AES 2013 Show Report
The AES show time came around again and I went off to San Francisco. I always enjoy writing about a pro audio show for a high end audio audience, and this year the show had more high-end stuff and fewer big consumer vendors than ever. I view that as a good thing, although a lot of people seemed to be upset at the continuing numbers of big box vendors pulling out of the show. I rather liked that.
The thing that makes the AES unique is that there is a scientific conference and a trade show taking place at the same time, and there is some limited interconnection between the two. Because of that, the AES is the only trade show where you can actually find the engineers that designed the products out on the trade show floor, and it's a place where the marketing people can come into contact with people who actually know how their products work inside.
This kind of cross-pollination is what makes it such a great show and that is continuing again this year.
One of the new additions to the show is the engineering briefs, and this is the third show at which it's been tried. Engineering briefs give people an opportunity to present research that may not be complete, or to give a recap on existing research in a field. It's a great way for people to present interesting research that may not warrant a full paper. While published paper preprints cost money, copies of the engineering briefs are available free to AES members in an attempt to publicize research.
Even with the reduced size of the show, I don't get a chance to see more than a tiny fraction of what is out there, and I don't get to write about everything I see because it wouldn't fit here anyway. So if I missed out on something wonderful and important, don't be surprised (but feel free to let me know so I can look for it next time around). If I got something wrong in my rush from booth to booth, definitely tell me so I can fix it.
It seems like everyone and his brother was introducing another new microphone, and most of them were more of the same old Chinese products in different new packaging.
One microphone that was very much not like that was the AEA KU4, which was a slightly altered version of the old RCA KU3A microphone, which was basically a custom ribbon mike built by RCA for dialogue recording on film sets back in the 1930s. Those microphones have become legends, but very few people have ever actually used them because so few were made, and the ones that were made were somewhat inconsistent because they were made by hand. AEA's new mike brings some degree of consistency and availability to the design.
Mark Fouxman from Samar Audio Design also had some interesting new cardioid ribbon microphones: one model was using a labarynth arrangement like the RCA BK-11 and 77DX, while another took the Altec track and used an omnidirectional capsule mixed with the ribbon. These folks are fairly new but have some very carefully built and well-thought-out designs.
Microtech Gefell, the former East German branch of Neumann which became independent when the Berlin Wall went up, is showing another new mike in their great line of reconstructed vintage designs. Their CMV 563 M7S is a reissue of the design they made for many years using the M7 capsule. The M7 uses a PVC diaphragm and as such sounds very different than any of the modern microphone designs, and the tube circuit uses a standard European EF86 pentode. Very much worth checking out.
A company from Finland, Sandhill, was introducing a new ribbon microphone that looked interesting. They're also using some manner of high tech new ribbon material that is designed to be difficult to damage, and an internal pre-preamplifier stage inside the microphone so it can be used with consoles that may not like working with conventional ribbon microphones. The Sandhill 6011A is definitely worth investigating.
Audio-Technica was introducing a new vocal condenser microphone, the AT5040.
They use four sub-capsules together to make something with the noise performance of a large capsule, but with the breakup modes of a smaller one. I wasn't able to listen to the thing properly so I don't have a sense of what this does to the pattern, and I wasn't able to talk to anyone who could really explain what was inside it. The electronics are interesting; it looks like there are two identical signal paths inside, but only one output and no fancy pattern control stuff that would require splitting the signal path. So after the introductory release at the show this microphone still remains something of a mystery, and it's probably the first thing I want to check out once things calm down. There were only a few papers about microphones this year, and I went into On the study of Ionic Microphone by Hiroshi Akino and Hirofumi Shimokawa of the Kanagawa Institute, and Tadashi Kikutani and Jackie Green of Audio Technica expecting little. But this presentation was actually fascinating. These folks generate a small cone of plasma in an inert gas using an RF source at 27 Mhz, and when sound waves alter the shape of the plasma cone the load on the oscillator is altered and the signal frequency changes. With an FM detector it is possible to get surprisingly good output. Frequency response and noise floor are still far poorer than conventional microphones but it's an ingenious device that may well have a future ahead of it. Preprint 8745.
The soundfield microphone is an array that permits various different pickup patterns to be synthesized after the fact, for a lot of control over the stereo image. The conventional tetrahedral array that is used, however, has some pattern control issues at high frequencies. In A second-order soundfield microphone with improved polar pattern shape, Eric Benjamin from Surround Research provides an alternative configuration that retains a good pattern up to 20 KHz. Very interesting for anyone doing recording work with the soundfield system or contemplating it. Preprint 8728.
There were a few papers on stereophony and various new stereo miking configurations. In A Microphone Technique for Improved Stereo IMage, Spaital Realism, and Mixing Flexibility, James Tagg from McGill University describes STAAG, a system using two near-coincident cardioid pairs. One is pointed forward at the sound source, the other pointed to the rear of the hall for ambience pickup and the two are mixed in post in order to give after-the-fact control of ambience. E-Brief EB71.
With the coming of surround, there are still a lot of new configurations for minimalist miking of 5.1 surround recordings. One of the approaches was described in Bellamy Baffle Array: a multichannel recording technique to improve listener envelopment. This employs fairly conventional near-coincident pairs pointed to the front and rear of the room, with a baffle in-between them and a center front microphone added. It seems reasonable and controllable. Engineering Brief EB59.
Not really a microphone paper per se, but a "where to put the microphone" paper, was Measurement and analysis of the spectral directivity of an electric guitar amplifier: vertical plane by Agnieszka Roginska, et al. Everybody knows that if you listen on-axis to a guitar amp you hear more high end than you hear out near the edge of the cone, because of the beaming of the high frequencies... but nobody has measured it. These folks measure it using a microphone array, in a way that provides a very useful and complete measure of sound with angle and distance. Preprint 8708.
SE Electronics was showing off an egg-shaped speaker called "The Egg." Every few years someone promotes elliptical or egg-shaped speaker cabinet designs and most of them run into serious problems with internal modes, but on a quick listen on the show floor this one sounded decent and it could be worth checking out in a better environment. The demo had some odd imaging and some resonances that may well have been the fault of the environment.
JBL was doing a demo of a new monitor speaker with a strange cross-shaped horn on the top end. It was billed as a horn speaker for people who don't like horns, but unfortunately the demo did not run on Monday and by the time I had a chance to get down there I missed it. Everyone who heard it told me how great it was, but the best thing I heard at the show was the ATC SCM25A Pro monitor speaker.
Sennheiser makes an amazing line of small array speakers called the K-Array, but once again they didn't actually have any on display. I'd really like to see Sennheiser bringing some of their more unusual products out to the show again.
A Chinese company called Xycad was showing off some sort of very thin and flat passive subwoofer they called the Transform 124SUB. Looked very interesting for installed sound applications where it can be difficult to physically fit into rooms, although nobody there seemed to really know much about it. The same folks talking about it also had some information about TASE Tech which seemed to make sound systems for large sports venues in China. I'd like to have known more about these folks.
Down in the paper sessions, Marshall Buck, David Graebner, and Ron Sauro had a paper called Long Distance Induction Drive Loud Hailer in which they described a horn and compression driver system that gave good vocal intelligibility nearly 700 feet from the source even in reverberant outdoor environments with reflective building surfaces. Not high fidelity at all, but a very interesting approach that anyone interested in horn systems should look at (along with the authors' earlier paper in 2009 on the compression driver design). Preprint 8747.
In Study of the interaction between radiating systems in a coaxial loudspeaker, Alejandro Espi and other folks from Beyma and the Universidad de Alicante in Spain create a finite element model of a coaxial speaker system and investigate various ways in which the mid and high frequency drivers can interact with one another. This includes displacement dependent linearities which are caused because the horn of the treble driver is the cone of the midrange driver which changes position with midrange signal. They came up with a lumped sum model to estimate the modulation effects between the two and built some hardware to verify the model. This is very important work; a lot of people have opinions about how interactions between coaxial elements work, but very few people have actually modeled them. The coaxial configuration is the only option when tight pattern control and flat response off-axis are both needed, as in many studio monitoring situations. Preprint 8724.
Some polyurethane compounds expand and contract when an electrical charge has been applied to them, and there has been some work done in using these materials for loudspeaker drivers. In Flexible Acoustic Transducer from Dielectric-Compound Elastomer Film, Takehiro Sugimoto and others from NHK and the Tokyo Institute of Technology improve the total output power of one of these devices by compounding the film with a dielectric material added to it. They evaluate a number of dielectric compounds and propose a mechanism. Preprint 8725.
Another curious sort of driver is described in Printable Loudspeaker Arrays for Flexible Substrates and Interactive Surfaces, in which Jess Rowland and Adrian Freed use planar voice coils made either by inkjet printing, plating, or machine-cutting, combined with flexible refrigerator-style magnets, to create small flexible planar drivers. Engineering Brief EB68.
In Experiments of sound field reproduction inside aircraft cabin mock-up, Philippe-Aubert Gauthier and other folks from the Universite de Sherbrooke and McGill University in Canada construct an array of very small transducers to provide an accurate three-dimensional soundfield inside a simulated aircraft cabin, for the purposes of testing and demonstrating interior sounds of existing and proposed aircraft. They provide objective evaluations of sound fields and basic preparation for future listening tests, and the configuration and discussion of sound fields is worth seeing. Preprint 8716.
Ambrose Thompson from Martin Audio presented an engineering brief on Parametric Horn Design in which he described a first cut of a Matlab model that allows design engineers to estimate performance of a horn system given a simple geometric model. Although not a high-fidelity simulation, it permits simple and rapid design iteration because it takes little time to estimate performance of a given design. Engineering Brief EB72.
In Spherical Sound Source for Acoustic Measurements, Plamen Valtchev, Dimitar Dimitrov, and Rumen Artarski describe an omnidirectional radiator using conventional coaxial drivers and waveguides. Is the response off-axis clean enough to use for listening? I don't know, but the response is far better than the conventional dodecahedron arrays that are normally used for making room measurements, and that's certainly an advance. Preprint 8706.
Juha Backman from Nokia talked on Midrange Resonant Scattering in Loudspeakers, describing scattering effects where the external sound field is reflected off ports, recesses, or horns. He describes a simple mathematical model for evaluating these effects and distinguishes them from simple cabinet diffraction effects. The claim is made that loudspeaker cabinet layouts can be improved by optimization of positions to reduce these effects. Prepring 8746.
Sean Olive and Todd Welti from Harman International gave a paper called The relationship between Perception and Measurement of Headphone Sound Quality, which explored some of the problems with measuring a transducer which is affected by the listener's physical body itself. They tested a number of headphones on a listening panel and found that most people preferred the headphones that had a modified diffuse-field response as measured either on the listener's head or with an IEC "standard ear." This is in conflict with the IEC standard which suggests headphones should be designed for a specific standard response to compensate for effects of the head. This work is definitely worth checking out and important; there are a lot of differences between listeners on most headphone tests and this makes headphone standards difficult. Preprint 8744.
A similar experiment was conducted from a different direction by Felix Fleischmann, Andres Silzle, and Jan Plogsties of the Fraunhofer Institute. They try various target response curves in a formal listening test using binaural signals and have listeners compare them with loudspeaker playback. The end curve they come up with isn't that much different than the one Olive and Welti do, but they approach it in a different way which encourages my belief that both of these groups are on the right track. Preprint 8740.
With the coming of the API 500 module standard, a lot of companies are now coming out with modular mixing consoles using API-format modules to allow you to mix-and-match bits and pieces from different manufacturers. I didn't have time to check out a fraction of the ones on the show floor, but there were some very impressive-looking ones out there.
One that did strike me was The Roots, an 8 or 16 channel 2-buss console designed by Steven Firlotte, the inventor of the Vac Rac modular tube gear, and a fine recording engineer in his own right. This is precisely the form factor needed to do conventional recording workflow with a digital workstation for small jobs, and Tree Audio which is making the system is definitely worth looking into.
Undertone Audio was showing off another one of their custom consoles, not designed around 500-series modules, but with some very interesting acoustical trapping intended to reduce reflections off the console surface.
Steven Slate was showing a digital console called the Raven X-1 with a user interface designed to work in the iPhone and YouTube world, and built with a touch screen control for most functions. Definitely a very different approach to mixing.
For studios without real consoles, Dan Lavry is introducing a device called the LK-1 Latency Killer. This is a simple mixing box that can be introduced between a mike preamp and a DAW to provide a headphone cuing mix with no converter latency between the performer's microphone and headphone. It is a very convenient little device, and even with a standard console being used in the studio it can provide a way for the performer to control what he is hearing in his own headphones a bit. Not a revolutionary gadget, but a useful tool to be used between a preamp and converter, from the guys who brought you some very fine preamps and converters.
It seems everyone and his brother has a new preamp. All of the old-line companies like Millennia Media and Great River were there showing their traditional gear. John Hardy was there, still promising a 500-series module but he still doesn't have it available yet. Maybe next year! Mr. Hardy was spending a lot of time trying to convert the Jensen JE110 discrete op-amp to a design that could be readily made with surface-mount parts, since the supply for larger through-hole parts is becoming much more expensive as they become specialty items.
One thing that was exciting was a company called MOON which was selling their 3500MP, an all-solid-state transformerless preamplifier using one of the traditional large transistor array front ends. Very interesting design, with great attempts to avoid all electrolytics in the signal path and to shield many of the large film capacitors that demand requires. Stepped attenuators for gain control for repeatability, and a very nice design that is worth looking into.
I didn't see a whole lot exciting on the show floor on the signal processing front. Probably the most interesting thing was that Chandler, Ltd., which has long been a manufacturer of high end preamplifiers and mastering consoles, has introduced a couple of guitar pedals. I know nothing about guitars but the Chandler people seem to and I'd be interested in the opinions of people who do and could try these.
On the digital signal processing front, Direct Out Gmbh was showing a variety of handy signal conversion boxes, including ADAT Lightpipe to MADI, MADI to USB, and so forth. All very much the kind of thing you don't think about until all of a sudden you need it urgently, and the kind of glue that holds studios together.
There were a bunch of papers on signal processing, and most of them were either totally out of my field or at some time in the day when I was unable to be there. I didn't see a single paper on dereverberation this year, which was kind of disappointing. But there were a couple of very interesting papers.
In Matching Artificial Reverb Settings to Unknown Room Recordings: A recommendation System for Reverb Plugins, Nils Peters, Jaeyoung Choi, and Howard Lei from Berkeley employ a statistical machine learning algorithm to match an unknown recording up with various standard presets of a digital reverb device, in order to suggest to a mixing engineer what options might best match the original track. This is a very handy thing for people like film sound mixers who might be adding effects into an existing room recording, and want the effects to match the original sound. Preprint 8700.
In Temporal Coherent-based Howling Detection for Speech Applications, Chengshi Zheng and co-workers from the Chinese Academy of Sciences describe a very easy-to-compute feedback detection algorithm, as could be used for automated tuning of PA systems and teleconferencing systems. They describe the algorithm and some measurements, but unfortunately, although they describe the detection time, they don't describe how far below the average speech level a ringing signal can be detected, and that one measurement is paramount for knowing how effective such a method could be. Preprint 8797.
John Lazzaro and John Wawrzynek from UC Berkeley were presenting a poster on A Tilt Filter in a Servo Loop. They had a very wide filter that made broad sweeping changes to the overall tonality, much like the "tilt" control on Quad preamplifiers. They implemented it in software and then had it controllable dynamically so it could be used to make broad sweeping changes based on program material. An interesting trick, it may have some useful applications in producing new sounds. Preprint 8692.
From my perspective the most exciting cable product was Belden's 1776 "Super Strong Microphone Cable" that was made as a reaction to some European rock bands that were destroying cables. These have jute strength members capable of holding several hundred pounds and a super-thick jacket, without being unpleasantly stiff. Lots of people talk about sound quality in cables, but it doesn't matter what it sounds like when it's broken. Here's a cable that won't break in bad conditions.
Gepco and Mogami were also selling some very high quality cable materials.
There was a booth that seemed to be seeing a variety of audiophile cables, including the XCellus cables by JIB Germany and the E-Spirit cables from Japan. I couldn't tell exactly who was running the booth, and the folks there seemed more interested in telling us how much the cables cost rather than why anyone would want to be using them. Long-term readers of my articles will recall my great skepticism in the face of the whole audiophile cable industry; it's easy to make cables that sound different, but in the end it's hard to tell if that means they are better or worse. That attitude is shared by many at the show, and that means if you're trying to sell audiophile cables into that market you have to do better than just say "look at this cable, it costs $7000, it must be great."
Electroswitch, NKK, and Schurter were all showing off various lines of switches and controls. Schurter also had some nice power input modules and fuse holders. Not high tech stuff, but absolutely essential for products that actually work.
THAT corporation was introducing another low cost microphone preamplifier chip, this one a little different. The THAT1583 is an inexpensive preamp front end chip that is designed to work with their digitally-adjustable gain control chips. What's interesting about it is that it can be used standalone without the fancy gain control, to provide a differential input, differential output device. The differential output is set up to directly drive typical sigma-delta A/D converter chips but could be used for all kinds of other things too (and yes, it has enough current to drive reasonably long cables directly).
Verina Fuchs and co-workers from the University of Siegen and Camco in Germany spoke on Investigating the Benefit of Silicon Carbide for a Class D Power Stage. Silicon carbide MOSFETs have lower on-resistance and lower gate capacitance than comparable silicon MOSFETs and the authors take a standard MOSFET class D output stage and replace it with a silicon carbide FET from Cree with some improvement. This is an interesting look at some of the newer devices coming out today. Preprint 8686.
In addition, V. Sala and some other folks from the Universitat Politenica de Catalinya talked about Evaluation of tff Distorting Effects Reduction in DCI-NPC Multilevel Power Amplifiers by using SiC Diodes and MOSFET technologies. Building multilevel diode-clamped-inverter amplifier stages has become popular in recent years in order to get very high power levels with reasonable distortion levels and good efficiency, but the reverse recovery time of the clamping diodes and switching FETs becomes a problem. These folks investigate using some of the new silicon carbide devices in these applications. Preprint 8720.
HIGH SAMPLE RATES
There wasn't a single mention of SACD anywhere on the show floor this year, or any DSD gear. There were some high resolution converters, but there was not the huge amount of promotion going on.
In How Can Sample Rates be Properly Compared in Terms of Audio Quality? Richard King and others from McGill University discussed the various possible errors that can alter the results of listening tests of sample rate conversion. They then used a Yamaha Disklavier player piano as a sound source to record the same piece several times at different rates. They did not find a conclusive difference in sample rates on a properly blinded test and discuss why that might be the case and how to more carefully check the matter. Engineering Brief EB77.
A couple of new vendors this year! Agilent was showing their U8903A audio analyzer, a handy device with a lot of signal generating and analysis capability in a single portable box. Rhode and Schwartz was selling a similar box, the UPV, as well as their PR100 radio monitor for interference analysis. The PR100 is a wideband receiver with a digitized IF, so you can see a 10 MHz wide slice of bandwidth at one time. This means intermittent sources that would be lost on a conventional spectrum analyzer can be seen.
Klippel and Etani were showing off the latest versions of their speaker analysis systems, and NTI had a number of small pocket audio analyzers.
A lot of inexpensive compression drivers and small microphones have diaphragms vacuum-formed from various thermosetting films. Victrex was showing off a new film material called Aptiv which uses a new PEEK polymer that is more rugged and stiffer, for reduction of break-up modes. It looks like a material that might also be usable for dynamic microphone diaphragms as well. It's a polyaryletherketone if that helps any of the plastics guys reading this.
Also on the plastics front, Globe Plastics, Inc. was showing some high end composite materials intended for speaker horns and cabinets. Very dense stuff for reduced body and cabinet resonances.
Materion Electrofusion was back again this year showing their Truextent beryllium diaphragms for compression drivers as well as other beryllium materials made with their patented process. They're able to make thinner and lighter diaphragms than anything made with conventional processes.
Many folks on the show floor were demonstrating acoustical materials, from sealed doors and standalone voiceover booths down to foam and diffusers. I didn't have any time to look seriously at any of that stuff, but I did see a couple papers.
In A Computational Acoustic Model of the Coupled Interior Architecture of Ancient Chavin, Regina Collecchia, Miram Kolar, and Jonathan Abel from Stanford's CCRMA present a model of the acoustical properties of an ancient Peruvian temple, designed as a tool for archaeologists studying the ceremonial uses of the hall. The authors treat the individual hallways as reverberant waveguides rather than modeling individual surfaces, which results in an easy to calculate reverberation algorithm that accurately matches measurements taken on site. Preprint 8696.
Lots of audio magazines brought reporters, staff, and circulation people to the show, and there's no time for me to mention all of them so I will just mention a few favorites.
First of all is the British magazine Resolution, with their large format magazine about audio production work, which was celebrating their tenth anniversary this year at the show. Sound on Sound, another great British magazine, was there as well. These two magazines seem to have taken the entire market for high-end technical magazines.
MIX magazine, the American publication about audio production was there as well, although they seem to be getting smaller and less technical with every issue. Recording magazine, a great magazine for amateur recordists (for which I occasionally write) was there, as well.
Broadcast Engineering magazine was there as well, and they also seem to be getting less technical every year, but then again so is the whole broadcast industry
Live Sound magazine was there, as well as FOH magazine, another good source of information on the live sound world. FOH was also giving out copies of EPD, the 2012 Event Production Directory which is a good source of info on suppliers of services for live event work.
AudioXPress, the magazine which once was Audio Amateur, was showing off their usual range of articles about DIY construction, but they were also showing off the Elektor and Circuit Cellar DIY magazines which they also distribute now (as well as Voice Coil, a magazine for the professional speaker design industry). They also distribute the Linear Audio series of books on audio electronics design, sort of the Evergreen Review of electronics.
Nonlinear distortion has been recognized as a problem since the 1930s, but the traditional measures (like THD and IMD) aren't very useful today for comparing low distortion levels of equipment with varying topologies. Lots of people have been at work at improved measurements that will correlate better with listening, and in Nonlinear Distortion Measurement in Audio Amplifiers: The Perceptual Nonlinear Distortion Response, Phil Minnick describes a system where random noise with one narrow band removed is applied to a device under test and the level of the signal in that band is then evaluated. This is repeated for 25 different bands, and each of these then weighted by audibility of each band. This method gives no ability to weight each individual harmonic, but it does allow the tester to excite a large number of possible distortion sources at once. The measure seems to correlate well with a short listening test. Engineering Brief EB69.
In Another View of Distortion Perception by John Vanderkooy and Kevin B. Krauel from the University of Waterloo, the authors describe a circuit that generates primarily second harmonic distortion that is independent of signal level. Although they don't do controlled testing of audibility, they lay the foundations for doing so, and they make some very interesting observations of the audibility of such distortion. Engineering Brief EB76.
A company called DSP4YOU out of Hong Kong was showing a very slick little standalone DSP box called the OpenDRC. Using the SHARC processor, it provides a standard hardware box that vendors can load with their own software, or can use as a prototype before building custom hardware. They also had some handy devices for streaming audio over Ethernet with the new AVB standard.
QES Labs out of Italy had some sort of mastering A/D converter. Transformer coupled, tube or transformer input stage that could be selected at a button, A-D conversion architecture that they say is new and different although nobody there seemed to know anything about it. Could turn out to be interesting, although it was hard to tell with the presentation.
Resinno was showing their interesting new piano legs at the DPA booth. These are legs that can be retrofitted onto many standard grand pianos for a very distinctive look and (they claim) a sonic improvement. Very cool-looking. Why is it here at an audio show? I'm not sure, but it's fun anyway.
Sound On Sound sponsored a project studio expo program with various presentations by Sound on Sound authors and advertisers. It really felt like an attempt for the AES to try and get some of the younger project studio crowd into the event. I hope Sound on Sound paid well for the privilege, because it didn't really feel like it belonged at AES; it was sort of a mixture of tutorial and marketing, without it being clear about where the line was.
Another thing that seemed like it was at the wrong show was the Burl booth where they had two scantily-clad young ladies handing out material. That's the sort of thing that you expect to see at CES or NAMM, but not at the AES show.
Jay and Jeffrey McKnight gave a talk called "Some Popular Misconceptions About Magnetic Recording History and Theory," in which they talked about various aspects of AC bias and gap compensation which aren't well-understood by the average analogue tape engineer.
Allen gave a lecture he called the "egg show" talking about the influences of sound on motion picture production, and discussing some of the various uses of soundtracks to create a given feel to a film. This was also given as one of the historical session talks because there wasn't a philosophical session to put it into.
This year's Heyser Lecture was given by J. J. Johnston from Bell Labs. The Heyser lecture is an annual invited talk featuring distinguished lecturers, and there is nobody more distinguished and fascinating than Mr. Johnston, who has done more to make audio coding as transparent as possible as anyone else in the industry. I had to duck out of the lecture early in order to attend another event but what I saw was witty, dry, and informative, all at the same time.
San Francisco was full of police. The Sunday night of the show was the last game of the World Series, and coming back from dinner our car was mobbed by huge numbers of people streaming down Market Street, jumping on our hood and yelling "Let's Go Giants!" We did miss the bonfire in the median, though, but were kept up by police helicopters circling until the wee hours.
The next day I was approached on the corner of Market and Jones by someone who wanted to know if I wanted to buy some weed. Since you can get a prescription card to buy marijuana legally and there is a dispensary on every corner in the city, I can't imagine why anyone would ever be trying to sell weed on a street corner there. My friend Kristen thinks he was a narc and if she's right, he wasn't a very smart one.
Then the day after that, my walk to the Ferry Building was blocked by a police tape cordoning off most of a city block because a suitcase had been found tied to a tree (probably by some poor homeless sod looking for a place to stash all his belongings). Easily a hundred police officers, lots of yellow tape, and the now-familiar police helicopters circling constantly.
This year was a smaller, quieter show up on the show floor, but there was more stuff than ever down in the paper sessions. The addition of more tutorials and the engineering briefs definitely improved the paper sessions, and brought a few more people in that might ordinarily have stayed upstairs. The series of vendor sessions helped provide some additional mixing of the two groups, which I think is a fine thing. I hear people constantly complaining that the show is getting smaller and that NAMM is taking all of the big vendors away, but to my mind it makes it more like the kind of show I want to attend.