My initial plan for this post was to continue a more or less chronological account of my long journey to develop the S1 loudspeaker. Upon reconsideration though, it's sufficient to just say that nothing earth-shattering happened there. The cabinet was much more effectively braced. It was better lined. It was to some degree stuffed, at the mid-cabinet. Brass risers were added to isolate it from the base and therefore the floor. The port tuning was further adjusted and modified for somewhat of an aperiodic character. The connections, the wiring, and the baffle step correction components were all significantly elevated in quality and execution. And that's kind of it, for that chapter, in a nutshell. The S1 is very refined, and very labor intensive, but it's simple, really. Every unnecessary complexity has been trimmed away from it, and that's just what makes it great, I believe.
What I would like to discuss is essentially the dividing line between the engineering of a loudspeaker, and the art of transforming it into one that connects with listeners. In my experience, voicing is the finishing step that can elevate a loudspeaker beyond just empty observations in the audiophile vernacular. When done sufficiently well, it enables a speaker to deliver not what we often think we want from music reproduction, but rather what we may not recognize we actually want from it.
When I watch a movie, let's say "Saving Private Ryan", what is the experience I'm trying to have? Well, I'm trying to enjoy the telling of a story, of course - that is most important, I believe. I might be looking forward to the stunning image quality on a new TV, as well, or bombastic sound and visual effects. I'm carrying expectations into that experience, and they are quite specific to movies, and likely nice televisions, as media. It may be that the portrayal looks and sounds very real to me, and that may often compliment the pleasure of viewing it, but that isn't at the center of the experience. The core experience is perfect voyeurism into the lives of other human beings, and in that catharsis not to have the slightest need for feeling self-conscious while doing so. Movies are a discrete artform unto themselves, which conforms to unintuitive ideas and measures of excellence unlike the portrait of reality that they contain, and what’s more, with the unapologetic goal of improving upon that reality. If I were watching a live performance of that same story, let's say as a play, some of my expectations would overlap but many would not, simply because the same content expressed through another medium carries its own set of expectations.
I believe that the hi-fi listening experience is much if not exactly the same. Am I trying to feel as though I am literally present at a live event in my listening room? In many cases that's clearly a piece of it, and it's very tempting here to say yes, but I don't think that's really the case. My system certainly may sound incredibly real to me, and that feature may often improve upon music, but it's only a feature of the experience. At bottom, I'm really looking for the same kind of catharsis, and the way to most fully have it is not necessarily through a starkly transparent lens of “hereness” or “thereness”, but rather to experience a view of it so artfully rendered that one can't help but be fully overcome by it. Some of the beauty of that portrayal may be owed, in fact, to its not being a neutral depiction of what it was, if witnessed live. Live music performances seem to be something apart from hi-fi in many ways, with the core experience much about sharing a love of craft and great works, with the artists themselves, and with each other.
What live and recorded media do not seem to share is much of our internal evaluation of them as we experience each. At a concert, you can bet I'm not carefully introspecting the sound of the guitarist's fingers plucking the strings. I'm not offended if the brass instruments are a little glaring occasionally, or if the bass response is slightly bloated in the venue. Thinness, boominess, weak transients, incomplete treble extension, lack of imaging and an endless list of other negative qualia are forgiven of what might be a wonderful live concert. On the other hand, any of these deficits are certain to be all but intolerable in my listening room. It's as if we have more than one self that experiences the exact same music in each setting. The experiencing self, as Daniel Kahnemann might say, is the self that appreciates the live performance. However, it is the remembering self, or some other facet of self, that experiences recordings in the hi-fi setting.
The potential takeaway is that a perfectly ruler flat on-axis response curve may not produce the best listening experience for hi-fi. That's an important realization for any loudspeaker maker. One can't help but think of the Fletcher-Munson curve in relation to this topic, and while I think our sensitivity to frequencies does align with good sound, it doesn't fully track with an ideal presentation either. Ear sensitivity alone doesn't consider the amount of content in music at any given frequency, or the value we assign to disparate elements of music. The frequency band that covers the human voice, and the storytelling at the heart of music, is paramount. It must be effortless to enjoy, regardless of how sensitive my ears are to that range. Tones at the frequency extremes are less subjectively valuable. Generally speaking we don't listen to music that consists of nothing but content between 8kHz and 20kHz, or music that consists of only content below, let's say, 80Hz.
So, it's a lot of weighting to take into consideration. Hyper-sensitivity of our ears to some frequency ranges is one factor. The subjective value we place on some parts of music, such as the human voice, is another factor. Yet another is the amount of content in music at any given frequency. I'd add that age-related hearing loss is a factor, too.
How does it all wash out, then? Well, I think we can agree that nobody prefers to listen to hi-fi one meter away, head on, from a single channel. That measurement is important to me as a reference, but I don't really begin to judge a design outcome unless I'm pulling measurements and subjectively assessing in a typical listening position. To be clear, measurements are essential for sound decision making at every phase of development. I've taken literally thousands over the years, and my work would be greatly impoverished for it had I not. Each of those measurements is a fixed point in space, and it is only in full knowledge of those fixed points in space that one can coherently navigate, in any direction.
In an on-axis measurement from about four feet in the middle of the room, I look for as much as a 6dB gain between about 60-30 Hz. When not overdone, in most environments it will sound like a full and yet balanced bass response from the listening chair, at least for me. I likewise look for a little dip afterward, between about 70-220 Hz. In this region, walls and ceilings in residential construction tend to resonate, and that excess of energy often makes bass sound muddy, boomy and slow. The region between 300 Hz and 900 Hz is my baseline, and is flat. The region between 1 kHz and 7 kHz is critically important, in my experience. I look for a modest dip in response here; perhaps slightly more than modest between 1kHz and 3kHz. If a speaker sounds fatiguing I find that it is very often due to an excess of energy, there. From 8 kHz and above, I want a fairly significant peak in response. None of us likes to think we suffer age related hearing loss, however using myself as proof in that regard, I find that a peak from 8 kHz through as much of the last octave as possible rewards me with sparkle, air and presence.
It took years to define and then achieve the results that I wanted. The effort has involved exhaustive testing of cone materials, weights, and shapes, driver suspension refinement, cabinet construction, lining, and port tuning, and an unconventional approach to baffle step compensation. Of course, it's been worth it. The validation over the past year and a half has been amazing, and humbling.
I'm left conflicted on an important point. Does Songer Audio want reviews inclusive of performance measurements? To be clear, I have nothing but deep respect for the reviewers and media that take the time to measure performance, and share the results. What gives me pause, nonetheless, is the presumed standard of merit by which the data are judged. That benchmark is a spatially averaged ruler-flat response from three or four feet, and any deviation from it is reliably an object of commentary derision, often as reliably from individuals with no first-hand experience of the loudspeaker. Do I want to either endure or litigate any of that? No, I don't suppose that I do.
Perhaps the standard of measure for loudspeaker performance may be overdue for reconsideration, with different goals in mind and a fresh vernacular to describe them. I can't speak for others, but I'd rather watch The Godfather on VHS from an outdated TV than I would watch one of the Transformer movies on the largest and finest, state of the art television money can buy. Again, that isn't to say that The Godfather won't benefit from the latter, it surely will. I'm just offering a reminder that the purpose of the project is not data or intellectual satisfaction, but rather to feed our innate desire for beauty and catharsis. We will surely benefit to judge excellence in hi-fi based upon how successful each effort is toward that end, and not how successful it is in delivering on metrics that have no direct relationship with our deepest human needs.
Alright, dear reader, I'll close there for now. I think we're going to have a pretty exciting announcement to make in the next installment... stay tuned.
What I would like to discuss is essentially the dividing line between the engineering of a loudspeaker, and the art of transforming it into one that connects with listeners. In my experience, voicing is the finishing step that can elevate a loudspeaker beyond just empty observations in the audiophile vernacular. When done sufficiently well, it enables a speaker to deliver not what we often think we want from music reproduction, but rather what we may not recognize we actually want from it.
When I watch a movie, let's say "Saving Private Ryan", what is the experience I'm trying to have? Well, I'm trying to enjoy the telling of a story, of course - that is most important, I believe. I might be looking forward to the stunning image quality on a new TV, as well, or bombastic sound and visual effects. I'm carrying expectations into that experience, and they are quite specific to movies, and likely nice televisions, as media. It may be that the portrayal looks and sounds very real to me, and that may often compliment the pleasure of viewing it, but that isn't at the center of the experience. The core experience is perfect voyeurism into the lives of other human beings, and in that catharsis not to have the slightest need for feeling self-conscious while doing so. Movies are a discrete artform unto themselves, which conforms to unintuitive ideas and measures of excellence unlike the portrait of reality that they contain, and what’s more, with the unapologetic goal of improving upon that reality. If I were watching a live performance of that same story, let's say as a play, some of my expectations would overlap but many would not, simply because the same content expressed through another medium carries its own set of expectations.
I believe that the hi-fi listening experience is much if not exactly the same. Am I trying to feel as though I am literally present at a live event in my listening room? In many cases that's clearly a piece of it, and it's very tempting here to say yes, but I don't think that's really the case. My system certainly may sound incredibly real to me, and that feature may often improve upon music, but it's only a feature of the experience. At bottom, I'm really looking for the same kind of catharsis, and the way to most fully have it is not necessarily through a starkly transparent lens of “hereness” or “thereness”, but rather to experience a view of it so artfully rendered that one can't help but be fully overcome by it. Some of the beauty of that portrayal may be owed, in fact, to its not being a neutral depiction of what it was, if witnessed live. Live music performances seem to be something apart from hi-fi in many ways, with the core experience much about sharing a love of craft and great works, with the artists themselves, and with each other.
What live and recorded media do not seem to share is much of our internal evaluation of them as we experience each. At a concert, you can bet I'm not carefully introspecting the sound of the guitarist's fingers plucking the strings. I'm not offended if the brass instruments are a little glaring occasionally, or if the bass response is slightly bloated in the venue. Thinness, boominess, weak transients, incomplete treble extension, lack of imaging and an endless list of other negative qualia are forgiven of what might be a wonderful live concert. On the other hand, any of these deficits are certain to be all but intolerable in my listening room. It's as if we have more than one self that experiences the exact same music in each setting. The experiencing self, as Daniel Kahnemann might say, is the self that appreciates the live performance. However, it is the remembering self, or some other facet of self, that experiences recordings in the hi-fi setting.
The potential takeaway is that a perfectly ruler flat on-axis response curve may not produce the best listening experience for hi-fi. That's an important realization for any loudspeaker maker. One can't help but think of the Fletcher-Munson curve in relation to this topic, and while I think our sensitivity to frequencies does align with good sound, it doesn't fully track with an ideal presentation either. Ear sensitivity alone doesn't consider the amount of content in music at any given frequency, or the value we assign to disparate elements of music. The frequency band that covers the human voice, and the storytelling at the heart of music, is paramount. It must be effortless to enjoy, regardless of how sensitive my ears are to that range. Tones at the frequency extremes are less subjectively valuable. Generally speaking we don't listen to music that consists of nothing but content between 8kHz and 20kHz, or music that consists of only content below, let's say, 80Hz.
So, it's a lot of weighting to take into consideration. Hyper-sensitivity of our ears to some frequency ranges is one factor. The subjective value we place on some parts of music, such as the human voice, is another factor. Yet another is the amount of content in music at any given frequency. I'd add that age-related hearing loss is a factor, too.
How does it all wash out, then? Well, I think we can agree that nobody prefers to listen to hi-fi one meter away, head on, from a single channel. That measurement is important to me as a reference, but I don't really begin to judge a design outcome unless I'm pulling measurements and subjectively assessing in a typical listening position. To be clear, measurements are essential for sound decision making at every phase of development. I've taken literally thousands over the years, and my work would be greatly impoverished for it had I not. Each of those measurements is a fixed point in space, and it is only in full knowledge of those fixed points in space that one can coherently navigate, in any direction.
In an on-axis measurement from about four feet in the middle of the room, I look for as much as a 6dB gain between about 60-30 Hz. When not overdone, in most environments it will sound like a full and yet balanced bass response from the listening chair, at least for me. I likewise look for a little dip afterward, between about 70-220 Hz. In this region, walls and ceilings in residential construction tend to resonate, and that excess of energy often makes bass sound muddy, boomy and slow. The region between 300 Hz and 900 Hz is my baseline, and is flat. The region between 1 kHz and 7 kHz is critically important, in my experience. I look for a modest dip in response here; perhaps slightly more than modest between 1kHz and 3kHz. If a speaker sounds fatiguing I find that it is very often due to an excess of energy, there. From 8 kHz and above, I want a fairly significant peak in response. None of us likes to think we suffer age related hearing loss, however using myself as proof in that regard, I find that a peak from 8 kHz through as much of the last octave as possible rewards me with sparkle, air and presence.
It took years to define and then achieve the results that I wanted. The effort has involved exhaustive testing of cone materials, weights, and shapes, driver suspension refinement, cabinet construction, lining, and port tuning, and an unconventional approach to baffle step compensation. Of course, it's been worth it. The validation over the past year and a half has been amazing, and humbling.
I'm left conflicted on an important point. Does Songer Audio want reviews inclusive of performance measurements? To be clear, I have nothing but deep respect for the reviewers and media that take the time to measure performance, and share the results. What gives me pause, nonetheless, is the presumed standard of merit by which the data are judged. That benchmark is a spatially averaged ruler-flat response from three or four feet, and any deviation from it is reliably an object of commentary derision, often as reliably from individuals with no first-hand experience of the loudspeaker. Do I want to either endure or litigate any of that? No, I don't suppose that I do.
Perhaps the standard of measure for loudspeaker performance may be overdue for reconsideration, with different goals in mind and a fresh vernacular to describe them. I can't speak for others, but I'd rather watch The Godfather on VHS from an outdated TV than I would watch one of the Transformer movies on the largest and finest, state of the art television money can buy. Again, that isn't to say that The Godfather won't benefit from the latter, it surely will. I'm just offering a reminder that the purpose of the project is not data or intellectual satisfaction, but rather to feed our innate desire for beauty and catharsis. We will surely benefit to judge excellence in hi-fi based upon how successful each effort is toward that end, and not how successful it is in delivering on metrics that have no direct relationship with our deepest human needs.
Alright, dear reader, I'll close there for now. I think we're going to have a pretty exciting announcement to make in the next installment... stay tuned.
Last edited:
