Hello, and welcome to the Sound On Sound people and music industry podcast with me, Sam Ingalls. Under normal circumstances, I would hope to be bringing you this episode from the NAMM show in Anaheim, California. Unfortunately, these are not normal circumstances. And so we're forced to do it via zoom instead, but it still gives me great pleasure to introduce Tony Agnello and Richard Factor from Eventide. 

Great to be here. Thank you for having us.

So the first thing I need to say is congratulations because 2021 actually marks half a century of Eventide.  That's an amazing achievement. Are you going to be doing anything special to mark it? So we're going to be celebrating this, uh, this anniversary all year.

Yeah, we were hoping to have, uh,  We're going to celebrate, but obviously with COVID, that's, that's going to be put off. So if we can all get, you know, be,  take off our masks and be together somewhere, December 31st, I'd be all for that. It would still be our 50th anniversary.  And if not, maybe we'll do it for our 51st anniversary. 

So let's talk about those early days because back in the 60s and 70s It was actually quite normal for studios to design and build their own equipment and very often There wasn't anything on the wider market that actually met their needs. Is that how Eventide got started?  

It's it's literally how we got started  Because, uh, my buddy Steve Katz, who's one of the, uh, founders, had, had this, uh, studio, which was very small,  and it had an MM 1000, which is a 16 track, enormous, physically enormous tape recorder,  and he had no room for an assistant,  so, uh, he, he came to me and said, Hey, you know, I, uh, I don't have room for an assistant, I want to be able to find something on the tape. 

And I can't just turn around and do it and waste time in the session. How do I do that? And it was just the very beginning of integrated circuits. And I put together a tape search unit. So the studio needed it. Um,  we built it  and the next thing you know, uh, we're in business making these things, uh, with, OEM for Ampex. 

My mother said to me,  you're going to quit your job. You got a job. You got a day job. You're you're like. I'm getting paid.  And I said, well, you know, what's the worst that can happen is this is gonna fail  and I'll get another job. Uh, I could, I could have done that. You know, I,  I had skills. So,  uh, I quit my day job in the military industrial complex, started Eventide.

My mom looked up in horror and, uh,  Here we are. Uh, no, we, we did not anticipate anything other than let's give it 

a try.  At this time, of course, the studio world in general was almost entirely analog, but even Tide started making digital products almost from the very beginning. What was it that made you see the potential for applying this new digital technology to audio? 

I, I can actually answer that very well. I was working in the military industrial complex.  And, uh, we made things with integrated circuits for, uh, for audio. We made spectrum analyzers and, uh, digital filters. And one day I just listened to the input and output of, of this very exotic device that, uh,  you know, nowadays would, would fit on, uh, on a fingernail.

But back then was a rack mount gadget. And I said, boy, that's really interesting. I've never heard anything like that. That was about a year or two before, uh, before Eventide started. And I said,  This is neat, because, uh, all we had before then were, uh, were analog components and maybe a plate echo machine. 

So, that inspired me.  

So 

this was a military 

device?  

It was intended for military. I don't even remember exactly what the final app was. In fact, it may have been secret. Nobody ever told me, but it was a thing called a matched filter and it had time delay and accidental reverb because you couldn't get rid of that then. 

So it was quite the interesting device.  

So your work on digital audio led you to create the product that was the first really big Eventide hit, the H910 Harmonizer. What was the key innovation there?  

Well, the key, I guess the key innovation was, was, um, incorporating a number of controls to take advantage of pitch change and time delay. 

Being able to set the length of time delay and feedback around it. Um, but Truth be told, what I thought I was designing was a musical instrument. So the prototype of the H 910  actually had an integrated keyboard.  Pressing middle C would always give you unison, but you could move up or down the keyboard to create harmonies around your part. 

And a number of musicians, I think Elton John toured with it, um, you know, to create harmonies in real time.  I don't really remember how we made a decision not to make a musical instrument, other than uh, packaging the electronics as a rack mount product was much easier for us. And so we did that and offered the keyboard as a control accessory.

And we did sell some of those keyboards, Richard, I remember. I remember building some of those, quite a few.  So that's, that's really what my intent was. Now, you know, I had the advantage, as Richard said, you know, we had a small studio in the same building. And when I got my started Eventide, I was doing a lot of testing of delay lines.

And, um, the ICs in that time period, uh,  were a bit dodgy. Um, they, they suffered from something called infant mortality. So we would have these units, we'd, we'd Produce them, but we need to keep them around burning them in for several days before we, uh, risk shipping them to actual customers. So, um, so I had the luxury, you know, I was in bands and I was doing some recording of, of having, uh, more delay lines, I think, than any, any one else on the planet at the time. 

And I could drag them up to the studio, put them on several tracks, use the patch bay to feed, feed around them. And, uh, just.  discovered a lot of cool effects and thought it all could be packaged in one neat unit, and that was the Harmonizer. Well, if you go back to, uh, the very early 70s, there really weren't, I can't really think of other Rackman effects units.

You know, as Richard said, ICs were still new. So what you would find in recording studios, you know, other than the console and mics and speakers, were,  uh, EQs, compressors, limiters.  Um, but, you know, I really can't think of effects devices, you know.  Dave Davies, you know, cut up his speakers to get effects. Uh,  there were a lot of mechanical effects.

Tape was, well, well, since the 40s, I guess Les Paul, uh, you know, was the pioneer there, you know, speeding up tapes, uh, editing and reversing segments of tapes, using tape delay. No, no question that  tapes, that the tape machine was used for effects, absolutely. But that's, you know, that, that's of, uh, a different sort than a, a box that's, uh, that's a, that's...

All electronics that sits  in a, you know, in a rack with knobs. 

Of course there was the Phaser, but, uh, that was us. Instant Phaser. 

Right, the Phaser was probably, yeah, I guess we could make the point that the Phaser was the first, uh, pro audio rack mount effects box writ large.  

So it sounds as though it was quite important for you that you had this kind of symbiotic relationship with a working studio where you could test out your prototypes and your new ideas on real projects.

Absolutely.  And playing music and recording music.  And it was really all about music. I mean, Richard was playing drums, I was playing guitar. 

I was pretending to play drums. I was, uh, 

yeah, yeah, yeah. I like to joke that we're, uh, you know, we play instruments poorly, but we play.  Still play. We also, um, at least a couple of sessions, Richard, we, uh, we used a recording studio to actually do some recording. 

And how did that real world experience inform the design of the early harmonizers, for 

instance? Richard designed three very early delay lines, and the first two were shift register based. But then random axis memory became available, and Richard designed a 1745M,  which had RAM.  And it was a modular device, and he, um, designed a module to do pitch change. 

So once again, I had access to these things and I was recording. So I quickly discovered that, gee, if you do a little bit of pitch change, you can really widen, you know, the vocal or a guitar lead, and maybe a little bit more pitch change and feedback around it and you arpeggiate the sound. So it was really using the, you know, what Richard had had designed very early on that informed me. 

In designing the Harmonizer. By the late 1970s,  

Eventide processors were incredibly popular and they were being used on practically every record ever made. Could you always hear them when you listened to those records?  

Only if it really stood out. You know, bear in mind that once these devices got into studios,  and I've had this conversation with, uh, with producers and engineers from, from the 70s, and if you asked them what they used...

Our devices on. It was pretty much everything. So you might use just a little bit of pitch change, you know,  to, uh, to have the vocal sound a little bit fuller. You might use a little bit of flanging on a snare or a hi hat. Um, so they were putting these, they were, they were putting these effects,  um, used in, in fairly subtle ways on many, many tracks.

And we certainly didn't hear that. Yeah.  Also bear in mind, you know, stereo was still fairly new and this was a way to spread the image. So if you look back at, uh, and we have some photos of some studios back in the day, they would just stack up racks of our, of our gear and not to use, use the gear or the effects in a dramatic way, but to use them in a, in a subtle way to sweeten the sound.

If someone used it noticeably like, um, Visconti used it on, uh, ashes to Ashes, David Bowie song where there's that quirky warbly keyboard sound. Um, you know, and that was, and that was pretty noticeable. Of course, he wouldn't tell anyone how he did it, but  

we kinda knew.  And did people ever come up with uses for the products that you simply never anticipated? 

A lot of, uh, TV broadcasters would use, uh, the harmonizer or the, uh, pitch changer or whatever.  To, uh, to get an extra minute of airtime or an extra minute of advertising in a, in a program.  And, uh, you could do it without, without pitch change because they'd speed up the tape and lower the pitch.  And, uh, that would, uh, mean that the program, uh, say a 30 minute program would be over in 29 minutes.

You could slip in a local spot or whatever, uh, and 

make 

some bucks.  

One of the technologies that has really changed pop music over the last 20, 25 years is automated pitch correction. And that's something that was possible up to a point with the earlier harmonizers. Um, but it's also something that you actually anticipated.

For 

this 50th anniversary, I've been opening up some, uh, boxes that I've had packed away for, for decades. And I came across a, uh, uh, a patent disclosure from, what was it, like 80, 80, uh, 79 81 about 10 years before autotunes was patented where we'd actually described, you know, where we proposed autotune. I think the patent, either the patent attorney was too expensive or we just thought it was a pretty obvious step.

So we'd never never pursued it. But yeah, we knew that was happening and we could see it coming and right around 1980 79 80 is when Apple and Sinclair We're first offering computers and obviously you would need a computer to truly do autotune to dial things in properly, but you could see it coming. It  really seemed obvious.

And I was pretty amazed when 10 years later,  the patent office actually granted a patent for what seems  so obvious.  

Another Eventide milestone product was the SP 2016, which I've seen described as the first programmable processor, but also as the first product ever to support plugins. Um, tell us a bit about that.

It was an array processor, so it was a general purpose programmable unit. It wasn't purpose built to do reverb or pitch change or anything in particular. It was completely programmable. So, you know, in today's world, you know, we have Modern, uh, CPUs, mo modern processors that have so much processing power that they can handle, uh, audio processing in real time.

That wasn't the case in the late seventies. Um, so the 2016 actually pre predates DSP chips, which came along, uh, for instance, like the H 3000 and, and every general purpose audio processor from then on, used DSP chips. But, uh, the 2016 was early on, uh, DSP chips were not available.  Um, I understood what it would take to do real time, uh, audio processing. 

And, uh, built an array processor, which required lots of, lots of ICs, fairly, uh, simple ICs. Um, and built it in such a way that, uh, people could plug in ROMs, which would have the code to run any, any audio effect. Um, it got pretty popular because we, we managed to develop some really great sounding reverbs for it, but it also scrambled time, uh, which is more like a granular synthesis effect today.

Uh, it had vocoder,  um, programs, um, you know, it had, it was a looper. 

Don't forget Band 

Delay. And Band Delay, which, uh, Suzanne Chani used, used famous, famously and others, you know, where you could, where you could create a, a number of, Um, digital band pass filters and delay each band by a different differential amount and recombine them.

So, it was a, it was a general purpose signal processor and, uh, you know, at least one, one third party, uh, created some effects for it. And people who own this box could buy, you know, buy a plug in, buy a, a, actually, it was physically a IC that you would plug in to add a new effect.  But, you know, it became obsolete once DSP chips became available.

It was much more cost effective to use a CPU that was specifically designed to act like an array processor. So 

as well as the processors that we all know and love, Eventide also made a lot of obscure OEM gadgets and things. What's the weirdest thing that Eventide ever made?  

A buddy of mine who was working at the New York State Health Department Wanted, uh, he had this notion that you could, uh, dispense micro liters of reagents using, uh, a vibrating, um, vibrating transducer and a static electrostatic field. 

And, uh, you could literally calibrate this thing in droplets, nanoliter droplets.  I was, uh, asked, could I build one of these things and we used, uh, I think it was, uh, like a Teflon, uh, Teflon ring.  Uh, I, I don't even remember. I remember this is 50 years, five, zero years ago, literally more than that, possibly. 

But, uh, anyway, uh, made this thing with a, with a thumb wheel switch that basically, uh, you could adjust how many drops. Nanoliter droplets of a chemical reagent could, uh, could go into a cup.  And  that was weird.  I never thought I'd be in, uh, chemistry. So 

Richard, tell me, was he making LSD? This is 

the New York State Health Department.

It was actually, you'd get checks from the state. So, uh,  maybe it was LSD, now that you mention it. I don't 

know. And of course, you guys were always experimenting with new ideas and new technologies. And sometimes those led to successful products. But I guess there must also have been times when a cool idea just never quite made it off the test bench for some reason.

Back in the, I think it was probably mid to late seventies, uh, TRW, which I think stood for, and maybe still does, Thompson Rameau Wooldridge,  uh, came up with a solid state multiplier chip. It had a heatsink on top, and you could literally multiply, I think, first a 10 by 10 bit number, and then maybe a 12 by 12, and maybe even a 16 bit  multiplier.

It dissipated a few watts.  Uh, but I immediately saw the possibility of making an audio product out of it.  But this was, uh, before the 1066, which actually had processing, all this was, was a multiplier and a bunch of coefficients  and, uh, whatever sampling rate,  and we built, oh, I don't know, two or three of them.

I was playing around with floating point converters at the time, which kind of worked because you couldn't go out and buy a really good A to D.  And I wanted better resolution than what we were using. I don't remember how successful that was, because  here's the interesting thing. We literally built, manufactured two of these products. 

Two.  And we sold one.  And it ended up going to Brazil. Where, you will recall, Brazil is than the  United States, physically. I have no idea who ended up with it. But somewhere on Earth.  Maybe still in Brazil, is the Eventide S1066 special effects unit. Wow, and what did this thing sound like? It sounded very much like a special effects unit without too much complexity.

It was just a, it was really a filter. Whatever you can do with coefficients, it could 

have done.  Nowadays, of course, Eventide's technology isn't just found in old school rack mounted hardware boxes. Uh, you can also buy Eventide plugins, and Eventide guitar effects pedals. In many cases, the same algorithms, but you have to adapt those for three very different user interfaces.

I imagine that's a bit of a challenge. It's 

crazy making.  Not, not, not only the user interface, but just trying to go to market, um, to, to this disparate set of customers.  You know, some of our, uh, our customers are in film and, uh, you know, film post production. Some are in recording studios.  And then there are guitar players.

Um, and then there are, you know, hobbyists who are downloading our plugins.  Now, we're not smart.  It really is, it's insane. So when  

you're thinking about adapting one of these algorithms for use in, say, a foot pedal, what are the considerations that you need to bear in mind?  Well, 

how a guitar player is going to use it.

Look,  our advantage, Sam, is that  I would say everyone who's involved in conceiving and developing these products  as a musician. We all play.  Most of us, many of us are guitar players. I mentioned Russell earlier. He's a bass player.  You know, he's been in bands that have toured. Christian and Patrick and...

Colin, uh, you know, the folks are product specialists. They all play. Uh, they're all in bands. So  we're basically designing these products for ourselves. We have a sense of how we we want to use them. We've also been very fortunate and having been successful that we have relationships with, uh, with many artists and each time we conceive of a new product.

We run that by artists that we've worked with in the past who we, who we know. So, um, I, I was being flippant, but, but the fact is a lot of, a lot of thought and work goes into that. Uh, asking advice, you know, blue skying, suggesting things. Um, you know, I know when we, when we released our, our first, um, guitar pedal, uh, the Time Factor, our concern was that it had so many knobs.

It had 10 knobs and we had some.  Early, uh, users, um, actually early testers who were concerned that we had so many controls on, on the box, but we felt that that was necessary really to exploit all, all of the possibilities of the algorithms and the effects that we were putting into that box.  So off we went and the product has been successful. 

We then followed that up with a number of other boxes, with virtually the same ui.  Um, and then we came to the H9 and we went from 10 knobs to one knob thinking  one knob.  We just had 10, we thought we needed 10. Now we can get away with one.  But we worked hard at, at, at defining that one knob in such a way.

That it would work for for a user. Um, we added Bluetooth so that the more the deeper control of it can be, uh, you know, can can sit as an app in your your iPad or iPhone or Android device. But, you know, in each case, it's it's starting from somewhere and then doing your doing your homework and have it having, you know, listening, having arguments.

Um,  and finally, making some decisions. Um, you know, obviously, we can't always get it right. But we do do our homework. We do listen. 

Many of Eventide's plugins package technologies that we're familiar with from rackmount units, but some of them actually incorporate new technologies. Uh, perhaps the standout there is the plugin Fission, which uses a technology you call structural effects.

What's that all about?  That was all 

about, uh, being fortunate to, uh, to be working with some, some brilliant,  um, DSP researchers, uh, Russell in particular, Russell Wedgley. And, um, I was just amazed when he actually  demonstrated this effect for me  because it, it throws a lot of signal processing at, at a problem that seems fairly simple.

You know, when we hear a sound, we can perceive that that sound has a timbre. It has a, has a component that's transient and a component that's tonal. And it's a bit of a magic trick, really. It's really how we perceive sound,  and he's come up with this technique where, um, the software  mirrors our perception in a, in a remarkably  accurate way.

And what was, what was most convincing for me is when he demonstrated for me and was able to split a sound into its transient tonal components and then recombine it. With virtually no artifacts, uh, I knew he was onto something.  And, you know, uh, uh, I can tell you that we're working on other,  on other plugins and, and other products that'll take advantage of this, this underlying technique.

It really is different. Uh, you know, in, in true Eventide fashion, we haven't done a, uh, a bang up job of, of marketing this technique. But I think people are starting to catch on. It's been used brilliantly in some places.  

So do you think that plug in processing opens up new possibilities for digital audio, which would have been harder to implement in hardware? 

Well, well, 

yes and no. You know, you say plug ins, but the plug in is running on a CPU somewhere. So really what's at the heart of this is the fact that these processors have gotten so screamingly fast. They have so much memory. That you can, you know, every year you can do more, there's more that you can do, you can run much more complicated algorithms.

And so whether it's sitting as a plug in or that CPU is sitting behind the front panel of a rack mount box, you know, like our H9000,  that's really what the driver is. And truth be told, that's always been the driver. As Richard pointed out, this um, this multiplier chip made it possible to, to think of something like the 1066.

That's, that's always been what our next step has been. Something new comes out that you can take advantage of and do something that you could never have done before. Whether it, whether it exists as a plug in or not, it's, what's, what's behind that is that plug in is running on a CPU somewhere. You know, either in an iPad or in your phone or in a computer or in a rack mount box.

It's just, it's the sheer processing power that's available to you. And what goes hand in hand with that is, you know, the, the software tools to build, uh, new effects have, have become much more sophisticated. Uh, the 2016, uh, actually had a development environment that we wrote ourselves, uh, which was its own, you know, assembler, compiler, specific for that one of a kind array processor.

And that was a fairly primitive tool. Tools today are light years ahead of that.  

At the other end of the scale, complexity wise from plugins and guitar pedals, your current flagship processor is the H9000, which you describe as a network effects platform. Do you think that audio over IP is going to become increasingly important in the studio world or is it going to remain limited to live and broadcast? 

It's,  

again, that's predicting the future. It seems to be. We know, we know of some studios deploying it. We know. In, uh, you know, in venues, if we ever get through this COVID situation and venues are active again,  it's a lot more cost effective to, you know, to run an Ethernet cable around than have to have audio snakes every which way. 

We know broadcasters are adopting network audio. Earlier I mentioned, you know, post production facilities, they use network audio. So it just seemed logical that if we were going to make a flagship processor from Eventide, That ran all of our effects that handled many, many channels of audio that it should it should be smart enough to sit on an audio network. 

So, you know, so we designed that box with some expansion cards so that optionally a user can say, Well, I'm running a Dante network or Maddie network, and I want to connect 

that way. I'm sure you're right that it's difficult and dangerous to predict the future, but equally I'm sure that all of our listeners will join me in wishing Eventide a very happy and successful and prosperous future, building on their first 50 years in the business.

So thank you, Tony. Thank you, Richard. If you'd like to know more, Richard has an episode of the... Gear Club podcast, where he talks more about the history of Eventide, and there's also plenty of materials available on the Eventide website. So thank you for listening, and thank you to my guests Tony Agnello and Richard Factor.

Thank 

you 

Sam, this was great. Thank you for having us.  

Thank you for listening, and be sure to check out the show notes page for this episode, where you'll find further information along with web links and details of all the other episodes. Oh, and just before you go, let me point you to the soundonsound.

com forward slash podcasts website page, where you can explore what's playing on our other channels.