d328eae7   
   From: No.Spam@No.Where.com   
      
   You're here to piss and moan over something Al wrote 5 years ago?   
   Maybe if you spent a day in Al's shoes you'd grow up a little.  But I   
   doubt it.   
      
   Fred   
      
   Bret L wrote:   
   > A pretty good article crapped on by Al:   
   >   
   >   
   > Listen Up, Horn Guys!   
   > By John Atwood   
   >   
   >>> "Geddes - Audio Transducers   
   >   
   > I'm posting this review of Audio Transducers written by Lynn Olson,   
   > since Lynn has been on vacation and has been busy with family matters.   
   > We wanted to wait until I had read the book and put in my two cents   
   > worth. I have recently finished it and my review is after Lynn's. Now   
   > for Lynn:   
   >   
   > Even though I'm a notorious horn skeptic, I have to admit they do some   
   > things well - dynamics most of all. Where I part company with horn   
   > enthusiasts is their persistent denial of horn coloration, or the   
   > claim that horns have better time response than other drivers (they   
   > don't), or the horn somehow damps diaphragm resonances (it doesn't).   
   > The most serious coloration are the modes, or resonances, induced by   
   > the horn itself - these colorations appear in the ripples seen in the   
   > impedance curves, reflections in the time domain, or variations in   
   > dispersion, as shown in Earl Geddes' landmark new book, Audio   
   > Transducers.   
   >   
   > To make horns, or any audio device, better, first we must understand   
   > how it works, and what causes the faults. Defending designs that date   
   > back to the late Thirties, and have seen little improvement since, has   
   > to give way to a better understanding of underlying principles, and   
   > most important of all, finding out if the underlying set of   
   > assumptions of the design are indeed accurate.   
   >   
   > The greatest impact of Thiele/Small LF design in the early Seventies   
   > was putting bass enclosure design on a rational and predictable basis   
   > for the first time in the audio industry. Prior to T/S, bass   
   > enclosures were designed by rules-of-thumb that could be off by as   
   > much as 6dB - in fact, there wasn't even agreement how to measure   
   > them! After Thiele and Small, modelling provided actual closed and   
   > vented-box systesm that were within 0.5dB - or better - of the model.   
   > That's the measure of how just much improvement a more complete   
   > understanding can provide. Similarly, Butterworth, Linkwitz-Riley, and   
   > modern Target Function Design (computer optimization) crossovers are   
   > far better than the grossly inaccurate "M-derived" filters of the late   
   > Forties, which were extremely complex and difficult to design - and   
   > worse, nowhere close to the performance of a real-word crossover.   
   >   
   > It's not widely known that horn theory hasn't progressed all that much   
   > since the pioneering work of Bell Labs in the late Thirties. This   
   > provided the foundation for Altec and JBL theatre horns that became   
   > the basis for the modern high-efficiency speaker of today. The   
   > constant-directivity horns used in theatres and prosound today,   
   > unfortunately, are not about fidelity, but delivering peak SPLs to   
   > every seat in the house, and using digital EQ as necessary to   
   > compensate for the lumps and bumps in the horn response. It's all very   
   > nice that modern horns for hi-fi are made of NC-milled wood, have   
   > Tractrix or other unusual profiles, and have titanium or beryllium   
   > diaphragms, but the underlying horn theory is little different than   
   > what Bell Labs was using in the late Thirties. So the departures from   
   > the theory - the above-mentioned ripples in impedance, time response,   
   > and directivity - are still there.   
   >   
   > The most serious problem with existing horn theory is the assumption   
   > that the horn is of infinite length, which conveniently sets aside the   
   > real-world problems of reflections from the abrupt termination at the   
   > horn-mouth, which then travel back into the horn, reflect off the   
   > phase-plug or metal diaphragm, and return back to the horn-mouth,   
   > setting up a standing-wave.   
   >   
   > Geddes' Audio Transducers is the first book I've seen that sets aside   
   > the obsolete Webster theory, with its severe departures from the real   
   > world, and uses finite-element analysis instead. With this more   
   > powerful technique, the standing waves, or "High Order Modes" as   
   > Geddes refers to them, are revealed within the horn. At the lowest   
   > usable frequencies, horns behave like open-ended pipes, with a series   
   > of pipe modes that start at a quarter-wavelength of the horn length.   
   > At frequencies below this quarter-wave, the horn ceases to act like a   
   > horn at all, and response and power-handling drop very fast.   
   >   
   > The profile (conical, exponential, Tractrix, etc.) controls the shape   
   > of the response at the cutoff (gently sloping vs sharp cutoff), but   
   > doesn't change the fact that the dominant mode occurs at the quarter-   
   > wavelength, and higher-order modes occur at a half-wavelength, 3/4   
   > wavelength, 1 wavelength, and so on. Since this pipelike shape has no   
   > internal damping, the Q of these modes are very high, and are the   
   > direct source of the previously mentioned ripples in impedance, time   
   > response, directivity, etc.   
   >   
   > In Geddes' commercial loudspeaker, the Summa, he takes the radical   
   > step of filling the horn itself with damping foam, sacrificing a bit   
   > of efficiency, but strongly damping the internal modes. The technique   
   > obviously works: the freq response vs directivity curves on the   
   > website are the best I've ever seen.   
   >   
   > The superb curves are aided by Geddes' thoughtful selection of   
   > drivers: the professional-grade B&C DE25 for the HF horn driver, and   
   > the B&C 15TBX100 15-inch driver for the bass. These are far better   
   > than the typical audiophile fare of 87 to 91dB/metre direct-radiator   
   > drivers from the usual high-end European sources. I did briefly   
   > audition the Summa at the last Rocky Mountain Audio Festival, but in   
   > all honesty can't tell you how they sounded. I'm not being evasive   
   > here - as my readers know, I'm as opinionated as anyone in audio - but   
   > the Summas were powered by a bottom-of-the-market $200 Pioneer home-   
   > theater receiver and a $100 Panasonic DVD player from Costco.   
   >   
   > If I understand Earl's comments in Audio Asylum's High-Efficiency   
   > Speaker group; correctly, he believes that mass-market electronics,   
   > DVD/CD players, and wires all sound the same, so the intelligent and   
   > thrifty buyer should save their money and buy home-theater electronics   
   > from big-box retailers. He was serious enough about this belief to   
   > rent a room for a thousand dollars at the RMAF show and demo his   
   > speaker with the electronics I mentioned above - I'm quite sure he was   
   > the only exhibitor using Costco-sourced electronics in the whole   
   > hotel.   
   >   
   > So what did I hear? A very accurately reproduced bottom-of-the-market   
   > Pioneer home-theater receiver and Panasonic DVD player. I listened,   
   > thanked Earl and his very gracious wife for appearing at the RMAF, dug   
   > into my pocket, and bought a copy of his book. If any of the readers   
      
   [continued in next message]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)