How will future audio historians judge this era and Dr. Bose's contribution?

[soundminded]

Once upon a time at least in my living memory, people used to imagine what the future would be like and how their own period in history would be seen through a retrospective lens of some decades or even centuries. People used to put artifacts of their era in sealed "time capsules" for future generations to one day dig up and study as a way of helping them understand their time and what they felt was important. Some have even been sent into space in case alien intelligence should happen upon it. Often the assessment of events in their own time is seen retrospectively as having been distorted, people in their own era not having seeing the forest for the trees and the future rarely turns out to be what most people imagined. Still it can be a lot of fun to conjecture.

So assuming the world will still be around in 50, 75, 100 years, technology will ultimately advance, and people will still be much as they are today, still enjoy listening to and playing musical instruments, still attend live concert performances, still enjoy recordings of music in their homes, what will it be like then, how will they view the current era of technology of recording and reproducing music, and how will the contribution of Dr. Bose be seen as part of the mosaic of this era?

I think I'll hold back on my own thoughts for awhile, I'd like to see what others think first.

[Steve F]

Although it wasn’t always this way for Bose, in the last 10-15 years or so Bose has made a conscious and intelligent decision to orient its products towards the non-hobbyist, non-enthusiast market. The various Acoustimass speaker/theater speaker models, their all-in-one theater systems, the Wave Radio, the Noise-Cancellation headphones, etc. have all been very successful, but also not aimed at the enthusiast market.

What Bose has done well is bring good quality audio to the mass market. They follow a “90-10” sales/marketing strategy, whereby they go after the 90% of the potential market that cares about good quality products (customers who like good clothes, name-brand items, nice cars, good restaurants, etc, in other words consumers who recognize and appreciate quality, but are not necessarily super enthusiasts about any one particular field). They let the other audio companies beat each other’s lights out competing for the ever-dwindling “audiophile” market, whatever that is (maybe 10%) these days. The Wave Radio, 301 and AM-5 may or may not be your personal cup of tea, but they work well, they’re reliable, they deliver the performance Bose says they will, and they do in fact sound much better than the mass-market junkerino sold at Walmart.

Look at the demise of the stand-alone audio retail specialty shop. They’re all but gone. Look at the total evaporation of the ‘audio’ enthusiast magazine industry. Stereo Review, Audio, High Fidelity—gone. Look at our beloved NE speaker brands—gone. But Bose—born in roughly the same time period as the other major speaker brands—flourishes profitable today, under its original ownership. Bose is not an empty marketing name, having been sold and re-sold a dozen times to nameless Asian corporations. Instead, Bose is still the original Bose, still in Framingham MA. All because they produce products that make sense to the 90% mass market, and they eschew the audiophile market, which they have correctly bet would disappear.

How will the component audio market be viewed 50-100 years from today? I think ‘audio’ systems will be non-existent, and all personal playback—audio, video, communications, whatever—will be through totally integrated systems. Most speakers today are part of integrated systems now—radios, ‘iPod docking stations, Theater-in-a-box systems, car systems, TVs, etc. I’d guess that the TOTAL speaker market—all the instances in the world where a loudspeaker is used—is probably 95% powered/integrated, 5% separate/component. Maybe that’s even being generous to the component side of things.

I just inherited some items from an older relative who passed away. There were a couple of 24”-tall vases, very ornately hand-painted and decorated. They’re beautiful, made of ceramic, obviously valuable antiques. I was told that my great-grandfather won them at a carnival game (!!) about 100 years ago. This was what the quality standard was for carnival prizes in 1910—beautiful hand-painted ceramic, absolutely gorgeous.

There will be nothing like the quality of an AR-3a fifty years from now. It will be priceless.

Steve F.

[RoyC]

I completely agree with you, Steve.

Bose products generally do not have the look and feel of something that will become a valued antique, but they have consistently been appealing to a great many people. That by itself will earn the Bose name a prominent place in audio history. The Model T was not the "best" automobile of its day but they were everywhere...

Based on the few specimens I have been into, Bose seems to squeeze relatively high sound quality out of relatively inexpensive parts, so manufacturing costs were probably low compared to manufacturers building fine, furniture quality cabinets with expensive drivers and sophisticated crossovers (like our much loved ARs). That had to be a competitive edge for Bose, and is a testament to good design and engineering. Bose was certainly the first to use non-wooden enclosures on a mass scale.

Bose was the first to recognize and capitalize on the fact that multiple speakers bouncing sound around a given space is generally very appealing to most people...and most people do not really care (or desire) to see where that sound is coming from. Last year a co-worker passed on his "antique" KLH system to me, which had been collecting dust in his basement. He remembered how carefully he had shopped for that system, and marveled at "how far things had come". He and his wife now rave about their Bose system consisting of tiny speakers blending into various corners of their residence, along with the concealed woofers. Unfortunately, for every old walnut 3a that becomes a valued "antique" there will be many, many more residing next to CRT televisions in landfills.

Something often overlooked is that Bose PA/sound reinforcement speaker systems are also very popular.

Roy

[Diamonds&Rust]

Soundminded, and to friends everywhere, Greetings.

Where all of the futurists usually go wrong is to completely miss the opportunity for complete abandonment of trends and account for leaps of technology, preferring to extrapolate the near-term present into the future, or doing some sort of regression into the past and projecting along a linear path. .

Nothing works that way. Spontaneous and lightning-quick mutations seem to be responsible for a great many things in evolution. Thus, while I’ll take a stab at this, I assume that I’m so wrong, so far off, that what follows is laughable.

In 200 years we will be able to generate the experience of music’s effect on the brain without a note’s ever being played.

I assume that in 100 years we will be able to be fed the entire experience of a live performance more or less directly into our brains - via what method I will not speculate. This will include all sensations of “being there” and “there” will be more or less wherever we choose and the performers might also be whoever we choose, but more likely, whatever has been calculated for us to enjoy it the most.

Whether we would recognize what will constitute a performance 100 years from now, I sincerely doubt. Why would anyone spend their lives learning to play a violin when a robot (of sorts - ethereal) can play it *perfectly* for you, emulating any real or imagined performer’s style, or maybe even with our own style synthesized from the best of every performer on record? Effortlessly. Musicianship, now a prized scarce resource, will become the birthright of everyone. Making sound with an instrument will be as quaint as making thread by hand is today - and the result of taking life-long lessons will be about the same as hand-making thread; inferior quality at 1000 times the cost.

Dr. Bose, et al, will be as interesting to that generation as the men who improved the siege engine are to us today. They will all be completely, utterly, forgotten, and their products will be recycled as kindling and illegal mechanisms for simulating erotic experiences via the xxx.net. A speaker’s presence in a room will only provoke disgust - like lighting a fire in a modern oven.

50 years from now, Dr. Bose’s inventions and products will be regarded with the same respect we currently afford quills that have to be dipped in an inkwell. Simulated tactile experiences will be a reality via a dorsal column stimulater of sorts. The very last generation of musicians will be born.

25 years from now Bose et al will be making personal entertainment products that are in a pair of glasses and will project directly into our eyes and ears making the experience of “you are there” as real as they can be without tactile reinforcement.

[RoyC]

Of course this is assuming we are not beating logs and imitating animal sounds, while dancing around the fire:-).

Roy

[soundminded]

I'm glad to see that there were at least some replies. For awhile, I was afraid nobody would venture to speculate about the future or reflect on the past.

First I'd like to reflect on the past and what I think is Dr. Bose's contribution to the art was. A paradigm is not just a model of how to solve a problem but a way of looking at the problem itself. In the problem of designing loudspeakers, Dr. Bose made a clear break with the existing paradigm of his day with the Bose 901. He looked at aspects of the problem from a different point of view and came up with radically different answers. My own analysis is that his paradigm is less than perfect and his execution of a loudspeaker based on it did not fully explore even the potential of that paradigm. Nevertheless, it overcame many problems other designers using the conventional paradigm never even thought about let alone tried to solve. Its limitations are different than conventional speakers and some listeners would rather live with those than what audiophiles consider acceptable. I would like to add that the basic paradigm for over 99.99% of all loudspeakers ever marketed including AR3a, and AR9 are built to the same basic paradigm and my analysis of it is that they cannot be made to sound like most musical instruments even as they would be heard in the same room except under the most contrived conditions similar to those in the live versus recorded demonstrations AR conducted. Normally, the way they are used it's not possible due to the limitations of the paradigm itself.

Dr. Bose also examined the question of understanding and duplicating concert hall acoustics in his white paper and while he contributed some valuable and thought provoking data, any pretense that his speakers could solve this problem was a complete failure. However, recognizing the importance of this may have nudged the industry to examine the question further and prodded them to develop quadraphonic sound in the 1970s. This was also a badly failed paradigm for its purpose however, home theater as we know it today is its lineal descendant. It has proven far more useful for putting the listener in the center of explosions, railroad trains, and jets flying overhead than for reproducing concert hall acoustics, something serious audiophiles give it no credence for any longer.

I have to disagree with RoyC having examined the 901s I bought. They are made from very high quality parts, the equalizer being manufactured with 1% and 5% filter components instead of the more customary 10% and 20% cheaper components common for the era and the circuit board looks to be glass epoxy, also rare and expensive for the era. The cabinets are very sturdy and well built and the drivers have held up remarkably well for almost 40 years with no sign of deterioration. The original drivers were manufactured by CTS and were sorted into three groups depending on their performance so that they could be assembled in closely matched sets. Aside from AR and KLH who manufactured their own drivers to their own specifications and very tight tolerences, I think these units were on the whole made as well as any other manufacturer's, probaby far better than most. Remarkably, 35 years after having purchased them and discussing them with Bose's service department when I asked about resealing the spaces around the hardened putty like material they used, they offered me a trade in with a 50% discount for the latest 901 Series VI. I consider that an amazing offer (which I declined because frankly, I had ideas about modifying the ones I had and I think their LF response is superior.)

'Diamonds&Rust'

"Where all of the futurists usually go wrong is to completely miss the opportunity for complete abandonment of trends and account for leaps of technology, preferring to extrapolate the near-term present into the future, or doing some sort of regression into the past and projecting along a linear path."

"Nothing works that way. Spontaneous and lightning-quick mutations seem to be responsible for a great many things in evolution. Thus, while I’ll take a stab at this, I assume that I’m so wrong, so far off, that what follows is laughable."

This is one area where I think we are in complete agreement. The unexpected breakthrough, the missed elements crucial in determining where events are really headed, the analysis skewed by the belief that recent trends can be extrapolated linearly indefinitely often leads to very wrong predictions by futurists, predictions which turn out to look silly but entertaining in retrospect. I love watching those old sci fi movies from the 1950s, wow did they ever get it wrong. And the professionals are often no better. It's interesting that among all the blunders, nobody seemed to see the coming sudden collapse of the USSR (except for one Soviet dissident Andrei Amalrik whose book, "Will the USSR last until 1984" is very difficult to find. He spent much of his life in a Soviet prison.)

"In 200 years we will be able to generate the experience of music’s effect on the brain without a note’s ever being played."

I think this is true but may come a lot sooner.

"I assume that in 100 years we will be able to be fed the entire experience of a live performance more or less directly into our brains - via what method I will not speculate. This will include all sensations of “being there” and “there” will be more or less wherever we choose and the performers might also be whoever we choose, but more likely, whatever has been calculated for us to enjoy it the most."

I think this is also correct. It conjures up frightening prospects for mind control. This technology would have the potential to do great harm to society as well as great good. OTOH this virtual reality could allow us to experience a range of events in our lives not even remotely possible in the real world.

"Whether we would recognize what will constitute a performance 100 years from now, I sincerely doubt. Why would anyone spend their lives learning to play a violin when a robot (of sorts - ethereal) can play it *perfectly* for you, emulating any real or imagined performer’s style, or maybe even with our own style synthesized from the best of every performer on record? Effortlessly. Musicianship, now a prized scarce resource, will become the birthright of everyone. Making sound with an instrument will be as quaint as making thread by hand is today - and the result of taking life-long lessons will be about the same as hand-making thread; inferior quality at 1000 times the cost."

I'm not so sure about that one. I saw one episode of Star Trek where Mr. Data played the violin in the style of Jascha Heifetz. I think people will still enjoy playing musical instruments because it is fun and will use it to express their individuality. Why do people still paint when we have photography? Why do they still use brushes, pigments, and canvas when they can work far more efficiently on computer screens? And why do people still prize the efforts of painters so highly? I think for the forseeable future, as long as we retain our humanity, the individual unique human element will be of some value to many.

"Dr. Bose, et al, will be as interesting to that generation as the men who improved the siege engine are to us today. They will all be completely, utterly, forgotten, and their products will be recycled as kindling and illegal mechanisms for simulating erotic experiences via the xxx.net. A speaker’s presence in a room will only provoke disgust - like lighting a fire in a modern oven."

"50 years from now, Dr. Bose’s inventions and products will be regarded with the same respect we currently afford quills that have to be dipped in an inkwell. Simulated tactile experiences will be a reality via a dorsal column stimulater of sorts. The very last generation of musicians will be born."

I know people who collect antique clocks, old radios, ancient instruments like sextants and primitive telescopes, music boxes and other mechanical music makers, old 78 RCA phonographs and 80RPM Edison phonographs. You can see them on PBS on "The Antique Road Show." I think there will always be a fascination with where we came from, how things once were, and marvel at how well primitive things worked even after acknowledging their limitations by the current standards. In the future world you paint, don't forget that if things continue to develop, what is cutting edge then will become the antiques of subsequent eras.

"25 years from now Bose et al will be making personal entertainment products that are in a pair of glasses and will project directly into our eyes and ears making the experience of “you are there” as real as they can be without tactile reinforcement."

Might be, the current crop of virtual reality products points the way. But that doesn't mean we will forget or ignore Bose's contributions. Einstein proved Newton wrong but we don't discard Newton. In fact, we study Newton long before we get to Einstein. There was an old joke when I was in engineering school. As freshemen we studied Newton and were taught F=m*a. As sophmores we studied Einstein and were taught F is approximately equal to m*a. As Juniors, we got deeper into Einstein and were taught F is not equal to m*a. And then as Seniors, when professors realize that we will soon have to go out into the real world and earn a living, we get back to being taught F=m*a. In fact, most things in this world are still analyzed and designed strictly according to Newton's laws.

I think the near term future for audio looks very bleak, the same tired failed ideas recycled endlessly with prices ever spiraling upward, each new slight twist or wrinkle described as a major breakthrough and lots of bogus products of no substantive value appearing to lure those who can be convinced to buy anything on a hope and a prayer swearing how well they work. The best of the current crop of those actually working on the problems of high fidelity sound reproduction IMO are utterly inadequate mentally for the task. Gordon Holt who wrote for High Fidelity Magazine in the 1950s and founded Stereophile as a 20 page newsletter in 1962 recently was interviewed by the current editor John Atkinson in which Holt expressed even far more pessimism now than he did in a speech he made in 1992 that was referenced in the article. Holt will write a more extensive article sometime next year. The reviewers and editors of TAS Magazine expressed similar thoughts in an article about a year and a half ago. I think these are rare and frank admissions of failure by those with a lifetime of experience and of dashed hopes people in this industry once had that there were solutions and breakthroughs just around the corner. In fact, the current paradigm of two channel stereophonic sound has been around in one form or another for probably around 80 years or so and made its first public appearance around 1940 in Walt Disney's full length movie "Fantasia" in which animation was accompanied by what was described as a concert like experience for the audience, the performers being the Philadelphia Symphony Orchestra under the baton of Leopold Stokowski.

Bleak as it now looks, I believe (based on more than pure speculation) that new and far more powerful understanding of sound as it relates to acoustics and hearing is possible and that much better sound systems for our homes than we can now conceive of are also possible and I don't think they will be impossibly expensive or difficult to use although the notion of take it home in a box and DIY for such systems is not likely in the cards. Whether they are ever commercially exploited or see the light of day on the market is for me the real question.

[RoyC]

I have to disagree with RoyC having examined the 901s I bought. They are made from very high quality parts, the equalizer being manufactured with 1% and 5% filter components instead of the more customary 10% and 20% cheaper components common for the era and the circuit board looks to be glass epoxy, also rare and expensive for the era.

Hi Soundminded,

I don't think we disagree. It was just bad wording on my part. By "inexpensive" I didn't mean to imply low quality or cheap. My point was that Bose seems very efficient in the production of very effective designs...and I overlooked the flagship 901's outboard EQ in the scheme of things. My only experience has been with the 201, 301, 501 and the PA/professional 802.

The 802 is the 901's cousin, also having the outboard EQ component, but has one less 4" driver than the 901. The remaining 8 drivers face outward, and are housed in a molded plastic, ported cabinet. (Bose seems to have pioneered the use of molded plastic speaker enclosures.) The 802 is easily portable and excellent for the intended purpose.

My experience with the other models showed simple, decently constructed cabinets, and very simple crossovers. It seems to me that Bose was able to simply integrate (good quality:-)) made-to-spec drivers to minimize the need for elaborate electronic modification...with, of course, the exception being the 802/901's active equalization of eight or nine 4 inch drivers. I have no experience with the company's later, module oriented systems.

In any case, it is difficult to deny Bose's consistent ability to provide very appealing sound reproduction to a fair number of enthusiasts, professionals and the masses.

Roy

[soundminded]

I guess it wasn't until I started surfing message boards on the internet a that I became aware that there are so many people who have strong emotional reactions just at the mention of the name Bose. It's apparently well known and the moderator of this board was afraid it might happen here. It seems entirely irrational to me. AFAIAC, Bose 901 like all other sound reproducing equipment is a machine and only a machine, it is not for example a musical instrument. I simply look at it in terms of its intended function, how well it performs that function, what its strengths and weaknesses are, if it seems well made, if it's a good value for money, and if it's reliable. I also look at the service available if it runs into trouble. I've analyzed this speaker along with many others and experimented on them as well. The shortcomings of this speaker (series I and II) strike me as relatively easy to understand and straightforward if not inexpensive to solve compared with some others, at least with the availability of parts and other equipment on today's market. (I wouldn't even know where to begin re-engineering the Martin Logan Summits.) It wasn't always so, making such modifications a few decades ago would have been far too expensive. I also like the fact that they are readily available used and in pretty good condition on the internet for around $150 to $200 a pair not including the pedistals so acquiring more of them for further experimentation is not a compelling take it when you can get it kind of thing. I've found a lot to be learned from them and I'm surprised that so many other people have dismissed them out of hand without considering their strengths and what they have to offer when properly installed. This is not an unqualified endorsement, far from it. In fact, from the point of view of someone who expects perfection as I do, when installed and used as manufactured they are fatally flawed. But then so are all the other speakers I've seen so it's just a question for me of how I approach re-engineering them.

[Mexicomike]

I was an avid audiophile in the mid 60's through the 90's and I auditioned just about every speaker around at that time. When the 901's came out they made a big point of the direct-reflecting sound. So naturally I auditioned them. I remember listening to them at Bill Case Sound in San Antonio, TX. There were 4-5 of us in the listening room. They were playing mostly classical music through them - Thus Spake Zarathustra - from the recently released 2001, A Space Oddessy.

Everyone was quite impressed with all the sound swirling around until somebody said, "Switch to those AR3As." The difference was amazing; The Bose, which had sounded quite good suddenly were revealed as having no bottom at all. To me they became "gimmick" speakers and for that reason, I consider Bose a marketing innovator, not a loudspeaker SOUND QUALITY innovator.

[soundminded]

I was an avid audiophile in the mid 60's through the 90's and I auditioned just about every speaker around at that time. When the 901's came out they made a big point of the direct-reflecting sound. So naturally I auditioned them. I remember listening to them at Bill Case Sound in San Antonio, TX. There were 4-5 of us in the listening room. They were playing mostly classical music through them - Thus Spake Zarathustra - from the recently released 2001, A Space Oddessy.

Everyone was quite impressed with all the sound swirling around until somebody said, "Switch to those AR3As." The difference was amazing; The Bose, which had sounded quite good suddenly were revealed as having no bottom at all. To me they became "gimmick" speakers and for that reason, I consider Bose a marketing innovator, not a loudspeaker SOUND QUALITY innovator.

I have to disagree. There are many reasons why the Bose 901 might come off as having much less bass than AR3a in any given listening test. One of them is speaker position and room acoustics which can be very critical especially for a loudspeaker which depends on room reflections for most of what is heard. Another is available amplifier power. Original Bose 901's power requirements for deep bass even at moderate levels are enormous. As I have said elsewhere, my listening tests confirm what was reported in ee Magazine, the 901 seemed to have a peak of around 7 db at around 250+ hz in my room and that is probably the system's intended resonance frequency (Bose said it had to be above 180 hz for its phase shift to be inaudible) below which response falls off at 12 db per octave but the equalizer only supplies a 6db per octave boost to the bass. Even without the "below 40" bass cut switch engaged, this would put the lowest octave quite a bit below the 1khz level. Add to that the low frequency cutoff of the room if it is not large and the speaker could seem bass shy with deep bass. However, my own experience equalizing the speaker for flatter response confirms what Bert Whyte of Audio Magazine and High Fidelity Magazine reported and that is Bose 901s enormous potential to produce substantial bass well below 30 hz. But the power requirements can be staggering. Whyte used a pair of 300 watt McIntosh amplifiers in his tests. My 138wpc solid state amplifier can be made to clip easily with these speakers even at modest levels at very low frequencies. At low levels, there is no problem rattling the windows and everything in the room including the one turntable I tried. Records in this particular setup were unplayable for this reason. Within its power handling capabilities and with sufficient power available, it will give AR9 a run for its money and will easily beat out AR3a. If one pair won't do it, two pairs surely will. I didn't attempt to get inside but I wouldn't be surprised if there is no damping material to lower the amplitude of the resonant peak. Has anyone opened an original or Series II up? Measured harmonic distortion is around 10%, double 5% of AR3a but still inaudible. In fact, HF's (or was it Stereo Review's) measurements were usable output to 23 hz at reasonable levels, and down to 26 hz at high levels. Within its power handling capabilities, its bass slightly beat out JBL Paragon D44000!

Besides these difficulties at low frequencies, a 4" midwoofer makes a very poor tweeter due to its high inertial mass. It's large diameter also precludes any but the worst dispersion, a critical factor since the first arrival of the forward driver creates "precedence" which bears heavily on the stereophonic effect which audiophiles call "imaging." The real improvement came when I added tweeters. The optimal modifications required a lot of experimentation which took me almost 3 years before I was satisfied. This speaker is not a gimmick or a marketing trick IMO although it's clearly not the ultimate or optimal embodiment of its concepts. Having read Dr. Bose's white paper, even though there's much I disagree with there is much innovative and provocative technology in the design, most of which has been overlooked or ignored by other manufacturers in the interim. Too bad for them, it's their loss but anyone who refuses to pay attention to what a professor of acoustics and electrical engineering from MIT has to say about speaker design has deprived himself of a valuable intellectual resource.

[Mexicomike]

I can't recall anything about the amplifier used when I heard 901s on that occasion or any other. Since Bill Case was one of the best audio spots in San Antonio, and spent a LOT of time with customers, I'm confident that they were using suitable amplification - since 3as aren't an easy amp load either.

Certainly room acoustics could have had a lot to do with it but they were in the same room and fairly close to each other - the 3a's were on some short stands - maybe a foot off the floor, the 901s were on their factory stands. The 901s were more toward the corners of the room, the ARs were several feet in from the same side walls.

I don't know that the setup was less than optimum but if so, had they been better set up, they might have made a different impression.

I remember this demo so well because we (wife and I) went specifically to hear (and probably buy - based on advertising) the 901s and came away with a set of 3As!

[soundminded]

I can't recall anything about the amplifier used when I heard 901s on that occasion or any other. Since Bill Case was one of the best audio spots in San Antonio, and spent a LOT of time with customers, I'm confident that they were using suitable amplification - since 3as aren't an easy amp load either.

Certainly room acoustics could have had a lot to do with it but they were in the same room and fairly close to each other - the 3a's were on some short stands - maybe a foot off the floor, the 901s were on their factory stands. The 901s were more toward the corners of the room, the ARs were several feet in from the same side walls.

I don't know that the setup was less than optimum but if so, had they been better set up, they might have made a different impression.

I remember this demo so well because we (wife and I) went specifically to hear (and probably buy - based on advertising) the 901s and came away with a set of 3As!

AR3a is a difficult amplifier load because its impedence at some frequencies is very low. Many otherwise fine amplifiers will not deliver acceptable performance and can even become unstable with AR3a. We blew up a 60 WPC HH Scott Receiver driving AR3s because of this. Scott refused to honor its warrantee saying it was the fault of the speakers, connecting them to AR3 constituting unreasonable abuse. Bose 901 is a difficult amplifier load because as frequency gets lower, its power requirements become rediculously high. A good quality 100 wpc receiver from the 1970s or 1980s should do very well with AR3a, not so with original Bose 901 and series II. To answer this challenge, Bose developed the 1801 power amplifier, one of the most powerful brute force designs I'd seen up to that time, a real challenger for Crown DC300. But the price for that amplifier was also very high. Today, a high powered amplifier is comparatively cheap, no more than $400 to $500 in today's dollars for a Crown. Bose 901 as I said also needs additional bass equalization. The effective total cone piston area is about 20 pi square inches for a 901, 16 pi for an AR3a but the AR3a has a much greater X max meaning it can push more air. Also, Bose 901 has a rated maximum capacity of 270 watts. This means that multiple units may be required for higher SPLs at very low frequencies. I drove the 901s back around 1970 with an AR amplifier and found that from my first floor apartment, I could shake dishes in my neighbor's 3rd floor apartment. I know that sounds crazy but it happened. My scope showed no signs of clipping at the amp output. Needless to say that apartment building was not well built. With the aquisition of AR9, my expectations for maximum bass capability entered a whole new realm. I drive mine with a very conservatively rated 60wpc high quality amplifier in a room about 4000 cubic feet. About a year ago, I was listening to the end of Bernstein's recording of Copland's 3rd symphony and when it ended and the next cut, Copland's organ symphony came on, I nearly simultaneously shattered all 28 panes of glass (including the french doors) in my music room. Scared the &*%$ out of me. Can Bose 901 do that? I think if I put enough of them together, say 4 to 8 pairs they could. Used, they're usually only about $200 a pair in good condition on e-bay.

I found that moving AR9 just a few inches could substantially change the bass level I hear in the rest of the room. AR3a is best placed with its back to a wall like AR9. Bose 901 requires from 12 to 18 inches in front of a wall. It's important that the walls behind and to the sides of Bose 901 are fairly reflective and solid otherwise they will soak up the bass along with a lot of the rest of the sound. Sheet rock does OK if it is solidly installed but it's less than ideal because it is not really solid with studs 16" on center. Most audio showrooms I've seen are designed to be acoustically on the dead side, not favorable for Bose 901.

IMO, one of Bose's contributions with 901 is that it was the first and still the only speaker I know of which was intentionally designed to be totally integrated with a room's acoustics, it's performance capabilities depending on the skill and circumstances of that integration. That may be one reason Bose does not publish specifications, they would be meaningless for comparison to other speakers. Often, the system did not perform as expected because the installation was flawed, sometimes the fault of the room where nothing could be done. Bose also failed to make adequate provisions to compensate for different room acoustics variables. Whether 901's unusual stereophonic presentation is your cup of tea or not and whether it starts out as a novelty you tire of or the speaker's other limitations become unacceptable is an individual choice. But given that in one form or another it has been available for nearly 40 years, it seems there are a lot of people who will take that set of shortcomings over others, usually audiophiles are not among them. In fact I think it's fair to say most audiophiles truly detest these speakers (great for me because the price used stays low.) Dr. Bose changed his strategy with Series III in the early 70s modifying the design to make it ported which sacrificed the bottom octave in favor of much higher efficiency and lower power requirements. So the economics of 1968 which dictated $500 for his speaker and $1500 for an amplifier now in 2008 becomes $1500 for his speaker and $500 or less for a receiver, a pretty clever marketing decision which has proven itself in high sales volume and profits. Four decades after its introduction it's an old idea now that's been around a long time, so long you can only buy them new on special order. So why do people get so worked up over it?

[soundminded]

The 901 is an interesting system, and one reason that it sounds so different is its rather unique radiation pattern, which is like no other systems I have encountered. I see it as a fairly directional speaker, with the front driver being rather beamy in the treble and the rear drivers (which, as mounted in groups of four on two angled panels, behave like two rather large drivers) being beamy in both the midrange and treble. If the wall behind the speaker is smooth and reflective those beams continue to be narrow dispersing as they reflect. A more broken up wall area would not have that problem, of course.

In any case, what I wanted to mainly comment upon is that while the wall behind may cause problems with the bass coming from the 901 if that wall is too flexible, the same situation will exist for any other speaker placed similarly, because fairly deep bass is omnidirectional. And if an AR-3a is located against the wall its woofer is still about a foot away. A 901 located typically has its bass source at about the same distance. Consequently, both systems will generate an "Allison Effect" notch in the mid bass (Allison himself measured this artifact with the AR-3a and even published a paper on it), and if either is on a stand that is not too high up that notch will be made even deeper. The room integration of the 901 does not approach the smooth mid-bass integration of systems like the original Allison Model One, Two, and Three, or the later IC-20 model, simply because of the distance relationship the system has with nearby room boundaries.

I once reviewed a pair of Series VI models for The Sensible Sound, and I will admit that I was impressed with what they could do with a small jazz combo recording in a large playback room. However, I also measured them and their room curve was considerably less flat (even with the active equalizer optimized) than what I had measured from a number of conventional systems. No matter how good the original 901 may have been in the deep bass range, later (ported) models will benefit from the addition of a good subwoofer.

Howard Ferstler

Interesting that this thread has come back to life after nearly a year. Do you have a link to your review?

Insofar as original Bose 901's bass is concerned, the design concept was unique but relatively straightforward. While most designers tried to push the system resonant frequency down as low as they could get it, Bose pushed it up with a small cabinet and then exploited the linear falloff below resonance by compensating for it with an inverse linear filter. But the 901 driver/enclosure falls of at 12 db per octave while the filter only adds 6. Not only that but it starts too high and when combined with the driver/enclosure's resonance peak results in a system peak which e/e magazine measured as 8db at 500 hz and I find in my room as 7 db at between around 250 and 500 hz. Not only that but bass falls below the 1khz output at around 100 hz so the deep bass 901 is capable of cannot be heard in sufficient measure without further boost or unless played at very loud levels for most recordings. Also power requirements at the lowest octave is staggering. My 138 wpc receiver will clip if the system is driven more than moderately on some recordings. LF cutoff of small rooms typical of home listening environments will only add to the problem. Julian Hirsch I think measured THD at low frequencies at around 10%, about double that of the AR1 family of systems. Output was useful to 26 hz at loud levels, 23hz at lower levels. This was in response to a question by a reader asking how 901 compared to JBL Paragon D44000, a top LF performer in its day. This puts 901 in the same league with AR1 and Paragon even if the electrical and space requirements are different. In my room, a turntable without a well designed suspension is not usable due to LF system gain. I can't find my AR turntable missing in action among hundreds of unopened boxes for 8 years and I haven't tried my Empire 698 yet. AR9 when equalized for the room it is installed in was able to knock my Denon 1520's laser right off the track at moderate levels on some recordings. A Toshiba DVD player I replaced it with fares much better. Systems with very wide LF resopnse capability can introduce many unticipated problems not experienced with lesser systems. It is surprising how may CDs have very low frequency disturbances recorded on them which you will not normally hear from most sound systems. Turntable rumble and acoustic feedback also become problems.

Bose 901's frontal hf radiation stinks. There is simply no other way to put it. How can a 4" midwoofer even if it could overcome its inertial mass propagate sound above 10khz off axis? This is not a matter of quality of manufacture but is inherent in the physics of it. That is why tweeters are small and why those with the widest dispersion are the smallest and have carefully worked out geometries. AR3 and AR3a are only 3/4" compared to the typical 1" and their hemispherical dome is not recessed into a horn. As a result, it has the widest dispersion of any single driver at 15 khz I have ever seen, only 5db down from its on axis response at 60 degrees off axis. This is why a replacement for it cannot be found, the current speaker design philosophy doesn't value wide dispersion. Instead of integrating loudspeakers into listening rooms, modern design tries to isolate it from listening rooms. This of course is doomed to fail. While 901 is difficult to integrate into a room, Monopoar box speakers are next to impossible. The best approach IMO was LST whcih recognized that even AR3a's tweeter had inadequate HF dispersion. Too bad this concept wasn't further exploited later on such as with AR9. Installing a speaker in a wall where the front is flush with the wall (in the old days audiophiles used to install them in closet doors) or in built ins where the sides and backs are surrounded by cabenetry, books etc. also helps.

My experiments are based on a mathematical model. This model incorporates dimension not taken into consideration in other models. 901 inadvertently stepped into one of those dimensions which IMO is a major reason why it was so popular despite its price and other shortcomings. But Bose has virtually abandoned this philosophy at least in his marketed products if not in his own research laboratories. 901 for me was a good starting point to explore how far that model could be pushed without building a new speaker system from scratch. I don't know what 901 series III and beyond offers in the way of bass response compared to series I and II but I must assume that as a ported design, it can't go nearly as low, certainly not given its size.

[Guest boseguy]

The 901 was an extraordinary design, and remains unique in its approach. As with all Bose products, it was/is manufactured to the very highest quality/uniformity standards. It's a fine product in terms of delivering what it says it will deliver.

But to clarify again: It has been available "for 40 years" only because Bose insists on continuing to make it available. Bose has become a direct marketer of its products, and all of them are available from the company as well as being sold through stores. Some are sold ONLY from Bose directly.

It's because of this flexible business model that the 901 continues to live. It is not a commercially viable retail product any longer, and hasn't been since I was there in the early 90's. With the limited (if not non-existent) retail demo/display facilities at Best Buy and the complete demise of the specialty audio retail chains, the 901 has no retail home. If Bose didn't arbitrarily keep it going themselves, it would have ceased to exist long ago. To think that it has "survived 40 years" is somewhat misleading.

Again, it's a fine, high-quality product. But it is not viable in the open marketplace any longer, and hasn't been for quite a while.

boseguy

[tysontom]

The 901 is an interesting system, and one reason that it sounds so different is its rather unique radiation pattern, which is like no other systems I have encountered. I see it as a fairly directional speaker, with the front driver being rather beamy in the treble and the rear drivers (which, as mounted in groups of four on two angled panels, behave like two rather large drivers) being beamy in both the midrange and treble. If the wall behind the speaker is smooth and reflective those beams continue to be narrow dispersing as they reflect. A more broken up wall area would not have that problem, of course.

In any case, what I wanted to mainly comment upon is that while the wall behind may cause problems with the bass coming from the 901 if that wall is too flexible, the same situation will exist for any other speaker placed similarly, because fairly deep bass is omnidirectional. And if an AR-3a is located against the wall its woofer is still about a foot away. A 901 located typically has its bass source at about the same distance. Consequently, both systems will generate an "Allison Effect" notch in the mid bass (Allison himself measured this artifact with the AR-3a and even published a paper on it), and if either is on a stand that is not too high up that notch will be made even deeper. The room integration of the 901 does not approach the smooth mid-bass integration of systems like the original Allison Model One, Two, and Three, or the later IC-20 model, simply because of the distance relationship the system has with nearby room boundaries.

I once reviewed a pair of Series VI models for The Sensible Sound, and I will admit that I was impressed with what they could do with a small jazz combo recording in a large playback room. However, I also measured them and their room curve was considerably less flat (even with the active equalizer optimized) than what I had measured from a number of conventional systems. No matter how good the original 901 may have been in the deep bass range, later (ported) models will benefit from the addition of a good subwoofer.

Howard Ferstler

I want to preference this by saying that I respect Howard Ferstler’s opinions on audio topics in general. He is a knowledgeable resource on Allison loudspeakers, and he has several articles and four hobbyist books published on audio/video topics. In my opinion, his assessment of the quality of the Bose 901 is flawed, and I take issue with his criticism.

Howard’s comments about the Bose 901 are quite similar to the criticisms of several other negative reviews on this speaker. It is a mild form of “speaker bashing.” As I have mentioned in the past, I have no preference for the Bose 901; but it goes without saying that this is a competent and accurate loudspeaker that has been highly praised over the years by many well-known critics and testers such as Julian Hirsch, Norman Eisenberg, Larry Zide, Bert Whyte and many others. These reviewers raved about the performance and quality of this speaker, but for some reason this praise irritates (and perhaps threatens) some audiophiles, manufacturers and critics, and the speaker has been under almost constant attack since its introduction. Yet, the 901 has had an enviable commercial track record, unmatched by most other loudspeakers in the history of audio -- and is still in production as possibly one of the longest-lived high-fidelity loudspeaker systems (along with the Klipschorn) ever produced. Nevertheless the 901 has been the focus of a great deal of criticism, not the least of which are comments by Howard Ferstler. Perhaps because the 901’s performance has been elusive and difficult to quantify (no legitimate objective-measurement curves ever published on the speaker) it has become an easy target for criticism and has been “bashed” by numerous audiophiles and reviewers, apparently with some sort of ax to grind. I suspect that Ferstler’s criticism is founded on a flawed measurement technique and his cockamamie notion of poor dispersion on the part of the 901. Ferstler is also an “Allison man,” and he might resent the notion that the Bose 901 offers a spacious sound presentation and an accurate facsimile of sound reproduction. Allison speakers are excellent designs; who is to say the Bose 901 isn’t an excellent design as well?

On the subject of the 901’s radiation pattern, Howard contends that the speaker is “directional” and “beamy” in mid- and high-frequency output. In a message to me recently, Howard was a bit more aggressive in this description:

“Because each four-driver array behaves acoustically as a single 9-inch or so square driver above the bass frequency where both arrays couple and radiate omnidirectionally each will beam as a larger driver, and each is beaming off at a slightly different angle from the other. If the wall is flat and smooth (no items there to break up the reflections) the beam will bounce, billiard-ball style, and remain beamed as it comes off of the wall. The beaming, of course, will change in angularity as the frequency varies, but at midrange and treble frequencies each array is beaming tight as hell. I continue to say that at midrange and higher frequencies the speaker generates three beamed arrays, with the two aimed toward the rear being way more beamy than even the single-driver radiation from the front.”

This claim is troubling to me mainly because it is not based on any legitimate scientific fact, but mostly on conjecture and assumption. I guess the “9-inch” dimension comes from the simple math of adding the diameters of each Bose 901 side-by-side driver (4½ and 4½ equals 9), and let’s disregard the actual radiating-area of a circle. Never mind the well-known fact that those eight small drivers couple only at low frequencies, and are not combined into one big panel at the higher frequencies. Howard’s descriptive “billiard-ball” style bounce of the sound off the back of the 901 sounds like the beam from a flashlight, so it would only stand to reason that you could never be in “line” to receive each of the focused beams of sound energy, thus the sound you would hear would be diffuse or full of gaps. Whole sections of the sonic spectrum would be missing to the listener, since the sound would not disperse off the back of the speaker, as in Howard’s model. Perhaps whole sections of violins would have been bounced, flashlight-beam, “billiard-ball” style over into another part of the room only to be completely missed by the main listener. I make fun of this notion because it is probably the most cockamamie explanation of the 901’s sound propagation I have ever heard!

As for the so-called “Allison Effect,” the notch in the mid-bass frequencies, there is certainly some logic with this, but I would add that since there is a driver mounted on the front of the 901, generating the same output as the rear drivers, some of the “missing” information is restored by that speaker alone, and the notch move further up the scale. The other drivers are on different, angled planes, with different distances from the rear wall, and thus the notch would be broadened a bit as well. Besides, who’s to say that Bose didn’t correct any anomalies in the frequency response of the 901 when he designed the equalizer? He optimized the equalizer for the 901 to be mounted specifically 12-18 inches from the front wall, and a certain distance above the floor or below the ceiling, so it would be very easy to correct any lack of energy -- if there was any. I suspect that the boundary “notch” is a non-issue with the 901. It is therefore pure assumption on Howard’s part to say that the speaker does not “approach” the smooth mid-bass integration of systems like the original Allison Model One, Two, and Three, or the later IC-20 model, simply because of the distance relationship the system has with nearby room boundaries. This has not been quantified by anyone.

Sometime back Howard wrote a review in The Sensible Sound magazine (I’m not sure of the status of this magazine at this time) on a set of Bose 901s, Series VI, belonging to a friend of his down in his home state of Florida. This well-written review was a quasi-scientific in-room test report on his friend’s audio room, and consisted of two pairs of 901s mounted at opposite ends in a room approximately 31’ x 21’ x 10’. For the review, Howard opted to measure only the front pair of 901s. Strangely, Howard’s friend completely disregarded Bose’s recommended 1-foot-from-the-front wall mounting arrangement, deciding that it was not to his liking, so he mounted the speakers about five (5) feet back from that reflecting wall. No effort was made to accommodate Bose’s recommended placement for this review, but this difference was clearly noted. Nevertheless, the 901 review was in my view “damned with faint praise,” to the extent that Howard made numerous references to the performance of the 901s, such as “enough to deliver surprisingly smooth treble response,” but he made a hand-drawn response curve that he has posted on, of all places, the Allison Yahoo chat group. This curve is from his test procedure using an AudioControl 1/3-octave RTA with ANSII Class II filters (good, but not as precise as the more expensive Class III filters used with the better RTAs).

For his measurements, Howard uses an AudioControl 3051A 1/3-octave real-time spectrum analyzer with pink noise to formulate a “room curve” of sorts (he calls it a “room/power” curve similar to what one might get by using multiple-microphone measurement the acoustic-power output throughout the listening area of a given room) that gives to him a “slice” of the acoustical performance in a small cube of space in that listening room. He uses a small window of about one cubic foot in which he moves the microphone around, to assess the sound at that point. The 3051 manual, however, recommends “multiple” microphone positions as well as an isolated mounting of the microphone to avoid any absorption or reflection from the tester himself. As far as I can tell, Howard ignores both of these recommendations and hand-holds the microphone in one small place in space.

The 3051A is certainly a well-respected tool, but it is designed specifically for pro-sound equalization adjustments for rock group concerts and so forth, to optimize -- with a graphic equalizer -- the sound from the big sound systems used by musicians and so forth. AudioControl recommends that an “average” of several measurements be made with resultant changes to the equalizer for the sound system, depending on the concert hall. The 3051 was not designed to make loudspeaker “frequency-response” measurements, of sorts, as Howard has indicated, and the company makes no such claims. As such, it is incapable of making steady-state frequency-response measurements, but if a concert hall or auditorium has serious acoustical flaws, it will show up on the graphic display, enabling the sound engineers to tweak the sound-system equalizers for flatter response. Yet from this instrument Howard has come to believe that he is making definitive measurements on the performance of a loudspeaker in a real room, and concomitant quality judgments about those speakers are made in his reports -- and this is flawed judgment. Furthermore, Howard insists that that above low-bass frequencies, all rooms basically sound the same, regardless of dimension and furnishing, such that what he measures in one room is essentially what he measures in another room, except that in the bass. He has said that his listening room at home is essentially the same as what he measured in the room of his friend’s 901s, even though his friend’s A/V room is 21 x 31 feet, and his home listening room is probably about 2/3rd that size. He contends, therefore, that high frequencies will be perceived pretty much identically from one room to another, but “all bets are off in the low frequencies.” Proof of this assumption on his part is his statement, “however, I also measured them and their room curve was considerably less flat (even with the active equalizer optimized) than what I had measured from a number of conventional systems.” This tells me that he feels that what he “measured” in his own home could be used to judge performance measured in someone else’s room, and he is stating that he has “measured” a number of conventional systems with superior performance than the 901. This is nonsense, of course, and I do not believe that any two rooms will sound exactly alike unless they have exactly the same ratio of dimensions, the same furnishings, the same number of windows and doors, the same number of openings to hallways and so forth. There are certainly similarities from room-to-room, but there are probably more differences than similarities.

Howard is entitled to his own assumptions, of course, but I don’t believe his criticism of the Bose 901 has any merit whatsoever.

--Tom tyson

[soundminded]

I want to preference this by saying that I respect Howard Ferstler’s opinions on audio topics in general. He is a knowledgeable resource on Allison loudspeakers, and he has several articles and four hobbyist books published on audio/video topics. In my opinion, his assessment of the quality of the Bose 901 is flawed, and I take issue with his criticism.

Howard’s comments about the Bose 901 are quite similar to the criticisms of several other negative reviews on this speaker. It is a mild form of “speaker bashing.” As I have mentioned in the past, I have no preference for the Bose 901; but it goes without saying that this is a competent and accurate loudspeaker that has been highly praised over the years by many well-known critics and testers such as Julian Hirsch, Norman Eisenberg, Larry Zide, Bert Whyte and many others. These reviewers raved about the performance and quality of this speaker, but for some reason this praise irritates (and perhaps threatens) some audiophiles, manufacturers and critics, and the speaker has been under almost constant attack since its introduction. Yet, the 901 has had an enviable commercial track record, unmatched by most other loudspeakers in the history of audio -- and is still in production as possibly one of the longest-lived high-fidelity loudspeaker systems (along with the Klipschorn) ever produced. Nevertheless the 901 has been the focus of a great deal of criticism, not the least of which are comments by Howard Ferstler. Perhaps because the 901’s performance has been elusive and difficult to quantify (no legitimate objective-measurement curves ever published on the speaker) it has become an easy target for criticism and has been “bashed” by numerous audiophiles and reviewers, apparently with some sort of ax to grind. I suspect that Ferstler’s criticism is founded on a flawed measurement technique and his cockamamie notion of poor dispersion on the part of the 901. Ferstler is also an “Allison man,” and he might resent the notion that the Bose 901 offers a spacious sound presentation and an accurate facsimile of sound reproduction. Allison speakers are excellent designs; who is to say the Bose 901 isn’t an excellent design as well?

On the subject of the 901’s radiation pattern, Howard contends that the speaker is “directional” and “beamy” in mid- and high-frequency output. In a message to me recently, Howard was a bit more aggressive in this description:

“Because each four-driver array behaves acoustically as a single 9-inch or so square driver above the bass frequency where both arrays couple and radiate omnidirectionally each will beam as a larger driver, and each is beaming off at a slightly different angle from the other. If the wall is flat and smooth (no items there to break up the reflections) the beam will bounce, billiard-ball style, and remain beamed as it comes off of the wall. The beaming, of course, will change in angularity as the frequency varies, but at midrange and treble frequencies each array is beaming tight as hell. I continue to say that at midrange and higher frequencies the speaker generates three beamed arrays, with the two aimed toward the rear being way more beamy than even the single-driver radiation from the front.”

This claim is troubling to me mainly because it is not based on any legitimate scientific fact, but mostly on conjecture and assumption. I guess the “9-inch” dimension comes from the simple math of adding the diameters of each Bose 901 side-by-side driver (4½ and 4½ equals 9), and let’s disregard the actual radiating-area of a circle. Never mind the well-known fact that those eight small drivers couple only at low frequencies, and are not combined into one big panel at the higher frequencies. Howard’s descriptive “billiard-ball” style bounce of the sound off the back of the 901 sounds like the beam from a flashlight, so it would only stand to reason that you could never be in “line” to receive each of the focused beams of sound energy, thus the sound you would hear would be diffuse or full of gaps. Whole sections of the sonic spectrum would be missing to the listener, since the sound would not disperse off the back of the speaker, as in Howard’s model. Perhaps whole sections of violins would have been bounced, flashlight-beam, “billiard-ball” style over into another part of the room only to be completely missed by the main listener. I make fun of this notion because it is probably the most cockamamie explanation of the 901’s sound propagation I have ever heard!

As for the so-called “Allison Effect,” the notch in the mid-bass frequencies, there is certainly some logic with this, but I would add that since there is a driver mounted on the front of the 901, generating the same output as the rear drivers, some of the “missing” information is restored by that speaker alone, and the notch move further up the scale. The other drivers are on different, angled planes, with different distances from the rear wall, and thus the notch would be broadened a bit as well. Besides, who’s to say that Bose didn’t correct any anomalies in the frequency response of the 901 when he designed the equalizer? He optimized the equalizer for the 901 to be mounted specifically 12-18 inches from the front wall, and a certain distance above the floor or below the ceiling, so it would be very easy to correct any lack of energy -- if there was any. I suspect that the boundary “notch” is a non-issue with the 901. It is therefore pure assumption on Howard’s part to say that the speaker does not “approach” the smooth mid-bass integration of systems like the original Allison Model One, Two, and Three, or the later IC-20 model, simply because of the distance relationship the system has with nearby room boundaries. This has not been quantified by anyone.

Sometime back Howard wrote a review in The Sensible Sound magazine (I’m not sure of the status of this magazine at this time) on a set of Bose 901s, Series VI, belonging to a friend of his down in his home state of Florida. This well-written review was a quasi-scientific in-room test report on his friend’s audio room, and consisted of two pairs of 901s mounted at opposite ends in a room approximately 31’ x 21’ x 10’. For the review, Howard opted to measure only the front pair of 901s. Strangely, Howard’s friend completely disregarded Bose’s recommended 1-foot-from-the-front wall mounting arrangement, deciding that it was not to his liking, so he mounted the speakers about five (5) feet back from that reflecting wall. No effort was made to accommodate Bose’s recommended placement for this review, but this difference was clearly noted. Nevertheless, the 901 review was in my view “damned with faint praise,” to the extent that Howard made numerous references to the performance of the 901s, such as “enough to deliver surprisingly smooth treble response,” but he made a hand-drawn response curve that he has posted on, of all places, the Allison Yahoo chat group. This curve is from his test procedure using an AudioControl 1/3-octave RTA with ANSII Class II filters (good, but not as precise as the more expensive Class III filters used with the better RTAs).

For his measurements, Howard uses an AudioControl 3051A 1/3-octave real-time spectrum analyzer with pink noise to formulate a “room curve” of sorts (he calls it a “room/power” curve similar to what one might get by using multiple-microphone measurement the acoustic-power output throughout the listening area of a given room) that gives to him a “slice” of the acoustical performance in a small cube of space in that listening room. He uses a small window of about one cubic foot in which he moves the microphone around, to assess the sound at that point. The 3051 manual, however, recommends “multiple” microphone positions as well as an isolated mounting of the microphone to avoid any absorption or reflection from the tester himself. As far as I can tell, Howard ignores both of these recommendations and hand-holds the microphone in one small place in space.

The 3051A is certainly a well-respected tool, but it is designed specifically for pro-sound equalization adjustments for rock group concerts and so forth, to optimize -- with a graphic equalizer -- the sound from the big sound systems used by musicians and so forth. AudioControl recommends that an “average” of several measurements be made with resultant changes to the equalizer for the sound system, depending on the concert hall. The 3051 was not designed to make loudspeaker “frequency-response” measurements, of sorts, as Howard has indicated, and the company makes no such claims. As such, it is incapable of making steady-state frequency-response measurements, but if a concert hall or auditorium has serious acoustical flaws, it will show up on the graphic display, enabling the sound engineers to tweak the sound-system equalizers for flatter response. Yet from this instrument Howard has come to believe that he is making definitive measurements on the performance of a loudspeaker in a real room, and concomitant quality judgments about those speakers are made in his reports -- and this is flawed judgment. Furthermore, Howard insists that that above low-bass frequencies, all rooms basically sound the same, regardless of dimension and furnishing, such that what he measures in one room is essentially what he measures in another room, except that in the bass. He has said that his listening room at home is essentially the same as what he measured in the room of his friend’s 901s, even though his friend’s A/V room is 21 x 31 feet, and his home listening room is probably about 2/3rd that size. He contends, therefore, that high frequencies will be perceived pretty much identically from one room to another, but “all bets are off in the low frequencies.” Proof of this assumption on his part is his statement, “however, I also measured them and their room curve was considerably less flat (even with the active equalizer optimized) than what I had measured from a number of conventional systems.” This tells me that he feels that what he “measured” in his own home could be used to judge performance measured in someone else’s room, and he is stating that he has “measured” a number of conventional systems with superior performance than the 901. This is nonsense, of course, and I do not believe that any two rooms will sound exactly alike unless they have exactly the same ratio of dimensions, the same furnishings, the same number of windows and doors, the same number of openings to hallways and so forth. There are certainly similarities from room-to-room, but there are probably more differences than similarities.

Howard is entitled to his own assumptions, of course, but I don’t believe his criticism of the Bose 901 has any merit whatsoever.

--Tom tyson

I managed to find Howard Ferstler's review on the web. I'm sorry to say I found it disappointing. For one thing, it was not the original Bose 901 but I think Series VI which had considerably compromised bass being a ported design in a small box versus the original acoustic suspension design. We've discussed this many times, no need to rehash it. More disappointing, the speakers were installed in a manner radically different from what the manufacturer recommended. I'm not sure what the signifigance of his review is if it means anything at all. I heard Series VI earlier this year at a Bose store in a shopping mall. It had the same dull treble it always had, the top octave being missing in my opinion. Dr. Bose said in his white paper that when he listened to other speakers, the more money they charged the more shrillness the customer got. This may still be true but avoiding shrillness by not producing any high frequencies is not the right answer either. I have no problem with speaker bashing or bashing any other products which fall short of their advertised claims whether explicit or implicit if those criticisms are justified. IMO, the implied claim of all "high fidelity" sound systems is to recreate the audible experience of live music with exact faithfulness. The across the board failure of the art to correctly analyze the problem and solve it would be bad enough if it weren't for the fact that not only have prices for some equipment soared to absurd levels but that each new variant is claimed as the magic silver bullet that finally gives what others have only promised. I've read a lifetime of this kind of ad copy and stopped paying attention to it a long time ago. I've also stopped paying attention to the equally misleading opinions of the pundits who reveiw them for magazines. I'm not saying they are necessarily crooked, the distinct possibility exists that most of them are tone deaf. I don't think anyone with normal hearing would be fooled for very long by any available audio equipment you could buy that they are hearing live music and not a recording. This was the unique experience of the live versus recorded demos Acoustic Research gave in the 1960s. I wonder how convinced I would by them today now that I am a far more critical listener.

The factual evidence that 4" drivers beam their midrange and whatever treble they produce is based on both well established theoretical considerations and on measured evidence. Just look at the off axis response of any 4" driver and you will see indication of its directionality usually starting at around 2 khz but rarely if ever much higher. It is a fact that as frequency increases and the wavelength of sound decreases to the point where it is becoming comparable in size to the diameter of the vibrating member, it becomes increasingly more directional. What he left out is that tweeters of any type are hardly much better. While they can produce sound above 10 khz, just look at the off axis response of any of them and you will see that even at 30 degrees, dispersion at 15 khz stinks, down often 10 db or more. That's a 90 percent falloff in just 30 degrees. AR's 3/4" tweeter was remarkable in that it was only down 5 db 60 degrees off axis at 15 khz. Even so, AR engineers recognized the inadequacy of even that which is why the 2 outside tweeters on LST are angled 45 degrees. Does this matter? For the front driver, it does without question. Not quite as much as Howard Ferstler would suggest with the rear drivers though. Aim a flashlight in a darkened room at an angle to a wall near the corner of a room so that it bounces off both walls the way 901 bounces sound. You will see that because of this angle, the further the beam travels, the more it disperses due to divergence. This is what happens with the outside rear drivers, the inside rear drivers are much the same. The sound does not reflect like a laser beam pointed at a mirror.

I do not think there is any FR notch in the mid or upper bass of 901 when properly installed. Quite the opposite, as I've said e/e magazine reported there is a broad rise in that area. They said they didn't hear it, I do in my room. This anomoly is easily filtered out today, it's no longer a problem. That Bose 901 like all speakers is sensitive to room acoustics as well as placement within a room comes as no surprise. All speakers are. The ultimate test of a speaker is whether it can produce sound from recordings that sound like real instruments and voices. I'm sorry to say that Bose 901 in any incarnation IMO cannot. But the same can be said for all of its competitors. Victor Campos of KLH once said words to the effect that what speaker you like depends on what form of distortion bothers you least. That still seems to be true.

Personally I have no axes to grind. I viewed re-engineering Bose 901 to see what improvements were possible as a technical challenge and an exercise in learning. It was also a lot of fun and I did learn a lot from it. Recent improvements to both it and AR9 have resulted in further improvements which I had only hoped were possible. I'm very pleased with both of them and they can be made to sound surprisingly alike but AR9 will still not create quite the effect that the direct/reflecting 901 does, at least not yet.

[tysontom]

I managed to find Howard Ferstler's review on the web. I'm sorry to say I found it disappointing. For one thing, it was not the original Bose 901 but I think Series VI which had considerably compromised bass being a ported design in a small box versus the original acoustic suspension design. We've discussed this many times, no need to rehash it. More disappointing, the speakers were installed in a manner radically different from what the manufacturer recommended. I'm not sure what the signifigance of his review is if it means anything at all. I heard Series VI earlier this year at a Bose store in a shopping mall. It had the same dull treble it always had, the top octave being missing in my opinion. Dr. Bose said in his white paper that when he listened to other speakers, the more money they charged the more shrillness the customer got. This may still be true but avoiding shrillness by not producing any high frequencies is not the right answer either. I have no problem with speaker bashing or bashing any other products which fall short of their advertised claims whether explicit or implicit if those criticisms are justified. IMO, the implied claim of all "high fidelity" sound systems is to recreate the audible experience of live music with exact faithfulness. The across the board failure of the art to correctly analyze the problem and solve it would be bad enough if it weren't for the fact that not only have prices for some equipment soared to absurd levels but that each new variant is claimed as the magic silver bullet that finally gives what others have only promised. I've read a lifetime of this kind of ad copy and stopped paying attention to it a long time ago. I've also stopped paying attention to the equally misleading opinions of the pundits who reveiw them for magazines. I'm not saying they are necessarily crooked, the distinct possibility exists that most of them are tone deaf. I don't think anyone with normal hearing would be fooled for very long by any available audio equipment you could buy that they are hearing live music and not a recording. This was the unique experience of the live versus recorded demos Acoustic Research gave in the 1960s. I wonder how convinced I would by them today now that I am a far more critical listener.

The factual evidence that 4" drivers beam their midrange and whatever treble they produce is based on both well established theoretical considerations and on measured evidence. Just look at the off axis response of any 4" driver and you will see indication of its directionality usually starting at around 2 khz but rarely if ever much higher. It is a fact that as frequency increases and the wavelength of sound decreases to the point where it is becoming comparable in size to the diameter of the vibrating member, it becomes increasingly more directional. What he left out is that tweeters of any type are hardly much better. While they can produce sound above 10 khz, just look at the off axis response of any of them and you will see that even at 30 degrees, dispersion at 15 khz stinks, down often 10 db or more. That's a 90 percent falloff in just 30 degrees. AR's 3/4" tweeter was remarkable in that it was only down 5 db 60 degrees off axis at 15 khz. Even so, AR engineers recognized the inadequacy of even that which is why the 2 outside tweeters on LST are angled 45 degrees. Does this matter? For the front driver, it does without question. Not quite as much as Howard Ferstler would suggest with the rear drivers though. Aim a flashlight in a darkened room at an angle to a wall near the corner of a room so that it bounces off both walls the way 901 bounces sound. You will see that because of this angle, the further the beam travels, the more it disperses due to divergence. This is what happens with the outside rear drivers, the inside rear drivers are much the same. The sound does not reflect like a laser beam pointed at a mirror.

I do not think there is any FR notch in the mid or upper bass of 901 when properly installed. Quite the opposite, as I've said e/e magazine reported there is a broad rise in that area. They said they didn't hear it, I do in my room. This anomoly is easily filtered out today, it's no longer a problem. That Bose 901 like all speakers is sensitive to room acoustics as well as placement within a room comes as no surprise. All speakers are. The ultimate test of a speaker is whether it can produce sound from recordings that sound like real instruments and voices. I'm sorry to say that Bose 901 in any incarnation IMO cannot. But the same can be said for all of its competitors. Victor Campos of KLH once said words to the effect that what speaker you like depends on what form of distortion bothers you least. That still seems to be true.

Personally I have no axes to grind. I viewed re-engineering Bose 901 to see what improvements were possible as a technical challenge and an exercise in learning. It was also a lot of fun and I did learn a lot from it. Recent improvements to both it and AR9 have resulted in further improvements which I had only hoped were possible. I'm very pleased with both of them and they can be made to sound surprisingly alike but AR9 will still not create quite the effect that the direct/reflecting 901 does, at least not yet.

Soundminded,

Regarding your comments about the live-vs.-recorded concerts and its realism, etc., you are correct that no speaker can accurately reproduce the full spectrum of live sound. Ensemble sound, such as what AR did back in the late-1950s and early 1960s, could be accurately reproduced if done properly with the right equipment. I'm not sure how the Bose 901 would have done do on this type of live-vs. recorded music, but on bigger groups, if might even do better than the original AR-3. The Jonathan Livingston Seagull music system was a good example of using several 901s to reproduce music of this sort.

Although highly criticized over the years as being “contrived,” Acoustic Research’s 75 AR-3 live-vs.-recorded concerts (AR, DYNAKIT and CONCERTAPES), begun in 1959, were very successful -- especially the ones with The Fine Arts Quartet. These were not P. T. Barnum demonstrations, but true comparisons. These concerts required difficult echo-free, top-grade recordings and manipulation of the preamp tone controls for each different playback venue. Sometimes, but rarely, differences could be heard. During the actual concerts, the musicians would frequently “lift” their bows but continue to stroke as though they were still playing, allowing the AR-3s to take over with a carefully synchronized transitions. This invariably fooled nearly everyone except for the people sitting right on the front row, where faint tape hiss was a factor in recognizing the reproduced sound through the AR-3s. Of course, there was always a statistical percentage of correct “golden-eared guesses” for when the actual switchovers took place from live to recorded, so Ed Villchur developed a plan to stop the guessing. In one of the early tapes (I have a dubbed DAT recorded copy from the original tape of one of the original 75 AR-3 live-vs.-recorded session tapes), and there was a trick played on the audience by The Fine Arts Quartet. In the audience were professionals, critics and general music lovers (Carnegie Recital Hall in New York), and at the beginning of the concert (Ravel’s Scherzo from Quartet in F and Bartok’s Allegretto pizzicato, from Quartet No. 4) lead-violinist Leonard Sorkin began by playing only reproduced music, but the quartet musicians continued to bow the strings as though they were playing. This went on for some time before the music ended. About midway through the music, Sorkin stopped and asked the audience, “By the way, how many of you could detect the switchovers between the musicians and the AR-3 system?” There was a big show of hands -- probably 50% of the audience. He said, “well, I’m sorry to tell you this, but except for the first bar, there were no switchovers, and the AR-3s played through to the end!” For the next several years of these many concerts, the “guessing” slowed down quite a bit. In the final analysis, these AR live-vs.-recorded sessions are still considered to be the only known truly “successful” live-vs.-recorded music reproduction ever done.

You mentione the "dullness" of the 901's treble output. AR’s Ed Villchur once said that some people complained about the “dullness” of the AR-3’s high frequencies, but his point was that high frequencies heard in a concert hall are significantly attenuated by the time they are heard at the listener’s ears; and once again, the real purpose of high-fidelity reproduction is to take one to the concert hall rather than attempt to reproduce the symphony orchestra in the home. Real music is rarely is a “bright-sounding” as reproduced music, so a loudspeaker such as the AR-3 or the Bose 901 might sound reticent when listening on axis, but will sound more natural (and accurate) well back in the reverberant field because the energy is well dispersed throughout the listening environment. This is what Bose talked about in his 1968 AES paper and in his patent. With the help of Bolt, Beranek and Newman, Bose did a lot of research (12 years of study) into the acoustics of concert halls, and his findings were well documented. As you know, the Bose 901 follows this concert-hall spectral-energy pattern, and the speaker has extended and smooth acoustic-power response in real rooms. Since the 901 disperses so much energy off the front wall through reflection, it has a serious advantage over most conventional speakers when it comes to realism of sound reproduction. Another example of this type of direct/reflected design would be the original Mark Davis designed Soundfield One speaker, which also relied on reflected sound as well as electronic equalization to correct acoustic anomalies. It was different, and it used a significant amount of DSP to get the desired results. I’ve heard the Soundfield speaker years ago, and I was impressed, but the price tag on that speaker was exorbitant, and it was too large. That speaker failed because of the lack of practicality; the Bose 901, however, was relatively inexpensive.

As for the 901's treble output, the drivers are nominally a 4½ inch-frame devices. The actual cones, however, are 3-inches in diameter; and if you consider one-half of the surround, the cone diameter becomes 3½ inches. This is about the same size as the original AR-4 tweeter (at 3½ inches also), and as we all know, the off-axis performance of this speaker is good if not excellent. At the 13-15kHz point (effectively the top end of the AR-4 as well), anechoic frequency response 30° off-axis is down about 7 dB, and about 15 dB down 60° off-axis. The laws of physics will treat the Bose 901 driver almost the exact same way it treats the AR-4 tweeter, as it is a function of the radiating-surface diameter and the frequency wave length. Since the transient response of the 901 is nearly perfect across the spectrum, there is little indication that the drivers are misbehaving well up into the upper treble range of 14 to 15 kHz. But more importantly, the dispersion of the front-firing 901 driver is purely academic: the speaker’s dispersion comes almost entirely from the reflection off the wall from the rear-firing drivers, and the front serves mainly as localization for the 11% or of direct sound normally heard at a concert hall. My big beef with Howard Ferstler’s “billiard-ball” reference is that the drivers on the rear of the 901 cabinet do not, as he says, couple together like one big “9-inch” woofer trying to reproduce all the frequencies from the low bass to the high treble. This is a flawed notion on his part. They do couple together below the point of ultimate-radiation impedance and stay coupled down into the deep bass, but they increasingly decouple above that frequency. There is some line-source effect from the configuration, but it is minimal. There are also interference effects from the multiple drivers, but all that is swamped in the reverberant field. Overall, the 901 disperses the high frequencies quite well from the back of the enclosure, as well as bounces this dispersed sound well into the room, better than most any other design. This is the reason that so many knowledgeable testers and critics praised the sound of the 901.

The new-style 901 with the non-sealed enclosure is probably capable of greater bass output than the original units, and using less power to do so. As for distortion, I haven’t seen a comparison of the sealed vs. the vented design, but the engineers at Bose certainly worked hard on the new design to improve it over the original one. In either case, it is known that the harmonic distortion from these speakers, while possibly measurably higher than the Acoustic Research counterpart, is still low enough to be largely inaudible. If very high power and low distortion is needed, one can always add another pair of 901s to run in parallel (but usually requiring a second high-powered amplifier).

--Tom Tyson

[soundminded]

Tom, I'm going to try to avoid writing a book in reply to your posting. As I've claimed, all of my experiements are based on an unpublished mathematical model I developed 35 years ago and still try to not fully reveal since I have hopes it may still one day have commercial value.

"Regarding your comments about the live-vs.-recorded concerts and its realism, etc., you are correct that no speaker can accurately reproduce the full spectrum of live sound."

I hope I did not say that. Many speakers can reproduce sounds from the lowest to the highest audible frequencies and loudness from the threshold of hearing to the threshold of pain. But there's more to sound fields than that and what that doesn't take into account they cannot reproduce in a home.

"I'm not sure how the Bose 901 would have done do on this type of live-vs. recorded music, but on bigger groups, if might even do better than the original AR-3. The Jonathan Livingston Seagull music system was a good example of using several 901s to reproduce music of this sort."

That would depend on the hearing accuity of the listener but to someone whose hearing is good and pays attention, Bose 901 hasn't got a prayer. This is because among other reasons, it cannot reproduce the sharp transient attack of most instruments especially percussive instruments and violins no matter how it is equalized. I agree with Gordon Holt's statement in his review that the inertial mass of the cones is just too great for those drivers to be effective tweeters. Also, if you recall the article, the 901s were not used in a direct reflecting mode but turnd around so that most of their sound was directed forward. It was not a live versus recorded demonstration, and reproducing instruments at a live venue says nothing about the same sound systems capabilities in a home because most of what you hear at a live performance, the reverberant sound field due to the acoustics of a hall is not captured on commercial recordings. This can easily be understood by the fact that microphones used in commercial recordings are placed far closer to the instruments than anyone sitting in the audience would be seated and their usual pickup pattern which is cardiod does not equally pick up sounds coming from directions away from the musicians. Beyond that, there is no known way to segregate the reverberant sound from the direct sound on a recording and no known way to recreate the sound fields they reproduced in anything resembling an adequate reconstruction.

BTW, I heard a difference in the live versus recorded demo of the guitarist. Surprisingly, the AR3 was slightly brighter. I was sitting on axis. I considered the difference though to be insignificant. I was amazed it was as close as it was. I don't think I was nearly as critical a listener in the Nickelodeon demo. I was kind of startled and overwhelmed by the novelty of the Nickelodeon itself to pay as close attention as I should have. Again I think I heard differences but the similarities were striking.

"You mentione the "dullness" of the 901's treble output. AR’s Ed Villchur once said that some people complained about the “dullness” of the AR-3’s high frequencies, but his point was that high frequencies heard in a concert hall are significantly attenuated by the time they are heard at the listener’s ears"

My model dismisses the integrated time averaged spectral transfer analagous to the frequency response of a concert hall as worthless. This is because as I explained in a previous posting that the timbre of musical instruments is a dynamic event, the first part of each note being either directly from the instruments themselves or from reflections off of nearby objects carries the greatest high frequency content giving the sound its distinctive definition and clarity (the attack at the front edge of piano key being struck for example) but the high frequencies disappearing at a relatively faster rate than middle and lower harmonics and the fundimental. For example, typical of RT60s, the time for sound to decrease by 60 db in most concert halls is between 1.8 and 3.0 seconds at 1 khz, and only 1.0 to 1.2 seconds at 8 khz, the limit of most tests. This makes the sound melower as the brain integrates the echoes with the direct sound to form a subjective impresson of it while retaining its clarity. The perception of the space you are in, the apparent power of the instruments, and their timbre are therefore inseparable, different aspects of the same phenomena. BTW, Vilchur's flawed idea is far from dead. BBC's current views on sound makes the same assumptions and mistakes and there is a "BBC FR shift" which deliberately attenuates high frequencies along these lines for the same reason.

Is a symphony orchestra bright sounding? Listen to one in a relatively confined practice room as I did in the basement of Queens College's Coulden Hall and you would be surprised just how bright it its. Cement block walls in a much smaller room than the hall itself. Unamplified in a concrete band shell outdoors works well too.

Bose 901 cannot possibly hope to reproduce anything like the sound of a symphony orchestra as heard in a concert hall in your home. As I said, the reverberation constituting most of the sound heard live is not on the recording. (vinyl phonograph records compress sound to squeeze it into the limited dynamic range capabilities insufficient for classical music making the tail end of each note or phrase during a rest much louder than when the instruments are playing. This makes the reverberation louder at that time which may in part explain why many people like vinyl better than CDs which do not need to employ dynamic compression.) A concert hall is much more than a place of assembly, it is a part of the music itself because of the way it modifies sound. This not only affects the way the audience perceives it, it affects the way the musicians perform and the kind of music that can be performed in different environments. This is why symphony orchestras usually sound better in their home halls in acoustics they are familiar with than on tour. The 5 to 8 second RT of a cathedral makes it unsuitable for all but organ concerts and large choral works usually at a slow tempo with mostly half and whole notes. Beranek's lecture at Georgia Tech in 2001 sadly no longer available on the internet traced the history of the science of acoustics from its beginnings early in the 20th century when professor Sabin of Harvard was asked to assist the design of a new home for the Boston Symphony Orchestra. The result, Boston Symphony Hall patterned after a hall in Leipzig is considered by many to be the best room for listening to musical performances in the United States and the second or third best in the world. Sabin discovered through his experiments that RT is proportional to the volume of a room and inversely proportion to the absorption of the materials it is made of. (wood unless at least 2 1/2 inches thick according to Beranek is poor.) My listening room is 4000 cubic feet. Carnegie Hall is 900,000 cubic feet, 225 times greater than my room. One concert pianist told me the acoustics of Carnegie Hall were ruined when the NYC Fire Marshall ordered that the oil based paint on the walls be removed and the place painted with latex paint. Beranek didn't think much of this hall, especially seats in the lower balconies where acoustic shadow makes hearing difficult. Beranek talked about his disaster at Philharmonic Hall at Lincoln Center where he was eventually fired. His side of the story is that the managers told him to design a hall that would seat over 2900 in the audience. He said the hall would have to be too big and acoustics would suffer badly. He was right and the place has been reworked including at least two total gut rehabs since it was built. As the result of solicitations for contributions of money for all of this work, it got to be called derogatorily "Philanthropic Hall." Eventually it was renamed Avery Fisher Hall after Fisher donated a large sum. It's still not a world class hall. BTW, the cost for designing and building a room of this type today is over 100 million dollars and is sometimes unsuccessful. How does your listening room compare? :-) Ever since, Beranek and his colleages in other acoustic engineering firms like Kirkegaard have been on a crusade to reduce RT as part of renovations. Data I have for a couple of hundred halls renovated between 1962 and 1982 published by ASA show this was always the case even in Concertgebouw where the engineers were specifically directed not to change the acoustics.

If Bose 901 and any other speaker/home sound system cannot reproduce the sound of live music at a concert hall, opera house, or cathedral, what is the best that can be expected of them? IMO to reproduce the sounds of individual acoustic instruments and small ensembles as they would be heard in your room if they were there. But even this is well beyond all home sound systems for a large number of reasons. For one, speakers don't propagate sound the way musical instruments do. Walk around a musical instrument and listen for changes in timbre. You usually won't hear any. Now walk around your speaker and you will hear great differences as you get of axis of the tweeter. The reflections off the walls of your room are entirely different and these play a major role in what you hear from a recording. So much so that moving the speakers just a few inches or aiming them in slightly different directions can have a substantial effect. In an effort to minimize these effects, audiophiles have moved speakers away from walls, added sound absorbing material to the walls, bought highly directive speakers but unless you listen on axis in an anechoic chamber, a very bad place to listen to music you will hear lots of room reflections especially at middle and low frequencies. Speaking about dispersion, a 7 db falloff for a 4" tweeter means that 85 percent of the sound is confined to a cone only 30 degrees in radius. Sadly, most 1" dome tweeters today aren't much better. Just look at the graphs in catalogs like Madisound. For a Bose 901 driver as a tweeter, you can see that the cone being recessed behind the suspension in a way similar to the UMR in AR9 restricts its lateral dispersion but unlike a dome driver, no part of it projects beyond that lateral restriction. A 15 db falloff BTW means its up to 96% of its energy within a 60 degree cone of the tweeter axis. By any measure that stinks and is vastly different from the way midrange and low frequency sounds are propagated. Interestingly, Leo Beranek and Floyd Toole came to the same conclusion about music lovers liking lateral reflections. Toole published that in his papers, Beranek's is in a paper on his web site trying to corelate 20 measured parameters of concert halls with preferences of golden eared conducters and other cognescenti. The number one correlation is what he calls BQI for binaural quality index which depends on early lateral reflections. This is why shoebox shaped halls work better than most other shapes including pie wedge shape halls. This is one thing Bose 901 does well, AR LST and some di-pole speakers do fairly well, and most other speakers do poorly. Another problem is the varialbes of recordings themselves. In my experience, the relative spectral balance of all recording media including CDs is highly varialbe, all over the map in fact. There are no standards for making recordings so each one has to be equalized individually. This is a very tedious and time consuming process requiring for me at least to listen again and again after rest intervals of hours or days. So without the means, experience and determination to do this chore, even if a sound system played one or some recordings accurately, it would not perform well on most others in that regard.

Ultimately all sound is subjective. Since it is an event in time, you cannot compare two sounds directly the way you could hold two photographs side by side and say which has the sharper image. Comparing sounds even in rapid fire AB tests relies on memory. In listening to the best recordings of Heifetz from a technical recording point of view, my friend and I have entirely different memories of what his priceless Guarnari del Jesu sounded like. We both heard it live only once at Carnegie Hall in the 1960s. Her impression is that it was far more shrill than I rememeber it. I attribute this to the time gap and her likely hearing loss at high frequencies and needing to compensate for it to hear what she remembered. We both have far more experience with a similar violin known as the Guanari del Jesu ex Kochanski which we've heard many times in many places in both our homes, the owner's home, and in many concert halls. No commercial sound system I've heard comes close to duplicating this sound. Neither did a fine violin maker who tried. While the rawness of the E, A, and D strings on his copy might have worked out in time with playing, the G string was a disaster, far too weak and would never have come around in all likelihood. If it were otherwise, we'd have paid the $4000 asking price immediately.

On a somewhat less pessimistic note, all of this does not mean that we cannot enjoy our recordings and electronic toys to listen to facsimiles of music. But what it also means IMO is that if you are a critical listener with good hearing and much experience listening to live music performed on acoustical instruments and expect your audio equipment to sound like live music no matter how much it cost or who made it, if you are honest, you are in for a disappointment. At this time that is still well beyond the state of the art.

[Carlspeak]

Here is an interesting sidelight to this discussion on Bose speakers. It relates to the sad state of critical listening evident now. It doesn't relate to any specific classic NE loudspeaker. But, I thought it was worthy of entry into some discussion area as CSP on it's own merits.

Here is a link to a round table discussion worth a view on U-Tube.

It's relatively quite long (2.5 hrs) but could be taken in piece meal.

http://uk.youtube.com/watch?v=SY5hI98HEi0

It was recorded at the Philoctetes Center for the multidisciplinary

study of imagination.

The basic idea is a round table discussion of the importance of "Deep

Listening" and high quality audio playback. The subject wanders a

bit, because the panelist are 3:2 music industry vs. audiophile.

On the panel are:

Greg Calbi (mastering engineer)

Steve Berkowitz, Senior Vice President of Sony Music's Legacy Records

Evan Cornog - Associate Dean of the Graduate School of Journalism at

Columbia University

Kevin Killen - recording engineer

Craig Street - record producer

Michael Fremer - writer for many audio publications.

The basic premise is that "Deep Listening" I.E. sitting down and

listening to the music, giving it your full attention is important.

And with a good audio system it can be transcendent.

Of all of these, only Berkowitz seems to say that the quality of the

system isn't that important to him. Though he does seem to enjoy it.

Calbi says that some of the greatest audio experiences he has had

were listening to Fremer's system.

Lots of good stories, too. The tales of remastering were very

interesting to me. What, in fact, IS the master?

[soundminded]

I listened to the entire link and while it was interesting, it didn't seem to me to have anything to do with what we discuss here. In fact it was about what I'd have to call "anti-high fidelity." It was not about recreating the sound of live musical performances. It was not about precisely capturing and recreating the sound of the world's best musical instruments and preserving the best performances of the world's greatest artists. It was about "manufacturing" sound as a successful commercial enterprise from what several at various times in the discussion frankly admitted did not sound very good in the studio. It was also about pleasing their direct customers with their end products. Did you catch the part about where one customer said that he didn't want a piano to sound like a piano and the guy thoght, hmm, maybe I should put paper clips on the strings. They talked about "good sound" without ever discussing what that meant to them. Also, their scientifc knowledge about sound was very weak and their occasional statements about it were usually wrong flying in the face of known scientific facts. BTW, I didn't like any of the music they played. If that was what high fidelity sound was about, I wouldn't have become interested in it.

[tysontom]

Soundminded, these messages are nauseatingly long and detailed. This is the last long message from me on this topic. However, one last time in response to your message:

>Tom, I'm going to try to avoid writing a book in reply to your posting. As I've claimed, all of my experiements are based on an unpublished mathematical model I developed 35 years ago and still try to not fully reveal since I have hopes it may still one day have commercial value<

If you developed this “mathematical model” 35 years ago, you best be getting on with it; audio is about dead. Why have you held back on doing something with your idea?

>I'm not sure how the Bose 901 would have done do on this type of live-vs. recorded music, but on bigger groups, if might even do better than the original AR-3. The Jonathan Livingston Seagull music system was a good example of using several 901s to reproduce music of this sort …That would depend on the hearing accuity of the listener but to someone whose hearing is good and pays attention, Bose 901 hasn't got a prayer. This is because among other reasons, it cannot reproduce the sharp transient attack of most instruments especially percussive instruments and violins no matter how it is equalized. I agree with Gordon Holt's statement in his review that the inertial mass of the cones is just too great for those drivers to be effective tweeters<

Soundminded, equalization has nothing whatsoever to do with transient response. There is nothing anywhere to suggest that the Bose 901 has poor transient response. The cones are small, light-weight and rigid and do not act like some heavy, lumbering diaphragm that lacks control from its motor circuit. The “inertial mass of the cones” theory you mention is pretty much bunk, by the way, especially coming from Gordon Holt; moreover, if the 901 moving assembly is rigid enough to prevent cone breakup, and the magnet structure is sufficiently strong to provide the necessary damping, there is no reason that the speaker can’t respond accurately to transients within its pass-band frequency range. Of course, it will not respond up to 40 kHz, but up to 14-15 kHz, it does just fine. In fact, Julian Hirsch (whom I respected far more than I ever respected Gordon Holt) commented on the 901’s excellent transient response in his review of the original 901: “…The Bose 901 had an utterly clean, transparent and effortless sound. Its clarity and definition when reproducing complex orchestral passages were, in the writer’s opinion, unsurpassed by any other speaker he has heard. This impression was confirmed by its tone-burst response, which was uniformly excellent across the frequency spectrum….”

>My model dismisses the integrated time averaged spectral transfer analagous to the frequency response of a concert hall as worthless. This is because as I explained in a previous posting that the timbre of musical instruments is a dynamic event, the first part of each note being either directly from the instruments themselves or from reflections off of nearby objects carries the greatest high frequency content giving the sound its distinctive definition and clarity (the attack at the front edge of piano key being struck for example) but the high frequencies disappearing at a relatively faster rate than middle and lower harmonics and the fundimental...<

What is all this gobbledygook? This does not make sense.

>BTW, Vilchur's flawed idea is far from dead.<

What are you referring to here? What flawed idea? Live-vs.-recorded sound?

>Is a symphony orchestra bright sounding? Listen to one in a relatively confined practice room as I did in the basement of Queens College's Coulden Hall and you would be surprised just how bright it its. Cement block walls in a much smaller room than the hall itself. Unamplified in a concrete band shell outdoors works well too.<

99.99% of all symphony halls are not as you describe, and every concert I’ve ever been to that was in a normal hall, the high frequencies are much more muted and, of course, spacious-sounding, but never “bright.” It is a proven fact that if you are sitting back in a symphony hall, you are getting well-over 80% reflected sound, and the high-frequency balance is tilting downward.

>If Bose 901 and any other speaker/home sound system cannot reproduce the sound of live music at a concert hall, opera house, or cathedral, what is the best that can be expected of them? IMO to reproduce the sounds of individual acoustic instruments and small ensembles as they would be heard in your room if they were there.<

No one has been able to successfully replicate a full orchestra in a symphony hall. G.A. Briggs of Wharfedale Wireless Works attempted full-orchestra live-vs.-recorded demonstrations in the 1950s using several refrigerator-sized Wharfdale full-range loudspeakers, and although it was a valiant effort, it wasn’t considered successful. The ensemble-size of a string quartet is far easier to reproduce, of course, and AR pulled this off effortlessly. AR did also reproduce a 5-rank Aeolian-Skinner pipe organ in Mt Kisco, New York in a small cathedral. But this notion of bringing a full orchestra or musical group into a house to attempt to reproduce it accurately has long-ago been disregarded as impractical nonsense. Again, the whole idea of accurate sound reproduction is not to bring the full orchestra into your house; the idea is to transport you to the concert hall (while you sit in your easy chair) and give you the illusion of the original sound. Accurate high-fidelity reproduction is an illusion and not an exact reinstatement of the instruments, musicians and podium.

>Speaking about dispersion, a 7 db falloff for a 4" tweeter means that 85 percent of the sound is confined to a cone only 30 degrees in radius. Sadly, most 1" dome tweeters today aren't much better. Just look at the graphs in catalogs like Madisound. For a Bose 901 driver as a tweeter, you can see that the cone being recessed behind the suspension in a way similar to the UMR in AR9 restricts its lateral dispersion but unlike a dome driver, no part of it projects beyond that lateral restriction. A 15 db falloff BTW means its up to 96% of its energy within a 60 degree cone of the tweeter axis.<

Once again, the 901’s front-firing driver is not the source of wide dispersion, extended and smooth acoustic-power response from the 901. That comes from the rear-firing drivers reflecting off a hard wall. The front speaker simply gives localization (the 11% that you would get in a concert hall), and it is probably good that the dispersion is not like a dome ¾-inch tweeter.

>On a somewhat less pessimistic note, all of this does not mean that we cannot enjoy our recordings and electronic toys to listen to facsimiles of music. But what it also means IMO is that if you are a critical listener with good hearing and much experience listening to live music performed on acoustical instruments and expect your audio equipment to sound like live music no matter how much it cost or who made it, if you are honest, you are in for a disappointment. At this time that is still well beyond the state of the art.<

People continue to look for the “holy grail” of sound reproduction, and many are willing to spend whatever it takes to get to that point. But that is not the purpose of accurate-sound reproduction, and chasing after that “holy grail” will only lead to frustration. But my argument is that the Bose 901 does its job well -- extremely well -- and it has had unwarranted (and largely opinionated and unscientific) criticism since day one. It’s far from perfect, but it’s much, much better than some people would have you believe.

--Tom Tyson

[JKent]

First, a belated welcome to Howard Ferstler! Howard is prominent over on the Yahoo! Allison forum and has been very helpful to me (and to many others). He is a noted author in the hi-fi field (do a google search).

Now: On topic. Dr. Bose's greatest strength may be his marketing genius. None of the other New England companies, so revered here, have survived. Bose otoh has survived and is doing very well.

Besides his infamous suit against Consumer Reports to save his speakers' reputation ( a move that ticked off many audiophiles and CU supporters ) and the endless late-night TV ads for Bose's "miracle" products, there seems to be a commitment to the customer. Recently a CSP forum member wrote about a pair of 901s that he had picked up at a garage sale. They had been bastardized with replacement drivers. When he contacted Bose, they offered to sell him a brand new set of 901 Series VI speakers for half price, as an "upgrade" for a previous customer, regardless of the fact that he was not the original owner. That's customer care!

Of course many (myself included) consider Bose products to be over-priced, over-hyped mid-fi stuff. But at least we can give credit where credit is due.

Kent

[soundminded]

"If you developed this “mathematical model” 35 years ago, you best be getting on with it; audio is about dead. Why have you held back on doing something with your idea?"

The mathematical model is my real invention. Everything else I have done, seen thought, built, considered is based on it. My problem is that I haven't figured out how to make any money out of it so far. When I received my patent for a sound system based on it, the twenty most likely candidates I approached I thought could best exploit it were not interested. Then it was in part stolen. Two patent attorneys I contacted advised me not to sue. The infringement was based on a minor loophole, a discrepency between the abstract, the first paragraphs that describes the invention, and the claims, the legal protections at the end of the patent. It shouldn't have mattered but they felt otherwise. A new patent attorney has reviewed some material I sent him but his problem is that he doesn't know how to get around my original patent. It took me a year but I figured out how. But given my prior experience, I'm not sure another patent is the way to go. I'm thinking about some other approaches. Publishing the model is the last thing I want to do. Once the cat's out of the bag...

"Soundminded, equalization has nothing whatsoever to do with transient response."

I'm not going to state this as a matter of opinion but as a matter of fact. Frequency response and transient response are two ways of looking at the same thing. It is required of every electrical engineering student to be able to calculate and plot either one given the other in either direction. I once had a heated discussion about this on another blog site with John Curl. He initially took your position. (He was trained as a physicist, not an electrical engineer.) Then when he realized his mistake, he took my position and ascribed his former position to me. I had fun at his expense for a long time juxtaposing his two postings which contradicted each other. BTW, this can be found in any advanced text in electronic circuit design and analysis such as the one I used that was written by my professor Dr. Paul Chirlian over 40 years ago. But every other text will have the same in it. Or just ask a professor of electrical engineering.

"There is nothing anywhere to suggest that the Bose 901 has poor transient response."

"The “inertial mass of the cones” theory you mention is pretty much bunk, by the way, especially coming from Gordon Holt"

Whether something is right or wrong does not depend on who said it. While I disagree with much he's said over the years, in this case I think he's right. Besides, you can hear it. What is your explanation for why audiophiles so overwhelmingly detest this speaker? IMO lack of a high end is a big part of it. That the speaker exhibited no signs of ringing in tone burst tests only demonstrates that it had no high frequency peak. This is one of the correlations between frequency response and transient response. I respected Julian Hirsch for many reasons including his thorough and objective measurements and impartiality. On the other hand, I do not respect his hearing when it clearly contradicts what I hear. In its day, 901's rolled off high end was not all that different from many other speakers. But it never moved beyond that. Meanwhile, my ability to listen critically has greatly improved even if my hearing is no better. Critical listening is IMO a learned skill that takes a great deal of time, determination, and patience.

"What is all this gobbledygook? This does not make sense."

One method of measuring concert halls is with what is called an ILG fan. It is a loud fan with sound output of known frequency spectrum. It is placed on the performing stage of a concert hall and a calibrated microphone is placed in various test locations. The spectrum of the fan's sound measured this way is compared to its known spectrum and the difference is the time averaged spectral transfer function of the hall, usually presented as ranging between an upper and lower curve based on the range of measurements. This is a steady state measurement. But sound is not heard that way. It is heard first directly from the instruments themselves or early reflections off nearby objects which have the highest relative high frequency content. As time passes, the listener hears echoes which not only diminish in amplitude but in relative high frequency content compared to the earlier arriving sound from the same note. This is why when averaged over time, the concert hall appears to have a high frequency rolloff. This is also reflected in the fact that the reverbertion time at higher frequencies is shorter than at middle and lower frequencies. But you can't duplicate the timbre of musical instruments by putting the signal throgh a filter which results in this time averaged curve. Thsy results in the muffled sound typical of speakers designed around that philosophy. Their clarity and brilliance are lost. A seeming contradiction for audiphiles, a sound which is both brilliant and mellow at the same time because of this phenomenon.

">BTW, Vilchur's flawed idea is far from dead.<

What are you referring to here?"

What I just said above.

"It is a proven fact that if you are sitting back in a symphony hall, you are getting well-over 80% reflected sound"

Actually it's usually well over 90%. Bose's white paper showed it as 89% only 16 feet from the stage at Boston Symphony Hall with the percentage getting progressively greater as you went further back in the hall. But microphones are placed close to the instruments picking up very little reverberant sound compared to what the audience hears. As a result, we play the 2. 5 or 10 percent we do have 10, 20, 50 times louder to make up for the 90, 95, or 98 percent of the sound we don't have, sound which is qualitatively very different from what is on the recording.

"But this notion of bringing a full orchestra or musical group into a house to attempt to reproduce it accurately has long-ago been disregarded as impractical nonsense"

"Again, the whole idea of accurate sound reproduction is not to bring the full orchestra into your house; the idea is to transport you to the concert hall (while you sit in your easy chair) and give you the illusion of the original sound."

This is what manufacturers of audio equipment have often claimed, if not explicitly then implicitly. Sound reproduction systems capable of this do not exist....yet. The technology is well beyond the current state of the art. Perhaps this is why audio is dead as you say. My model and experience demonstrates that it is not impossible, it is entirely possible but not by any methods marketed or in evidence of research today. It also isn't at all easy.

"Once again, the 901’s front-firing driver is not the source of wide dispersion, extended and smooth acoustic-power response from the 901. That comes from the rear-firing drivers reflecting off a hard wall. The front speaker simply gives localization (the 11% that you would get in a concert hall), and it is probably good that the dispersion is not like a dome ¾-inch tweeter."

The first arrival of high frequencies is a key to sound localization. Lack of adequate high frequency dispersion of the front driver is one key to understanding the shortcomings of this loudspeaker system. I got interesting results when the rear firing components were high frequency enhanced but not the front. A strange effect like deliberte electronic phasing of sibilants and other transient attack components.

"People continue to look for the “holy grail” of sound reproduction"

Why shouldn't they? That is what advertising hype in this industry has promised for over 100 years. But the reality doesn't come close to the hype. One important question few people ever ask is WHY does sound reproduction need to be accurate? The answer is that when properly reproduced as an excellent facsimile of the original, the reproduced sound of many great musical instruments performed by great artists is truely a beautiful thing to hear offering far greater enjoyment than people not familiar with live unamplified music can possibly imagine.

I'm sorry but Bose 901 does not do its job well of reproducing the sound of musical instruments in a home well IMO. But then again neither do any of its competitors... for entirely different reasons.

"Soundminded, these messages are nauseatingly long and detailed. This is the last long message from me on this topic. However, one last time in response to your message"

I'm sorry you feel that way. I think that anyone who thinks a posting is too long should not feel obligated to read it let alone respond to it.

[tysontom]

The mathematical model is my real invention. Everything else I have done, seen thought, built, considered is based on it. My problem is that I haven't figured out how to make any money out of it so far. When I received my patent for a sound system based on it...

"Soundminded, equalization has nothing whatsoever to do with transient response."

I'm not going to state this as a matter of opinion but as a matter of fact. Frequency response and transient response are two ways of looking at the same thing. It is required of every electrical engineering student to be able to calculate and plot either one given the other in either direction....

One method of measuring concert halls is with what is called an ILG fan. I

"It is a proven fact that if you are sitting back in a symphony hall, you are getting well-over 80% reflected sound"

Actually it's usually well over 90%. Bose's white paper showed it as 89% only 16 feet from the stage at Boston Symphony Hall with the percentage getting progressively greater as you went further back in the hall. But microphones are placed close to the instruments picking up very little reverberant sound compared to what the audience hears....

"Once again, the 901’s front-firing driver is not the source of wide dispersion, extended and smooth acoustic-power response from the 901. That comes from the rear-firing drivers reflecting off a hard wall. The front speaker simply gives localization (the 11% that you would get in a concert hall), and it is probably good that the dispersion is not like a dome ¾-inch tweeter."

The first arrival of high frequencies is a key to sound localization. Lack of adequate high frequency dispersion of the front driver is one key to understanding the shortcomings of this loudspeaker system.....

I'm sorry but Bose 901 does not do its job well of reproducing the sound of musical instruments in a home well IMO. But then again neither do any of its competitors... for entirely different reasons.....

I'm sorry you feel that way. I think that anyone who thinks a posting is too long should not feel obligated to read it let alone respond to it.

Soundminded,

I didn't realize you had a patent on your design. What is the patent number? It would be in the public domain if a number was granted, so this will help avoid a search. Send it to me off-line if you wish. I think it is great (and I congratulate you) that you are pursuing your idea, and at some point you will find someone willing to build it for you.

I didn't say that frequency response and transient response were unrelated; on the contrary, they are closely related, and poor transient response will almost always show up as a frequency-response aberration. However, I said "equalization" has nothing to do with transient response, and I know that is true. You can shift up or down the balance of output from a transducer using an equalizer, but if the driver "rings," you cannot tune that out -- it will only be increased or decreased by the equalization amount. My point is that there is no evidence anywhere that the Bose 901 drivers "ring," which is what is inferred when saying they have too much mass to reproduced the frequencies they are assigned. This is incorrect.

I am very familiar with the ILG fan, used for calibration in anechoic chambers and so forth. I just wasn't able to follow your line of reasoning in that previous paragraph.

I'm glad we agree on the direct/reflected ratio of energy in a concert hall. By the way, if the microphones were placed well back in a symphony hall to pick up the "reverb" (instead of over the top of the musicians), the sound when reproduced would be (and has shown to be) muddy, distorted, diffuse and nearly unlistenable in the playback environment. The reverb effect is therefore lacking, as you suggest (this can be replaced with a good preamp with the proper delay settings), but the direct/reflecting part of the experience is restored with the use of wide-dispersion loudspeakers such as the Allison: One, AR-LST and the Bose 901.

The delay in "arrival" in a listening room of the output from the front-firing and the back-firing speakers in the 901 is miniscule, and barely noticeable. Again, the front speaker is used mainly for "localization" of instruments, and the rear eight drivers are used to disperse the signal, therefore, the dispersion of the front-firing driver is not that critical. It is actually not as bad as people say, however. The end result is that the 901 -- with whatever faults it has -- has extremely extended acoustic-power response in a typical listening room, something that the majority of front-firing loudspeaker (save possibly the LST, Soundfield One and Allison: One) struggle to do. In contrast, the acoustic-power response of an AR-9 is not as extended or uniform as the Bose 901, whereas it outpoints the 901 in many other respects, such as deep-bass power and on-axis high-frequency output.

As far as the 901 doing its job well in a listening room, this is of course subjective and opinion on the part of the listener. And as for the lengthy messages, I'm referring to mine as well as yours. It's very hard for most people to follow lengthy, detailed messages, and I'm very guilty of this as well. I will say, however, that there have been a lot of people (almost 3,000) to read this dialogue since it began, so there is interest out there.

--Tom Tyson

[tysontom]

First, a belated welcome to Howard Ferstler! Howard is prominent over on the Yahoo! Allison forum and has been very helpful to me (and to many others). He is a noted author in the hi-fi field (do a google search).

Now: On topic. Dr. Bose's greatest strength may be his marketing genius. None of the other New England companies, so revered here, have survived. Bose otoh has survived and is doing very well.

Besides his infamous suit against Consumer Reports to save his speakers' reputation ( a move that ticked off many audiophiles and CU supporters ) and the endless late-night TV ads for Bose's "miracle" products, there seems to be a commitment to the customer. Recently a CSP forum member wrote about a pair of 901s that he had picked up at a garage sale. They had been bastardized with replacement drivers. When he contacted Bose, they offered to sell him a brand new set of 901 Series VI speakers for half price, as an "upgrade" for a previous customer, regardless of the fact that he was not the original owner. That's customer care!

Of course many (myself included) consider Bose products to be over-priced, over-hyped mid-fi stuff. But at least we can give credit where credit is due.

Kent

Kent,

I agree about welcoming Howard Ferstler to this forum. As you say, Howard is prominent over on the Allison Yahoo group, and he has contributed a lot to that forum. He is extremely knowledgeable about Allison speakers. He is a long-time friend of mine, too, and I think everyone (including me) respects his wisdom when it comes to Allison speakers. Where he has gone astray, however, is in his judgement of the Bose 901. He has some unscientific and misconceived notions about the performance -- especially the dispersion -- of that speaker.

Certainly each is entitled to his own opinion on the 901 -- clearly the most controversial speaker in history -- but few can argue with its success. You can only "market" for so far; at some point you have to have a good product, and the fact that so many of these speakers have been sold is testimony to the quality of this product. Also, I might add that most of the criticism on this speaker comes from people who do not own them and who have not lived with them for any length of time.

--Tom Tyson