I agree we have.
My only point was that I was agreeing with Lynn that 1/4 inch is audible.
Hi Kindhornman
Sorry to have intruded in your thread but since you guys were talking about something (where the sound goes) which is vitally important in large scale sound and I had thought folks might have been interested in what governs that. Yes in hifi land, a horn is not a flat baffle but more because of the names and mindset BUT NOT because there is a difference in the way the sound propagates from a small source.
For what I do, I have found that latter view to be a useful way to approach acoustic problems.
So far as Don’s contributions to Audio or perhaps what important contributions are, maybe we see things a little differently. If you choose to ignore out of hand a usable thumb rule because some past writing didn’t sit well, that is one way to approach it..
For me, that paper was a turning point in horn design and still today, to avoid the narrowing at pattern loss, the last part of the horn has to have a wider horn wall angle as he identified.
For me after a loudspeaker life confined to below 100Hz and work life in acoustic levitation above 20KHz, In 1999 It was sort of a realization about why that kind of horn had poor lf loading that lead to the full range Unity horn and now the Synergy horns our business is based on.
Tom
Sorry to have intruded in your thread but since you guys were talking about something (where the sound goes) which is vitally important in large scale sound and I had thought folks might have been interested in what governs that. Yes in hifi land, a horn is not a flat baffle but more because of the names and mindset BUT NOT because there is a difference in the way the sound propagates from a small source.
For what I do, I have found that latter view to be a useful way to approach acoustic problems.
So far as Don’s contributions to Audio or perhaps what important contributions are, maybe we see things a little differently. If you choose to ignore out of hand a usable thumb rule because some past writing didn’t sit well, that is one way to approach it..
For me, that paper was a turning point in horn design and still today, to avoid the narrowing at pattern loss, the last part of the horn has to have a wider horn wall angle as he identified.
For me after a loudspeaker life confined to below 100Hz and work life in acoustic levitation above 20KHz, In 1999 It was sort of a realization about why that kind of horn had poor lf loading that lead to the full range Unity horn and now the Synergy horns our business is based on.
Tom
Tom,
I wasn't knocking Keele, Not really, I just haven't read anything from him since some very early PA sound book years ago that did have some earlier misinformation. I am not disregarding what he has said subsequently since then, I just literally haven't seen or read anything by him since then. I didn't ban his work or have a grudge, just haven't read more current papers he may have written.
I am not discounting your approach to looking at a flat baffle as a horn either, though it is not the common practice or definition that most would relate between flat baffle and horn. I have a few ideas about some of what goes on and some of that I just keep to myself, not looking for any arguments here, I just think sometimes some issues are overlooked and other effects attributed to a phenomena.
I wasn't knocking Keele, Not really, I just haven't read anything from him since some very early PA sound book years ago that did have some earlier misinformation. I am not disregarding what he has said subsequently since then, I just literally haven't seen or read anything by him since then. I didn't ban his work or have a grudge, just haven't read more current papers he may have written.
I am not discounting your approach to looking at a flat baffle as a horn either, though it is not the common practice or definition that most would relate between flat baffle and horn. I have a few ideas about some of what goes on and some of that I just keep to myself, not looking for any arguments here, I just think sometimes some issues are overlooked and other effects attributed to a phenomena.
Hi Pooh
I have had several of our products from work in my listening system room while they were in development. I even had a J-2 for a while (a large full range horn) while I was working out an internal bug, but being 5 feet tall horn, it took up too much space to have 2. Most of the time I have had a pair of old SH-50’s for above 70Hz and a pair of TH-50 prototypes for low frequency.
Before these, I had the older Unity type horns one of which you heard.
Both radiate as if they were one source (no lobes or nulls) but the Synergy horn eliminates the remaining phase shift normal crossovers impose, they radiate as if they had one very wide band driver in time and space.
Best,
Tom
I have had several of our products from work in my listening system room while they were in development. I even had a J-2 for a while (a large full range horn) while I was working out an internal bug, but being 5 feet tall horn, it took up too much space to have 2. Most of the time I have had a pair of old SH-50’s for above 70Hz and a pair of TH-50 prototypes for low frequency.
Before these, I had the older Unity type horns one of which you heard.
Both radiate as if they were one source (no lobes or nulls) but the Synergy horn eliminates the remaining phase shift normal crossovers impose, they radiate as if they had one very wide band driver in time and space.
Best,
Tom
I think somewhat similarly.. though the last time I mentioned it was in relation to terminology, specifically:
A waveguide is guide for waves - anything that limits directivity is essentially a waveguide. This includes the baffle, and depending on freq. - even a driver's diaphragm.
A horn on the other hand is a subset of a waveguide, more specifically a waveguide that limits dispersion expressly for gain. ie. Horns are all about gain, not dispersion control though they necessarily have that as well.
Some waveguides utilize absorption and are not horns (exhibiting no gain as a result of the dispersion control). Other waveguides exhibit gain only incidentally (often not utilizing it, or "padding" it down), and really aren't "horns" either.
interestingly enough there is a philips motional feedback system on ebay that closes tonight. i just happened to find it when looking at another auction.
Philips RH541 Motion Feedback MFB Powered Speakers Superb | eBay
Good question. It's very, very good, certainly the best I've ever heard from a high-efficiency system. The image, or rather, impression of acoustic space, is quite different than conical-type horns, which seem to beam the image right at you. The combination of the large-format compression driver and the AH425 result in a very spacious sound that comes into the room ... I heard it stop short about 2~3 feet from my knees, and go behind the speakers about 30 to 50 feet. Very deep image, the deepest I've ever heard from any horn system, although there is definite kinship with the big Altec multicells.
I should mention that when I speak of "image", I'm not using the word in the Stereophile or Absolute Sound meaning. For me, the "image" is the entire impression of acoustical space, with or without instruments playing. On some recordings, there's a moment of silence before the orchestra starts, and you can hear the space suddenly fill the room. That's the "image" I'm speaking about, which is quite different than the studio decision where to pan-pot the instruments and singers.
Comparing the Ariel and the new speaker is difficult, since the listening room and electronics were different. Nevetheless, there were impressions that stood out. As mentioned before, the image is very deep, but also very precise, something that's new to me. You can easily hear the musicians leaning back and forth as they play, a rather startling impression, and probably due to the very quick time-decay response of the system.
Extra-width? Hmm, hard to say. There's plenty of extra-width impression, but that strongly depended on the playback electronics and what's on the recording. Height? Yes, plenty of that, even though I'm aware that's purely an artifact of two-speaker stereo replay and funny things going on around 8~9 kHz (pinna artifacts).
There's a lot of Ariel character in the new loudspeaker, but there's Altec bass in there too, which is not something I've heard before. I sense the Ariel is a more forgiving loudspeaker ... the new speaker has resolution in the Stax electrostat range, but it's a loudspeaker, so you get loudspeaker/room imaging and bass quality.
But then again, although the Ariels are forgiving, they sound pretty awful on poor-quality amplifier. Only one or two transistor amps are remotely suitable, and are not really a good choice. PP pentode amps with a lot of solid-state gimmicks sound pretty harsh and grainy, but the better-designed PP pentode amps are pretty listenable. But the Ariels are really all about DHT amplifiers; that's when they come alive and sound they way should.
I haven't heard the new speaker on transistor amps, but I would expect the sound to be really gruesome. It's like the amplifier defects are on display on a table, sitting there right in front of you, while the music floats serenely above it, ignoring the defects.
If the recording engineer is being cute, you can actually hear the singer go back and forth between triple-tracking and double-tracking in the middle of a song; weirdly enough, it sounds just like three people all standing in row, right behind each other, with the rearmost singer disappearing and re-appearing as they sing.
From what I heard, the new speaker is a combination of the Ariel, the bigger Altecs, and Stax phones, with an image quality all its own. Dunno if this description helps or just confuses.
P.S. It's not a sound I think transistor-amp listeners would necessarily enjoy, where you keep your distance, sit back and analyze things. DHT enthusiasts are into emotional engagement, and this speaker, like the Ariel, is optimized for that, with intensely vivid tone colors, and a very strong in-the-room impression. Pianos, for example, sound like they're about three feet from you (not surprising, since they're usually close-miked). By contrast, the Ariels would put the piano slightly behind the plane of the speakers.
I should mention that when I speak of "image", I'm not using the word in the Stereophile or Absolute Sound meaning. For me, the "image" is the entire impression of acoustical space, with or without instruments playing. On some recordings, there's a moment of silence before the orchestra starts, and you can hear the space suddenly fill the room. That's the "image" I'm speaking about, which is quite different than the studio decision where to pan-pot the instruments and singers.
Comparing the Ariel and the new speaker is difficult, since the listening room and electronics were different. Nevetheless, there were impressions that stood out. As mentioned before, the image is very deep, but also very precise, something that's new to me. You can easily hear the musicians leaning back and forth as they play, a rather startling impression, and probably due to the very quick time-decay response of the system.
Extra-width? Hmm, hard to say. There's plenty of extra-width impression, but that strongly depended on the playback electronics and what's on the recording. Height? Yes, plenty of that, even though I'm aware that's purely an artifact of two-speaker stereo replay and funny things going on around 8~9 kHz (pinna artifacts).
There's a lot of Ariel character in the new loudspeaker, but there's Altec bass in there too, which is not something I've heard before. I sense the Ariel is a more forgiving loudspeaker ... the new speaker has resolution in the Stax electrostat range, but it's a loudspeaker, so you get loudspeaker/room imaging and bass quality.
But then again, although the Ariels are forgiving, they sound pretty awful on poor-quality amplifier. Only one or two transistor amps are remotely suitable, and are not really a good choice. PP pentode amps with a lot of solid-state gimmicks sound pretty harsh and grainy, but the better-designed PP pentode amps are pretty listenable. But the Ariels are really all about DHT amplifiers; that's when they come alive and sound they way should.
I haven't heard the new speaker on transistor amps, but I would expect the sound to be really gruesome. It's like the amplifier defects are on display on a table, sitting there right in front of you, while the music floats serenely above it, ignoring the defects.
If the recording engineer is being cute, you can actually hear the singer go back and forth between triple-tracking and double-tracking in the middle of a song; weirdly enough, it sounds just like three people all standing in row, right behind each other, with the rearmost singer disappearing and re-appearing as they sing.
From what I heard, the new speaker is a combination of the Ariel, the bigger Altecs, and Stax phones, with an image quality all its own. Dunno if this description helps or just confuses.
P.S. It's not a sound I think transistor-amp listeners would necessarily enjoy, where you keep your distance, sit back and analyze things. DHT enthusiasts are into emotional engagement, and this speaker, like the Ariel, is optimized for that, with intensely vivid tone colors, and a very strong in-the-room impression. Pianos, for example, sound like they're about three feet from you (not surprising, since they're usually close-miked). By contrast, the Ariels would put the piano slightly behind the plane of the speakers.
Last edited:
Sounds like a well written description to me. Not like most reviews you read at all in the hi fi world which is really refreshing..
How would you guess my Decware amp would do on such a system? It's the 6w SEP Taboo model. Sounds pretty nice on most everything so far (mostly Fostex fe108ez), and I imagine better at least than most SS amps would be for this design?
How would you guess my Decware amp would do on such a system? It's the 6w SEP Taboo model. Sounds pretty nice on most everything so far (mostly Fostex fe108ez), and I imagine better at least than most SS amps would be for this design?
Last edited:
Lynn, I think one of the solid state amplifiers you could tolerate was the one designed by the late Dan Banquer - the LNPA 150 by RE Designs. I'm wondering whether you have had the chance to audition the currently-much-heralded Hypex Ncore amplifiers available cheap to DIYers:
Hypex Electronics BV - NC400
designed by the esteemed Bruno Putzeys.
Also, are you aware of a new design speaker which seems to adhere to all the key design features of the Bastanis which you mentioned in post #1, and at a bargain price - the Trilogy T1 and 2 by Clayton Shaw: Speakers
On the surface of it, these would seem to be a gift to the non-DIYer.
Clayton, as you may know, has a track record as the designer of the Emerald Physics line of speakers.
Hypex Electronics BV - NC400
designed by the esteemed Bruno Putzeys.
Also, are you aware of a new design speaker which seems to adhere to all the key design features of the Bastanis which you mentioned in post #1, and at a bargain price - the Trilogy T1 and 2 by Clayton Shaw: Speakers
On the surface of it, these would seem to be a gift to the non-DIYer.
Clayton, as you may know, has a track record as the designer of the Emerald Physics line of speakers.
Last edited:
Lynn- when you say these new speakers are not suited for SS amps, - are you sort of designing for amps with higher output impedance and lower damping factor? Or is there something else that escapes me??
Pending the sale of our second house, I'm planning to do something new....
I've always wanted something horn-based, - at least in the mid/highs,,,,
any chance of a picture or two....?
Pending the sale of our second house, I'm planning to do something new....
I've always wanted something horn-based, - at least in the mid/highs,,,,
any chance of a picture or two....?
Hi buzzforb
I am not Lynn, but I suspect one of the reasons is the alnico magnet. But I will let Lynn answer this completely.
Deon
PS. Lynn, I would also like to know if you had a chance to hear the Hypex NC400 yet. If not, had you heard any of the other digital amps, and if so, what was your conclusion? Some of those small Tripath amps might go nice with the LTO, but might also not. My worry is that the LTO will highlight their flaws, as small as they may be.
Wow, what a lot of questions!
Altec vs TD15m: actually, a pretty close call. The biggest downside to the AE speaker was the really long wait-time for my pair ... something like a year, or more. You can get the GPA's with Alnico magnets right away. I also have some reservations about the sonics of ceramic magnets vs Alnico (or neodymium). AE uses an effective flux-shielding method for the pole-piece, but nothing is 100%, and some of the ceramic sound may still get through. Both drivers are really good, and different than high-power prosound drivers with heavier cones, which have a duller, grayer sound.
Not a big fan of Class D for full-range use. True, they are free of the grit-n-grain of transistor Class AB amplifiers, but they remind me of sigma-delta DACs ... note-perfect, but not as vivid and engaging as a good Class A design. In less diplomatic language, Class D sounds dry ... which to me, is not "accurate" at all, but an electronic artifact that you never hear with acoustic music in a real-world acoustic space.
What separates real acoustic music from typical high-end sound is the vividness, sweetness, and dynamic swings of the real thing, along with a completely natural and realistic spatial impression. This is why I focus on optimizing the intrinsic linearity, flat response, and freedom from diffraction of the physical transducers and amplifying elements of the playback system. This is a personal belief, but I feel once switching artifacts and other nonlinear behavior makes its way into the playback chain, it cannot be undone by clever algorithms, just moved around, so the problem is swept out of sight.
For amplifying elements, I look for active devices that have the lowest level of high-order harmonic generation, and loading techniques optimize linearity. For transducers, I look for flat response, underhung voice coils (so the field lines that cross the voice coil are straight) and low-diffraction cabinet construction and horn design.
Class D seems to go in the other direction. The signal is chopped up in time and amplitude, using a sigma-delta technique with dither and noise-shaping, and not reassembled into analog until it reaches the final MOSFETs.
The final reassembly process relies on complex, high-order feedback systems to assure the end result matches the input signal, and extra care needs to be taken so reactive (and nonlinear) loudspeaker loads do not compromise overall loop stability. There's an enormous amount of signal-processing going on with these amplifiers, for the relatively trivial task of 10~20x linear voltage amplification and 100~1000x linear current multiplication. It's only possible because silicon-based processing power (and algorithms) are cheaper than hand-assembled triodes and transformers. But is it better? I'm not so sure.
I expected sigma-delta converters to sound better than ladder/R2R converters. To me, they don't sound as good, despite better measurements. I then discovered that reviewers with Class AB transistor amps were writing the positive reviews about the sigma-delta DACs, while the triode listeners were coming to different conclusions. That led me to the two articles I recently wrote for Positive Feedback magazine:
Mountains and Fog, Part One
Mountains and Fog, Part Two
and sort of a joke article as a followup:
How to Make PCM sound like DSD
The two PFO articles stirred up the "What's Best Forum" around page 32:
DSD vs PCM, Page 32
There are a lot of good comments from real DAC designers, like Rene Jaeger (HDCD and Berkeley Audio Designs) and the designer of the Phasure DAC.
Thorsten Loesch has some excellent comments on the subject here:
Thermionic Valve Analogue Stages for Digital Audio; A Short Overview of the Subject, by Thorsten Loesch
As mentioned earlier, I really expected modern sigma-delta converters to sound better than my fairly dated Burr-Brown PCM-1704-based DAC. They didn't. The colorations I heard from the sigma-delta converters were completely new to me: a sort of flat, monotone quality to the music, that made the performers sound bored and not interested in what they were doing. This was really weird: I'd never heard analog electronics or loudspeakers do that particular trick before, which is what led to the research for the two articles.
Anyway, what I think is going on with sigma-delta converters and Class D amplifiers are new types of coloration we can't describe just yet.
Altec vs TD15m: actually, a pretty close call. The biggest downside to the AE speaker was the really long wait-time for my pair ... something like a year, or more. You can get the GPA's with Alnico magnets right away. I also have some reservations about the sonics of ceramic magnets vs Alnico (or neodymium). AE uses an effective flux-shielding method for the pole-piece, but nothing is 100%, and some of the ceramic sound may still get through. Both drivers are really good, and different than high-power prosound drivers with heavier cones, which have a duller, grayer sound.
Not a big fan of Class D for full-range use. True, they are free of the grit-n-grain of transistor Class AB amplifiers, but they remind me of sigma-delta DACs ... note-perfect, but not as vivid and engaging as a good Class A design. In less diplomatic language, Class D sounds dry ... which to me, is not "accurate" at all, but an electronic artifact that you never hear with acoustic music in a real-world acoustic space.
What separates real acoustic music from typical high-end sound is the vividness, sweetness, and dynamic swings of the real thing, along with a completely natural and realistic spatial impression. This is why I focus on optimizing the intrinsic linearity, flat response, and freedom from diffraction of the physical transducers and amplifying elements of the playback system. This is a personal belief, but I feel once switching artifacts and other nonlinear behavior makes its way into the playback chain, it cannot be undone by clever algorithms, just moved around, so the problem is swept out of sight.
For amplifying elements, I look for active devices that have the lowest level of high-order harmonic generation, and loading techniques optimize linearity. For transducers, I look for flat response, underhung voice coils (so the field lines that cross the voice coil are straight) and low-diffraction cabinet construction and horn design.
Class D seems to go in the other direction. The signal is chopped up in time and amplitude, using a sigma-delta technique with dither and noise-shaping, and not reassembled into analog until it reaches the final MOSFETs.
The final reassembly process relies on complex, high-order feedback systems to assure the end result matches the input signal, and extra care needs to be taken so reactive (and nonlinear) loudspeaker loads do not compromise overall loop stability. There's an enormous amount of signal-processing going on with these amplifiers, for the relatively trivial task of 10~20x linear voltage amplification and 100~1000x linear current multiplication. It's only possible because silicon-based processing power (and algorithms) are cheaper than hand-assembled triodes and transformers. But is it better? I'm not so sure.
I expected sigma-delta converters to sound better than ladder/R2R converters. To me, they don't sound as good, despite better measurements. I then discovered that reviewers with Class AB transistor amps were writing the positive reviews about the sigma-delta DACs, while the triode listeners were coming to different conclusions. That led me to the two articles I recently wrote for Positive Feedback magazine:
Mountains and Fog, Part One
Mountains and Fog, Part Two
and sort of a joke article as a followup:
How to Make PCM sound like DSD
The two PFO articles stirred up the "What's Best Forum" around page 32:
DSD vs PCM, Page 32
There are a lot of good comments from real DAC designers, like Rene Jaeger (HDCD and Berkeley Audio Designs) and the designer of the Phasure DAC.
Thorsten Loesch has some excellent comments on the subject here:
Thermionic Valve Analogue Stages for Digital Audio; A Short Overview of the Subject, by Thorsten Loesch
As mentioned earlier, I really expected modern sigma-delta converters to sound better than my fairly dated Burr-Brown PCM-1704-based DAC. They didn't. The colorations I heard from the sigma-delta converters were completely new to me: a sort of flat, monotone quality to the music, that made the performers sound bored and not interested in what they were doing. This was really weird: I'd never heard analog electronics or loudspeakers do that particular trick before, which is what led to the research for the two articles.
Anyway, what I think is going on with sigma-delta converters and Class D amplifiers are new types of coloration we can't describe just yet.
Last edited: