What the Japanese learned about emotional impact was that, in general, it was a fleeting thing. It was not something that was consistant, kind of fashionable. In other words what had an emotional impact today may not tomorrow and that the emotions could be easily swayed by external influences. In short, emotions did not correlate very well with anything real, like handling or noise level, etc. but the superficial things like color and styling and seat material. But its exactly the superficial things that we come to take for granted in the long run and pretty much ignore.
The car company with the strongest customer loyalty is Honda. Now anyone will tell you that Honda cars are about dead last in emotional impact (except the NSX of course!). They do not go in for fashionable styling (more so now than in the past) and they base their designs more on function than aesthetic considerations.
I find that speakers are quite like this. In general, a strong initial emotional impact is very likely to be exactly that which sounds so bad tomorrow, etc. It's the speakers that kind of just sit there without being imposing in any way that end up delighting the most in the long run. Because, it's the music, not the loudspeakers, that bring the real emotion.
The car company with the strongest customer loyalty is Honda. Now anyone will tell you that Honda cars are about dead last in emotional impact (except the NSX of course!). They do not go in for fashionable styling (more so now than in the past) and they base their designs more on function than aesthetic considerations.
I find that speakers are quite like this. In general, a strong initial emotional impact is very likely to be exactly that which sounds so bad tomorrow, etc. It's the speakers that kind of just sit there without being imposing in any way that end up delighting the most in the long run. Because, it's the music, not the loudspeakers, that bring the real emotion.
I would wonder about that. There must be some emotional impact before becoming a loyal customer. For me, in cars, it's the features cleverly implemented and arranged in whole that really reaches out to me. For audio, when I was young, the sound and price was an impact; few of which really hit home. Now, it's the overall looks, sound, and value. Really none on the market hits home. The ones that look good are expensive and sound terrible; the ones that sound good, look terrible. To make things more complicated, men and women have different buying habits.
gedlee said:
In 100% agreement with all of the preceding points. At the risk of stirring the pot, I'd like to add one more subjective quality: Resolution.
Now, let's get one thing out of the way immediately. Most audiophiles, and most of the reviewers I've met, confuse HF tizz and overshoot with resolution, just as inexperienced photographers confuse oversharpening as increased detail. Advanced amateurs and professional photographers look deeper, at rendition of textures, which only large-format media and high-resolution digital files can deliver. They notice the "flat" look and excess edge enhancement as undesirable artifacts of a low-res image that has been overprocessed, and appreciate the near-3D look of an inherently high-resolution image, whether it's a 4x5 Tri-X negative, or a 25 megapixel RAW file with 12-bit depth per color.
The same applies to audio. Recording studio professionals consider it essential that a monitor speaker/amplifier system reliably distinguish between 16, 20, and 24-bit masters, differentiate between different ADC/DAC conversion systems, or different flavors of microphone preamps. A monitor/amplifier that does not discriminate between different quality grades of digital recording is going to have translation issues when the final recording is made. If a monitor/amplifier doesn't let you hear the artifacts of lossy compression, that's a major issue on the recording end.
Audiophiles and reviewers who are used to Red Book as their quality standard are not going to be well-attuned to the subtler qualities of a higher-resolution system, and may seek out electronics and loudspeakers that conceal the limitations of 44.1/16 PCM. My guess is that most of the so-called "advancements" of the Eighties and Nineties in the audiophile world were nothing more than elaborate tone controls to subjectively minimize Red Book shortcomings, aided and abetted by the magazines and retailers shifting their focus to high-margin (90%) audiophile jewelry like cables, dots, suspension gimmicks, et al. A good rule of thumb reading any audio review is to figure out the retail profit margin of the product being touted - it can range from 40% (on the low side these days) to more than 90% (common for the gimmicks).
But I would like to suggest that "Resolution" is still useful, although grossly abused by mainstream audiophiles. Speakers are not the same in this respect, with home-theater-in-a-box at the low end, and the highest rank of professional monitors at the top (although many pro monitors have surprising levels of coloration). Audiophile products, of course, are all over the place, with an awful lot of tizz-and-boom dominating the market, right up to silly price levels of $150,000 per pair of speakers. When you spend that kind of money on a car, at least you get a good car. With audio, there's no assurance of that.
Lynn Olson said:
Hi Lynn,
In Your original Post #1 You mentioned large scale dipoles. Isn't it so that dipoles tend to emphasize HD and intermodulation due to their working principle? I'm not shure what You experienced when listening to Bastianis or Linkwitz' implementations. From both I never saw data on the item of non linear distortion. Not even S.Linkwitz shows measurements from the finished gear as a whole - as far as I know. The preliminary investigations are of little use. To deploy the drivers in a dipole makes the worse of them. I'm curious what comes out in the end then.
Resolution is associated with low intermodulation to some extend. Is there any trick in sight to overcome the contadiction between "dipole" and "resolution"? Besides sheer size of course.
(slightly back on topic, sorry ;-)
so long
My biggest reservation about any dipole is the EQ boosting many of them require. The dipole portion of the Bastanis is not equalized - and I know, since I was in the room when Gary Pimm connected his amplifier directly to the speaker and the source. The EQ, somewhat counterintuitively, is applied to the self-powered closed-box subwoofer, which runs all the way up to about 220 Hz (as measured by Gary).
The Orion is another matter entirely. No EQ per se for the tweeter, which sees a conventional half-space load, but plenty of EQ for the mids, and most of all, the pair of push-pull woofers. I was expecting lots of distortion from the woofers, considering their moderate area, low efficiency, and massive equalization, but it didn't jump out at me as all that noticeable. And I turned it up louder than most of the people in the room wanted me to, including Siegfried.
Exhibitors are often shocked how loud I listen at trade shows, but all I'm trying to do is get the low-level details above the noise floor of the show - you can't tell much about a speaker if everything below 50~60 dB SPL is lost in rumble from other rooms and crowd noise from the hallways. Typical show demo conditions limit the dynamic range to somewhere between 55 and 95 dB SPL (or less), which is barely AM radio or non-Dolby Compact Cassette fidelity.
The tizz I was referring to in the previous post wasn't so much about distortion, but upper-mid breakup in the midwoofer or midrange driver, and/or ragged HF response at or above 12~15 kHz. Many audiophiles mistake these colorations as "detail" or "resolution". But then they never listen to pink-noise and hear the colorations as the colorations they really are.
On moderately loud sparse-spectrum little-girl-with-guitar or carefully-selected audiophile-flavor blues and jazz (Diana Krall, Jazz at the Pawnshop, et al), a speaker with rough response in the 3~8 kHz region gives an illusion of detail. This illusion completely falls apart on wide-dynamic-range choral or symphonic music, which is why I've been kicked out of demo rooms at trade shows more than once. Kind of spoils the sales atmosphere when a $60,000 $tereophile top-reviewed speaker sounds like tin cans falling down a stairway.
The Orion is another matter entirely. No EQ per se for the tweeter, which sees a conventional half-space load, but plenty of EQ for the mids, and most of all, the pair of push-pull woofers. I was expecting lots of distortion from the woofers, considering their moderate area, low efficiency, and massive equalization, but it didn't jump out at me as all that noticeable. And I turned it up louder than most of the people in the room wanted me to, including Siegfried.
Exhibitors are often shocked how loud I listen at trade shows, but all I'm trying to do is get the low-level details above the noise floor of the show - you can't tell much about a speaker if everything below 50~60 dB SPL is lost in rumble from other rooms and crowd noise from the hallways. Typical show demo conditions limit the dynamic range to somewhere between 55 and 95 dB SPL (or less), which is barely AM radio or non-Dolby Compact Cassette fidelity.
The tizz I was referring to in the previous post wasn't so much about distortion, but upper-mid breakup in the midwoofer or midrange driver, and/or ragged HF response at or above 12~15 kHz. Many audiophiles mistake these colorations as "detail" or "resolution". But then they never listen to pink-noise and hear the colorations as the colorations they really are.
On moderately loud sparse-spectrum little-girl-with-guitar or carefully-selected audiophile-flavor blues and jazz (Diana Krall, Jazz at the Pawnshop, et al), a speaker with rough response in the 3~8 kHz region gives an illusion of detail. This illusion completely falls apart on wide-dynamic-range choral or symphonic music, which is why I've been kicked out of demo rooms at trade shows more than once. Kind of spoils the sales atmosphere when a $60,000 $tereophile top-reviewed speaker sounds like tin cans falling down a stairway.
None. That's why I (very slowly) design my own. Although it might sound like it, it isn't arrogance - when I designed the Ariel, Amity, and Karna, there was nothing remotely close to them on the market. (Well, you could buy HT speakers that looked the same, but they sounded pretty awful.)
If I completely lost in interest in the project and also came into a bunch of money (or the damn stock market finally recovers to pre-Millenium levels), I'd probably seriously consider the AudioKinesis speakers that Duke makes, or Jean Hiraga's 604's. Those line up with my current tastes.
If I completely lost in interest in the project and also came into a bunch of money (or the damn stock market finally recovers to pre-Millenium levels), I'd probably seriously consider the AudioKinesis speakers that Duke makes, or Jean Hiraga's 604's. Those line up with my current tastes.
Lynn Olson said:
The equalization boosts the cone movement and too supraharmonic distortions are emphasized by the inherent amplitude vs frequency characteristic. With dipoles (and cadioids too) all evils are squared as to say.
BTW a "jag" is a FORD nowadays.
I still wonder why I haven't seen any farfeald measurement of a dipole with respect to distortion. Mine - a cardioid - was horrible. 10s of % of 3rd HD when at the same level a closed box, same driver did o/k. And it was not off limits regarding excursion. Perhaps Mr. Geddes is right with his assumption that HD isn't so much of a hassle. Your subjective evaluation might be conformal to the objectivists expectation.
Did You ever try Klein&Hummel monitors? They should be as neutral as can be if used along their specification. Some top models allow time domain optimization with a switch!
so long
Rudolf said:
Any interest in the distortion measurements of HH and K+T for some of their dipoles? It´s K2, K3 and K5 at 90 dB. Not sure about the distance, but I think 1 m.
I´m afraid I could not find one with "10s of % of 3rd HD".
Hi Rudolf,
O/k. The best You could do is measure it Yourself. That rise in distortion is inevitely a consequence of destructive interference, which latter gives the well known -6dB/oct tilt towards lower frequency.
If You can't measure You should stay away from magazine xperts even more. 1 meter = 3 feet is no way a farfield distance with a bass dipole. Farfield starts at about 1/2 wavelength minimum (70Hz => 7feet / 2.5meters). For the sake of completeness, that is the reason bass dipoles won't work at all in living rooms. For sure not as "dipoles".
You knew that all before ...
panomaniac said:
?????
"Dipole" is by definition some special kind of directivity, isn't it? By mathematics this special behavior is obtained in the farfield and NOT in the nearfield of the device. Far- vs. nearfield depends on the wavelength of interest.
A common living room (in Europe) is not bigger than 6 x 7 meters^2 = 20 x 25 feet^2. If a dipole resides in the middle of the room then the distance to the walls is about 3 meters = 10 feet. That is 1/2 wavelength of 55Hz. If the listener sits direct at the wall then some "dipole" behavior would be apparent. More close to the dipole its constituting 2 source have to be encountered each for themselves.
Additionally above that frequency "dipole" directivity gradually comes to some effect. But below that the dipole more or less works as a quadrupole or the like - totally unknown, hence uncontrolled by the most critical listener. That quadrupole or the like builds up by virtual image sources from the walls.
This is no "meaning" of mine but very simple maths. Take it for granted. It's all in the "theory" of dipoles that many DIYer didn't read to thoroughly. Each concept has it's limitations. With dipole one of them is distance.
This is a caveat against making dipoles bigger to overcome the limitations from emphasized non linear distortions. The bigger the dipole the bigger the room has to be and hence the more dBs one needs to fill that room soundwise.
so long
xpert said:
Nothing new in that. As most people around here I know how a dipole works.
Anything wrong in trusting the magazine experts more than a forum alias "xpert"?
My OBs are never wider than 0.5 m (physical size). So I can´t speak for wider baffles. When I measured my OBs or H-frames at 1 m I quite consistently got about 4db/oct loss. Full 6dB loss at 2.5 m. Thats my usual listening distance. So I am confident that 1 m HD measurements are not THAT far off:
This is the CIARE CH250 in half space - k2, k3, k5 at 90 dB
An externally hosted image should be here but it was not working when we last tested it.
And this in the Dacapo! open baffle
An externally hosted image should be here but it was not working when we last tested it.
BTW: The Dacapo! is designed to be crossed over to a dipole sub at 200 Hz.
Bass dipoles work as much as "dipoles" in a room as bass monopoles work as "monopoles". And we all are well aware that dipoles don´t work below the lowest room modes. So nothing really new about that too. You are truely not the only "xpert" in this forum.
Rudolf said:
As far as You can't evaluate Your system by yourself I expect propagated rumors, not more. We can't discuss a magazines measurement as we dont know how it was performed. As the magazine didn't care about near- vs farfield I don't even look at it. Under the aspect of generating knowledge those measurements ain't worth the paper they are printed on. They have to look good to satisfy the customer (reader), no complications please!
I made own measurements, I dropped dipole.
so long
"xpert": you spectacularly make no sense at all. To me. But I'll give you that - you sure have self-confidence.
Since other people showed data which you criticized. Can you now share, to support your points with evidence, your own measured data on HD and IMD vs. measurement distance (please, especially that one) vs. SPL for same driver in different setups, say dipole vs. cardioid vs. closed box, vs. excursion, vs. radiation patterns, with detailed description of measurement apparatus, software and competence in use thereof, since you critically examined all of the above in your last 2, 3 posts and found existing published data wanting?
Since other people showed data which you criticized. Can you now share, to support your points with evidence, your own measured data on HD and IMD vs. measurement distance (please, especially that one) vs. SPL for same driver in different setups, say dipole vs. cardioid vs. closed box, vs. excursion, vs. radiation patterns, with detailed description of measurement apparatus, software and competence in use thereof, since you critically examined all of the above in your last 2, 3 posts and found existing published data wanting?
Lynn Olson said:
Hi,
I just reactivated my pro sound gear. Some kind of Geddes approach. Two way bass/mid plus serious waveguide/horn with digital corssover/controller. Tremendous. Schonberg choral works may sound harsh with some notes, Xenakis may be sharp sometimes, Bjoerk may over do her eagerness with to much pressure refrain for refrain, but kinky intermezzos. Rolling Stones sound like cream on Let It Bleed, lots of sound alike blues. Provocing pseudo "black" background voices. I am convinced I can trust the system. It is most probably correct anytime. What better can be said about a stereo set?! It helps to check out music for itself, enlightment somehow.
so long
MBK said:
Do it yourself! Don't trust phoney commercials on what a deviated pragma does better than common stuff. Check it out and stop speculating. You are in question for the sake of Your own satisfacton. What a difference does it make what I did with my dipoles/carioids? Do it yourself!
ciao