Positive Feedback ISSUE 58
november/december 2011



AES 2010
by Scott Dorsey


The Audio Engineering Society conference has been going on for many decades now, and what makes it a unique and wonderful thing is that it's a trade show and a scientific conference, one upstairs and one downstairs. The interrelationship between these becomes very interesting as the engineers designing equipment on the trade show floor show up at the paper sessions, and people doing research downstairs get a chance to go up on the show floor and see practical hardware.

There has been an increasing cultural divide between these two halves in the twenty-five years I've been going to the conference, probably something that reflects the pro audio community at large. This year, though, there was also an increasing physical divide; the two halves were located at the opposite ends of the Javits center in New York with two other conferences in-between them.

The good news is that although the trade show floor was somewhat smaller this year, it was very tightly packed. Most of the better exhibitors were there, just with slightly smaller booths, and it seemed like attendance was actually up. On Saturday afternoon it was almost impossible to walk around on the floor for the crowds. Still, I got to see most of the show upstairs and a good bit downstairs.

I'd like to point out that this show review and the accompanying awards mostly consist of things that I personally find interesting, so if something you find exciting is missing, don't feel left out. It's a big show and I'm only one person and have to pick and choose. I tend to choose to spend more times looking at transducers, microphones, speakers, and the like and not much time at all looking at digital editing systems and plug-ins. There were lots of plug-ins at the show but if you're interested in those you'd do better to read Mike Metlay's show review in Recording Magazine. I try and write about the show from a high-end audio perspective which is a little different.

I was very disappointed in a few things, for example the House Ear Institute wasn't there doing ear testing this year. That is one of the things that a lot of people look forward to every year at the show and they were sorely missed. Beyond that, here's what I saw...


GRAS Sound and Vibration was showing off their line of high quality measurement capsules based on the classic nickel-diaphragm B&K designs, as was ACO.

MicW from China was showing some nickel-diaphragm mikes which appeared to be adaptations of the original B&K line intended for recording applications.

However, they didn't have any for audition, and when I opened some of the display units, they all appeared to be completely empty inside with no capsule or electronics.

On the other end of the scale, Udo Wagner from Microtech Gefell was demonstrating their M960TS microphone, which is a transformer version of their classic M930 nickel-diaphragm studio mike. They also had another new mike, the M1030 which is a similar design in a different case and grille style, which changes the sound somewhat. These are also nickel-diaphragm capsules.

Gefell was also re-introducing the classic CMV 563 microphone that they made back in the 1950s.

Josephson Engineering wasn't showing off their ingenious new pickup for wind instruments, but they did have the new Josephson C716 microphone which uses a foamed metal material for the grille which claims to be far more transparent than the repeating pattern of a mesh.

Martin Ucik from Plus24 was showing off the Bubblebee lavalier mike windscreens... these look like furry little animals but really seem to do wonders at reducing wind and clothing noise on lav mikes.

A new company called Ronin Applied Sciences was showing off a very interesting new tube condenser microphone using a Siemens C3g low-noise tube intended for long-distance wideband telephone circuits. Because it's so quiet and intended for hundreds of kHz wide circuits, it's a great pick for a microphone and not one I've seen before. This could be interesting.

Mark Fouxman of Samar Audio Design was showing off some very interesting new ribbon microphones, which sounded clean and had a good null. He also, interestingly enough, had a line of ribbon mike transformers as well as condenser mike output transformers built around iron toroid cores. Good stuff and I hope to see more of these folks in the future.

Down in the paper sessions, there were a lot fewer microphone papers than in previous years. This report may give you the opposite impression since I tend to go to the microphone papers at the exclusion of others, but it did seem like there was a big shortage this year.

Mark Zaim from Audio-Technica gave a good "engineering brief" on "Phantom Power the Modern Condenser Microphone Part II: The Effects of Load Impedance on Microphone Performance" which evaluated roughly how loading affected basic measurements of condenser microphones. Needless to say he found lower impedance loads reduced the maximum SPL available, but he did not measure distortion at lower levels. He found the transformer output mikes tested to be less sensitive to loading as far as maximum SPL goes, although again many transformers require fairly narrow impedance loads for lowest distortion. So this was interesting but nowhere near detailed enough. "Engineering Brief 19" is available for free on the AES website.

In "Proximity effect detection for directional microphones", Alice Clifford and Joshua D. Reiss develop a digital algorithm that can evaluate a vocal signal and detect low frequency boost caused by being close to a directional microphone and correct for flat response. This is nothing that can't be done manually, but it can be an interesting technique for dealing with people who move constantly on and off mike. Preprint 8475.

There was a very interesting concept that could be quite valuable, discussed in "Multiple Microphones Speech Enhancement Using Minimum Redundant Array" by Kwang-Cheoi Oh from Samsung. In this paper, he takes the concept of the minimum redundancy linear array of transducers, something developed back in the sixties for phased array radar systems, and applies it to a fairly narrowband (voice grade) array of microphones in order to create a steerable beam forming array that is able to identify and track one person speaking, for teleconferencing applications. It was sometimes difficult to understand Mr. Oh and the paper seems somewhat difficultly-worded but the concept is worth exploring. Preprint 8482.

Totally at the other end of the spectrum as far as multiple microphone arrays goes is "Computer Assisted Microphone Array Design" by Michael Williams. He demonstrates a Matlab program which allows the user to evaluate the potential quality of imaging with various microphone configurations played back over various multichannel speaker configurations.

This could be a useful tool for both research and teaching. Preprint 8461.


Gabriel Sound was showing off speaker re-coning services with one of the huge 3-foot Electro-Voice woofers sitting in the middle of their booth. They also were showing off new drivers from Faital Pro and RCF... and I'd like to point out that both companies make some drivers that have good applications in high end home systems including some fun coaxial horns from Faital that seldom are seen in the US. The Faital 8HX150 would make a great home system.

Renkus-Heinz had a demo of their I2C digitally-steerable speaker systems, a very small line array set up in a small room, but with really shockingly good sound, especially for PA speakers. Very clean top end, not super detailed but not spitty, and the pattern control was excellent with very good control over reflections from the poor room. These things use internal Class D amps with no feedback around the amps at all, which is a very peculiar design but one that seemed to sound good.

After three years of talking about the K-Array, the Sennheiser folks actually had one out on the show floor to demo. This is a line array of very small driver speakers designed for very tight pattern control. Unfortunately the one they had available to demo was the KB1 portable system and none of the larger arrays were available. The demo seemed kind of spitty-sounding but since they were only able to play back from a pocket MP3 player it's hard to tell much about the speaker. It really does seem like Sennheiser is trying very hard not to actually sell these things.

Professional Audio Design was showing off the Augsburger studio monitors, which are big horn-loaded speakers with wide wooden horns for very wide dispersion. They don't sound like horns, but sounded quite clean although it was hard to tell on the show floor.

Focal was showing off their line of fine monitor speakers, as well.

And, Phoenix Audio was showing off a very clean-sounding monitor, using a Scan-Speak tweeter and custom internal electronics. This is a powered monitor, so an active crossover and amplifiers are built into the box.

Down on the paper sessions a lot of people were discussing reducing distortion on compression drivers for horn loudspeakers. In "Dual Diaphragm Compression Drivers", Alex Voishvillo from JBL Professional talks about a compression driver using two light plastic film diaphragms in place of the conventional heavy metal diaphragm; the mass of the individual diaphragms is much less but the ability to move air is similar. Anything that reduces distortion and widens frequency response should be encouraged. Preprint 8502.

Everybody knows speakers change in sound a bit after a break-in period, and lots of folks have noticed that some drivers change in sound after they have aged for many years. In "Mechanical Fatigue and Load-Induced Aging of Loudspeaker Suspension," Wolfgang Klippel presents a good empirical model to predict these effects. Preprint 8474.

Another paper on a similar subject was Finn Agerkvist's "Nonlinear Viscoelastic Models" in which he describes some refinements to the conventional mass-spring model normally used to describe the behavior of rubber materials like speaker surrounds and spiders. In the real world, the stiffness of the surround can vary with the test frequency, and his model demonstrates this and other effects. Preprint 8500.

In 1972, the Thiel-Small model was a huge advance in speaker design because it allowed one to predict the response of a vented box speaker without building it. However, it fails to work with transmission line designs and other configurations where the vent is extremely long and may be a large portion of a wavelength away from a given driver. To deal with that, Juha Backman from Nokia recently came up with a good model of transmission line systems, but there is some argument in the speaker design community about how accurate it is. In "A Computational Model of Vented Bandpass Enclosure Using Transmission Line Enclosure Modeling", Jongbae Kim, Gyung-Tae Lee, and Yongje Kim from Samsung discuss using Backman's model to predict and optimized response of an oddly-shaped speaker designed to fit into leftover space in a flat-panel television. They find the model very accurately matches their measured response. Preprint 8499.

Large concert speaker systems today are almost all moving toward line arrays: long arrays of individual cabinets, because they allow the radiation pattern and response to be adjusted on the fly. This allows one speaker system to be usable in a wide variety of halls. Rather than just throw them up and tweak until it sounds right, though, Ambrose Thompson, Jason Baird, and Bill Webb came up with a Matlab model which uses successive optimization techniques to come up with an optimal setup for a given room. In "Numerically Optimized Touring Loudspeaker Arrays-- Practical Applications", they describe a very powerful system which appears to be primarily limited by the accuracy of the speaker measurements and the accuracy of the room plan used. Preprint 8511.

Also in that field was "The influence of the directional radiation performance of the individual speaker modules, and overall array, on the tonal balance, quality and consistency of sound reinforcement systems" by Akira Mochimaru, et al. This was an excellent overview of the various possible array configurations and the advantages and disadvantages of each with respect to controlling the radiation pattern (so more sound gets to the audience directly and less reflects off of walls and ceiling) and the frequency response (so it remains flat in all the seats in the hall, or as much so as possible). Preprint 8470.

Another fun optimization problem was presented by Holger Hiebel in "A Parametric Study of Magnet System Topologies for Miniature Loudspeakers". Working on miniature loudspeakers for cell-phone systems, many of the individual design parameters are far more integrated than with conventional speakers. A small change to one parameter can change others dramatically, so designing a speaker becomes a problem in optimization of multiple parameters. Preprint 8498.

In rooms with low frequency problem, one possible workaround is to use multiple subwoofers. The problem is, they interact with one another as well as the room. In "Subwoofers in Rooms: Equalization of Multiple Subwoofers", Juha Backman from Nokia talks about various methods for adjusting equalization to make speakers coexist, although unfortunately he doesn't mention the serious phasing issues between them and how the group delay in typical IIR equalizers changes them (not always for the better or worse), or how all-pass networks can help. Preprint 8471.

Conventional woofers have the cone centered and sealed by a surround around the edge, while the voice coil is kept in the center of the magnet gap by a "spider" at the base of the cone. But if your voice coil is very long, in order to permit wide cone excursions, you may need two spiders in order to keep the voice coil centered, and then a lot of other design parameters change. In "An approach to small size direct radiation transducers with high SPL", J. Martinez and others from Beyma in Spain explore these changes. Preprint 8467.


Lots of folks had headphones on display but really I didn't get a chance to do much listening to them.

John Gresko from Direct Sound was showing some studio headphones designed for extreme isolation... it was hard to tell what they sounded like with the demo music, but the isolation was very good and in a tracking situation that is paramount.

Down on the papers floor, Richard King, Brett Leonard, and Grzegorz Silora were talking on "The Effects of Monitoring Systems on Balance Preference:

A comparative study of mixing on headphones versus loudspeakers." These folks presented test subjects with a stereo backing track and a mono solo track and asked them to balance the two, and they consistently found differences between people working on headphones and on speakers. It's clear than panning decisions and reverb setting decisions are affected severely by speaker vs. headphone listening, but in this study they found even fundamental level decisions were affected. Preprint 8566.

Another great headphone paper was "Inter- and Intra-Individual Variability in Blocked Auditory Canal Transfer Functions of Three Circum-Aural Headphones" by Florian Voelk. He found that the differences between actual frequency response measurements of headphones made on human heads varied substantially from trial to trial, especially above 6 KHz. Not just variations between subjects, but major variations between trials on the same people, because minor positioning changes on the headphones made substantial differences in the actual response at the ear canal. An eye-opener for anyone auditioning headphones. Preprint 8465.

Acoustical Stuff

Acoustacorp was showing their AcouPop, a portable gobo that folds down into a small box for convenient storage and transport, but can be assembled quickly into a very effective sound-absorbing and blocking panel. Very handy for studio use but I could see something like this even being useful in the home or for sound reinforcement applications to shadow loud sources on stage.

GK Acoustics was showing some very nifty diffusers built into perforated canvas panels that could be produced with your own artwork. Send them a photo, they'll send you a nice panel with the image on the front which is actually a diffuser. Great for the back walls of listening rooms or studios where you want to deal with reflections but don't want to make it too dead.

Likewise, Jose Carmo from Jocavi in Portugal was showing off some very brightly colored and very cool-looking diffusion panels. They are interesting enough to fit in well with a variety of modern styles and, of course, they provide diffusion. Wife acceptance factor 100.

Down on the paper sessions there were a lot of great papers addressing acoustics. One of the most interesting I saw was "Acoustical modeling of gunshots including directional information and reflections" from Robert C. Maher at Montana State. He points out that the sound of a gun differs a lot depending on where it's pointed, since it's anything but an omni-directional sound source, and that this affects any impulse response measurements taken with a gun. He does a good job of showing how identifying a weapon by sound is not as easy as some vendors claim, but that identifying positions and location is feasible in some cases. Preprint 8494.

Active sound absorption has been increasingly interesting as more and more people have been working in very small boxy-sounding studios with low end response problems. The same problems exist in smaller listening rooms, but an active sound absorption system can be built in far less space than would be required for a passive bass trap of similar effect. In "New Thoughts on Active Acoustic Absorbers", John Vanderkooy from the University of Waterloo describes a physical model of acoustical absorbers and compares measurements made in small rooms with those predicted by the model. Some mysterious differences exist, and so there is some discussion of where they might possibly come from. Very interesting work, and although there are only a few commercial products available in this field, better theory may make it possible to build even better units that may increase the popularity of the idea. Preprint 8458.

Another thing that happens when rooms get small and wavelength get long is that conventional ray-tracing models stop working, and acousticians have to rely on more crude methods of estimating the response of a proposed room design. In "Accurate Acoustic Modeling of Small Rooms", Holger Schmalle, et al. detail an effective finite element modeling system that can be used to estimate room response before a room is built, even at low frequencies in small rooms. Results correlate well with measurements. Preprint 8457.


Lots of new digital console technology out there, but I have to admit that I missed most of it. Spoke with the people at Calrec who have now moved from the custom-built large scale analogue console business into the custom-built large scale digital console business... they say they have parts to make one smaller analogue console in stock and once that's gone it's all digital.

The Awesome Transistor Amplifier Company was showing off a channel amplifier in API 500 format. This is basically the make-up-gain section that comes at the end of the summing buss on a mixing console, and by using it you can turn a bunch of individual mike preamp modules (and maybe equalizer and compressor modules) in API 500 format into a mixing console. That's a cool and very useful thing.


Sound Devices was showing their 7-series digital recorders, wonderful little portable devices that can record to hard disk and/or compact flash. They have convenient stripped-down 2-track models for reporters on up to a big 8 channel unit, and they really do seem to have become the Nagra of the new era.

Nagra was also there, showing off a new low-cost (well, low cost for Nagra) handheld 2-track recorder.

Joeco makes the "Black Box" recorder which is a 1U high 24 channel hard disk recorder aimed at the live concert recording market. They introduced a new version with a MADI multichannel digital interface which allows it to be plugged into a lot of the newer digital PA consoles for seamless concert recording without going through any extra converters.

Incidentally MADI seems to be increasing in popularity still, in spite of being a very old interface design. Direct Out from Germany was showing a wide variety of MADI devices, from their M.1K2 routing system which basically provides a virtual patchbay for multichannel digital signals (and which can be re-patched from a web browser), their Ma2chbox.xt MADI-input headphone amplifier (which allows you to select any two channels and listen to them, a great tool for studio diagnostics), and a new 32-channel A/D and

DA Converters

Lynx was showing their new Hilo converters, with two channel A/D and D/A in a box with respectable metering (and the metering is a big part of that).

Antelope was showing off some new converters as well, but what is most interesting to me is their Integrity line which appears to be aimed at the high end home market. Very good preamp designs, combined with the features you'd want for a home system (such as a gain control on the preamp, multiple inputs that can be switched in and out, and even an optional remote control).


Lots of new stuff out there but I didn't get a chance to check much of it out. Rob Roy Campbell from the Electronaut company was showing off a very interesting little tube preamp.... conventional design, 12AY7 in, SRPP out, and up to 63 dB of gain in the middle, beautifully handmade.

John Hardy was about to show the Twin Servo 500, basically their traditional Twin Servo preamp but shoehorned to fit into an API 500 rack. He didn't have one to ship yet but it also looks like a potential winner.


Lindell Audio in Sweden was showing off the 17X compressor, a FET-VCA-based design with a number of interesting features like an adjustable high-pass in the side chain to prevent pumping without having to patch external equipment in.

Rupert Neve Designs is making a thing called the 5045 Primary Source Enhancer which they are selling through Yamaha. Both Neve and Yamaha had cut sheets on the thing but nobody had one to demo... it is an ingenious little noise reduction device intended to reduce feedback problems in sound reinforcement applications but I could see it also being very handy to reduce leakage into vocal and solo mikes in the studio too.

Neve was also introducing a new stereo buss compressor in the Portico line, which looked very interesting and could have some mastering applications as well.

Moog was showing "the ladder," a synthesizer-style filter built into the increasingly-ubiquitous API 500 module format. It's based on the original Moog 101 low-pass filter module and can be used to create a wide variety of aggressively weird sounds.

A new company called "dsp4you" was selling various dsp modules and audio network modules that can be integrated into products, as well as a very nifty device called the "pwr-DSP" which consists of a dsp processor and from one to three power amplifier modules, along with built in software to perform many different crossover functions. This is a great tool for anyone building and designing speakers, if they are making one or a million, because it allows you to tweak the dsp parameters on the fly and optimize the crossover.

TAC System was showing their NML RevCon-RR reverb reduction software. It doesn't really work all that well, but the amazing thing is that it works at all, because nothing else does. It really is a revolutionary gadget especially for dialogue production work because it really does improve intelligibility of vocals made under live conditions.

Sonoris Software was showing various software intended for mastering studios, including an equalizer and compressor. What was very good, though, was a package to create DDP files for mastering ordinary red book CDs in a convenient way, minimizing the chances of anything going wrong on the way to the pressing plant. DDP files used to be very common before cheap CD-R recorders became available, but they remain a much better way of getting disc images out.

Martin Holters and some other folks from the Helmut Schmidt University in Hamburg have been investigating digital models of various guitar pedal circuits for some time, and have talked about various Roland pedal models at the past few AES shows. In “A Digital Emulation of the Boss SD-1 Super Overdrive Pedal based on Physical Modeling” they divide the pedal circuit up into three stages, each of which they simulate individually with a transfer function and then they put the three functions together to create a composite model. Doesn't look into interactions between the three stages caused by nonlinear loads, but it's a good first-cut and seems to sound good. Preprint 8506.

There were a lot of papers on Class D amplifiers as usual, but the only one I checked out was "High Order Analog Control of a Clocked Class-D Audio Amplifier with Global Feedback using Z-Domain methods." Class-D amplifiers tend to have linearity problems from various sources, and you would think adding global analogue feedback would help a lot of that. The problem is that the latency through the modulator and integrator in the amplifier is very high, making feedback network design difficult if you want to avoid oscillation. Peter Kemp, Toit Mouton, and Bruno Putzeys used an ingenious method to do so. Preprint 8468.


Steve Lampen and David DeSmidt from Belden were there, showing off some new stuff including their 2221 miniature microphone cable, which looks like a very rugged cable for connecting to mini phone, LEMO connectors, and similar miniature connectors. Gepco and Mogami were also showing high grade bulk cables.

Vovox was showing their line of audiophile cables, while Wireworks was showing a variety of different premade cable assemblies.

The folks from ESP were there on the show floor showing off some very well-made power cables, and I think this was a good thing.

Unfortunately one of the fellows from ESP also gave a tutorial on power cable design, downstairs as part of the sessions. I do have to give him credit for not making outrageous claims about his cables, but the claim she did make, he didn't support. He spent half of the presentation giving degreed engineers a third-grade lecture about power supply loads being nonlinear, and at the point where he started confusing load impedance and characteristic impedance (which are two totally different concepts that shouldn't have the same name) it became too painful for me to sit there. Presentations like this do not actually help the audiophile cause; if anything they dissuade engineers from investigating possible effects. I'd feel much better if people would just sit down and say "some people hear a difference and we don't know why, so you should try it" rather than doing a lot of hand waving.

On the other side of things, in "Cable matters: instrument cables affect the frequency response of electric guitars," Rafael Cauduro Dias de Paiva and Henri Penttinen from Brazil showed some typical cable designs represented as lumped-sum networks with a 1M load from a guitar amp, as well as actual measurements of the same cables with a Fender pickup, to show how relatively small amounts of cable shunt capacitance can make a huge difference in the system response when a high impedance input is used. This isn't news or in any way innovative but it was an excellent demonstration of the effect nicely presented. Preprint 8466.

Playing Records

AT and Shure had some phono cartridges on display but that was about it for the whole phono thing this year. Still, there were a lot of papers on the subject. In "Practical Digital Playback of Gramophone Records using Flat-Bed Scanner Images", Baozhong Tian, Samuel Sambasivam, and John L. Barron scan 78 rpm records on an inexpensive desktop scanner (making two scans for each side of the record and pasting them together), and manage to get remarkably good playback considering the difficulty of identifying the full shape of the groove given just a shallow image. Not a replacement for proper stylus playback yet, but still a very worthwhile technique for high-speed digitizing of large record collections so they can be heard well enough to identify what records need proper transfer. Processing times will have to improve for that but it's a worthwhile technique. Preprint 8545.

On the other extreme, that of high quality playback, Jason Hockman and friends talked on "Discrimination between Phonograph Playback Systems" in which they described some blinded studies between various high quality turntables and decided that telling the difference between good playback chains is difficult. Sure, that seems obvious on the face of it, but they quantized it in a useful way. Preprint 8547.

Connection Gadgets

Sescom was at the show for the first time in perhaps twenty years. If you don't know these folks, they make small audio transformers of reasonable quality, and lots of gadgets that use them, as well as a wide variety of standard problem-solving cables. These guys make all kinds of nifty connection gadgets that have become standard in the industry.


Electroswitch and NKK were there showing off switches, and NKK has some little pushbuttons with small LCD video displays on them so you can change the image on the button on the fly, while Electroswitch has some excellent high reliability rotary, toggle, and conventional buttons.

The Burr-Brown division of Texas Instruments was showing off some very high performance op-amps intended for audio use, the OPA1662 and OPA1652 families.

Very low noise, very low distortion although only available in surface mount packages. When I asked about old-style DIP packages I was told that was a limited market today.

Oliver Archutt from AMI/Tab Funkenwerke was showing a very wide variety of replacement components for classic German microphones, including transformers, body parts, and even a well-made replica of the classic CK-12 capsule.

Per Lundahl from Lundahl transformers was showing their usual line of transformers for input and output stages, but what is really interesting this year is that they now have a shielded transformer enclosure for their big power and output transformers for tube amps. Using these enclosures not only reduces magnetic leakage but also makes mounting the big c-core transformers on a chassis much easier. This is definitely a handy gadget to make using good audio transformers more convenient.

EV/Eminence was showing off a really nifty speaker protection device called D-Fend, which actually comes out of the SLS Audio laboratories. This is a very small board that contains a microcontroller, powered by the audio signal, and some power transistors which provide a series resistance that increases when the speaker comes close to damaging itself. It can be programmed externally over a USB port to set the device set point as well as the shape of the cutoff curve. Unfortunately although the literature claims it is" patented technology," the board is marked "patent pending" and they do not yet have a patent that describes the exact technology. Still a very interesting protection idea and one that would have been impossible a short time ago before the low-power microcontrollers became available.

Twenty years ago, Marshall Leach from Georgia Tech (who sadly just died recently, and for whom there was a short memorial at the show), came up with a SPICE model to emulate a triode for the purposes of circuit simulation.

In the decades that followed, various people have improved on those models and in "A Triode Model for Guitar Amplifier Simulation with Individual Parameter Fitting", Kristjan Dempwolf, Martin Holters, and Udo Zoelzer from

Helmut Schmidt University in Hamburg extended the current physical models to include grid current. This allows the model to remain faithful under various abnormal circuit conditions including clipping. Although they did so in an attempt to model guitar amp distortions, a more accurate tube model makes tube system modeling more faithful for everyone, from RF folks to hi-fi amp designers. Preprint 8507.

Another paper, which has some problems but is worth looking at, is the peculiarly-named "Why do tube amplifiers have fat sound while solid-state amplifiers don't" by Shengchao Li. The author makes some odd assumptions about amplifier design, lumping a lot of different designs together, making the invalid assumption that hi-fi designs can be evaluated by the same standards as guitar amplifiers, and suddenly jumping from push-pull designs to triode designs without pointing out that they have totally different sound. However, his math model of the single ended triode amplifier is worth looking at anyway even if it bears little connection to the presumed subject of the paper or the conclusions. Preprint 8536.


RME, Dorrough, and TSL Professional Products group were all showing off digital metering systems. TSL was also showing off some rack mounted monitoring units with calibrated metering and some little speakers for basic checking applications in the video industry.

Down on the papers floor, Fabian Begnert, Hakan Ekman and Jan Berg from Sweden was talking about the "Difference between the EBU R-128 meter recommendation and human subjective loudness perception". People have been looking for a meter that actually measured subjective volume since before the original VU standard of the 1930s, and the most recent measuring standard is EBU R-128. They find there is good matching between the measured value and the subjective loudness level found by their listening panel, but that there were some consistent differences. Some highly compressed material was felt to be louder than the R-128 meter indicated, but overall the correlation was very high. Preprint 8489.

Etani Electronics from Japan was demonstrating their ASA-10 Mk II measurement system, a small box that connects up to a laptop to allow you to perform swept sine measurements, impulse response measurements, FFTs, impedance vs. frequency plots, all of the things you'd want to do in the field for checking a room or speaker system out and a very handy Swiss army knife kind of tool.

R&D Team was showing VACS, the "Visualizing Acoustics" software that provides a lot of interesting different ways of looking at datasets of acoustical measurements.


Lots of fine work being done on listening tests, and correlating listening tests with objective measurements to make both of them more effective.

Probably the most interesting paper I have seen in years was "Distortions in Audio Op-Amps and Their Effect on Listener Perception of Character and Quality" by Robert-Eric Gaskell, Peter E. Gaskell, and George Massenburg. In this test, a number of similar op-amps were tested in a two-stage moderate gain circuit, and in a blind listening test, listeners were able to tell the difference between them. Good correlation existed between the subjective preferences and the distortion spectra, but in a complex way that actually favored higher distortion op-amps. Well worth checking out for anyone interested in listening tests. Preprint 8503.

In an attempt to see how accurate the recent media reports on listening habits of teenagers was, Sean Olive from Harman International did a short study called "Some New Evidence That Teenagers May Prefer Accurate Sound Reproduction." He played various tracks in various ways to a group of 18high school students enrolled in the performing arts program (admittedly not an average population) and found most of them did prefer better sound. Preprint 8583.

Another interesting paper was "The Effect of Head Movement on Perceived Listener Envelopment and Apparent Source Width" by Anthony Parks and Jonas Braasch at Rensselaer. Many listening tests ask listeners to hold the heads without moving them, but is this a realistic way to listen to anything?

Parks and Braasch find that the perceived image width is greater if you are able to move your head, but that the basic sense of envelopment doesn't change appreciably. This both shows that tests need to account for these effects and secondly it says something about how the human brain detects position. Preprint 8567.


I'd never really thought much about this, but there seemed to be a lot of interest in some of these panoramic record and playback systems.

Visisonics from College Park, MD. was showing their Real Space Audio Camera. This is a small ball, smaller than a basketball, with 64 microphones and 5 video cameras mounted around the circumference, for recording audio and video in all directions.

On the other end of the chain, Pavlos Tripodakos was showing video of the Soundsphere, a system that allows presentation of multichannel audio in a spherical arrangement. He had pictures of the system being demonstrated but wasn't able to get the device shipped in from Greece for an actual demo.


This year there were two different companies that set up booths on the tradeshow floor looking to hire engineers, and that's a good sign considering how many folks are out of work these days. Gentex was looking for people to do all sorts of design work for automotive safety products, while TV Globo, a huge television network out of Brazil, was trying to hire English-speakers to come down and work in Brazil as they continue to expand.


The three big audio magazines: Recording, Mix, and EQ all sent representatives to the show. Radio Magazine was there too.

Vance Dickason from Voice Coil was there, and I might add that Voice Coil also does a great AES show preview which goes into details on the paper sessions regarding loudspeakers, and is well worth checking out. Audio Amateur, published by the same outfit, was there, as was Elektor, a European magazine that they are now distributing in the US.

Of course the usual Journal of the AES staff was there, and I want to specially thank Marry Ellen Illich, the associate editor of the JAES, for helping me get press stuff straightened out.

Folks from Steve Ciarcia's Circuit Cellar magazine were there, and that's a fun magazine that started out mostly as digital electronics and now is almost entirely devoted to embedded computer systems and applications... and there's a lot of embedded control stuff in typical audio gear these days.

And the two big British magazines, Resolution and Sound On Sound had representatives there and booths. I continue to be struck by how much more serious-minded and technical the British audio press is than the American.


Studioshare had a booth.... these folks are audio.studio.share.org and they provide a way for studios to safely and reasonably exchange equipment and services. Wish there was something like this for the high end community so you could hear other equipment and get a sense of what it was like.

Runnur was selling a really nifty little over-the-shoulder bag that seemed very convenient when you're also lugging a lot of production audio gear. It's interesting, because they were actually selling them on the show floor and I thought that was forbidden by the convention contract. Even so, it was definitely a neat gadget with a clear market in the audio industry.

Syncheck.com was selling a really ingenious gadget for testing video and film system synchronization and finding out what the video and audio coding latency really is. In a world where synchronization seems all to have gone wrong and nobody gets clean lip-sync any more, this is a thing whose time has definitely come.

There was a workshop entitled "Neodymium: Coping with the Consequences of Supply And Demand Elasticity" that looked very interesting since Nd is a critical ingredient in making most modern permanent magnets and can sometimes be a big component of total transducer cost. I wasn't able to attend it, sadly, due to time but I'd like to mention it here just as an example of some of the really interesting side topics at the show.


It was a good show. It was smaller than in previous years but there was still way more material than one human being could see. I'd actually like to see the show floor upstairs continue to shrink and continue to be concentrated on smaller companies with newer products; I worry the huge growth in the show floor in the 1980s and 1990s kind of detracts from the main purpose of the show. But it's still a unique and wonderful thing and I encourage all to visit it.


pf logo