Navigation for These Hopeful Machines

phill at home Phill Macdonald live

Photos courtesy Phill Macdonald

New Zealander Phill Macdonald was the keyboard player in the band Marginal Era. After moving to the UK, he became product specialist for Korg UK and as one of Korg’s elite ‘MIDI Patch Boys’, worked on voicings for many of their synthesizers including the Trinity, Prophecy, Radias and MS2000 and their Legacy plug-ins. He later moved to Novation where he programmed factory sounds for their products including the Nova, Supernova and Super Bass Station. Now resident in New Zealand once more, he still produces sound for Novation but also has his own studio facility, Taufactor.

Taufactor website


Phill Macdonald interviewed by James Gardner, 29 March 2013. Corrected and revised by Phill Macdonald, June 2014.

James Gardner: What are some of the instruments you’ve done the programming for?

Phill Macdonald: Well, it started pretty much with the [Korg] Wavestation.[1] I was working for Korg UK as the product specialist there. And we developed a UK sound set for the Wavestation when it came out, which was subsequently incorporated in the Wavestation library and ended up in the Wavestation SR Rack. That lead to me joining what they call the MIDI Patch Boys [laughs]. It’s a collection of people that Korg have around the world for voicing purposes. And they tend to be from the major markets. So there was Michael Geisel from Germany, Michele Paciulli from Italy, Steve McNally from Canada, Jack Hotop, Andrew Lubman and Skippy Lehmkuhl from America and quite a few Japanese people and occasionally other people—some of them were artists as well.

So I joined them and did some work for the Trinity—I did a bunch of stuff for that, an interesting project called the Prophecy, which was one of the first physical modelling things. It had picked models and brass models and a whole lot of different stuff. I did some stuff for the X3, the Triton SE, the Triton Studio, MS2000, MS2000B. And some other products like the Legacy plug-ins. And then when I joined Novation I did the Bass Station Rack, I did the Drum Station and I did the Super Bass Station, the Supernova, the Nova, the Supernova II, and the Supernova II and Nova II keyboards.

Were you the sole programmer on those or was there a team?

Well in those particular [Novation] ones I was pretty much it, although there were contributors—we had Novation distributors that we used. But I’d say 90-95% of the programs, the performances and the arpeggiator patterns and all the associated gubbins.

There’s quite a lot of work in developing a preset. You know, people just go through the actual sounds and go “oh yes, yes, yes” or whatever, but there’s a lot of work involved in actually making sure all the modulation assignments are assigned and all that kind of stuff.

How did you get into that side of things in the first place?

Well, probably like yourself, I was brought up in the era where there were no memories! [laughs] And the only way to change a sound was to actually manipulate it on the fly. So you had to learn what everything did.

Who first said “would you like to program for us”?

I did do a little bit of programming for Casio over here, because I had a relationship with the Casio distributor here. But I actively pursued getting into the MIDI Patch Boys at Korg. That was my plan. You know, there are a lot of very, very experienced programmers in that team. It was like being thrown in the deep end!

Did you submit a portfolio of patches to show what you could do?

Well, two of the engineers at Korg and I did a bunch of sounds for the Wavestation and they heard them and then decided that they wanted to include them in their library, so...

How did you end up programming on the Wavestation, with those Korg engineers?

Well, as you know yourself, England is a major player in the music industry and the dance scene was really just beginning to hit in the early ‘90s and we felt that it wasn’t reflected in the voicing at that time...

You were already working for Korg at this time?

Yeah, I was their Head Product Specialist.

What did that mean, practically?

Well one aspect of the job is “I can’t get it to work—how do you make it work?” There’s that aspect of it. The other side of it is doing the exhibitions and trade shows; product training for dealers, and as it progressed, as I became involved in the patch development, it also became product training for distributors—for their product specialists. So that’s how it came about.

So you became aware that there was a gap—that what the customers wanted wasn’t being delivered by the product at the time.

Yeah, because there was no UK input at that point, and the House scene was just exploding at that point.

This would be the late ‘80s, then?

Yeah. So I kind of pushed to get that stuff in there. Subsequently they put me on a lot of projects that related to that market

You said before that there were different teams in different territories...

No, not different teams—there’s one team, but there’s people from the different territories in it.

Every market has its distinctive influences. Germany, for instance, has a great history of synthesizer music and America has a great Rock and R&B heritage.

Cultural things do have an influence, and indeed the dance scene is a cultural thing. Those programmers were also extremely good musicians and capable of most styles as players, and you were after a fair representation of the total market. Although there is always some common ground in different markets, at that particular time, House was huge in the UK.

Given that most people are going to flick through the presets in an instrument quite quickly, what was your brief for programming?

Ah, well there isn’t actually a brief. But you’re obviously influenced by your own market. So to some degree you become your own brief. Once you’ve done the sounds, the job is then to convince all the rest of the team that what you’ve actually done is actually relevant and sometimes you’ve got to fight your corner. 

It’s a bit different when you’re working on something like a workstation, because that caters for everything. So there’s loads of different styles from Arabic to whatever, and much more variety than there is in a synthesizer. So a synthesizer does tend to be more dance orientated, generally. But it’s a competition—we’re all quite competitive as well, so you end up kind of overdoing it to try and get as many of your sounds in as you possibly can, because it’s democratic—it’s voted on by all your peers. At least that’s how it was at Korg.

The process for a workstation would be—do all the sampling, then collectively make a decision about which samples go in it. Then do the programming using those samples, for the core sounds, and then make a selection of those collectively, from what’s there. And then go off and make combinations, or performances—whatever you like to call them—using those sounds. It’s a bit simpler in a synthesizer.

Initially that was all done face-to-face, but as the Internet kicked in it migrated to that method.

Were you under any pressure from the marketing department to angle the presets in an instrument in any particular direction—towards a particular genre?

Well, for Novation I was the marketing manager, so I chose the direction. For Korg I was given pretty much a free hand, but it wouldn’t work if I didn’t get results. I mean if none of my sounds were selected, I wouldn’t be there the next time. Generally, sound creation is usually left up to the programmer. Common sense prevails—for example a big-band brass sound wouldn’t be that useful in a trance bank. Sometimes there will be a specific style requested. It varies from product to product and from manufacturer to manufacturer.

Talk me through that selection process—programmers have submitted a bunch of sounds, and then they go through a selection panel. How does that work?

With a spreadsheet. You just have all the sounds. The way that we did it is really thorough because everyone comments on collectively get all the sounds, you share it amongst everyone and then you collectively comment and even suggest edits to each other, or pick up things like aliasing or something that you might not have noticed. And then gradually you hone them all, and then it’s just down to a vote. Then there’s another aspect, which is Quality Control—you’ve got to make sure there’s no clipping or aliasing or anything of that nature, and you’ve got to take full advantage of all the modulation possibilities.

And this is among the programmers themselves, it’s not steered or driven by a marketing department, but I’m sure every company is different. But yeah, it’s very democratic. And everyone wants the power-up position. [laughs]

How is that decided, you know, ‘Sound No.1’—the sound you hear when you first turn the machine on?

Well I call it the pole position.  It comes down to votes on the spreadsheet. Obviously, for any member of that team, that’s where they wanna be [laughs] and is competitive. But it just makes the product better.

Because that sound has got to have some ‘wow’ factor...

Oh yeah, absolutely—it’s got to knock your socks off. The first 16 have to knock your socks off. Preferably the whole lot, but there will always be better sounds than others in the bank. You want the most impressive, startling, stop-you-in-your-tracks kind of sound in the power-up position. And when you’re programming, you’re thinking about that all the time. [laughs]. You’re continually trying to think of different ways to wring its neck to make sure you can get the last drop of blood out of it. It is a competitive environment, that’s for sure.

Would you be working on a prototype model, or a circuit on a breadboard or what?

It varies. It could be a simulation running on a card in a computer or it could be just software. For Korg it was usually a complete, fully functional piece of hardware. Occasionally it’s a prototype in a box, or a circuit board on a plank. Sometimes you’d be thinking “have I done some weird modulation thing, or is it a bug?” You’ve got to wrestle with that sometimes. But generally it’s fully functional by the time you get your hands on it.  Either way, there will usually be updates as you’re developing the sounds. And additionally, you’ll be kind of making suggestions, perhaps, to changes to the operating system or parameters back to the manufacturer during that process.

Presumably you’re working without a manual, too?

Yeah. That wasn’t so much of a problem for me at Novation, because I wrote the manual!... [laughs] Usually there is a Draft Manual for the software engineers.

So you were brought in at a much earlier stage with the Novation stuff than you were with the Korg products.

Well not necessarily. They [Korg] obviously did take input from the guys putting the voicing in, you know: “if we could control this parameter it could be good”, so it is a collaboration, but generally with Korg stuff it was 100% done by the time you got hold of it.

What’s the most challenging aspect of coming up with new sounds?

That’s the $64,000 question, that one. The most challenging is trying to dream up the sound in the first place. Everyone’s got their own way of programming, so I wouldn’t dare to tell anyone how to do it, but from my perspective there’s two ways of programming. One of them is the Happy Accident, where you’re just messing about and you go “ooh - that sounds a bit interesting”. But nine times out of ten that generally ends up sounding like someone standing on a cat, because you haven’t done it in a methodical way. What I tend to do is either dream up a name and make a sound that reflects that. Or I’ll actually think “ah ­ if I did this, and this and this and this, it would possibly do something really quite interesting”. Those are the two methods that I mainly use to try and devise something. But the difficult thing is really coming up with the idea. And I’ve always found that the best way to come up with the idea is to walk the dog. And for that reason I gave Poppy, our dog, a credit in the Novation manuals. [laughs]

If there’s a new kind of sound-generating capability in a synthesizer, do you start by trying to show that off with your programmes?

Yeah, you try, but it depends on the product. With something like a straight virtual analogue synth—“Dowww”—you know, really that’s about it. But we’re always trying to find ways to show it off in a unique way. The [Korg] Prophecy was easy at that point[2] because it was one of the first mass-market modelling machines. The [Novation] UltraNova, although it’s a virtual analogue, it’s got a whole load of wavetable stuff—like a PPG—in it, so I went straight for that.

Sweeping through the wavetables and so on.

Yeah. And you can actually morph from one wave to another. It’s not abrupt like a PPG was. You would go for that, but you’ve got to have all your meat and potatoes in there as well. I mean obviously there will always be a certain amount of plagiarism.

Because you’ve got to have a Rhodes and a synth bass and so on...

 Yeah. So...the hardest thing is to actually just to come up with an idea for a fresh one. And lately the way I’ve been doing it is this combination of different things, like you’ll have wavetables, arpeggios and synchronized LFO modulation all doing different things at the same time, but as a team, to really manipulate the sound in a unique way. It’s like anything – the more stuff you have, the more possibilities there are.

With the way things are now, with things like Native Instruments’ Massive, and stuff like that...well I don’t envy your position as a teacher, saying “Here’s Massive”...

And then, three years later...

Yeah, you know—“Go off and read the manual!” I mean you’ve obviously been through the same thing, so it’s been an evolutionary process for us. So for someone to learn it in a year and become proficient is a challenge. And the other thing I’ve found is that writing patches has become like a production job, because of the kind of sounds you have to include. So it’s also important to include all the meat and potatoes – the stuff that really works. A really great [Roland] Juno 106 bass, with that legendary kind of bottom end. That’s got to be in there. All those really great bass sounds, classic pads and strings etc. ‘cause you know they work.

I’ve found that in England a lot of the producers are around 40 or 50. I think that’s because you have to be able to draw on your experience to be able to say “well, no that’s rubbish” – not necessarily just the song but also the sound. I’m not sure whether you can teach that. Some people have just got that knack for the production side of it. Whereas a lot of really talented musicians are so focused on their technique that they’re not actually focusing on what their tone is like.

What kind of changes have you noticed since you started programming in the late 80s? Obviously the gear has evolved but have the expectations also changed?

When I joined Novation, we had a specific goal to make one of the best virtual analogue synthesizers you could get. So the Supernova had a much better architecture than a classic analogue synthesizer, and we included multiple effects in each part, separate arpeggiators in each part, and masses of voices.  And a whole lot of knobs, because I hate menus. And it was a great success. Things like the Access Virus had taken it a step further by adding more polyphony and a whole lot of sampled waveforms. The UltraNova’s got a whole lot of sampled waveforms in there as well. And with so much at hand you’re like a kid in a candy store, and the temptation is to kind of go mad. And with the Supernova you could. That’s cool for huge kind of evolving film score pads, but sometimes you have to pull in the reins. But having said that, I can’t see anything startlingly new.

It was interesting in the early ‘80s. When the [Roland] Jupiter 8 came out, everyone’s jaw dropped, you know. And then when the [Yamaha] DX-7 came out the same thing happened, and then when the [Roland] D-50 came out the same thing happened, and then when the [Korg] M1 came out the same thing happened. And we’re still moving forward in that way. I’m just not sure how much further we can actually go.

But at the same time we have so much more power at our fingertips with a laptop than we did 30 years ago.

Yeah. I mean you look back at it, the DX-7 in the ‘80s was bad enough for complexity but nowadays, Massive, the Massive plug-in, is truly massive, and that’s the nature of the beast. The more we can make it do, the more complex it will become. But I’m talking about something that’s small and portable and easy to understand, yet startlingly different. I don’t think we’re going to see quite that. I think it’s going to be more an evolution and a better way of dealing with the complexity of it.

Have you seen a shift in the degree to which real-time performance skills are part of what the product is or is it still as it was in the 80s, say?

Are you talking necessarily about knob-twiddling?

The hardware interface, yes.

Yeah, I mean I personally do like “tweaking the knob”—you feel more involved, rather than trying to do it with automation on a DAW. There have been some interesting kind of innovations recently—the MiniNova has a whole lot of pads in it like the Launchpad, so you can kind of engage different modulation matrixes with the buttons, which is really good for Dubstep.

But the hardware will evolve and to some degree I think will be driven side by side with the kind of DJ side of things. The two things are kind of merging.

Do you think more controllers will be along those lines, where you have something that isn’t a keyboard, which doesn’t look like a traditional musical instrument?

Oh definitely! Jordan Rudess has got a company[3] that makes apps for iPad and Windows 8. They’re touch-screen-driven—they don’t have a keyboard. So they’re already out there.

Although the technology’s been around for a long time to make other kinds of controllers, they’ve never really caught on. But maybe the time is right for those non-keyboard devices to take off.

Yeah, possibly. Possibly. But it’s like anything—you never really can tell. I mean Jordan’s a fantastic player—absolutely gifted beyond belief—but then again some other players that, like him, have been classically trained—they may not embrace that. So I think it might be more in line with someone who’s perhaps like a DJ who has musical aspirations rather than trained players.

But surely a matrix of pads is more in line with how you think about making ‘DJ-type music’ than a conventional keyboard, anyway.

Yeah, I mean Logic’s particularly good at pattern things, so yes—mute, mute, mute; unmute, mute, unmute. Yes it’s definitely the way to go, though I guess it depends on which market you’re aiming at.

Given how retro things have become, do you find that people want to use old sounds more than new ones?

That’s an interesting question, because there’s two ways of thinking about this.

You can say “OK, I want a nice big fat chorused bass, like a Moog bass” but if you speak to some producers they might say “Nah. No. It’s got to be one oscillator, because you get better punch” And they’re right. But if you listen to it on its own it’s really unimpressive. [laughs] So sometimes that’s a bit of a compromise—something that you know works, but doesn’t stop you in your tracks. So there’s always a kind of balancing act with that kind of stuff. And also you risk alienating certain sectors if you’re going for one particular market. The other thing, of course, is that some sounds, like the “ai-ee, ai-ee, ai-ee” thing that’s happening nowadays in Dubstep—it’s really interesting now, but will it be interesting in two years’ time?

Is the development cycle for new products quicker than it used to be?

No. I’d say it’s longer, because the modulation matrixes and modulation possibilities and all that, is more complex and there’s more of it.

And some technologies aren’t as forgiving as analogue. Digital clipping, for instance. You can have a little bit of grunge happening in the old Jupiter, but you really don’t want it in a digital synthesizer. So you’ve got to be very meticulous about actually kind of going through it. The more you add, the more quality control you’ve got to do. And there are more mundane things like trying to make sure that the level of the patches in a given bank are the same. There’s a lot to consider. The more the merrier, but the longer it takes. But the better it is.

You were saying a minute ago that a one-oscillator bass sound might be very useful in a production but it doesn’t sound very impressive on its own. Conversely, some sounds that “knock your socks off”, as you put it, are practically unusable in any actual song because they take up so much space. Do you find that?

Absolutely, yes, and that’s the danger of the “pole position”! [laughs]. Yeah, you’ve got to be careful about that because... I mean something like the Radias has four parts that you can have in one patch, so you can have a bass, a drum kit...and you can put one finger down and out comes a song, virtually. That’s how big it can get. But then again, if it sounds good...



[1] Introduced in 1990.

[2] The Korg Prophecy was introduced in 1995.

[3] Wizdom Music. <>