January 2005
Just Around the Corner

My AES Highlights—and there were plenty!

I don't know who and what I have to kiss to convince the powers-that-be at the Audio Engineering Society to hold its biennial West Coast convention in San Francisco all the time. Since they moved the New York show to the Javits Center, which is about as far away as you can be from anywhere in Manhattan you'd ever want to be without getting wet, the San Francisco location has become the only place where the AES meets that is actually fun. Hotels, restaurants, transportation, places to buy clean socks — everything is right at hand and you don't feel as if you're trapped in some kind of high-tech space capsule from which the only escape is by taxi. The weather during this fall's show was particularly delightful and many attendees — even those who don't smoke — took advantage of the beautiful, recently enlarged Yerba Buena Gardens right above the Moscone Center to get away from the show's din and stale air for a while.

Even the talks and sessions, which tend to be incredibly dull, were somehow more lively this year: The “Afternoon With Bob Moog” was delightful, the “Pioneers of Electronic Music” session, although a much longer walk from the Moscone Center than any of the literature would have you believe, was jam-packed and fascinating, and the Heyser lecture by Walter Murch, in which he showed us a sound film made in 1894, was a jaw-dropper.

There were only two things not to like about this year's AES. One was that there were simply too many events for one person to get to — but, of course, that's almost always the case. The other was that Digidesign decided to hold its own little show, “Digiworld,” for itself and its development partners at company headquarters in Daly City. It's not very far, and Digidesign provided shiny buses for everyone. But somehow it felt unfair, and I don't think it's something that Ampex, MCI or even Harman, all of whom have at one time had the same kind of 600-pound gorilla presence on the AES show floor, would have ever considered. I couldn't make the journey because of time restrictions, so not only did I miss the product demos, but I also missed interacting with the many people I know at the company whom I like checking in with every year.

Virtual Mixing Company Virtual Mixer

But enough complaints. It was a great weekend. The Red Sox had just pulled off a miracle, the Patriots were still undefeated and we didn't yet know how powerfully the politics of fear and loathing were about to express themselves. I got a lot of backslaps on the show floor for my blatantly partisan pre-election “Insider” column and only a couple of vicious hate e-mails. And I had the most fun, as I usually do, avoiding the big-name press conferences and hunting down the innovative and clever little things (the majority of which cost little money) that most reporters pass over. So that's what I'm going to tell you about.

For example, the Virtual Mixing Company has been a mainstay at AES shows for about 15 years, but the company rarely gets much ink. You've seen these guys — they have a tiny booth with a video display that shows the various elements in a mix as floating multicolored spheres, which change size and fly around as you play with the faders, pans and effects. It's called, not surprisingly, the Virtual Mixer, and I've always found it an interesting concept, but it's had a hard time catching on, largely because implementing it in hardware has been tricky — every mixing platform would need to be approached differently. Well, this year, the company showed a Pro Tools plug-in, and all of a sudden, it makes perfect sense: It can be hardware-independent because there is no hardware. The plug-in is $400 and a touch-screen display that interfaces seamlessly with Pro Tools' mixer is another $500. But even more interesting, the company was talking about adding 3-D glasses and virtual reality head trackers. It is also considering integrating it with other workstation platforms, but for now, if the company can get enough Pro Tools users interested, it may finally find its market.

Another cool workstation controller was at Frontier Design Group's booth. This company is best known for low-cost format converters and PCI audio cards, as well as helping Tascam design its FW-1884 DAW control surface. Frontier's TranzPort is a 1-pound, 6×7-inch wireless box with about two dozen buttons, a two-line LCD screen, a plug for a foot switch and a big wheel. “When you're sitting at the drums and ready to do a take, the last thing you want to have to do is get up and go over to your computer and put it in Record,” said one of the demonstrators. “So now you can do it with this.” It uses 2.4GHz radio signals for communication, not infrared light (and not Bluetooth, either), which means it can go through walls, and claims a typical range of better than 30 feet. Data flows in both ways, so the unit will display things like timecode numbers and track names. Support for Pro Tools, Logic, Cubase, Digital Performer and Sonar is already onboard, with other applications coming soon (Reason and Live were mentioned) and there's some customize-ability built in. The receiver unit, which is smaller than a business card, hooks up to the computer via USB. TranzPort uses four AA batteries, and its makers boast that it's so energy-conservative, it doesn't need a power switch, although there is an automatic power-down sequence. The company hopes to ship it by the end of January at a price of around $200.

A different kind of workstation interface was on display at the Open Labs booth. With soft synths and samplers taking the place of hardware in so many studios and even live applications, it was only a matter of time before someone built a powerful soft synth engine into something that looks more like a musical instrument than a computer. Harkening back to the days of the original Fairlight and such dinosaurs of the pre-MIDI era as the Con Brio and the McLeyvier, this Austin, Texas, company's NeKo 64 sticks into a single 98-pound box a PC with five PCI slots; a touchscreen; a QWERTY keyboard; three control surfaces with rotary encoders, sliders, a touchpad, a joystick and a whole bunch of buttons; an impressive array of inputs and outputs; and a 61-note piano keyboard. Available in configurations ranging from a single 1.4GHz AMD Opteron CPU to dual 2.0GHz processors, and with other options such as 250GB hard drive, up to 8 GB of RAM, an ADAT interface, a DVD burner and more, this sucker can handle as many as 100 plug-ins at a time and produce up to 500-voice polyphony. Windows XP is built into the system, with a custom simplified music-oriented user interface laid on top of the OS. Because the hardware is all-modular and the sound engines are all software-based, the system is, theoretically, upgradable without limits. It ships with some 19 soft synths and nearly as many effects modules, as well as Tracktion, the low-cost multitrack software that's now part of Mackie's product line, and Karsyn, a plug-in organizing program described as being similar to Steinberg's V-Stack but with more options and channels. If that's not enough, adding your own host software and your favorite VST or DirectX plug-ins is as easy as loading them into any old computer. It's not exactly cheap — prices start at $5,500 and climb fast — but for someone who needs a lot of power and a really comprehensive hardware interface, and who moves around a lot, NeKo could make a great deal of sense. And compared to the original Fairlight, it's a steal.

Speaking of plug-ins, a new set for Pro Tools called UltraFX, from the Vermont company SoundToys (which used to be called Wave Mechanics until its owners decided there were too many companies out there with “Wave” in their names), have an interesting twist: They share a feature called the Rhythm Pattern Editor, which lets you set up complex cyclical patterns up to 32 bars long to control processing parameters. A pattern can use any reasonable time signature and division of the beat, and can assign different levels to individual beats. Patterns can free-run or be locked to MIDI Beat Clock. Best of all, they can be stored in a library where the other modules in the group can get access to them. So you can have your resonant filter do a 9/8 jig, your delay line play “Take Five” and your phaser work out on a 14-bar shuffle, and then switch them around. The set also includes Crystallizer, a “granular echo synthesizer” that is “inspired” by a well-known preset on the old Eventide H3000, a device that many of the company's engineers helped develop. In the months to come, expect new flanging, panning and “disturbingly abusive fidelity reduction” modules.

Frontier Design TranzPort

In the area of surround mixing, many of the larger companies were showing new concepts and products, but some of the most intriguing were at the small booth of Immersive Media Research (IMR), a software company that grew largely out of work done at the University of California San Diego. IMR offers a suite of tools for surround work, starting with a software mixer designed to “lower the barriers to experimenting in surround.” The basic mixer program, which sells for less than $300, uses ReWire to hook up to a host application, allows real-time control via MIDI and USB, and comes with a library (which you can add to) of complex spatial paths and loops and algorithm-based setups.

Then there's the Immersive Designer software, which uses a graphic interface for time and vector control over volume and position, as well as a host of integrated DSP functions including reverb, Doppler Shift and air absorption to let you spatialize a mono or stereo signal in a zillion different ways. The motion information can be saved separately from the audio files so that a move can be applied to other audio tracks. The company's favorite demo was taking a simple subway train sample and manipulating it to make it sound like it's coming at you from five different directions simultaneously.

Finally, the company's Immersive Encoder takes any group of up to 18 sound files (even if they're in different formats) and combines them into a single multichannel file in most common formats, as well as a very impressive binaural format for true surround monitoring on headphones. Interestingly, the company claims that this mix can be successfully encoded as an MP3 file, but I didn't get a chance to check it out. A Pro version of the encoder includes DTS encoding, and the company hopes to license other proprietary codecs in the future.

One aspect of AES that many attendees from the studio world sometimes miss is new developments behind the scenes of the audio manufacturers and designers whose gear they use. Despite the fact that these developers' booths tend to be quiet and not have a lot of flashing lights, they deserve attention as their products will determine how the gear we will use in the not-so-near future will look and feel. One example of this was a booth in a corner of the show floor that didn't have throbbing monitors, just a couple of posters and a British-accented engineer describing his company's work. The engineer was Michael Page, and he works at the Oxford, UK, research labs of Sony. His topic of discussion was the proposed AES-X140 standard, which Sony developed originally under the name of “Super Multichannel Audio Connection,” or (of course) “SuperMAC.” When it is officially adopted by AES, which may be a done deal by the time you read this, it will become known as “High-Resolution Multichannel Audio Interconnection,” whose acronym is unfortunately far less attractive.

Open Labs NeKo 64

What's amazing about AES-X140 is that it crams a huge amount of audio data onto Cat-5 Ethernet cable. No longer will we need expensive MADI, fragile fiber or clunky TDIF cable to interconnect our various digital hardware devices. We will now be able to use cheap and flexible cable you can buy at Staples and have runs up to 100 meters, with predictable latencies of less than 80 microseconds. How much data? How about 24 channels of DSD or 96kHz 24-bit PCM, or 48 channels of 48kHz, 24-bit PCM? That's in addition to a 5Mbit/second data channel, which can be used for timecode, remote control, TCP/IP, metadata or about 160 MIDI streams.

Also out of plain sight are those companies that make the chips that go into the gear we use. What they like to show at AES are how they can make life easier, better and cheaper for their customers — the manufacturers themselves. Among these is Analog Devices, a major player in the audio industry from Norwood, Mass. (which, as anyone who has seen a car commercial on Boston TV knows, is pronounced “NAAH-wud”), whose SHARC Audio ICs have for some time been at the core of products from Mackie, Tascam, Studer, Harrison and many other pro audio makers, as well as consumer and car audio manufacturers. Besides announcing the latest line of more powerful chips, the company revealed a slick new software development tool called VisualAudio, which uses a graphic object-oriented interface to create code for its processors. The software should help speed up development time for companies using the chips through all four phases of the process: design, code generation, tuning and testing. It includes a large library of audio function modules, a customizable user interface to optimize control and testing functions, and support for a broad range of codecs. Users can write their own modules using C or assembly language and save the modules and sub-systems they develop for use in their next generation of products.

Along the same lines, but with one critical difference, is a new company called Anadigm, with headquarters in Cheshire, England, and Tempe, Ariz. “The idea for the company came out of Motorola,” co-founder and CTO Ian Macbeth told me. “They stopped doing programmable chips, so we took it over.” Anadigm is selling field-programmable chip sets that operate in the analog domain and has created an extensive set of development tools for the chips. “The silicon is software-configurable, with the modules defined by the code,” explains Macbeth, “so you can build your own circuits and control interfaces and keep everything in the analog domain, avoiding the need for all those converters.” The whole development kit, including the software, sells for a mere $199. “The goal is to bring down the cost of DSP design for manufacturers of smaller systems. We make our money selling them the chips.” Initially, the company will focus on speaker systems, guitar effects and synthesizers, and will include a library of pre-packaged code for filters, crossovers, oscillators, and dynamic and time-based effects. But Anadigm is also looking to the auto sound, architectural acoustics and home theater markets, and further down the road, it hopes to cross over into areas beyond audio.

So thanks San Francisco, and thanks AES. What say we do it again in two years?

Paul Lehrman writes from a blue state.

Copyright ©2005 by Paul D. Lehrman