Without disgorging the entire critical vocabulary, it may be worth exploring several aspects of coloration that relate to one listener's perception of neutrality. Anyone who prefers a boomy cabinet may go at it. It would be particularly interesting to hear from designers of boomy cabinets.
Movement in the direction of neutrality implies flatter frequency response.
Neutrality implies CONTINUOUSNESS, in the sense of consistency of musical expression throughout the frequency range. A COLORIST may argue that continuousness demonstrates little more than seamlessness of coloration. To this I reply that coloration necessarily manifests itself discontinuously across the frequency range, and necessarily through a distribution of undesirable colorations in addition to desirable colorations. Eliminating an undesirable coloration is always progress toward neutrality. Even avowed colorists will express this preference. More on this further down. For the moment consider continuousness a virtue.
Relating to continuousness, movement toward neutrality implies a more organized presentation. The notion of ORGANIZATION is not far removed from Bryon’s notion of distinctness. Improved organization of sound is likely the consequence of small corrections to pitch and timbre, improved transients and decay against a quieter background-- that may result from reduced distortions and interstitial resonance peaks in frequency response as identified by Cbw723. However, in terms of how one hears a better organized presentation qua neutrality, the overall gestalt is that the system “settles down” and sounds more balanced and unforced. The example that Cbw723 cites of adding an aftermarket clock falls nicely into this category.
One aspect of an organized presentation is that dynamics are more precisely expressed through instrument bodies. Absent this natural sense of embodiment, dynamics tend to travel on their own envelop apart from instruments. This seeming dislocation of dynamics from instruments can be a bumpy & disorganized ride. In contrast, with NATURAL EMBODIMENT there is a sense of heightened control and containment of dynamics within the three dimensional boundaries of instruments. Neutrality in this sense is to be distinguished from imaging, insofar as the precise embodiment of dynamics adds characteristics from the time domain to imaging. Neutrality is also indicated by good downward dynamic range, by which I mean that the sense of natural balance and organization is preserved with low-level information or as the volume is lowered.
There is linguistic austerity in notions of continuousness, organization, balance, and control, that may be germane to popular usage of the word “neutrality”. I have intentionally omitted visual metaphors, the absence of which helps differentiate neutrality from aspects of sound best described in static terms.
Add the requirements of CLARITY and high frequency extension, which provide framing for dynamics within resolving detail, while also contributing to neutrality in the frequency domain. Conventional wisdom considers “warm” and “analytical” to be mutually exclusive; this is where the majority of colorists make their compromise. Anyone who has experienced treble edge is naturally inclined toward a padded treble in the service of euphonic warmth. However, many such problems stem from disagreeable colorations that are the unintended consequences of more agreeable colorations. IME there is really no such thing as too much resolution(with the possible exception of digital processes such as upsampling), provided that careful attention is given to engineering and quality piece parts. Resolving problems in this way invariably improves both resolution and musicality. Finally, clarity and HF extension are critical to clearing a transparent soundstage through which instruments emerge fully delineated and in correct proportions. The rock solid STABILITY of delineated images across the time domain may also signify neutrality, as this reinforces the sense of continuity, organization, and control discussed above. |
"My original post was intended... to operationalize the term 'neutral.'"
"Neutrality by definition is 'without difference'."
"...by comparison to other systems using the same source material." (Emphasize "same source")
Apart from the usual judgments regarding faithfulness to source, if neutrality is defined as a lack of difference, then neutrality requires that analog and digital sources converge. Otherwise we would have separate definitions of neutrality. Elimination of differences between analog & digital sources does not of itself prove neutrality. However, elimination of differences between the two may be considered a reasonable condition for neutrality.
To "operationalize" neutrality(at least regarding selection of source equipment), one should listen closely for differences between the formats and strive through upgrades toward convergence. |
"the strange idea of "neutral"
Neutrality is about balance-- the notion of nothing more and nothing less, nothing added and nothing substracted. In this sense neutrality is more descriptive and useful than transparency.
"all components are inaccurate... I must admit I am in the subjective camp..."
This position is at least consistent.
"there is always going to be some designer bias"
Unfortunately the idea of designer bias is more commonly rooted in cost constraints than designer bias.
"a better term to use is a flat frequency response"
Ignores correct pitch & timbre, the hallmark of a high-end system.
"No real performing space is "neutral"...the room's acoustics always have a huge effect on the musician's sounds...The reference point should be what you want the music to sound like."
Does the room in which the playback system resides really affect sonics as much as the original recording space? IMO not true of any high-performance system-- particularly at lower volumes. |
"When you remove a bit of system distortion, different things sound more different because a common element has been removed from everything you hear."
Exactly my experience through hundreds of internal modifications to components. I'm in the camp of the Objectivists-- except for slight reservations about where the truth lies in LF. A few recent speaker designs like Emerald Physics and Bamberg use Class D digitally EQ'd bass amps to deliver a qualitatively different kind of LF extension and control. This is all fairly new, and I feel more subjectivity judging the shifting paradigm for neutrality in LF than in other areas of FR. |
Incidently, in forums & industry market-speak the notion of "voicing" as representing the pure expression of a designer's original intent is probably as misapplied as "neutrality" to forgive all sorts of shortcomings. |
Cbw723, I can't extend my experience of close convergence between modded analog & digital sources to downstream components. I can only report that mods to downstream components were IMO technically sound and moved the system closer to neutrality as I sense neutrality. The leap of faith necessary to believe this is whether you accept that piece parts made by Duelund, Mundorf, ClarityCap, V-Cap, VSE Superregulators, etc., result in superior performance or merely alternative colorations. However dropping some of these exotic pieces into standard commercial designs is if nothing else an interesting test of the subjectivist viewpoint. |
It would be complicated for non-audiophiles to judge neutrality or any other metric of high quality playback. My wife does well in this regard despite a disinterest in audio, because her ears have been educated through long exposure to the “second-hand smoke” of my audio habit. The general population-- widely exposed to norms of Ipod and boom box listening-- has insurmountable biases against neutrality. Finally, the non-audiophile exposed to live music outside of a controlled studio environment or a cat-bird seat in a concert hall, may have a good sense of realism without having heard sound free of hall and pro audio affects. The best hope is that experienced audiophiles can gravitate to a convergence of opinion—which is unlikely given that most are hunkered down in private listening. Web reviews of RMAF and CES are unreliable, as room sounds at shows tend to morph over the several days of the show. For me the local Audio Club meeting is the only reliable venue to explore a possible convergence of opinion. In this month’s meeting in my area, about 45 members heard a presentation of second-from-top ATC active speakers—which are widely consider “neutral.” I know how I felt about these speakers, but will wait for the club president to compile separate reviews from all members before rejecting the theoretical possibility of consensus in a congress of audiophiles.
Seems to me the discussion is ultimately a dispute about the hierarchy of adjectives used to describe the listening experience. What are the properties(whether described directly in aural terminology, or indirectly by analogy drawn mostly from visual processes) that are most fundamentally descriptive of playback? What other less significant descriptors lie underneath? Does a "better" component reveal the interconnectedness between all the perceived properties of playback, or do the properties just float around independently inside a component like monads or seasonings?
Personally I agree with Bryon that resolving, neutral, and transparent are three of the best audiophile adjectives. But allowing even for wide disagreement among audiophiles regarding the meaning and significance of these descriptors, his three are in the final analysis somewhat static. They fail to account for time-domain factors like microdynamics, macrodynamics, pitch, timbre, and of course timing itself. Of the three static terms, perhaps neutrality is the broadest and most appealing, as it is free of precise visual metaphor and therefore available for a wide range of interpretations and contexts. For me the idea of neutrality refers mostly to flat frequency response.
As a modifier, in the course of making small changes inside a component I can hold control variables constant and listen to the effect of a single variable change such as a low-noise resistor or an improved power supply rectifier. Some of the biggest surprises occur around one's perception of neutrality in the sense of flat frequency response. For example, one of the most difficult things to eliminate in a tube system is loose bass. You can be absolutely convinced that your tube component is resolving and transparent and has realistic tonality in the midrange and treble, while rationalizing away a lack of bass control as warmth, embodiment, involvement, whatever. Now make a single improvement to PS, and bass control and perceived neutrality markedly improve. But note as well that dynamics have improved, and midrange & treble pitch, timbre, etc. have all improved. The funny thing is that with the vast majority of circuit changes made on solid technical grounds, all the descriptors of listening are dragged upward. Solving the most stubborn & obvious problem like sloppy bass usually ameliorates a range of lesser, even unrecognized deficiencies. In the rare case where something goes wrong when something else goes right, some further incremental change will usually redress the situation.
Assuming one accepts the possibility of continuous improvement (and what restless audiophile does not), as remarked several times in the thread, we still need an external reference point. For most this reference point is live music as recalled from memory. My experience is that this conviction that one has “golden ears” is mostly a conceit about the power of auditory memory. Auditory memory for most of us (and probably most pro reviewers as well) is problematic and more flawed than we understand. Hence I argue that the reference point needs to be present in the room. For me the side-by-side comparison of source formats—vinyl & RBCD— provides the best cues about neutrality and other metrics. I’ve had the same CDP and vinyl rig for many years. In stock form they were respectable but sounded very different. The more technical improvements poured into each down unrelated analog & digital paths, the closer they converge on the same sound. And this convergence may be as good a demonstration of neutrality as any other.
It's not for nothing that philosophy has been fascinated by the close relationship between music and mathematics. If there is anything to this view of music, it is reasonable to hope for consensus regarding terminology in audio. |
Bryon, so much of this is about convergence. Interesting speculation that as we move up the food chain of components, instead of arriving at a convergence of opinion, distinctions continue regarding nuances. This is understandable, as once the worst of common coloration is removed, the more small differences are revealed for scrutiny.
In the past few years we increasingly read magazine editors & reviewers remark upon the accelerating pace of sonic improvement. The context for these remarks is usually highly-engineered solutions that stand out from previous designs. I'm inclined to agree with some of this-- particularly in the area of loudspeakers and piece parts. As a generalization, the best equipment is beginning to sound more alike, and the few remaining differences between the best components are more challenging to articulate. Yet even as a rising tide lifts all boats, small differences still tend to jump out at you. Reviewers continue to make distinctions using the same words, but the words describe a changed reality closer to convergence. Does this arise merely because of the all too human need to make distinctions(and to sell magazines), even without meaningful differences? If the pace of change continues, at some point ALL components will join Stereophile Class A Recommended Components, even as the magazine's reviewers continue to try to communicate uniqueness in the full reviews.
The aging of the high end consumer has also brought nostalgia into tastes and purchases. Vinyl takes us back, MM/MI takes us back, under-engineered SET amps takes us back. In the context of aging, coloration becomes a virtue.
Finally, through the economic bubble there has been a huge accumulation of high-end inventory in the marketplace-- much of it falling into the bland middle ground of the bell curve of performance. During this period boutique manufacturers and costly components proliferated all out of bounds, and with this the challenge to review and compare equipment. In some instances subjective opinions about the advantages of certain colorations may be used to advance hidden agendas or retroactively to justify unwise purchases. |
Learsfool, there is a remote possibility that the gradual convergence between modded analog & digital sources occurred by coincidence or was guided by a common bias operating separately through two qualitatively different mod processes. However, as analog and digital sources approach each other AND coloration becomes nearly undetectable, then perhaps sufficient conditions for neutrality have been satisfied. Not that listening is a perfect science, or that there are no differences of opinion between listeners regarding neutrality. However once and awhile one hears a system that sounds very much like real music free from coloration, and IMO this should set the particular listener's expectation for neutrality. IMO in terms of flat frequency response, correct pitch and timbre, resolution & transparency, the current SOTA gets quite close to live music. Where all systems seem to fall short is in the dynamics of live music. Or if they communicate excellent dynamics, then they tend to fail by other measures. In any case the quality of dynamics should probably be distinguished from neutrality. |
This simile of a filtering ski goggle is interesting, as is the water analogy. Perhaps audio components are analogous to brightness and contrast controls on a TV. With such controls it is possible to vary saturation and to whiten or darken the visual palette. Visual "neutrality" lies near the middle of the range of both controls. Perhaps the stereotypical SET has over-saturated contrast, while SS amp is under-saturated with brightness turned up. However to develop useful indices of audio neutrality, one should probabably avoid reasoning by analogy and describe aural phenomena directly. But this is more difficult(and perhaps less interesting)than analogies. |
Bryon wrote: "Admittedly, my operationalization is only a way to judge the RELATIVE level of coloration/neutrality of a system, not its ABSOLUTE level of coloration/neutrality."
Maybe it's possible to move the discussion in a slightly different direction with observations about reference points and relative vs. absolute measures. (On this tack I return to the notion of neutrality in the broad sense.)
Distinctions about coloration may be made relative to an external reference point of live music, or to an internal reference point of a previous or alternate iteration of one's system. Unfortunately fidelity to an external reference point will be debated ad nauseam, owing to endlessly varying opinions about live sound, as well as human frailty in reconstituting performance from memory.
So what internal references points are reliably available? Also, if one is to exclusively adopt internal reference points, any improvement is by definition relative to one’s current system rather than to an absolute. Bryon initially suggested two ideas to operationalize one’s aural judgment of neutrality—ideas about distinction and difference. I suggested that convergence was also a meaningful marker-- particularly the convergence of vinyl and digital sources through independent mod processes. Such convergence at least demonstrates CONSISTENCY between internal reference points. However the question arises as to whether such consistency merely reflects the bias of PREFERENCE rather than increased NEUTRALITY. Since all mods were made as single-variable changes on scientific grounds, I am inclined to view the progress as demonstrative of neutrality rather than personal preference. However others may reasonably disagree.
But more interestingly, can the scientific method be applied generally to the notion of subjective preference? Preference in this sense may be defined as movement toward one's PARTICULAR idea of live music. I believe the answer is yes to science, if the notion of personal preference is operationalized by the test that EVERY vectors of the listening experience must either be subjectively improved or remain unchanged. The basis here is to abandon the notion that colorations are a soup of isolated variables and combinations of inevitable compromises. If a system becomes more like that which one prefers in every sense (without a single shortcoming relative to prior iteration), then one may reliably conclude that neutrality is improved.
In practical terms, neutrality in this sense is the sum of positive vectors such as pitch, timbre, dynamics, frequency extension and control, spatiality, quiet background, and such others as one may consider important. The important thing is to develop an exhaustive list of significant variables and to listen carefully for each one. The test is made with respect to both the sum of variables and their differences. Even though these variables are not like terms, their differences can be measured with respect to the notion of preference, and positive must always be considered better. For the test of improved neutrality to be satisfied, no variable may go negative relative to another. A difficult test that few systems will pass—-but a test that forces the audiophile into a critical and uncompromising analysis of coloration. |
The process of incremental modding leads me to the hypothesis that most of the qualitative measures of sound are linked variables depending upon the independent variable of a circuit change. When a circuit is changed in small ways according to rational engineering, all the variables of listening tend to move together in a direction of preference. This is particularly true when frequency response is unaltered, as anomalies of frequency response may confuse the perception of other qualities. Holding frequency response constant is desireable when evaluating other determinants of neutrality.
The experience of hearing disparate variables of listening moving in opposite directions after an incremental internal mod, is the exception rather than the rule. However in a complete system of multiple components the situation can be much more complex, with coloration occuring as the sum of complementary defects as much as a combination of strengths. Nonetheless, this is no reason to rationalize away some perceived zero-sum cul de sac as a preferred coloration. |
Mrtennis, For me the interesting aspect is deconstruction of the notion of coloration, which is often left unexamined by subjectivists focused purely on aesthetic enjoyment. Unlike a live acoustic performance, audio playback is at the intersection of art and engineering, and as such may deserve its own vocabulary to translate between art and science as precisely as possible. In addition, there is a certain tension between art and science that may be best examined through philosophy. |
Byron replied to my last post with questions:
(i) What about situations where you change a component and you get the result that SOME characteristics have improved and SOME have stayed constant. Is that an improvement in neutrality?
(ii) Can you say more about the phrase “carried in the direction of preference”? That seems like a Subjectivist thing to say, but you are an avowed Objectivist with respect to neutrality. I am not “holding you to” your previous posts. It’s that I get the sense that you are STILL an Objectivist, and so I am confused by the distinctly Subjectivist wording of your operationalization.
The answer below includes consideration of some ideas from the last few posts. To address question (ii) first:
The Objectivist defines neutrality as an absence of coloration, yet may refrain from absolutism by observing that neutrality is in practice an incremental process toward an unattainable goal. Perhaps we should call the Objectivist who is also an Absolutist with respect to believing that it is feasible to arrive finally at the goal of neutrality(wire with gain, absolute sound, etc.) an EXTREME OBJECTIVIST. In the opposing corner, the Subjectivist takes on more diverse incarnations. His principal certainty is that colorations are inevitable. He believes that there are good and bad colorations. While he knows what he likes, he may believe that making a fine distinction between good and bad that is generally acceptable to others, is difficult or irrelevant in view of varying listener tastes & priorities and the overwhelming complexity of systems variables. Finally, he may adopt a wholly relativistic POV by asserting that even a basic distinction between good and bad is impossible-- in which case he is an EXTREME SUBJECTIVIST. At such an extreme he may become hopelessly vague about distinctions and mystical about equipment. Such Extreme Subjectivism may reasonably be banished from this forum with a boom box and Ipod as a parting gift, if not a 100 lb. solid copper chassis with a foot tall KR tube...
A close analysis of preference behaviors may be used to escape the irreconcilable differences of objectivism and subjectivism. For the Subjectivist, any choice of preferred coloration, if examined critically, is by definition a compromise that accepts both appealing and unappealing colorations. The truth of this assertion follows from the inevitable limitations of electronics. The propagation of a desirable coloration is necessarily accompanied by the propagation of undesirable coloration. Objectivists and subjectivists can agree on this point, as both experience restless compromise in the selection of audio components. On the other hand, for the Objectivist it becomes apparent that a choice reflecting one's preference for a relatively less colored component by definition includes fewer factors that war against preference. The two view the same problem from different perspectives. The Objectivist seeks to eliminate coloration, the Subjectivist seeks to increase desirable coloration without adding undesirable coloration, and both choose components that move them in the direction of preference. However both have the burden of careful reflection upon compromises made at particular stopping points along the continuum of coloration. The subjectivist may bear an additional intellectual burden to avoid dismissing or rationalizing away some undesirable coloration in order to justify a preferred coloration.
Assuming that the many variables of the listening experience are numbered and accounted for, an answer is suggested to Bryon’s question (i): All that is necessary to signify increased neutrality is for one variable of the listening experience to advance toward one’s preference without the retreat of another other variable. If this condition is met, for the Subjectivist no coloration has been compromised by any other, and for the Objectivist a coloration has been reduced or eliminated. This UNITY of improvement evidences an ADVANCE in engineering. It is not important for the Subjectivist and the Objectivist to agree, provided that the Subjectivist is careful to watch all his variables carefully and notice when he is being betrayed by his electronics. The Objectivist is en garde to colorations by definition.
From a practical point of view, it may be argued that this condition can never be met. However my own experience in making single-variable changes inside components suggests otherwise. In practice the majority of single-variable mods made on the grounds of solid engineering practice tend to move all variables of listening in a positive direction. This reinforces the idea of LINKED VARIABLES. Admittedly however, it is difficult (though not impossible) to extend this idea to a complex system of multiple components with a greater number of engineering variables. |
The little blue pill may be a case in point regarding preference choices relating to neutrality and coloration. Of course there are more Subjectivists than Objectivists in this particular coloratura, so consensus tends to move in an upward bias toward the contention that the more color the better. But if this argument is to stand up to the close analysis, we must consider whether the pill adds coloration or removes it, in the sense of removing the corrupting veil that separates us from faithful reproduction of the “original performance”? But by removing this veil, have we achieved a realistic performance that is neutral in the manner of real life, or something more colored and larger than life? Before answering too quickly, consider that too many pills make one literally “see blue” as a side effect—and btw she is not fooled and may possibly even become uncomfortable by the stilted & exaggerated performance of TOO MUCH BLUE. But short of choking on this extremity, all vectors of experience tend to move together in the direction of preference, and for both sexes. However if after the deed is done a woman has been satisfied, then we may reasonable conclude that the analogy to high end audio has in the end gone limp.
Somewhat more seriously, Learsfool, granting your point, general agreement on a definition of neutrality is pinned to agreement about coloration—which is why I’m starting to think of neutrality in the broadest terms as “absence of coloration.” Also as you suggest, the Objectivist may fall into traps as easily as the Subjectivist. But if at an extreme the Objectivist emphasizes analysis over aesthetics, the Objectivist is guilty more of intellectual hubris than demonstrative of a failure of aesthetic appreciation. Similarly the Subjectivist may believe that by emphasizing taste and appreciation over analysis he has a monopoly on aesthetics—which in turn evidences his own kind of intellectual hubris. The two see the same phenomena from opposite directions but with similar human foibles. But make no mistake about it, in selecting his component according to preference, a cultivated Subjectivist has analyzed (even if selectively or subliminally) many listening variables, just as a cultivated Objectivist makes choices that are aesthetic as well analytical. That is why I began to explore preference choices as common ground for both parties, and to look at the dichotomous viewpoints of these observers on coloration as a possible basis for a deeper convergence.
Bryon questioned why I feel “The propagation of a desirable coloration is necessarily accompanied by the propagation of undesirable coloration.” The thought is simply the obverse of practice in modding of making a single-variable change inside of a component based on solid technical grounds. After making such a change I generally observe that all variables of the listening experience tended to move in the direction of preference—which as an Objectivist I consider a less colored presentation closer to neutrality. Had the same mod been done in reverse, varying kinds of coloration would have emerged. I think I could demonstrate to any experienced audiophile of either persuasion that some of these emergent colorations were undesirable to all, while other colorations might be considered by some as benign or even preferable—were they not accompanied by the collateral damage in other areas. So in view of this last point, my operationalization of conditions required for movement toward neutrality requires the absence of retreat of any listening variables from preference. Admittedly this requires that the Subjectivist become more objective in terms of exhaustive definition and analysis of listening variables. Admittedly the experiment is harder to operationalize in the context of a complex system involving whole component swaps. The approach may have particular appeal to DIYers, and also to owners of designs like Merlin and Atma-Sphere, which have been refined in small steps over many years. IMO practicing slow incremental change trains the ear to connect the art to the engineering differently than swapping through a succession of so-called “break-through” products.
Bryon, I like this idea of coloration-independent characteristics and loosely connect it to my idea of professed subjectivists and objectivists whose different aesthetics may each progress toward separate but valid senses of neutrality defined in the broadest sense. In the end this approach allows each to settle on a different sound that is in its own way close to colored, or colored with the minimal number of undesirable colorations. However my mod experience suggests it is very challenging to separate desirable from undesirable colorations, in the sense that a single engineering change does not affect everything through linked variables.
Reading a recent audio club review of a speaker that was demoed to around forty educated listeners, I was struck how the generally negative reviewers accentuated a wide array of perceived defects, while positive reviewers focused on the few strengths that they considered paramount in a component. Maybe Objectivists & Subjectivists divide along these lines when thinking about audio. |
Which part didn't you like, the Viagra or the sex? |
Tolstoy: "Truth, like gold, is to be obtained not by its growth, but by washing away from it all that is not gold."
Similar to Barnard Malamud's Roy Hobbs in the "The Natural", who tried with futility to hit a hectoring dwarf (troll?) in the grandstands with line drives from his Wonderbat.
Unfortunately not a single rational idea may be attributed to Ayn Rand, particularly in view of the acolytes of Objectivism(having nothing to do with Audio) who nearly brought us to a second great depression-- which for some readers continues in this thread. |
Bryon wrote, "by improving the way you think about things, you improve the way you perceive things."
Similarly, as an English teacher once told me, if you can't communicate clearly in words, you are not thinking clearly. This thread evidenced clear articulation of at least several new constructs. It was a refreshing change from the repetitiveness of many subjects posted to forum. The vocabulary of audio was expanded a bit, and perhaps as a result, some will think about about their systems and biases more clearly than before.
|
Newbee, Sorry I was inhospitable. Your speculation that the entirety of this thread constitutes trolling is interesting. Even more interesting, the derivation of internet trolling comes from the practice of TROLLING FOR NEWBIES, as popularized by usenet veterans of the early '90s who enjoyed drawing gullible "newbies" into circular discussions. I suppose Bryon could theoretically be a troll, albeit a kind of PHILOSOPHER TROLL. In any case, your choice of the moniker Newbee places you at personal risk of being considered TROLL BAIT.
Bryon, it may be appealing to view neutrality as a phenomenon of both time and frequency domains. I tend to think of imaging and resolution mostly in the frequency category, and dynamics in the time category. However, these are linked characterics, insofar as a well-resolved image frames dynamics within a precise boundary. Taken together these characteristics communicate embodiment. A dislocation of dynamics from imaging in this sense may be considered coloration and therefore a failure of neutrality.
IMO the discussion is ultimately down to an enumeration of the taxonomy of coloration. My last post was about changes in sound that I hear when improvements are made according to generally accepted engineering principals(e.g. Schottky rectification, discrete voltage regulation stages, galvanic separation, low-noise resistors, low-resonance capacitors). Others will likely have different perceptions of coloration.
|
A fuller extract of ideas from G. Holt’s audio glossary seems relevant in context:
1. Frequency response: Phase shift and distortion can sound like frequency-response aberrations.
(Note 1: Perception of neutral frequency response may be subjectively similar to perception of low distortion and correct phase. One is therefore be tempted to use the term neutrality in all three contexts.)
2. Balance: The subjective relationship between the relative loudness of the upper and lower halves of the audio spectrum; "tonal balance."
3. Coherent: seamless from top to bottom… no audible evidence of different… colorations in different frequency ranges.
4. Continuity: Uniformity of coloration (across the operating range).
5. Discontinuity: A change of timbre or coloration due to the signal's transition (across the operating range through) dissimilar coloration.
6. Seamless: Having no perceptible discontinuities throughout the audio range.
(Note 2-6: He identifies colorations as shifts in tonality across the frequency range. My previous reference to neutrality as “continuousness of musical expression across frequency range” was an attempt to describe this. I question whether there can really be such a thing as continuous coloration, positing instead that inevitable variations in coloration across frequency range indicate unsolved problems in the playback system. To distinguish problems in playback from problems in recording, the trained listener merely needs to listen to a wide variety of recordings on the same playback system.)
7. Pitch resolution: The clarity with which pitch…is perceived. Poor pitch resolution makes all notes sound similar…
(Note 7: Here he touches both on Bryon’s original notion of coloration as a failure to differentiate, and on Cbw723’s remarks on Shannon entropy.)
8. Fast: Reaction time, which allows a reproducing system to "keep up with" the signal fed to it. (A "fast woofer" would seem to be an oxymoron, but this usage refers to a woofer tuning that does not boom, make the music sound "slow," obscure musical phrasing, or lead to "one-note bass.") Similar to "taut," but referring to the entire audio-frequency range instead of just the bass.
9. Smooth: Not necessarily a positive system attribute if accompanied by a slow, uninvolving character.
10. Control: The extent to which a (system) sounds as if it is "tracking" the signal being fed to it. The sound is tight, detailed, and focused.
(Note 8-10: Here he identifies fast & controlled dynamics as essential to an uncolored presentation. Absent precise dynamics, differentiation is lost (e.g. “one-note bass.”))
11. Definition (also Resolution): That quality of sound reproduction which enables the listener to distinguish between, and follow the melodic lines of, the individual voices or instruments comprising a large performing group.
12. Detail: The subtlest, most delicate parts of the original sound, which are usually the first things lost by imperfect components.
13. Veiled, veiling: Pertaining to a deficiency of detail and focus, due to moderate amounts of distortion, treble-range restriction, or attack rounding.
14. Focus: The enhanced ability to hear the brief moments of silence between the musical impulses in reproduced sound.
15. Hangover: A tendency for reproduced sounds to last longer than they should.
16. Ringing: The audible effect of a resonance: coloration, smear, shrillness, or boominess.
(Note 11-16: Higher resolution defined in this sense a good thing, and is connected by Holt to Bryon’s original notions about the distinction of differences. Interestingly, the related quality of Focus entails removal of additive colorations so as to reveal interstitial silences. Also, Holt connects compromised resolution to “attack rounding”—or failed dynamics.)
17. Error of commission: Signal degradation due to the addition of sounds that were not present in the original signal. Distortion and coloration are examples of errors of commission.
18. Error of omission: Signal degradation due to the loss of information that was present in the original signal. Smearing and treble loss are examples of errors of omission.
19. Gestalt response: The evocation of a complete memory recognition by an incomplete set of sensory cues. A gestalt response to the few things an audio system does outstandingly well can make imperfect reproduction seem more realistic than it actually is.
(Note 17-19: Here he gets into interesting territory similar to Bryon’s ideas about coloration as additions, subtractions, or alterations. I think we may conclude that Holt regarded coloration primarily as an error of commission. The idea about Gestalt response is fascinating insofar as he suggests that absent errors of commission, a less than fully resolving playback system may be convincing and perhaps even uncolored in the strictest sense of the word.)
20. Neutral: Free from coloration.
21. Uncolored: Free from audible colorations.
Finally, Holt used the terms Subjectivism and Objectivism in a particular sense. Subjective reviewing is his term of art for critical observation based on controlled listening, psychoacoustics, and a precise vocabulary to evaluate colorations. He considered Objectivists to be the so-called “meter men” test bench-oriented reviewers of Julian Hersh school who tended to ignore the science of listening. Holt’s Subjectivism is not inconsistent with objective analysis. As he says in the glossary, given a precise definition of terms “there is no longer any excuse for an audio reviewer saying, ‘I can hear a difference, but there's no way of describing it.’ Now, there is a way.”
So Holt's view was that audible colorations can be precisely described, and if inaudible, do not exist. |
To whip the horse's eyes with one more water analogy, "You can bring a horticulture but you can't make her think."
Too much abstraction, Dorothy Parker? |
"Sounding better and sounding neutral are two completely different things."
Assuming that one accepts common usage in critical vocabulary that "neutral" means "uncolored," can you suggest a better term than neutral to describe the experience of "better"? |
Kijanki, my view is that analytic & sterile err at the opposite extreme of unresolving warmth. Both kinds of extremes are colorations and as such, represent deviations from neutrality. There are many instances in which internal mods to electronics improve resolution while also preserving and even enhancing warmth. A move in this direction is an uncompromising step toward neutrality. Granted there are no absolutes, but as long as a playback system bothers one even a bit with an undesireable coloration, I do not rationalize away the negative coloration as an inevitable compromise. For me the notion of a step toward neutrality is a useful construct to describe resolution of what in the conventional wisdom of audio may appear on the surface to be unreconciliable characteristics. Resolutions of this type may be heard as the system getting out of the way of the music. |
Such descriptions of personal experiences regarding coloration and neutrality as related to specific components, may serve as helpful illustrations of abstractions explored elsewhere in the thread.
My personal preference (and also my perception of neutrality) is to have as much warmth & embodiment as can be obtained without compromising transparency, high resolution, and realistic pitch & timbre. I suppose this reveals a romantic bias. However merely reversing the same adjectives in order of importance would tend to indicate an analytic bias. Regardless of one's starting point on a scale of preference, neutrality tends to occur toward the middle. For me "better" is about finding that middle while doing a bit more of everything right at both extremes of the scale. When this occurs it becomes clear that this is not a zero-sum game. |
Learsfool, as one progressively raises the bar in the hobby through a process of acquiring or hearing successively improved components, what is the nature of the "reinforcement of preference" that you propose is occuring? Assuming that break-throughs in technology and improvements in price/performance ratio are actually obtainable and that one welcomes this process, is the result of change merely to cement the listener to a prior preference of coloration, or to liberate him to reassess and refine preference in the context of a newly available perceptions? Seems to me that the very notion of progress implies change in the latter sense. |
Bryon, your constructs are interesting, and in conclusion testament that "what we observe is not nature herself, but nature exposed to our method of questioning." Heisenberg |
Cbw723 wrote, "All that has been suggested is that components may have either more or less coloration, and that it may be possible to distinguish one of those conditions from the other."
My point is not too different from yours. Movement along a continuum toward a relative lack of coloration is possible. An advance in engineering and neutrality has been indicated when all variables of the listening experience are carried in the direction of preference. But if after making a change in your system you find that some of the variables of listening have moved opposite to preference, this indicates a lack of advance toward neutrality. Whether or not you prefer to stop along the way and accept a relatively colored presentation is your business. But given the way electronics function, the chances are good that a preferred coloration is accompanied by an unpreferred coloration, and one must make grudging compromise between the two. What colored system is without defect recognized even by its owner... |
Learsfool, all good points. Regarding whether "convergence" on some ideal of neutrality or transparency is possible or even desireable, the question should not be what "most" audio systems fail to do(whether through shortcomings in dynamics or other nuances of playback). The spirit of the OP was more about how to describe or operationalize the improvement that one hears when the veil is lifted by a superior component. Is what one hears from the better component closer to an idealized "neutrality", or merely some more preferable coloration? In the end we are down to the familiar subjectivist/ objectivist debate on whether the merry-go-round is nothing more than an endless trade of one coloration for another, or presents occasional glimpses of real improvement.
IMO audio components are very different from instruments, halls, etc., each of which exhibits its own indelible character. There is no mistaking a terrible child's violin played in an echoing bathroom for anything but a real instrument played in a real space. By definition all violins are real violins, regardless of "voicing." In contrast, the notion of "voicing" an audio system is problematic. In "most" components voicing is the sum of built-to-cost compromises and major or minor deficits in design-- affectations that may have little to do with pure concepts like designer's original intent. In any case the result in audio is nearly always a sound that is not mistaken for a real violin. In audio components as in all other things the exception to the rule is rare and more interesting to contemplate.
|
Bryon, several posts back you raise interesting points by distinguishing between perceptions and preferences, while acknowledging their interrelatedness. You suggest that perception and preference are fungible characteristics that both evolve through educated listening, and that preference follows perception, albeit at a slower rate of change. You maintain that there are fewer differences in perception between educated listeners than is generally remarked. Through educated listening, divergences of preference may ultimately represent shared but differently weighted perceptions.
My sense is that listeners who have moved through the arc of lots of equipment generally experience "progress" in their evolution, and look back on past components with an admission that the process was a journey of perception and taste, rather than a random sensory experience. One would like to believe that perception is on the leading edge of preference. There is enough logical positivism in the process to suggest science, and enough metaphysics to justify continued purchases. |
Kijanki, Warmth that lacks correct pitch definition is yet one more example of an undesirable coloration-- often indicative of weaknesses in rectification, power supply, or coupling caps. IME it is one of the most difficult problems to eliminate in a tube component, but it is not inherent in tubes per se, and once resolved, warmth is preserved with improved "neutrality" as a sense of full & natural embodiment, but without overhang, mid-bass bulge, or congestion.
At the opposing extreme of coloration that some would describe as an overly analytical presentation, the Benchmark is an OK DAC that has been substantially improved by an active community of modifiers. Rather than use a budget DAC to illustrate the point about an opposing coloration in constrast to warmth, it might be more interesting to think about something like a Martin Logan full range electrostatic. In this case at least, we have an absolutely clear window into high resolution, with loss of embodiment and warmth as perhaps a necessary design compromise. But I would submit that within the limitations of what this speaker DOES it is operating in a neutral (i.e uncolored) manner. This assertion follows earlier distinctions made by Bryon concerning errors of commission compared to errors of omission. |
