustments to any combination of modulatable parameters using the SQ1's Data Entry slider.
The SQ1's ROM Sounds give you a very good sense of what the synth is about, portraying its versatility and character and playing to its strengths. Play the first few Sounds in Bank 0 and you'll get a good idea of the breadth of the SQ1's sonic capabilities. Here you'll switch from a bright acoustic piano to an atmospheric and mysterious choral sound to a rough 'n' dirty guitar to a realistic bassoon-cum-oboe. The SQ1 can give you rich pad sounds, with or without breathiness and tinkling bells mixed in, it can give you spiky metallic techno sounds, reedy pipe organs, hard and punchy synth basses, warm and rounded fretless basses, funky percussive guitars, bright and punchy analogue-type brass, lugubrious clarinets, breathy flutes... You get the idea. As with the VFX and VFX SD, I'd say that the SQ1 handles the "real" sounds quite well, but doesn't match the quality or the number and variety of such sounds to be found on Korg's M- and T-series instruments - especially given the ability of the Korg range to access further samples on PCM sample cards, something which the SQ1 doesn't have. If you want your very own symphony orchestra in digital memory, there are other instruments which are more appropriate. The SQ1's forte is synthesis, and in this respect it allows you to achieve plenty of sonic variety.
Ensoniq's latest synth allows you to program a second type of Sound configuration, namely a "drumkit"-style arrangement of up to 17 Voices across the keyboard. In this arrangement, each Voice can be given its own key range, the only proviso being that Voices aren't allowed to overlap one another. ROM Sound Banks 8-9 contain 20 drumkit Sounds which have been given names like 'Big Ol' Rock Kit', 'Top Forty Kit', 'Fusion Kit', 'Synth Kit' and 'Out Kit'. In contrast to the limited number of drumkit memories provided by some synths, the SQ1 allows you to program any RAM Sound as either an instrumental or a drumkit Sound.
Ensoniq have provided the SQ1 with a healthy if not a comprehensive range of acoustic and electronic drum and percussion samples - but then it's hard to see how anything short of a sampler can provide the full range and diversity of rhythmic sounds which we've become accustomed to hearing nowadays. There again, Ensoniq have sensibly allowed you to incorporate almost any of the SQ1's Waves into your own drumkit Sounds (only Transwaves and the Multiwave are disallowed), and to pitch each Wave/Voice up or down across a wide range, play it forward or reversed, feed it through a four-pole low-pass filter (with a fixed cutoff point which can optionally be controlled from velocity), and give it gate and release times together with volume and pan settings.
Other Voice parameters govern velocity control of cutoff point and amplitude, and the output buss routing of the Voice, as well as whether or not its pitch will track the keyboard. All in all, you can program some pretty strange 'drumkits' if you want. However, it's a shame that there's no Exclusive facility whereby you can get one sound to cut off another (the open/closed hi-hat syndrome), while the scope for controlling how much signal processing is applied to each Voice is quite limited for the drumkit context.
The SQ1 offers you a choice of 13 24-bit digital effects: Concert Reverb, Hall Reverb, Room Reverb, Warm Chamber, 8-Voice Chorus, Chorus & Reverb, Flanger + Reverb1, Flanger + Reverb2, Phase Shifter, Phaser + Reverb, Rotary Speaker + Reverb, Distortion + Chorus + Reverb, and Compression + Distortion + Reverb. Wot, no DDL? Unfortunately not.
With between four and eight programmable parameters per effect, and the ability to dynamically modulate selected effect parameters from any one of 15 sources (pitch and mod wheels, CM pedal, keyboard position, velocity, aftertouch via MIDI, assignable MIDI controller, sustain pedal, timbre slider and six "ramp" envelopes), there's no shortage of flexibility in the SQ1's effects processing. In fact, the character and versatility of the SQ1's effects bring home the fact that effects processing can be as integral a part of sound creation for keyboard players as it is for guitarists. Be experimental: don't just put distortion on guitar samples, try it on other sounds too.
Each Voice within a Sound can be routed to one of three stereo busses: FX1, FX2 or Dry. In addition, you can set separate FX1 and FX2 mix levels per Voice. For single effects, FX1 and FX2 are both routed through the same effect, with the two mix levels allowing you to apply separate amounts of, say, reverb to different Voices. Combined effects in conjunction with the FX1 and FX2 mix levels allow you to route some Voices through reverb only while others are routed through a more specific effect - distortion, say - and from there either to the reverb or directly to the stereo outs. So for that power trio number you want to work on, you can distort the guitar while applying reverb only to the drums and bass.
Talking of distortion, the Compression + Distortion + Reverb effect should appeal to anyone who likes to get nasty with their sounds. To my mind it's the most natural-sounding distortion I've heard from a digital processor. The signal from the processor's FX1 input is routed through a flanger, compressor, distortion (with a maximum level of 11!), hi/low EQ and - after being mixed with the direct signal from FX1 - reverb. There's a great deal of versatility in this combination of effects, but to top it all you can get into some ridiculous feedback effects by utilising the Reverb-to-Compression Feedback parameter, which routes the post-reverb signal from both the FX1 and FX2 inputs back into the compression stage. Not one for the faint-hearted.
Another effect that needs to be approached with a certain amount of caution is the 12-pole Phase Shifter, which has a stereo cross-feedback parameter that can do some serious damage to your speakers if you're reckless with the feedback level. But that's rock 'n' roll.
Where two Sounds have been assigned different effects (a) the SQ1's output is momentarily but noticeably muted while the new Sound's effect is copied into the processor, and (b) the sustained Sound takes on the effect of the new Sound. So while the SQ1 follows in the Ensoniq tradition of allowing you to overlap sounds - sustain a note playing one Sound while selecting and playing another Sound - the company have included a System parameter called Voice Mute which allows you to switch off the sound overlap feature if you'd rather not have it.
THE SQ1'S SEQUENCER will, in many respects, be familiar territory to anyone who's used an Ensoniq sequencer before. Anyone who is new to the Ensoniq modus operandi will find a sequencer that is versatile yet straightforward to understand. An onboard, dedicated sequencer should provide a certain operational immediacy, and this is something which the SQ1's sequencer manages to do very effectively with its dedicated transport controls and track buttons and an unfussy approach to sequencing.
Essentially, the SQ1's sequencer allows you to record up to 70 eight-track Sequences (Banks 0-6) which can be chained together in up to 30 Songs (Banks 7-9). A maximum of 99 steps can be programmed per Song, with each step containing a Sequence plus track mute and transpose settings and a repeat number (0-99).
However, following in the footsteps of the VFX SD, the SQ1 allows you to record a further eight song-length tracks in parallel with each Song chain, giving you a total of 16 tracks (I always was good at arithmetic). The purpose of the Song tracks, as they're called, is to allow you to record extended parts while the Sequence tracks handle repetitive accompaniment parts in familiar fashion. However, you can record into Song tracks sectionally, while Sequences can be up to 999 bars long (memory permitting, of course), so there's no one way to go about using the sequencer.
"Presets not only allow you to program a wide variety of keyboard textures, they turn the synth into a rather useful MIDI master keyboard."
Each track within a Sequence can be assigned one Sound, together with key range, transposition, MIDI channel, MIDI patch number, internal/MIDI status, aftertouch reception and sustain pedal on/off settings. With the appropriate key ranges set, Sounds can be layered and split in Sequence mode by clicking on one track button and then double-clicking on the others; if you really want, all eight tracks (up to eight Sounds) can be layered and split. Of course, not all tracks need play SQ1 Sounds: some can be set to play via MIDI only, allowing you to incorporate external sounds into your SQ1 sequences and keyboard textures. Individual track status can be set to Local, MIDI, Both or Ext, giving you plenty of flexibility in deciding how to combine SQ1 and external sounds.
Each Sequence can also have its own programmed effect. This effect is common to all tracks within the Sequence, overriding the effects programmed for individual Sounds. However, the Voices within each Sound can preserve their output buss routings, or alternatively you can force them to a particular buss; also, in order to avoid clashes you can specify a particular track/Sound as the only source of control data for modulating effect parameters.
When you create a Song, you can also program an effect for that Song; the SQ1 allows you to choose whether you want that effect to override the Sequence effects, or whether you'd rather be able to change effect from Sequence to Sequence. Bearing in mind the brief but noticeable muting of the SQ1's output level when you switch from one effect to another, you need to treat effect changes with caution - even more so, perhaps, when that held note in one of the Song tracks is interrupted by a Sequence effect change.
If you don't record any sequence data into a Sequence, but program all the above-mentioned parameters, you've got what Ensoniq call a Preset. In this way you can store 70 Sequences or 70 Presets or any combination of the two, which is a pretty flexible arrangement. Presets not only allow you to program a wide variety of keyboard textures using the SQ1's own Sounds, they turn the synth into a rather useful MIDI master keyboard as well.
The SQ1's sequencer memory is non-volatile, which means that when you switch the synth off you don't lose your sequence data - or, looked at another way, when you switch the synth on, your sequences are straight away available to you. In case you're wondering, sequence data can also be stored to RAM cards or via MIDI SysEx, but I'll discuss this later.
Real-time recording on the SQ1 can be set to Remove, Add, Loop or Punch in/out. Although you can layer/split two or more tracks for recording, the SQ will only record into one track at a time. Remove is ordinary recording, where you wipe over any previous data each time you record into a track.
Loop, as you might imagine, allows you to record drum machine-style within a track (looping in overdub), while Add is overdubbing without looping. With Punch in/out set, you can manually punch in to record at any point in a Sequence, then Stop the Sequence at the point you want to finish recording. However, if you also have Auto Punch enabled, you can get the sequencer to punch in and out of record automatically at predetermined in/out points.
A Sequence's length is determined by the length of the first track you record, though subsequently you can lengthen or shorten the Sequence as well as append one Sequence to another (or to itself). After you've hit Record for your initial track, you get a beat from the inbuilt metronome but the SQ1 doesn't begin recording until you start playing; this allows you to set a tempo (25-250bpm) and get a feel for the beat before you start playing. For subsequent tracks you're given a one-bar count-in, though you can switch this out if you wish.
Each time you finish recording, the SQ1 allows you to audition the take before deciding whether or not to keep it, while if a previous take exists you can compare old and new and decide which one to keep - a familiar and welcome Ensoniq practice. Another useful function is Goto, which allows you to locate the sequencer to any position within a Sequence or Song.
Each Sequence can be set to looped or one-shot playback, and Sequences can be selected and played like patterns on a drum machine. Sequence edit functions include track quantise, copy, erase, merge, transpose, shift, scale and filter. Some of these functions allow you to work on a section of a track and on a specific note range. The latter is a particularly welcome inclusion, as it allows you to do things like quantise or erase a specific drum Voice in a drumkit Sound, transpose one note only so that it plays a different drum Voice, or erase one of the Voices in a split Sound texture.
Quantisation on the SQ1 is post-record only, and includes straight and triplet values up to 1/64th-note triplets (the sequencer's maximum resolution is 96ppqn). Notes are simply shifted to the nearest beat of the quantise value. Shift allows you to slip a whole track forward or backward in time in 96ppqn steps, up to a quarter note either way, while Scale allows you to globally increase or decrease the level of various types of controller information in a track. You can also filter out note data and various types of controller data, and copy selected data to a different track.
The SQ1 allows you to edit individual event data within a track. To my mind the synth's small LCD does you no favours here, especially when compared to the large event lists and grid-type event edit displays offered by computer-based sequencers. Also, any event editor which doesn't allow you to edit the position of a note (you have to delete and reinsert at a new position on the SQ1) just isn't trying to be friendly.
Available for the first time on an Ensoniq sequencer is Step-time recording, which allows you to enter notes and controller information. This can be with or without auto stepping (without allows you to record chords per step), and allows you to set note gate times as step length, any fixed duration or a duration determined by the number of steps you hold down a note for. With a minimum duration of 1/64th-note triplets, you can enter some impossibly fast runs, while the fixed note duration facility can be useful for getting that mechanical feel.
The SQ1 does have one other type of recording, called Mixdown, which comes into play, so to speak, when you're in Song mode. In addition to note and controller data, the Song tracks can be used for recording real-time "mixdowns" of volume, pan, envelope release-time and timbre data for both Sequencer and Song tracks. Using the Data Entry slider and Up/Down buttons, you can mix down one track and one type of data at a time.
"Whether or not Ensoniq like to use the term themselves, with the SQ1 they've come up with the best argument yet in favour of the workstation."
One thing you can't do in either Sequence or Song mode is record real-time track mute and solo settings, which is a great pity - especially as you can solo and mute tracks in real time from a screen in the Mix Bank. You select a track by pressing the relevant track buttons, then use the edit controls to set the track to S(olo), P(lay) or (M)ute. While it's nice to have this facility, I would prefer to see muting and soloing done off the track buttons only - with, say, a double-click to solo a track and successive single presses to play and mute it. If you were able to press more than one button at once, so much the better.
The SQ1 does allow you to record initial track on/off settings as part of each Sequence/Preset. If you're working with short Sequences, you could try making several copies of each Sequence and giving each one different track mute settings, then selecting them as appropriate for the moment.
Step-specific track mutes can be recorded at the Song chain level, where you can also transpose selected tracks per step. Chaining Sequences together to form a Song is a tedious business at the best of times. I'd like to see Ensoniq add an "auto-compile" feature whereby you play and select the Sequences in real time and the SQ1 automatically compiles a Song chain for you - surely a fairly trivial programming task. I'm not suggesting that this would always be a superior way of working, but it does have the virtue of allowing you to work at the musical rather than the "technical" level. And why not have the facility to record track mute settings at the same time as your Song is being compiled? This sort of spontaneous approach to recording/(re)mixing is something manufacturers should be paying heed to.
The SQ1 can, of course, act as master or slave for MIDI syncing purposes, and transmits and receives MIDI Song Position Pointers in both Sequence and Song mode. The synth can also be used as a multitimbral slave for playing back sequences recorded into an external sequencer (if you set the synth to Multi reception) or as a multitimbral sound source for MIDI guitarists (if you set the synth to Mono mode A or B reception).
SQ1 Songs can be called up via MIDI (using Song Selects 00-29), but, unfortunately, Sequences/Presets can't. I'd like to see the SQ1 adopt the Control Channel method used by Roland, whereby you reserve one MIDI channel for selecting Sequences/Presets using MIDI patch changes, while other MIDI channels select Sounds in the usual way. Not only sequencer users but MIDI guitarists and other MIDI instrumentalists would find this a useful feature, I'm sure.
THERE ARE TWO ways of storing the SQ1's Sounds, Sequences, Presets and Songs: to a plug-in RAM card (MC32 or MC64) or to a remote storage device via MIDI SysEx. The RAM cards can be used to store Sounds only, Sequences/Presets/Songs only or a mixture of the two. A maximum of 160 Sounds can be stored on either card (as two banks of 80), so the extra memory of the MC64 allows you to store more sequence data than on the MC32.
The SQ1's card slot can also accept ROM cards. Apparently there will be two categories, both offering 160 Sounds per card: SC Series cards will provide factory-programmed Sounds, while ISC Series cards will provide Sounds "programmed by leading sound developers from around the world".
With 3.5" disks being a significantly cheaper storage medium than RAM cards, sooner or later it's going to make a lot of sense to invest in a generic SysEx filing device, even if it does go against the grain of the "all-inclusive" workstation ethos. If you've bought an SQ1 because you don't want to get involved with computers, your best bet is probably an Alesis Datadisk, which will be able to handle the sort of file sizes required by SQ1 sequencer memory dumps.
Sounds and sequencer data are transferred separately via SysEx, with bulk and individual options in each case, and providing SysEx reception is enabled, the SQ1 will receive data automatically. Saving to and loading from disk via MIDI is inevitably going to be slower than using an onboard disk drive, not to mention a RAM card, so where you need speed (when playing out live, for instance) it's still going to be worth having one or two RAM cards to hand.
ENSONIQ DON'T LIKE to use the term "workstation", because, to paraphrase one of their adverts, they've taken the work out of their instruments. So instead they've called the SQ1 a Personal Music Studio - a description which conjures up images of portable karaoke machines in my mind. Rest assured, the SQ1 isn't about to assault your ears with versions of 'My Way' or 'Tie a Yellow Ribbon Round the Old Oak Tree', nor do you have to sing along to it if you don't want to. And while a karaoke machine encourages you to be recreative by taking all the hard work out of music-making for you, the SQ1 encourages you to be creative by taking all the hard work out of music-making for you. For once, the advert writers may have hit the nail on the head instead of shooting themselves in the foot.
The SQ1's front-panel is easy to comprehend and straightforward to operate, yet this hasn't been achieved at the expense of flexibility and sophistication. At the same time Ensoniq have provided a very effective integration of programming and sequencing on the SQ1 - which is, after all, what you should expect on a workstation. The sequencer allows you to get your ideas down quickly and conveniently, while doing its best not to limit you in how you put your music together. All in all, the SQ1 encourages your creativity to come out by allowing you to put your ideas into practice with the minimum of fuss.
Of course, first and foremost you have to like not only the sounds that the SQ1 is capable of producing - and here it's a satisfactorily versatile instrument - but also its sound quality. There's definitely an "Ensoniq sound", and it's one which I'd loosely characterise as being grainy, gritty and gutsy while striking a happy balance between warmth and incisiveness. As such it's an effective antidote to the polite clarity which tends to characterise the sound of Japanese synths. The same sort of thing can be said about the SQ1's effects, notably the distortion - I just wish Ensoniq had included a DDL (they would've probably made it break up on regeneration).
Ensoniq have brought the SQ1 in at a good price for what you're getting, though if you take into account (or perhaps that should be take out of account) the extra cost of upgrading the sequencer memory and adding on some form of SysEx remote storage capability (both of which are advisable, it seems to me) then you need to think in terms of spending several hundred pounds more at some stage.
There are all sorts of arguments which can be made against the workstation instrument and in favour of the computer-based/modular approach. To some extent their validity is a personal thing, dependent on the way that you want to work and the requirements of the music that you want to record. At the same time, when certain workstation instruments seem to represent little more than an opportunity for flogging tired technology in a new guise, can you wonder that the workstation concept has acquired a tarnished reputation?
Ensoniq have always thought of the workstation as an integrated instrument rather than a collection of components which have to be made to fit together.
Whether or not they like to use the term, with the SQ1 they've come up with the best argument yet in favour of the workstation.
Prices SQ1, £1189: SQX70 sequencer memory upgrade. £165: MC32 RAM card. £TBA. MC64 RAM card, £TBA. All prices include VAT.
More from Ensoniq GB. (Contact Details)
Review by Simon Trask
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!