Photos: Richard Ecclestone
Tens of thousands...nay, hundreds of thousands of words have already been written about the OASYS, over 2000 of them in this magazine (see the preview in March's SOS at www.soundonsound.com/sos/ mar05/articles/korgoasys.htm). Quite apart from Korg's humongous brochure, the Web offers numerous discussions, videos, and a host of support services that tell you everything you need to know about it.
Except that they don't. You know that OASYS is expensive, has three modes of synthesis, is a powerful sampler, incorporates wave sequencing and KARMA, has a big sequencer, built-in effects, and offers audio recording to make it as close to an all-in-one production system as is currently possible. But even if you know all that, do you know what it feels like to use one? Let me answer that: if you haven't used one, you don't.
So, over the next two months, I'm going to try to tell you what it's like to program and make music on the OASYS88. What I'm not going to do, however, is tell you everything about it. Why? Because the manuals are well over two inches thick, that's why. If four issues of Sound On Sound are piled on each other, they're still not as thick as the OASYS's parameter guide! And do you know what...? These huge tomes still don't tell you what it's like to use an OASYS!
But now I'm getting ahead of myself. Let's take a step back and begin, as all good stories should, at the beginning...
Back in June 1995, Korg asked me whether I would be willing to exhibit some vintage keyboards at an event they were planning to hold in the Science Museum in Kensington, London. They wanted to demonstrate the progress they had made since their earliest days, so I took a Minikorg 700, a Trident, a Polysix, and a PS3200, and sat back to enjoy whatever it was that the company had lined up for the evening. Soon, I wasn't leaning back, but leaning forward, intent on gleaning everything that I could from the presentation. The event was, of course, the dual launch of the Trinity and Prophecy, and it was riveting.
But dramatic though the launches were, the presenter, Korg's Steve McNally, kept referring to two mythical products that appeared simultaneously to predate and supersede the Trinity and the Prophecy. Both had been developed at Korg R&D in California. The first, called Synth Kit, was a software environment within which the physical models in the Prophecy had been developed. Based on a Mac computer with additional DSP hardware, it was perhaps the first system capable of crunching the numbers needed to perform physical modelling in real-time.
The other was a keyboard designed using Synth Kit, intended for thee and me, and which was mooted to cost £10,000. This huge sum of money was justified by the instrument's ability to run several synthesis engines simultaneously. There was even an example of it on show. It was big, it was blue, it was called OASYS, (the Open Architecture SYnthesis System) and it looked the business. But we couldn't touch it, we couldn't play it, and we most certainly couldn't hear it.
The OASYS concept was an appealing one and, in 1995, quite radical. To quote Mr McNally, whom I subsequently interviewed, "OASYS is basically a computer, rather than a hard-wired keyboard synthesizer. If you want to change how it works you can load a completely different synthesis system from a hard drive. It's multitimbral as well as polyphonic, and can also be multitimbral in the sense that different types of synthesis can be positioned under different areas of the keyboard. For example, you can have a single patch which, when you play softly at the top of the keyboard, gives you an FM sound layered with an analogue sound, but gives you a physical model of a saxophone when you play a little bit harder, and a PCM sample of a pipe organ when you bring in the ribbon controller... It's all completely controllable."
Umm... no, it wasn't. The 'Blue Bomber' (as it was later called within Korg) never appeared. The reason was that it didn't work. Apart from a few limited demos, it had never worked, and it was never going to work. The technology available in 1995 simply wasn't up to the task.
Four years later, the second incarnation of the OASYS appeared. The OASYS PCI card incorporated many of the ideas that had formed the bedrock of the OASYS ideal. It boasted a 12-channel mixer at the centre of the system, and this provided access to some superb physically modelled synth engines, more than 100 excellent effects, a PCM-based synthesis engine with multisample loading and manipulation, audio streaming from hard disk, 32 MIDI channels of control, plus 24-bit audio I/O. The difference was that, whereas the original OASYS had been planned as a synthesizer with a computer inside, the OASYS PCI was a synth on a card designed to go inside a computer.
In some ways, OASYS PCI was a triumph. Its synthesis was superb and, when Korg released it, they announced that Synth Kit itself would soon be made available, thus allowing users to design their own models. Unfortunately, it had a fatal flaw. When used to the full, the polyphony dropped to as few as four or even three notes. So, while OASYS PCI was very close to McNally's description of the mythical OASYS keyboard, you couldn't do anything useful with it. Furthermore, while Synth Kit eventually appeared, the anticipated slew of third-party synth models and effects algorithms did not.
OASYS PCI was not a success, and it passed through the world causing barely a ripple on the consciousness of music makers anywhere. But the promise of 1995 seems finally came to have come to fruition with the appearance of the third product to bear the OASYS name, the Open Architecture Synthesis Studio. The first production OASYS (serial number 000001) is in my studio as I type this, and it's the unit you can see in these pictures. As you can see, it's beautiful.
Bringing OASYS To The Market
Korg faced a dilemma when developing the OASYS. Unlike previous workstations, whose capabilities and limitations are decided when their various VLSI chips are fabricated, Korg have stated that the OASYS specification is a work-in-progress, limited only by programming and marketing decisions. You want some more of this? Sure. Some more of that? No problem.
This has led to an enormous wish list that encompasses the sound engines, effects, sequencing and recording capabilities. Korg couldn't incorporate every idea in the first release version of the OASYS's operating system, so they had to decide what was a priority, and what could be the subject of later upgrades. In the end, they decided that sound generation, sound quality, and the KARMA engine were the most important aspects of OASYS, so they concentrated on finishing the three synth engines that it boasts today.
However, that's not to say that other aspects of the OASYS, such as the sequencer and recorder, are unfinished. I've had access to them for two months, and can attest that they are fully functional, and seem robust in operation. However, I think it's safe to say that we will see a great deal of development in these areas over the coming months.
Every once in a while a keyboard appears that captures your imagination even before you start to play it — the ARP Quadra, the Roland Jupiter 8, and the Yamaha GX1 all spring to mind — and the OASYS is one of these. But unlike the colourful Quadra and Jupe, and the physically imposing Yamaha, the OASYS is slightly understated. When you sit in front of it, it says, "Yes, I'm bloody impressive, but in an unassuming sort of way. If you're a serious musician, I'm exactly what you need."
Touching the OASYS confirms what your eyes are telling you. Take the keyboard as an example. Korg has pursued a policy for nearly two decades of building two families of synthesizers based on any given technology: a flagship line that incorporates expensive keyboards, and a more affordable line with low-cost keyboards. It's clear which one they've used in the OASYS; the keyboard is a pleasure to play.
More imposing yet is the screen. Everybody must know by now that the OASYS is essentially a PC at heart, and that it is edited primarily via a 10.4-inch colour, touch-sensitive screen, yet this wasn't always the plan. Apparently, OASYS started life with a screen the same size as a Triton's. But, at one of the early development meetings, Tsutomu Katoh, the founder and President of Korg, entered the room, saw the drawings and said simply, "Too small... screen too small... make screen bigger". Soon afterward, an email was circulated to announce that the synth was to have the screen that it now boasts.
So, what's it like to use a 10.4-inch colour, touch-sensitive screen on a synthesizer? The answer, in my best American accent, is "awesome, dude". As regular readers of Sound On Sound will know, I still use my Trinity Pro as both a major sound source and as my controller keyboard, and one of the most significant reasons for this is because the touchscreen works so well. Although small by modern standards, it is responsive, has no faulty pixels or lines across it, and it's angled by just the right amount to make it useable in almost all situations. On OASYS, it's simply excellent to have a large, bright, touch-sensitive screen that you can place at the angle of your choosing. What's more, by avoiding Microsoft's Windows and Apple's Mac OS X, Korg have been able to write application software that is perfectly tailored for the screen, and I am convinced that it would not have been possible to equal its clarity within the standard operating systems. I do, however, have a reservation. Having owned a laptop whose hinge mechanism sheared on more than one occasion, I'm concerned about whether the clutch mechanisms supporting the OASYS screen are strong enough. The screen is supported only by its clutches, and that's a lot of weight to be bearing backward week in, week out. What's more, unlike the screen on a laptop, which suffers no poking or prodding, you're going to be stabbing at the OASYS screen from the day that you start to use it. I have discovered that the hinges are only guaranteed for 2000 movements, which may be fine in the studio, but which may well prove inadequate for long-term, regular live use. I'm surprised there's no support or brace, like the one that supports a Minimoog's control panel. I wonder whether Korg might consider this.
As for the rest of the controls, it's extremely rare for everything on a synthesizer or workstation to fall to hand so correctly. How many times have you felt that it would be nice if this switch were over there, or that fader were over here? I find few keyboards to be ergonomically satisfying, and the last with which I was really comfortable was the Trinity. Well... it's taken 10 years, but there's now another, and it's the OASYS.
If you haven't yet used an OASYS, and are just looking at these pictures, it's important to realise that (with the exception of the value slider and up/down buttons) you edit it using the touchscreen and the switches to the right of the screen, and you control it using the performance controls and the knobs and switches to its left. This means that, for example, if you want to create a huge filter sweep in the middle of a song, you should first assign the cutoff frequency to a knob or fader, not try to control it directly from the screen. You can do the latter, of course, but why would you? It's the wrong way to go about it.
Once you've got to grips with this basic philosophy, everything falls to hand: not just to the extent that editing sounds becomes a doddle, but putting together whole songs using synthesized sounds, samples, sequences and audio recorded to disk is remarkably straightforward. I would venture that, notwithstanding some criticisms (of which more later, and next month), there has never been a keyboard that is closer to being a complete and useable music production system.
OASYS — Or POCKY?
I work in Japan for about 10 days each year, and I am a sucker for the snacks available from the all-night convenience stores spotted every hundred metres or so along every major street in Tokyo. There are cute little sandwiches, surprisingly good sushi and bento boxes — and irresistible chocolate biscuit sticks which go by the name of 'Pocky'.
Well, it seems that Korg employee Steve McNally shares my taste for Pocky sticks, because (so the story goes) some years ago, he was nibbling away at a packet in a development meeting when Korg's president, Tsutomu Katoh, walked in and grabbed one of them. Apparently, Steve slapped his hand away and said something (probably through a mouthful of chocolate) that boiled down to, "Oi, get your own". Katoh growled, everybody laughed, and Katoh then left the room, only to reappear half an hour later with about 150 packs of the things, which he dumped on the table, before leaving again.
Now fast-forward several years, to the rebirth of OASYS. As you may know (or not), Korg products are developed by individual country groups that work in specific areas of the music industry. For example, Korg Italy is responsible for domestically oriented instruments such as the PA1X, while Japan has traditionally developed workstations such as the Triton. Anyway, in 2001, Korg R&D (the Californian group responsible for Synth Kit, the original OASYS, and OASYS PCI) presented a proposal to Korg Japan. This suggested that the concept of an updateable, multi-synthesis, open-architecture keyboard was now feasible, and could be developed provided that they and Korg Japan worked in parallel to design and build it. It was a radical proposal, because no other major instrument had crossed geographic boundaries in that way, so the senior management asked for a Proof Of Concept for Korg (Japan) to be presented by the end of the Year. In short, a POCKY.
I have picked up hints that — for a while, at least — this may have been an informal way of referring to the new project amongst chocoholic Korg employees. However, by the time the world heard of it, the product was once again named OASYS. In my view, that was a very good decision!
Although Korg like to think of the concept of open architecture synthesis as their own, the 10-year delay between the dream of OASYS and the reality has meant that the company has been beaten to the market not once, but several times, by very different systems that offer the same promise.
One of these was the Roland VariOS, which Roland call an 'Open System Module'. This is a configurable module that — in principle — will allow you to load new synthesis engines, effects, and production tools as and when they become available. Unfortunately, VariOS was launched in 2003 with three modules (V-Producer, VariOS 303 and VariOS 8) and, two years later, these are still the only modules available. The blame for this lies entirely with Roland, who never released a third-party developers' kit. Not surprisingly, few people have chosen to investigate the potential of VariOS.
More significantly, the past few years have seen the appearance of the keyboard/computer; most notably the Open Labs Neko, which was reviewed in Sound On Sound in December 2004 (see www.soundonsound.com/sos/jan05/articles/openlabsneko.htm). Unashamedly a marriage between a hardware control surface in a keyboard-shaped case and a PC running Windows (albeit with a shell operating system to hide the Windows core from the user), you might think that this offers everything that the OASYS does.
Some have taken this line of thinking further. Surely, they say, you could emulate the OASYS's sound generators with a PCM-based software synth, a virtual analogue software synth, and a Hammond software synth? Add a software sampler, a competent MIDI sequencer with audio recording capabilities, and a bunch of VST effects, and you have the lot, at a fraction of the cost. But as you can probably tell, I don't think this argument stands up, and here's why.
The heart of the OASYS is indeed a 2.8GHz Pentium 4 PC with 1GB of RAM. However, instead of running Windows, the boot system is a cut-down Linux core (Linux was derived from the Unix operating system, and is often used for critical applications where stability and reliability are of paramount importance). In the OASYS, there is just enough Linux present to make the computer boot and load Korg's proprietary application software. This confers a huge benefit. You have to understand that neither Windows nor Mac OS X was designed to be a real-time operating system. If, for example, Windows decides that it wants to check whether there is a modem present, or to update some of its internal parameters, it will interrupt whatever applications are running and do so. This is of no consequence if you're performing a spot of word processing or browsing the web, but if you're sending eight channels of 24-bit, 96kHz audio out of the PC in question, it can be disastrous. The only way around this is to disable as much of Windows as you can (which is one of the major elements of optimising PCs for audio applications) and to use software that has large enough buffers to bridge the gap when the processor goes off to gaze at Bill Gates' navel.
So, what's the big deal about OASYS? It should now be obvious. If you ask a PC to play more voices, or use more effects, or output more audio tracks, there will come a point where the CPU can no longer cope with the demands being placed upon it by numerous disparate applications. Anyone who uses such a system knows what happens next. At that moment, it takes just the tiniest push to shove the whole thing over the edge, and you get a glitch, a freeze, or the 'blue screen of death'. Sometimes you can defer the point of no return by increasing the latency of the system, but that's often not acceptable in a musical environment. On the OASYS, you can continue to ask for voices, effects, samples, and streamed audio, and when you decide to add another channel of VA synthesis, or another Hammond (or whatever), the synth simply says, 'I can't do it all, mate...' and steals a voice from somewhere. Because the whole system is integrated, it can cope with excessive demands in a way that a Mac or PC cannot.
Integration confers other benefits, the most significant of which is speed. Let's say that you're the arranger and keyboard player for a West End production, you do 'spots' on a TV show, and you play in a rock band for fun (if so, congratulations!). Each of these has different requirements, and the modern way to cope with them is to couple a portable PC offering sequencing and software-synthesis capabilities to a controller keyboard and a small selection of rackmount modules. It's not the most reliable of setups, and most players live in fear of glitches and crashes. On the OASYS, you hit a button or two, and everything is done; the appropriate synths and sounds are loaded, the sequencer is primed and ready to roll, and all the pre-recorded audio is sitting behind the internal mixer ready to be triggered. To be fair, this is not new — a handful of workstations offer similar facilities — but none promise to cope with everything so elegantly.
Returning to the issues at hand, can we really say, therefore, that the OASYS is an open system? There is as yet no third-party developers' kit, and it cannot handle software that conforms to VST, MAS, DXi, or any of the other common 'open' application formats. Korg's stated view is that the OASYS is as open as (for example) Digital Performer or Pro Tools, which require developers to write to the MAS or Audiosuite standards respectively, but I fear that this is a tad disingenuous. There are thousands of DP systems and hundreds of thousands of Pro Tools systems in use worldwide, which makes it viable for small companies to develop for them. It's daft to imagine that Korg will sell many thousands of OASYSs, so it's much less likely that third parties will develop for it, whether it's 'open', or not. This places a heavy burden upon Korg to provide new synthesis engines and other upgrades, or they may find that the lessons of the Roland VariOS come back to haunt them.
Big Screens, Little Screens
For the moment, however, all this is moot, because the OASYS's operating system does not support it. However, when I asked Korg whether they intended to provide support for external screens and control devices, the answer I obtained was encouraging. They said, "If enough people want this, we'll look into it". Time will tell.
As things stand, there are three sound engines in the OASYS. The first is a PCM-based engine called HD1 which, contrary to some of the uninformed comments you may have read on the Web, is far more than just an expanded Triton. The other two engines are expansion instruments, or 'EXis': the AL1 analogue modelling synth, and the CX3 organ. Despite the fact that they come pre-installed from day one, Korg's use of the term EXi for the AL1 and CX3 demonstrates that they see HD1 as the heart of the OASYS. This makes sense; workstations are invariably based on S&S, no matter what other facilities they host.
Currently, HD1 lives inside 1GB of RAM and, although there's a slot on the motherboard for a further Gigabyte, the OS is as yet unable to access it. Apparently, this upgrade is high on the list of Korg's priorities, and that's just as well, because the current arrangement imposes some unhealthy constraints on how you can use the system.
When you switch on OASYS, it loads approximately 303MB of PCM samples into its memory. Korg refers to this as the 'ROM' but, of course, it's not. The samples are held on the internal 40GB hard disk, so it's possible for the company to update or replace them at any time it chooses. Next, the OASYS can load one of two choices of EXs (EXtended sample) memory. EXs1 is another selection of multisamples occupying 314MB, while EXs2 is a grand piano (see the box over the page) that takes up a whopping 504MB. Ah, you're ahead of me... If you try to load all three sets of samples, you need approximately 1.1GB of RAM, and the OASYS has only 1GB. This means that your options are (i) the 'ROM' alone, (ii) the ROM plus EXs1, or (iii) the ROM plus EXs2. Any unallocated RAM (up to a maximum of 500MB) is then made available to the sampler, which we'll discuss next month.
While this system doesn't sound too bad, there's a problem. Firstly, if you want to switch between EXs1 and EXs2, you have to make your selection in the Global menu, switch the OASYS off and then back on again! Now, let's say that you've create a set of Programs that use the samples in EXs1. And let's say that you've placed them all in program Bank E so that you can find them quickly and easily. And let's say that you now want to load EXs2 to play the grand piano. Inevitably, every patch in Bank E is 'broken'. Likewise, you could program your piano sounds using the samples in EXs2, but these will go horribly silent if you invoke EXs1. Let's face it... that's clunky.
At some point, someone within Korg must have decided it was unacceptable to have the piano sounds mutually exclusive from much of the rest of the OASYS, so the designers took EXs2, shortened the loops and used fewer samples to create a lesser piano that occupies about 133MB of the 314MB in EXs1.
This, Korg claim, is a very respectable piano in its own right. In the version I am reviewing, 1.0.0, it loses one of the velocity splits (only the ff, f and mf samples have survived), and occupies just a single layer with no sympathetic resonance, so you would expect it to be less realistic than EXs2. Nevertheless, the difference is small. What's more, if you inspect the list of multisamples in EXs1 you'll find that the mp samples and the so-called 'Damper' sample are still available, so you can reconstruct something that sounds remarkably similar to EXs2, gain access to the other 200MB of EXs1 multisamples, and release 200MB for your own samples.
This brings us to the issue of editing Programs in HD1. As I stated at the start of the review, I'm not going to describe the details of this, nor am I going to trot out the specification. But perhaps I can give you an idea of what it's like to create sounds on the OASYS.
Put simply, it's great. You edit OASYS using the touchscreen to select things, and any combination of the Value slider, Up and Down buttons, or the data-entry wheel to adjust them. I found myself using the soft plastic end of a ballpoint pen to select items on screen, because this was more precise than my fingertips. Then, depending upon whether I was holding the pen as a 'righty' or a 'lefty', I could manipulate the slider and buttons to the left of the screen or the data-entry wheel to the right with the other hand. Very neat, and very comfortable.
Sound design will be second nature to anyone who has programmed any previous Korg workstation. In short (but in a long sentence), two oscillators feed two filter sections that feed two amplifier sections that feed an insert effects section that feed the master effects, the end result of which is sent to the outside world for your pleasure. If that sounds remarkably like an M1, that's not surprising. It is, although the power of each section — oscillators, filters, envelopes, LFOs and so on — is hugely greater. So, while there may be a lot of synthesizer here, it's not mind-numbingly complex.
You select pages (in effect, synthesis modules) using the two rows of tabs running along the bottom of the screen, and select parameters by touching... well, the parameters themselves. As you can see from the Program screen on the opposite page, the voice structure is surprisingly clear, and it's equally clear that a lot of thought went into the interface. For example, if you want to adjust the nature of Filter B in Filter1 of a dual-oscillator patch, you press the Filter tab, then the Filter1 tab that is thus revealed, and then the Filter B 'Type' control, selecting from the drop-down menu that appears. This may sound complex, but the beauty of the system is that, on a 10.4-inch touchscreen... it's a beautiful system!
If I had to single out one aspect of HD1 that's worthy of special mention, it would be the AMS Mixers (shown in the screengrab above). The AMS (Alternate Modulation Source) concept first appeared on the Trinity, and in essence replaced the patch leads and CV inputs of a modular analogue synth, allowing you to specify which source modulated which destination, and by how much. On the OASYS, Korg have taken the concept a step further by adding two AMS Mixers in each voice (or, to put it another way, up to four AMS Mixers in each Program). These offer six modes: A+B, AxB, Offset, Smoothing, Shape, and Quantise. I particularly like 'AxB' mode, because this allows you to control the amplitude of one modulator using another, just as you would using a VCA on a modular analogue synth. Given that the outputs from the two AMS Mixers are themselves available as AMS sources throughout the voice structure, this suggests no end of programming possibilities.
I would also like to compliment the OASYS on its patch-selection mechanism. Click on the Category area at the top of the Play screen and, rather than see an almost endless list of patches, you'll find a list of 17 categories (strings, brass, organs, lead synth, and so on) each with up to eight sub-categories. Click on one, then the other, and the thousands of Programs in the OASYS are narrowed down to exactly the type of sound you want, making selection painless. Bravo!
EXs2 — Piano Lessens
What constitutes a 'good' piano sound is a very personal matter, but few people would suggest that Korg have ever done other than lag behind the major players in the digital piano market. First, there was Kurzweil, and then there was Roland, and now... well, there's still Roland. In comparison with these, Korg's pianos have always sounded a little two-dimensional, and they have never achieved the same commercial success as other manufacturers' offerings. So, when Korg Japan were discussing what sounds to place in the OASYS, they decided that a superb piano was a top priority, and that it could occupy as much space as needed to get it right. The developers found that the required samples occupied more than 500MB, and thus was EXs2 born.
Faced with this, I removed all the effects from the Programs, inverted Korg's 'smiley face' EQ, and tweaked the curves and lengths of the envelopes a little. The sound improved considerably. Of course, the use of a conventional monitoring system ensured that I was not fooled into thinking that I was sitting at a nine-foot Bösendorfer (nor even my mum's Broadwood) but the modified Programs — although not up to the standard of the very best of the competition — now worked well. I am sure that all but the most demanding listeners would be happy to hear them in a mix and, with sympathetic playing and recording, many — myself included — could be fooled into thinking that they were listening to a 'real' piano. Of course, you may like Korg's Programs without having to go through these kinds of tweaks. As I said, piano sounds are very personal.
As you are probably aware, the OASYS offers extremely advanced effects, wave sequencing and KARMA to further spice up its sounds, and we'll discuss all of these next month. But I'm sure that you're dying to know how the HD1 Programs sound without all this fairy-dust being sprinkled across them because, unethical though the practice may be, the deficiencies of many a dull synthesizer have in the past been masked by built-in effects.
In short, HD1 can sound superb. Traditionally, I have used four brands of synths for specific characters of sounds. Kurzweils have provided orchestral sounds and the warm, low-end timbres for which they're famous, Yamahas have provided precise, percussive sounds, Rolands have provided the pianos, ethnic instruments and pads that sit unobtrusively in a mix, and my Korgs have always been used for the wonderful washes and textures that they do better than anything else. The OASYS breaks the mould. Its orchestral sounds are a revelation: more authentic and with more character than any previous Korg. Its percussive sounds are brighter and more precise than any previous Korg. It is capable of ethnic sounds and pads that work better than similar sounds on any previous Korg. As for the spacey textures and effects... well, you know what I'm going to say.
In part, this is because many of its PCMs are stereo, uncompressed and longer than those of other workstations. In part, the high internal processing speed means that the PCMs can be manipulated with high bandwidth and high precision, so they are clearer at high frequencies than ever before. In part, it's because its huge polyphony consigns note-stealing to history when using Programs. Furthermore, Korg's claims that the OASYS generates very little aliasing appear to be well-founded. Take something with a rich, high-frequency spectrum, and play it at the top end of the keyboard. Now apply an LFO to the pitch, and an octave of upward pitch-bend. On most synths, you would now be hearing all manner of unwanted frequencies polluting the sound. On OASYS, you hear little or none of this, and that's impressive.
However... much less impressive are the obvious multisample points on many of the sounds. To demonstrate this to yourself, select patch INTA024: 'Pipe Organ Mixture' or INTA032: 'Clav1/Mure Sw1', and play middle 'C'. Now play the 'B' a semitone below it. As you will hear on both Programs, the timbres change considerably. If you want to shock yourself, isolate multisample 0072: 'Pipe Organ Mixture2' in the former Program. Alternatively, play A4 and A#4 using 0116: 'Tin Whistle' — the two notes sound as if they come from two different types of instrument.
To be fair, generating large libraries of multisamples is difficult, and there are many examples of other manufacturers' instruments that suffer from the same problems. The silver lining here is that, because they are disk-based, the OASYS's sample sets can be replaced by improved versions. I hope that Korg will consider doing so.
Another criticism concerns Korg's misleading claims that the oscillators (and I quote from the brochure) "support four-way layering, switching and crossfading". This is not true. While you can load up to four samples in a single oscillator, you can't layer them four deep; only two can sound at any given moment. On a more positive note, the mechanism for inserting samples and crossfading or layering them and the way that these operations are represented on the touchscreen are object lessons for other designers.
While Korg can't claim to have invented wave sequencing (it's an extension of the wavetable synthesis made popular by PPG) or vector synthesis (this first came to prominence on the Prophet VS), it's fair to say that both these technologies reached maturity in the Korg Wavestation and its descendents. Happily, both these technologies are present in the OASYS, and in much improved forms.
Similarly, the Korg M1 may not have been the first synth with splits and layers, but it cemented the concept of the Combi (in other words, a selection of sounds distributed across the keyboard as required) whereupon other manufacturers adopted it as the ideal way to create complex layered sounds. Again, this technology is present in the OASYS, and again, it's more advanced than before.
However, I'm not going to say any more about these just yet because they are not used only by the HD1 engine, but also by the AL1 and CX3 engines. So this is where we'll pick up the story next month. Until then...