During a fleeting visit to SOS headquarters, Paul White managed to soften up DOD/DigiTech's Product Development Manager, Randy Thorderson, with a tasty curry before pumping him for info about new products and future developments.
Not so long ago, digital effects units were considered to be seriously exciting and esoteric devices, and the introduction of affordable digital reverb was arguably the main factor contributing to the erosion of the boundaries between professional and project studio output. But now every music shop is stocked to the ceiling with effects boxes, all of which offer similar combinations of reverb, delay, modulation and pitch shifting — so where can effects go next? Will we see new effects treatments or will designers concentrate on adding digital interfaces and providing more effects per dollar? I put these questions to Randy Thorderson, Product Development Manager of DOD/DigiTech during his recent visit to SOS HQ.
Digital Interfacing
"Digital interfacing is an obvious consideration in today's products, and one of the things we're busy with now is guessing which of the various interfacing implementations will become the standard in the future. There's AES/EBU, the Alesis ADAT interface, and one or two others, so we don't want to introduce yet another system. As we start to see more digital desks coming along, digital interfacing will have to become more important.
"At DOD/DigiTech, we concentrate mainly on the middle and lower parts of the marketing pyramid, and unfortunately that's where digital standards are least certain. At the top of the professional market, the standards seem to be more settled. Ideally, we need an interface that's going to work with just about anything, but I don't know if that's going to happen. I think it makes most sense to start with just the S/PDIF spec first, but it can be quite expensive adding this kind of interface for just two channels, and once you do add a digital interface, the world wants the more professional AES/EBU. For multiple‑channel systems, back panel space starts becoming a problem and when you move from a 1U to a 2U package, price starts to become an issue again. Unlike the consumer market, companies like us don't have the same benefits of mass production. Even if we buy the same parts, we have to buy them via a distribution chain, and that pushes the price up."
Virtual Effects?
Has the evolution of computer‑based workstations had changed DOD/DigiTech's view of where they should be heading? If the world is moving towards 'virtual' computer‑based project studios, will DOD/DigiTech have to move into selling 'virtual' software‑based effects units?
"I think we'd be foolish not to be looking into these areas. Part of me believes that the sound engineer in the studio isn't going to want to embrace an entirely 'soft' approach, because people are used to working with physical mixers where you can grab a fader. Obviously, you could eventually do everything inside a computer and control everything from an interface the size of a couple of books, but I don't know if engineers will be entirely comfortable with that.
"One of the things that worries us is that, as a manufacturer of signal processors, we're seeing signal processing being incorporated into other products such as keyboards and mixing desks. This could cause our customers to be less inclined to buy stand‑alone processing units, unless we can continue to offer more for less. Our current DSP processing engine, the S‑DISC chip, is only used in our own products, though some of our older technology (the HISC chip) was previously used by other companies in things like synths and small mixers.
"We are looking into the software control side of our products, especially via PCMCIA technology, and we are also in a position to look at processors on computer cards, such as NuBus or PCI. The question right now is how big is that market? If you're using the computer in your studio, it's usually tied up doing some specific things, and there's the multitasking question of how many things can run efficiently at any one time. In my own studio, my Mac is tied up with sequencing most of the time, and if you're running something like Digital Performer, it's a busy box. That may indicate that stand‑alone processors will be around for quite some time yet."
Do you think there's room for a halfway house, where external hardware is controlled from within a software sequencer, and if so, should it be via MIDI or some other protocol?
"That's a good question, because as you're aware, the MIDI protocol has limitations. One of the things we've done with new products like our Studio Quad is to make sure you have the tools to do better MIDI interfacing. The SysEx implementation makes it easier for a third‑party software designer to get involved. We also believe that we'll have to develop some software interfacing in‑house so that it can be released with the product."
Will this mean closer ties with the big four or five sequencer manufacturers in the USA and Europe?
"We have a good relationship with most of them anyway, but I think we'll try to interface with them much earlier on. The graphical interfaces that computers can offer will become more important, and that's something that computers can do pretty 'task free'. It's very different to having an S‑DISC card running inside the computer. We recognise that today's boxes are very powerful, which means they have a learning curve and the graphical interface is one way to make that easier.
"Another thing we're doing is looking to increase our product support via the Internet. We'll be setting up a full Web site in early November and you'll be able to pick up product information on any of our products. There are also guys who write public domain software that supports our products, so we want to put that on a public FTP site. That way, people will be able to get hold of new programs and patches to help them get the most out of a box. Arbiter Music Technology, our distributor in the UK, has already got a site running on the Internet."
New Directions
How about the effects themselves? We've had reverb, pitch shifting, and modulated delay effects for a long time now — have you seen anything that inspires you to go off in new directions, such as effect morphing?
"We're looking at some aspects of that in the new Studio Quad, where we have some dynamic morphing things that respond to playing intensity. But new effects? That's a hard one!
"We're more concerned with accurately recreating real room environments, but I don't know how excited people will get about that, because reverbs are already getting pretty good at this stage. But there are always things that come up when we're designing effects, such as our 4‑way panner for use in 4‑speaker systems. The other thing people always want is more flexibility for less money — they don't want to have to spend lots of money just to be able to choose whether to put the chorus before or after the delay. We think boxes will continue to do what they do, only better and cheaper — but as these products grow more sophisticated and more powerful, we have got to continue to simplify the user interface. We've taken the first steps with that already, and you can learn a lot from consumer products like VCRs. In many cases, the power of a product exceeds the user's ability to approach it effectively.
"In the Studio Quad we have what I think is one of the best sounding reverb algorithms we've ever created, but even so, it's basically still a reverb. There's nothing earthshakingly new about the effects in the box, but where it is different is in the ease of access and the fact you can do more things at the same time for less money. But you can't just keep putting more power and flexibility into a box unless you really make it accessible to the user."
Over the past couple of years, we've seen quite a few MIDI controlled filters which you patch after your synth to recreate the old analogue sound. Is this something you could easily add to a multi‑effects unit?
"Yeah, we're looking into that kind of thing. For me, one of the most impressive devices out there is Roland's VG8 guitar modelling system, which uses a lot of filter modelling. Our engineers keep knocking on my door to ask when they can start work on something like that. The implications of modelling in a signal processing context are vast, but at the same time it's a big piece of technology that would take some time to build. That kind of modelling technique could also come in quite useful in our quest to make a room algorithm sound as real as possible."
Something that has always intrigued me is the possibility of a processor that would allow me to sample the acoustic properties of a physical room, then model that room electronically. Is this possible? Will we ever see a unit that includes samples of the top dozen live rooms in the music world?
"I see that as a definite future area of research, but it would require a new technology. We're in a precarious situation insomuch as we're the number one manufacturer of guitar signal processors in the world and we're making a pretty good run at the middle and low end studio and live sound market. When you're number one you have to worry about market protection, and it's much harder to go forward and develop new technologies when you're protecting that number one spot. But you still have to do it, otherwise..."
...someone comes along and bites your backside? I would imagine that there are still lots of ways you can use existing technology. Filters are a known quantity, yet if you explore what Emu have done in their Morpheus synth, you can see that there's a whole world beyond bandpass resonant filters. By using vowel filters and suchlike, you could create a whole new range of sounds as well as model old standard effects such as the guitar voice box.
"We have a couple of DSP filter experts already looking at that sort of possibility. We also have a guy with a background in voice recognition, and that too has interesting implications for sound development. We have a large group of extremely bright guys at DOD/DigiTech, and the majority of them are musicians. It's a rare combination to find people who can use both sides of their brain really well.
"That's another reason the Internet thing is so exciting — there are lots of ideas floating around out there. We can build up a database which represents people's wish lists."
Still on the subject of filtering, I'm surprised that nobody has come up with a box that lets you sample not actual sounds, but the shifting spectral content of sounds so that you can impose the resulting filter characteristics on other sound sources, almost as you would do with a vocoder.
"Our company is devoting more time to that kind of thing, and it's an area that excites me a lot. There's plenty to be learned from synthesis‑type techniques, but our main concern is how to keep the operating system under control, since we are trying to bring the power of our products to the masses."
Can you go for a dual‑complexity operating system like you get on some VCRs? On the surface is a fairly basic set of operating parameters, but if you flip back the cover, there's a whole range of more complicated stuff underneath.
"We've taken a little bit of that approach in the new Studio Quad, which uses libraries containing skeletons of algorithms into which you slot your effects. To use it on this level, you probably don't even need to pick up the manual. And once you select the effect, you have a library of settings available for that effect. Should you want to go further, then you can delve deeper into the operating system."
I'm glad to hear that you think that way, because the impression I get of most products is that the designers think it's the only piece of gear anyone has got to think about. In my own studio, for example, I have a shelf over 10 feet long containing nothing but manuals, and there's no way I could ever find time to read through them all, let alone memorise their contents. To my mind, if a typical processor requires more than a handful of pages to guide you through it, it's not going to get used to anything like its full capacity. And professional users have even less time to experiment than home studio owners.
"A lot of the answers may lie in computer interfaces, and in context‑sensitive, on‑line help systems, but it is something we're very concerned about at DOD/DigiTech, because we really don't want our users wasting time with their manuals. We are sure they've got more important things to do.
"We have a large group of extremely bright guys at DOD/DigiTech, and the majority of them are musicians. It's a rare combination to find people who can use both sides of their brain really well."
Software For Hire?
Have DOD/DigiTech looked at the possibility of renting effects software, where the user buys a black box, possibly with a basic feature set, and then anything special that's needed is downloaded for a small fee and runs for, say, 48 hours before erasing itself? Financing software support for existing hardware has always been a tricky problem for manufacturers, so maybe this would be one way to give the software a tangible value?
"Possibly. Once a product has been released, the value of any continuing software development is very difficult to measure. But I don't know that any of our sales network would want a system like you describe in the very near future, at least not until it becomes more established in other areas of our lives. To make that happen, I think that your television set would need to incorporate a credit card swipe slot, so you could order products and services directly from your TV. Once that starts happening, then I think we can start to look at that kind of thing. The technology already exists to build a generic signal processing box, so renting software via modem might be one way to go."
Beyond Stereo
Have you looked at psychoacoustic algorithms for producing wider effects with more sense of space?
"We used to make OEM surround sound products so we know quite a bit about how they work, but we haven't put them into any of our products just yet. From a personal standpoint, I still haven't heard anything in the market that's made me jump up and down with excitement."
I imagine that the designers of the existing systems are hampered by the need to minimise tonal changes and to maximise mono‑compatibility, but when you're talking about processing only the effect, I would expect you to be able to use more dramatic processing without the risk of upsetting the overall mix.
"Yeah, I think there's quite a bit to be explored in that area, and looking longer term, it could be that there will be some surround aspect to everyone's normal music listening now that Dolby Surround TV has been accepted. We might even see a resurrection of quadraphonic recording in another form. Then we'll have to come up with quadraphonic effects processors..."