• Ensoniq EPS Update
  • Ensoniq EPS Update
  • Ensoniq EPS Update
  • Ensoniq EPS Update
  • Ensoniq EPS Update
  • Ensoniq EPS Update

Magazine Archive

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

Ensoniq EPS Update

Paul Wiffen takes an exclusive look a Ensoniq's latest offering in the field of sampling


A change in the distribution of Ensoniq products has resulted in the retail price of the EPS sampling keyboard dropping by over £250. In addition, the various updates and add-ons promised since its release last autumn are now all available at last and just released is a rack mount version which includes all these extras as standard. In the light of these developments, it seems a good time to re-evaluate the EPS's position in the market place.


There is a rather worrying tendency in the music business, particularly prevalent at the sampling end of the market, to dismiss all previous machines as passe, as soon as something new comes out. To read some reviews, you would almost think the arrival of a new machine had somehow magically caused all the rival units out there to suddenly stop working.

Yet there are some manufacturers who instead of bringing out a new machine each year, which makes their previous model obsolete, try to keep their instruments up to date with the latest developments by means of software updates. Unfortunately, these do not generate as much interest in the press as something in a brand new shiny box, even though they may be giving you the equivalent of a new machine in the software.

Indeed, there seem to be several Japanese manufacturers at the moment repackaging the same old stuff in new boxes without adding anything new, to try and squeeze a few more sales out of redundant technology.

Mindful of this Ensoniq have cleverly steered a middle course. For press interest and general public exposure, they have released a new product, the rack mount EPS-M with this year's amount of memory (1 megaword) and this year's options: SCSI Interface and Separate Polyphonic (N.B. polyphonic) Outputs. However, this doesn't mean that people who bought last year's model need pull their hair out, swear and then having calmed down, place an ad in their local free rag to try and sell the old machine.

All the improvements in the EPS-M are available as updates for the EPS, in the form of the 4X memory expander, the SCSI interface and the 8 output expander. So if they need any or all of these features, they can simply buy them individually, bringing the machine they have had the use of for the last 6 months up to the latest spec. Nobody loses out, nobody is left with redundant gear they can't sell.

The Story So Far



The Ensoniq Mirage was the first polyphonic keyboard sampler under £10,000 (and incidentally under £2,000) and as such revolutionised the use of sampling in music. No longer was the realism of sampled sounds just for rich pop stars and flash producers with more money than sense. The Mirage made it possible for any one to get into sampling.

However, American and Japanese manufacturers were not far behind with 12-bit machines and although the "grungy" quality of the Mirage was appreciated by some (and was in fact better quality than the Fairlight Series II, which had been the £30,000 State-of-the-Art for many years), serious fidelity was becoming the order of the day. This combined with the fact that operating the Mirage needed a reasonable fluency with hexadecimal, led to a decline in sales (although people kept using the excellent sound disks that had been developed by the chaps at Ensoniq).

Ensoniq's response to the challenge from other manufacturers was the EPS, which went further towards making sampling a performance-oriented technology than ever before. Whereas preceding samplers had concentrated on the greatest possible fidelity to the detriment of performance considerations (the large amounts of memory needed for high quality meant longer and longer loading times, during which the machines were inoperative, and the sounds were rarely set up for a player to get the most out of), the EPS went out of its way to make live use a pleasure instead of the usual nightmare.

It was the first machine to allow playing while disks were loading, the first to make patch selection into a performance control and the first (and still to my knowledge the only) sampler to feature polyphonic key pressure (so that different amounts of effect can be introduced to each note sounding).

This last feature is available because the same keyboard is on their SQ-80 synth. This keyboard does feel extremely nice to play once you are monitoring at a loud enough level to cover the physical noise of the keys hitting the pressure sensors, which can be a little off-putting at low levels.

In terms of sound quality, the EPS is perhaps not in the same league as the Fairlight Series IIIs, Emulator IIIs, or Akai S1000s of this world but then it is not in the same price league, costing less than half of the price of the cheapest of those. The EPS is also unique in that it lets you decide how important fidelity is to you in a trade off with the number of voices available. It does this by increasing the sample playback rate.

The default setting is 39kHz playback rate and this gives you 16 voices, but you can increase this significantly to a 52kHz (higher than CD or even DAT rates) with the loss of 4 voices. And if the amount of polyphony is your highest priority, then there is a 20-voice mode which uses the still acceptable playback rate of 31.2kHz.


It is also possible to use a higher polyphony setting whilst creating sequences so that you can hear all the parts at once, and then when transferring to tape, sync the sequencer to tape, solo each track and record it with the highest fidelity setting selected.

Although there is noticeable compromise in the input section of the EPS (when sampling through its analogue-to-digital stage) as against the much more expensive machines mentioned earlier, the output stage more than competes with the sort of oversampling you normally find on the pricier CD players. This is a process referred to as interpolation, whereby the machine plots extra sample points on playback to smooth out the higher frequency distortion caused by lower playback rates.

Experiments I have conducted recently, transferring S1000 and EMI samples to the EPS via the Alchemy program on the Mac and Universal Sound Designer on the ST have convinced me that the output stage of the EPS is every bit as good as those on these considerably higher-priced units, and the sounds are much more playable on the EPS.

Bearing in mind that Ensoniq themselves are now sampling sounds on the Dyaxis hardware (an American sample-to-hard disk recorder that works with the Mac) and then porting them over to the EPS via Alchemy, this means you are getting the same sort of fidelity as the big boys for a fraction of the cost.

The performance facilities on the EPS bear closer examination. The thing which more than anything made me prefer my S1000 sounds on the EPS was the ability to put related sounds on the different patches in a program and then switch very quickly between them whilst playing. This is achieved by two patch select buttons conveniently placed near the pitchbend and mod wheels. Pressing the left, right or both buttons each brings up a different patch to the main one, and each of these can be edited versions of the samples in the main patch or can be built from completely different samples.

I had great fun with my S1000 string sounds, putting the pizzicato on the left button and the marcato bowings on the right and the lot doubled when you hold down both buttons. Ensoniq have an excellent saxophone which has different blowing strengths, fall-offs, and glissandos on each patch and this gets you much closer to reproducing an authentic sax performance. Of course, you can also edit factory programs to produce custom patches of your own to add in to your playing.

There is not really space in this overview of the original EPS to go into the editing features of the machine in any great depth, but take it from me that it stands up well in comparison to any other machine out there at any price. The looping facilities include auto-loop (which seeks out zero crossings to cut down the time spent looping manually) and six different types of crossfade looping (with undo - a godsend as the best crossfade loop implementation rarely works well the first time, and without undo you would need to reload the sample from disk... if you had remembered to save it first). The envelopes are of the modern flexible rate/level type and there are three, defaulting to pitch, digital filter and amplitude.


One of the principle reasons for initial interest in the EPS was the internal sequencer as this was also the cheapest sampler with an onboard sequencer, taking it into the category of this year's jargon term of "workstation". Now whilst I hate the term and the concept behind it (which can best be summed up in the old phrase "jack of all trades, master of none"), I have to admit that the EPS sequencer is one of the best I've ever come across within a keyboard.

It is built around an "undo"-type philosophy, which allows you to keep either the "new" or "old" version of the piece after each operation (recording a new part, quantizing something, entering something in in step time, etc), letting you hear the result before you have to make the executive decision.

Of course, for computer owners the built-in sequencer may not be of too much interest, as you are probably running a sequencer of some sort or other on your computer (although you might consider recording your tracks across to the on-board sequencer once they are finished to save lugging your computer around when you go out on a gig or want to play your stuff to someone elsewhere than at home).

Of more interest is the Multi Mode where the EPS can respond to multiple incoming MIDI channels. This is perfect for multi-timbral control of the EPS from an external sequencer. There are a couple of other pluses when using the EPS with an external sequencer; although the play while load feature does not apply to the on-board sequencer (ie you cannot get new sounds to load in the middle of sequence playback), you can use MIDI program changes on the appropriate channel to load a new sound in the middle of a song.

Now Read On



Though there were many excellent features of the EPS which attracted a lot of interest when it first came out, as samplers have appeared with four times as much memory, multiple polyphonic outs and hard disk interfaces, this has tended to draw interest away from the EPS. The EPS is however one of the few samplers on the market which can be retro-fitted with separate polyphonic outputs, memory expansions (up to 4X the original) and hard disk interface. Let's look at each of these upgrades in turn.

Memory size is one of the principle factors in the quality of sounds you can get from a sampler. If you only have a few seconds sample time available, it doesn't matter if you have the highest recording quality possible. You will not get accurate representations of acoustic instruments (the principle raison d'etre of samplers for most people) if you can't fit the multi-samples necessary to create a realistic range into the machine.


The guys at Ensoniq have done some splendid sounds using a minimum of memory, but there are some instruments you will never fit into a half a meg of memory however hard you try. The reason that instruments like the Akai S1000 or the Emulator III sound so good in demonstration is that their whole memory (2 and 4 megabytes respectively) is given over to getting one sound as faithfully as possible. The other advantage of increased memory space in a multi-timbral keyboard like the EPS, is that you can have more sound available simultaneously for sequencing.

To these ends, there are numerous options for memory expansion. Unfortunately, as anyone who has anything to do with memory chips knows, ever since the event of last year, memory has been much more expensive than anybody could have guessed (instead of following the general downward spiral of the cost of electronic components, it is actually more expensive than 18 months ago).

The official Ensoniq products are a 2X memory expander and a 4X memory expander, retailing at £299 and £599 respectively and can be fitted fairly quickly by an Ensoniq service centre. Because these expanders took a fair amount of time to arrive, several different third party companies decided to make their own versions and there are now quite a few being offered for sale.

However, not all of them are exact replacements for the official Ensoniq ones, as I have seen versions which do not fit properly into the unit and others which do not allow the SCSI interface to be fitted as well. Make sure before you buy one of these substitute products that it can be properly installed and that it doesn't preclude the later addition of SCSI if you think that you may need it (see below for the uses of SCSI).

The only make which has the official Ensoniq stamp of approval is from PS Systems of California and this is a 4X expander which definitely does allow for the SCSI interface to be fitted, and this also retails for £599. Generally speaking, there are no great savings to be made on any of the other makes (the expanders are just memory chips on a circuit board and memory is still pretty pricey), so you are probably better off sticking with one of the officially-backed ones which won't lead you into warranty problems.

Separate outputs have been a major selling point ever since it became clear to everyone that multi-samplers are by their very nature multi-timbral. When people first began using individual outs for things like drum sounds and bass-lines, then they were quite happy with monophonic outputs, but as memory has been expanding to allow things like piano, strings and brass to be held in samplers at the same time, the demand for polyphonic outs has been increasing.

Indeed, the sales of high end machines like the Fairlight Series III and the Emulator III have suffered recently because their sixteen separate outputs are monophonic, as opposed to the eight polyphonic outputs of the Akai S1000 and Roland S330.

The eight output expander for the EPS puts it in the latter category, with the EPS capable of dynamically allocating the full 20 voices between these outputs (and the stereo left and right). The expander plugs into a D-connector on the back panel and sits neatly on the far left of the control panel. It retails for £195.

SCSI is perhaps the most recent arrival on the sampling scene. It allows for much higher data transfer speeds than the previous interfaces like RS-232, RS-422 or MIDI, which have been used to send sample data in the past. Its first and perhaps most important benefit is that it works at a speed which makes interfacing to mass storage media like hard disks and removable cartridge drives significantly faster than floppy disk storage.

The actual SCSI connector on the retro-fit is a 25-pin D-connector of the same type as on the Mac (as opposed to the 50-pin), so you can usually plug a standard Mac peripheral straight to the EPS. There is a list of hard disks which are officially recommended, available from Ensoniq, but as usual the best rule of thumb is to take your instrument, with the SCSI interface fitted, into the store where you are planning to buy your hard disk (most music stores have not got round to stocking them yet) and see it working with your own eyes.

However, there is a better option than hard disk which SCSI now gives you. Removable media drives which have access times similar to hard disks are becoming popular in computer applications but they are even more perfect for use with samplers. The reason is that sounds stored as sample data take up an enormous amount of room. You would not believe how quickly you can fill up even 60 meg of storage (which would last forever in most computer applications) with sampled sounds.

Now when even the biggest hard disk is full, to get more storage, you have to buy another (let's not even begin to talk about tape-streaming the amount of data we are talking about here). But with these removable media drives when the 45 meg cartridge (the most popular size currently) is full, you just eject it and insert a new one (at a fraction of the cost of a new hard disk).

True, the price of the drives is currently a bit more than a 40 meg hard disk but as soon as you need to store more than one hard disk's worth of sounds then the removable media drives become much more economical. Another advantage of the removable media drives is that if you own (or ever think you might own) another instrument or a computer which is SCSI-equipped, all you have to do is change to another cartridge formatted to the required type and you have a mass storage device for that piece of equipment too.

The removable media drive I borrowed from Argent's to try with the EPS, I also got working quite happily with the Mac, Emulator III, and Akai S1000. Its retail price is just under a grand (plus the dreaded VAT.)


Another major advantage of SCSI (assuming you are able to stretch to a computer which is so equipped), is that the high speed of sample data communication can be exploited with sample editing packages like Alchemy. Even running an excellent program like this, if you are communicating with your sampler via MIDI, then sample editing can be a great cure for insomnia. However, via SCSI, samples fairly zip across from computer to sampler, and suddenly doing a small edit on the computer seems like a couple of minutes invested and not a life's work. Chances are you are more likely to make use of that facility! The EPS is virtually a new machine with all these add-ons and there is definite attraction in being able to buy the basic unit much more cheaply and then being able to add memory outputs and SCSI as you need them. Of course if you want all the facilities straight away, then the rack-mount EPS-M does give you everything from the word go, and saves you a few quid on the composite price of the EPS and all its add-ons (although probably no more than you could save yourself with a little shrewd negotiation on the package).

True, the rack version is not as cheap as some people may have wished to see it, but it does cost significantly less than other machines with the same amount of memory, polyphonic outputs and SCSI. All in all, the EPS family is going to be with us for some time to come.

The new distributors of Ensoniq products to whom all Ensoniq enquiries should now be addressed, are: Ensoniq G.B., (Contact Details)

Prices:
EPS keyboard £1399
Ensoniq 2X Expander £295
Ensoniq 4X Expander £549
EPS SCSI Interface £129
EPS 8 Output Expander £149
EPS-M Rack Mount £2199


Also featuring gear in this article



Previous Article in this issue

Quadraverb

Next article in this issue

Patch Works


Micro Music - Copyright: Argus Specialist Publications

 

Micro Music - Jun/Jul 1989

Donated & scanned by: Mike Gorman

>

Should be left alone:


You can send us a note about this article, or let us know of a problem - select the type from the menu above.

(Please include your email address if you want to be contacted regarding your note.)

Review by Paul Wiffen

Previous article in this issue:

> Quadraverb

Next article in this issue:

> Patch Works


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!

If you're enjoying the site, please consider supporting me to help build this archive...

...with a one time Donation, or a recurring Donation of just £2 a month. It really helps - thank you!
muzines_logo_02

Small Print

Terms of usePrivacy