Peavey SP Sample Playback Synthesiser
Sampling bit by bit.
Looking for your first sampler or aiming to expand an existing sampling setup? Simon Trask gives you the latest SP from Peavey...
The sample replay unit isn't a new concept. Back in 1987, Oberheim introduced the DPX1, a rack-mount unit which could read the samples and program data from Ensoniq Mirage, Prophet 2000 and Emulator II disks, as well as samples in via MIDI in Sample Dump Standard format. More recently, SDS has become the standard means of transferring samples into the small but growing number of synths which include onboard sample RAM alongside the inevitable sample ROM.
Easily the most affordable of these synths at £1499 is Peavey's DPM3 SE Plus, which comes fitted with 512K of battery-backed sample RAM upgradeable to 1Mb. Samples loaded into the DPM3's RAM can be looped, trimmed and combined into multisamples, which can, in turn, be assigned to the synth's oscillators. In a commendable move last year, the company introduced an inexpensive mono 16-bit sampling module, the SX (reviewed in MT August '91), to act as a sampling 'front end' for the DPM3's sample playback capabilities.
At a little over £300, the SX represents a convenient and affordable way of getting your own samples into the DPM3, the DPM V3 rack-mount, or any instrument which can read samples in SDS format. It's also extremely easy to use - in fact, you could hardly wish for a friendlier, more straightforward introduction to sampling.
But of course, no-one would argue that the SX/DPM combination seriously rivals the power and flexibility of a dedicated sampler. So, having established the concept of separating sampling and sample playback into two discrete units, Peavey have produced another partner for the SX, one which puts the company into serious contention in the sampler market - despite the fact that they've chosen to call it a Sample Playback Synthesiser.
The 'synthesiser' tag is intended to highlight the SP's ability to offer more than simple replay of samples - it can, in fact, process them through its own integral, full-function synthesis section. However, the SP's approach to combining samples and synthesis in this way is very different to that of the DPM3 - as we shall see...
The combination of SX and SP makes for a powerful, flexible and very competitively priced self-contained sampling setup at £1228. At the same time, as a playback-only unit the SP is potentially a useful addition to an existing sampler - again very attractively priced at £899.
The SP's onboard sample RAM (2Mb as standard, upgradeable to 32Mb) can hold up to 200 Waves, 200 Tones, 128 Maps, 128 Presets and 16 Multis - collectively known as a Bank. You can store one Bank per floppy disk, and multiple Banks on a mass storage device such as a hard disk; loading in a new Bank can completely change the configuration of the SP.
So what do all the above terms mean? Well, Waves comprise the raw sample data, Tones are Waves routed through the SP's various synthesis functions, Maps are Tones mapped across the MIDI note range in multiple Zones, Presets allow you to layer and crossfade between two Maps and are assignable to individual MIDI channels, and Multis are MIDI multitimbral assignments of Presets to up to 16 MIDI channels.
The SP is 16-voice polyphonic and dynamically allocates its voices across the active Parts when in Multi mode. In addition, it can make use of up to eight Groups to define voice allocation priorities for different musical parts. Each Zone within a Map can be assigned to one of these Groups, while within a Preset you can assign each Group to one of four priority levels: low, medium, high or absolute. If you wanted to guarantee that certain parts would always sound, you could assign their Zones to a particular Group and set that Group to 'absolute' priority.
At the Zone level you can set either 'normal' or 'exclusive' for each Group. 'Exclusive' is an extension of the feature commonly found on drum machines which allows one sound to cut short another. On the SP, two different sounds assigned to the same Group will cut one another short, but repeated notes will use different voices providing the Group's polyphony is set to more than one voice. So, for instance, you could have repeated open hi-hat notes overlapping one another, but cut the open hi-hat short with a closed hi-hat. It's a subtle feature which typifies the kind of thought that has gone into the SP.
Although the SP has been fitted with a high density (1.44Mb) 3.5" floppy disk drive, you'll soon find yourself having to save samples across two disks with just the standard memory. You'll probably find that after upgrading the memory it'll be time to think about adding a fixed or a removable hard disk or even a magneto-optical drive for mass storage. This can be hooked up to the SP via one of the SCSI ports on its rear panel; alternatively, DAC sell an 80Mb hard drive which can actually fit inside the SP's 1U 19" casing.
One potential disadvantage of splitting sampling and sample playback across different units is the time needed to transfer samples between them. Sample transfer via MIDI is a notoriously slow business, and simply impractical once you start talking megabytes. Peavey needed a fast medium for data transfer between SX and SP, and they found it in the SCSI buss. SCSI is a parallel buss which is specifically intended for transferring large amounts of data as quickly as possible. Typically, it is used to link up computers, hard disks, CD ROM drives and the like.
The SP comes with two SCSI ports fitted as standard on its rear panel, while the SX can be upgraded with an SCSI port for £40; an SCSI lead (25-pin male to male) to link up the two units will set you back in the region of £20. However, it's well worth the extra; for example, a MIDI SDS dump of the full default SX memory (128kword) takes two minutes 25 seconds with handshaking (MIDI cables connected in both directions), but a mere four seconds via SCSI! And you wouldn't even want to think about SDS transfer without handshaking (...would you believe five and a half minutes?).
Clearly, MIDI transfer just isn't viable when you're routinely passing samples between the SX and the SP. As it is, MIDI SDS sample transfer into the SP has been implemented, but you can't transmit samples from the SP via MIDI, only via SMDI. SCSI transfer of samples isn't unique to Peavey, but the protocol they've used is. Basically, the company have developed their own protocol called SMDI, or SCSI Musical Data Interchange, which allows sample data to be transferred via SCSI in MIDI SDS format.
So far, apart from Peavey themselves only Kurzweil have adopted SMDI, on the K2000, though it's rumoured that Ensoniq are considering implementing it. However, as a 'standard' it has (so far) got precisely nowhere - with no proposal having been put before the MMA for consideration as of writing. But one thing's for sure: rapid sample transfer between instruments and between software and instruments is highly desirable, and MIDI's modest baud rate is never going to deliver it.
"Having established the concept of separating sampling and sample playback into two discrete units, Peavey have produced another partner for the SX, one which puts the company into serious contention in the sampler market"
The SP's Util mode allows you to set up SMDI communication between SP and SX by setting an appropriate SMDI Slave Device SCSI ID. You can then call up screens which tell you the current SX firmware version and the amount of memory installed in the sampling module, and set the sample rate (continuously variable from 12-48kHz) and sample length to be used by the SX. Other screens let you see a bargraph representation of the signal level appearing at the SX's audio input, read a sample clip count once you've taken a sample, select a destination Wave number and remotely initiate sample transfer from SX to SP. You can also send a Wave out via SCSI to the SX (or whichever SCSI device you've defined as the slave).
The process of sampling on the SX and getting the samples into the SP's Wave locations is incredibly easy, so long as you've got your SCSI connections right. In fact, you could almost believe that the SX and SP are one instrument.
The name SCSI Musical Data Interchange hints at a wider application than just sample transfer, but it's unlikely that it will become a kind of 'fast MIDI' beyond specific applications like this. It's too expensive, too confusing, too inflexible and just not rugged enough for general musical use.
One way of avoiding MIDI and SCSI altogether is to read samples directly off floppy disk. Peavey have been a bit slow off the mark here - they're still in the process of implementing S1000 and EPS16 Plus disk reading. Ironically, DPM3 owners are doubly disadvantaged - not only are they unable to enjoy the benefits of SMDI (because the DPM has no SCSI option), but also, the SP can't read DPM disks, owing to the fact that it uses high density floppies while the DPM only uses double density.
The SP's clear front-panel layout and the logical and consistent organisation of its LCD software pages make comprehension and operation very easy. Wave, Tone, Map and Preset buttons reflect the SP's architecture, the Disk button provides ready access to all disk operations, while Global allows you to program at the Multi level, and Util governs SMDI communication.
Operationally, you use the left/right arrow buttons to move the cursor around the LCD window, and the Inc/Dec buttons and infinite rotary Data knob to select LCD pages and parameters and to edit parameter values - depending on the location of the cursor. Additionally, the 2 x 20-character backlit LCD sometimes prompts you to press the Exec button in order to drop down to another level of LCD pages or to initiate a command such as Create Tone.
In addition to the pair of SCSI sockets mentioned earlier, the SP's rear panel provides MIDI In, Out and Thru ports, a small screw (which you can turn with your thumbnail) to adjust LCD contrast, and main and sub pairs of audio outs.
Once you have a sample in a Wave location, you can dial up 16 different kinds of information about it, including type (mono or stereo), loop (forward or off), size and length, start and end locations, loop start and end locations, and sample rate. As well as providing functions for deleting, copying and naming Waves, the SP allows you to alter the sample start and end points, define loop start and end points and a loop length fraction (basically a 'fine-tuning' control for loop length), trim excess samples, set an original key for the Wave, and fine-tune it.
You can also audition any pair of Waves as a stereo sample, and get the SP to perform a stereo merge operation on them if you like what you hear. If a stereo sample has been imported into the SP as two mono samples rather than an interleaved stereo sample (as generated by generic sample editing software like Sound Designer and Alchemy), the stereo merge function provides the means of reuniting the two halves.
The SP can be set to select only zero crossing points as you scroll through a sample looking for loop points, but there are no other aids such as crossfade looping. Also, samples can only be looped in the forward direction - though apparently forward/reverse looping will be provided in the next software upgrade. One helpful feature is Wavepoint Audition, which sustains a note at original pitch while you're searching for that perfect loop.
Once you've got your sample as you want it, you can assign it to a Tone as a DCO sound source and route it through a DCF (cutoff point but no resonance) and DCA. In addition to the DCA envelope there are two freely-assignable ADSR envelopes which can be routed to the DCO, DCA and the panning module, as well as LFO modulation of DCO, DCF, DCA and panning. You can choose from eight LFO waveshapes, including Grunge (!) - a noise-like waveform which can be effective at adding a touch of roughness to a sound when assigned to the DCO.
Each synthesis component can also be modulated by velocity and by an auxiliary modulation source selected from mod wheel, channel and polyphonic aftertouch, and three user-selectable MIDI continuous controllers. Always a welcome inclusion is a Wave startpoint modulation parameter, which allows the amount of attack transient on a sound to be varied in response to a modulation source - velocity is the most commonly used (eg, you could vary the amount of attack on a snare drum by how hard you hit a key), but keyboard position can also be useful.
A more unusual parameter is Wave Start Modulation Quantisation, which allows you to play different sections of a sample depending on the value of a selected control source. For instance, you could switch between different drum sounds in response to velocity strength, or play different sections of a performance loop in response to velocity or to mod wheel position.
"The combination of SX and SP is almost a marriage made in heaven - given the SP's ability to play stereo samples, it would be nice to have a stereo sampling front end as well"
When you have a group of Tones that you're happy with, you can progress to the Map level and assign them across the keyboard in multiple, non-overlapping Zones. This is where you set up a multi-sampled instrument or create a drumkit. Each Zone can be given its own transposition amount, volume level, pan position and output routing, and, as discussed earlier, assigned to one of eight Groups and set to either normal or exclusive mode.
The next level up, the Preset level, allows you to progress one stage further in that you can layer two Maps if you wish, and optionally switch or crossfade between them in response to one of nine control sources (keyboard position, velocity, random, modwheel, channel or polyphonic aftertouch, or one of three MIDI continuous controllers).
Additionally, you can set a switch/crossfade pivot point and a crossfade region width. Beyond this you can assign up to four Maps to each of the two Layers, and switch between them using, for instance, velocity or keyboard position - so in effect you can switch between up to eight different samples.
Other Layer-specific parameters include Transposition, Detune Amount, Pitchbend Range, Output Assignment, Pan Position, Volume Offset, Filter Cutoff Offset, DCF and DCA Velocity Sensitivity, DCF Envelope Amount Offset and DCO Mod Range. And you don't need me to tell you that's a lot of parameters and a lot of flexibility.
Finally, you can assign Presets to MIDI channels at the Multi level. The SP can be set to Omni, Poly or Multi response. When in Multi mode, Presets can be assigned to all 16 MIDI channels; also, if there are certain channels you don't want to use because other instruments are assigned to them, you can disable the SP's response on those channels. As mentioned earlier, you can assign a priority voice allocation level to each one of eight Groups as a means of ensuring that some musical parts have voices stolen from them before others - while others never have voices stolen from them at all.
Don't be fooled by its sample playback role, it's modest dimensions or its modest asking price - the SP is an impressively powerful and versatile instrument with plenty of depth and detail. You're unlikely to exhaust its potential in a hurry. This is an instrument to be explored and absorbed gradually. At the same time, I found I was able to get into the basic structuring, from Waves through to Multis, with great ease.
I would like to see more help provided for sample looping, and the addition of timestretching (...apparently this feature is already planned), though if programs like Sound Designer and Alchemy adopt SMDI there's always the option of graphically-based sampled editing away from the SP.
The combination of SX and SP with SMDI communication via SCSI is almost a marriage made in heaven - given the SP's ability to play stereo samples, it would be nice to have a stereo sampling 'front end' as well. As for the viability of the SP as a stand-alone sample playback instrument complementing an existing sampler such as an Akai S1000 and drawing on an existing sample library, a lot depends on SMDI's broader acceptance - plus, of course, the SP's ability or otherwise to read samples directly off disk.
Peavey have placed themselves firmly on the hi-tech map once again.
Prices: Peavey SP £899, Peavey SX £325, SCSI upgrade for SX £40, DAC SP-80Mb hard disk drive £580, Peavey sample library £19.95 for each two-disk set (All prices include VAT)
More from: Peavey Electronics (UK) Ltd, (Contact Details).
Gear in this article:
Review by Simon Trask
Previous article in this issue:
Next article in this issue:
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!