Objectivist or Subjectivist? Give Me a Break

[Phelonious Ponk]

Tim,

First it would be nice if you could quote others posts properly. The whole sentence was :

I feel the industry is afraid of this aspect - almost every manufacturer claims his amplifier is more neutral and faithful (without being too specific) than his competitor

and you reduce it to

almost every manufacturer claims his amplifier is more neutral

BTW, WBF is a niche. I know about it. Most of its members pursuit higher sound quality, and enjoy it.

Sometimes an entire post only has one statement I care to respond to. I don't hit the "Reply With Quote" button and don't get the name of the OP dropped in. Folk remind me from time to time to put the name in there. I keep forgetting. I guess I don't really think it's all that important or I'd remember. Sorry.

Frankly, Micro, I apologize for taking you out of context, but I reduced your statement to the part that I thought warranted a response. I think most amp manufacturers are trying to produce neutral, accurate products. I don't think almost every manufacturer markets his products by comparing them to everyone else's, and claiming his are more neutral. I'm curious now. I guess I'll have a look around and see if I'm wrong.

...Just checked Bryston, Krell and Mark Levinson. Went straight to the high-end monoblocks in every case. They all said something about accuracy. None said they were more so than all other amps. Maybe that claim is somewhere else on the sites....

Tim

 

[microstrip]

(...)
...Just checked Bryston, Krell and Mark Levinson. Went straight to the high-end monoblocks in every case. They all said something about accuracy. None said they were more so than all other amps. Maybe that claim is somewhere else on the sites....

Tim

Tim,
For our information:

sentences taken from the Bryston site:

Although the description of circuit innovations can indicate the research and commitment we bring to the design and manufacturing of our products, it’s not the whole story. The purpose of all this technology is to place you within the scene of the movie, or to transport you to the actual musical event.

Whether you can experience all the drama, movement, and emotions intended is the one truly meaningful measure of any audio/video system.

Its superbly low noise floor yields an inky black silence between notes for hauntingly realistic reproduction. The annoyances previously ascribed to ‘transistor sound’ are absent, and the amplifier seems to combine the seductive lushness of tube equipment with the focus and accuracy of solid-state.

sentences taken from the Mark Levinson site:

As one of the founding companies of high-end audio, Mark Levinson was instrumental in defining the concept – that audio components could achieve a level of performance far beyond the “high-fidelity” conventions of the day to reproduce music with astonishing realism. High-end audio components do not simply play back sound – they convey the essence of the artist and the emotion of the performance.

In 2001, Harman International purchased Mark Levinson and today the company is part of its Luxury Audio Group based in Elkhart, Indiana. The move enabled Mark Levinson to combine its decades of singular expertise with the research and development and manufacturing resources of global technology leader Harman International. The decade saw the introduction of ultrahigh-end products including the ? 53 Reference Monaural power amplifier, ? 502 Surround Sound Preamplifier and others, all designed in the Mark Levinson tradition of uncompromising quality.

With the technical hurdles of switching power amplifiers behind us and with the internal layout finalized, the No53 development team headed back to the listening room. We listened with different types of music. We listened with different speakers, electronics and cables. We listened in different rooms. We compared the No53 to other Mark Levinson power amplifiers. We compared it to the power amplifiers of our competitors. We listened blind, we listened sighted and we flew in customers from around the world to listen with us. Only then, when we were convinced that we had coaxed the last ounce of performance from this revolutionary design, did the No53 go into production.

No one says they compared their amplifiers with a straight wire with gain!

 

[NorthStar]

Sometimes an entire post only has one statement I care to respond to. I don't hit the "Reply With Quote" button and don't get the name of the OP dropped in. Folk remind me from time to time to put the name in there. I keep forgetting. I guess I don't really think it's all that important or I'd remember. Sorry.

-----See, I quoted your name above! :b ...And the entire quote, but I highlighted only the part that I was interested in.

Frankly, Micro, I apologize for taking you out of context, but I reduced your statement to the part that I thought warranted a response. I think most amp manufacturers are trying to produce neutral, accurate products. I don't think almost every manufacturer markets his products by comparing them to everyone else's, and claiming his are more neutral. I'm curious now. I guess I'll have a look around and see if I'm wrong.

-----The full context is there, but again only the part that I'm interested in is highlighted for me, you and everyone else to see.

...Just checked Bryston, Krell and Mark Levinson. Went straight to the high-end monoblocks in every case. They all said something about accuracy. None said they were more so than all other amps. Maybe that claim is somewhere else on the sites....

Tim

-----No more comment. :b ...Was it hard? Absolutely not!

 

[jkeny]

Just to change the sub-topic completely & perhaps add some extra datapoints to this discussion - here's an interesting recent paper (2012) that is well worth reading with the intriguing title ""Human Time-Frequency Acuity Beats the Fourier Uncertainty Principle" [url]http://arxiv.org/pdf/1208.4611.pdf[/URL]

Just to explain the title & give a bit of a teaser - there is the well known Heisenberg Uncertainty Principle that applies at the quantum level & asserts that there is a "fundamental limit to the precision with which certain pairs of physical properties of a particle, such as position x and momentum p, can be known simultaneously. The more precisely the position of some particle is determined, the less precisely its momentum can be known, and vice versa"

This uncertainty principle also applies to Time & Frequency & is referenced in the paper's abstract

The time-frequency uncertainty principle states that the product of the temporal and frequency
extents of a signal cannot be smaller than 1/(4pi). We study human ability to simultaneously
judge the frequency and the timing of a sound. Our subjects often exceeded the uncertainty limit,
sometimes by more than tenfold, mostly through remarkable timing acuity. Our results establish a
lower bound for the nonlinearity and complexity of the algorithms employed by our brains in parsing
transient sounds, rule out simple \linear lter" models of early auditory processing, and highlight
timing acuity as a central feature in auditory object processing.

I recommend all to read it if they are interested in how & why hearing might be more accurate than measurements & indeed redirect their focus to the time domain rather than frequency domain which tends to get the greater part of attention as is evidenced in this & many other audio forums!

 

[JackD201]

 

[Phelonious Ponk]

Tim,
For our information:

sentences taken from the Bryston site:

Although the description of circuit innovations can indicate the research and commitment we bring to the design and manufacturing of our products, it’s not the whole story. The purpose of all this technology is to place you within the scene of the movie, or to transport you to the actual musical event.

Whether you can experience all the drama, movement, and emotions intended is the one truly meaningful measure of any audio/video system.

Its superbly low noise floor yields an inky black silence between notes for hauntingly realistic reproduction. The annoyances previously ascribed to ‘transistor sound’ are absent, and the amplifier seems to combine the seductive lushness of tube equipment with the focus and accuracy of solid-state.

sentences taken from the Mark Levinson site:

As one of the founding companies of high-end audio, Mark Levinson was instrumental in defining the concept – that audio components could achieve a level of performance far beyond the “high-fidelity” conventions of the day to reproduce music with astonishing realism. High-end audio components do not simply play back sound – they convey the essence of the artist and the emotion of the performance.

In 2001, Harman International purchased Mark Levinson and today the company is part of its Luxury Audio Group based in Elkhart, Indiana. The move enabled Mark Levinson to combine its decades of singular expertise with the research and development and manufacturing resources of global technology leader Harman International. The decade saw the introduction of ultrahigh-end products including the ? 53 Reference Monaural power amplifier, ? 502 Surround Sound Preamplifier and others, all designed in the Mark Levinson tradition of uncompromising quality.

With the technical hurdles of switching power amplifiers behind us and with the internal layout finalized, the No53 development team headed back to the listening room. We listened with different types of music. We listened with different speakers, electronics and cables. We listened in different rooms. We compared the No53 to other Mark Levinson power amplifiers. We compared it to the power amplifiers of our competitors. We listened blind, we listened sighted and we flew in customers from around the world to listen with us. Only then, when we were convinced that we had coaxed the last ounce of performance from this revolutionary design, did the No53 go into production.

No one says they compared their amplifiers with a straight wire with gain!

No, but someone said this:

Micro: I feel the industry is afraid of this aspect - almost every manufacturer claims his amplifier is more neutral and faithful (without being too specific) than his competitor.

None of your quotes from amp manufacturers say their products are "more neutral and faithful" than others. I couldn't find that in the web content of three of the top high-end SS amp manufacturers and, evidently, neither could you. Just to avoid going back and forth a couple of more times, those quotes don't imply it either. They talk of the quality of their products without putting them above competitive products.

Tim

PS: Bob? How was that quoting technique?

 

[Phelonious Ponk]

Just to change the sub-topic completely & perhaps add some extra datapoints to this discussion - here's an interesting recent paper (2012) that is well worth reading with the intriguing title ""Human Time-Frequency Acuity Beats the Fourier Uncertainty Principle" [url]http://arxiv.org/pdf/1208.4611.pdf[/URL]

Just to explain the title & give a bit of a teaser - there is the well known Heisenberg Uncertainty Principle that applies at the quantum level & asserts that there is a "fundamental limit to the precision with which certain pairs of physical properties of a particle, such as position x and momentum p, can be known simultaneously. The more precisely the position of some particle is determined, the less precisely its momentum can be known, and vice versa"

This uncertainty principle also applies to Time & Frequency & is referenced in the paper's abstract


I recommend all to read it if they are interested in how & why hearing might be more accurate than measurements & indeed redirect their focus to the time domain rather than frequency domain which tends to get the greater part of attention as is evidenced in this & many other audio forums!

Damn, John, I got a headache just reading the paragraph you quoted and you want me to read the paper? That's just plain mean! I don't doubt a bit that timing errors can be a more important impediment to reproduction realism than frequency errors if they are near the same magnitude. But they're not. Once you get to the end game -- speakers, even really good speakers, playing into a room, frequency variation are impossible to miss. Timing? not so much, unless you're playing a bad turntable. The difference between a Klipschorn and a Magico is obvious to the half deaf and uninterested. The difference between jitter levels from a $99 DVD player and a SOTA DAC? Not so much. The (jitter, not analog output) difference between a really good $500 DAC and SOTA? Guys like Amir have to train themselves to hear that.

So yeah, all things being equal, time domain errors might be as important to audio reproduction as frequency errors. But all things don't appear to be equal. Or at least that's my story. Unless, of course, all that Mr. Science stuff up there meant something else altogether and I just gave you an answer to a question you didn't ask. It could happen.

Tim

 

[Ethan Winer]

I will accept the Carver challenge at face value, -70db NULL!

An important point that's been overlooked is that -70 dB was the best null Bob could achieve when matching his high quality amp against another brand whose specs are clearly inferior. I imagine the null between two amps that are both high quality could easily exceed 70 dB if he didn't have to add obvious distortion etc to make his amp match.

Another thing I noticed is that even though -70 dB = inaudible has now been settled with absolutely no room for disagreement*, nobody here has acknowledged changing their opinion. I already listed two times I changed my opinion after learning something, and nobody acknowledged that either. Nor has one person emailed me their choices about the various Wave file comparisons on my web site to show they can hear stuff at a lower threshold than my claims. I'm not trying to be a jerk here, but jeez guys, can't you man up and admit you were proven wrong?

*Even John Atkinson at the end of his article admitted he had been wrong:

We're still a little bewildered around here about how all this turned out. Not the way we expected. But that's the way it was.

--Ethan

 

[jkeny]

Damn, John, I got a headache just reading the paragraph you quoted and you want me to read the paper? That's just plain mean!

Let me help, if you are serious?

I don't doubt a bit that timing errors can be a more important impediment to reproduction realism than frequency errors if they are near the same magnitude. But they're not.

Sorry? You know this how?

Once you get to the end game -- speakers, even really good speakers, playing into a room, frequency variation are impossible to miss. Timing? not so much, unless you're playing a bad turntable.

You have a simplistic view of what timing in audio actually means, I think!

The difference between a Klipschorn and a Magico is obvious to the half deaf and uninterested. The difference between jitter levels from a $99 DVD player and a SOTA DAC? Not so much.

I disagree but even if this were so, what is it proving?

The (jitter, not analog output) difference between a really good $500 DAC and SOTA? Guys like Amir have to train themselves to hear that.

Hmm, it's not jitter we are talking about!

So yeah, all things being equal, time domain errors might be as important to audio reproduction as frequency errors. But all things don't appear to be equal. Or at least that's my story.

You are correct, it's a story but you should have started it with "once upon a time"

Unless, of course, all that Mr. Science stuff up there meant something else altogether and I just gave you an answer to a question you didn't ask. It could happen.

I didn't ask a question - I suggested to read a paper that goes to the heart of something that is fundamental to this debate - can hearing reveal more than current measurements? Answer = yes (according to this study)! Are you not interested in this revelation?

 

[treitz3]

...Nor has one person emailed me their choices about the various Wave file comparisons on my web site to show they can hear stuff at a lower threshold than my claims. I'm not trying to be a jerk here, but jeez guys, can't you man up and admit you were proven wrong?

Hello, Ethan. I have seen you mention this a couple of times. Would you be so kind as to provide a link? I don't know which website you are referring too.

Tom

 

[JackD201]

I've said this before but will say it again. We need to learn more about the perception side of the equation if we want to have even MORE meaningful measuring techniques. Many have gotten themselves stuck with borrowed studies from the telecom industry from over half a century ago! Meanwhile Universities and Labs around the world are learning more and more about how we perceive everyday. This is a very good example of a study that if proven may help give us better instrumentation to aid in better application.

A true objectivist never stops asking questions. To assume to have it all down is well, not very objective.

 

[Gregadd]

I recall the amp used by Carver nor it's specs were never revealed. Can anyone name it or provide the specs even before it was modified? I think it is important because members keeping making representations about it that I do not remember being revealed.
Greg

 

[DonH50]

I barely skimmed the paper. It appears to target time delay, and it has been shown that our ability to locate objects uses pretty small time delays (on the order of a few us IIRC). For components, the only place I see it being a significant issue is speakers, and that is part of what led to the effort to time-align the wavefroonts from speakers (and my love of planar speakers, but that's another story). Phase shift and group delay is easily measured over frequency for electronics, a little harder for speakers, but should certainly be measurable. I think it's another of those things we can measure but usually don't. Consumers seem to have a hard time undestanding some of the basic specs that are provided, and there seems to be little incentive to provide more (and a lot of reasons to not, including cost and time of testing, educating customers about why it matters, etc.) Seems to me most specs are there because they are required, either by regulations or because "everybody does them". The few added specs usually crop up when the company feels there is a marketing advantage (which may have nothing to do with actual audible advantage).

I do not know how much delay we can sense in say a musical selection if the highs and lows are delayed differently in time by a small amount. I would be curious if the paper (or any paper) has addressed that. I suspect it would alter the image somewhat.

My 0.000001 cents (microcent) - Don

 

[DonH50]

Gregadd -- I remember reading in a follow-up issue it was revealed as a Conrad-Johnson but do not recall the model.

 

[JackD201]

I barely skimmed the paper. It appears to target time delay, and it has been shown that our ability to locate objects uses pretty small time delays (on the order of a few us IIRC). For components, the only place I see it being a significant issue is speakers, and that is part of what led to the effort to time-align the wavefroonts from speakers (and my love of planar speakers, but that's another story). Phase shift and group delay is easily measured over frequency for electronics, a little harder for speakers, but should certainly be measurable. I think it's another of those things we can measure but usually don't. Consumers seem to have a hard time undestanding some of the basic specs that are provided, and there seems to be little incentive to provide more (and a lot of reasons to not, including cost and time of testing, educating customers about why it matters, etc.) Seems to me most specs are there because they are required, either by regulations or because "everybody does them". The few added specs usually crop up when the company feels there is a marketing advantage (which may have nothing to do with actual audible advantage).

I do not know how much delay we can sense in say a musical selection if the highs and lows are delayed differently in time by a small amount. I would be curious if the paper (or any paper) has addressed that. I suspect it would alter the image somewhat.

My 0.000001 cents (microcent) - Don

If this study is correct, something to supplant FFT analysis is a possibility.

 

[DonH50]

Perhaps. What is usually ignored in FFT analysis is the phase; only magnitude plots are normally shown. You need phase to correlate time differences. Since few are likely to correlate the phase with timing errors by inspection, a better measurement would be the impulse response. However, it should not be hard to present a plot of the phase response and group delay (the derivative of the phase, a simple mathematical operation for a computer).

 

[Gregadd]

Gregadd -- I remember reading in a follow-up issue it was revealed as a Conrad-Johnson but do not recall the model.

Thanks Don. I was refering to the clone not the CJ.
Greg

 

[DonH50]

What clone? The amp Bob started with? It was an ugly bread-board affair, not sure which of his then-current amps he used as a starter.

 

[Gregadd]

What clone? The amp Bob started with? It was an ugly bread-board affair, not sure which of his then-current amps he used as a starter.

Yes they amp he statred with. The one he made sound like the CJ. Members continue to talk as though they know what it was.
I read all the articles, a discussion forum on Sterophile and looked at some videos on U-tube. I do not recall it's identity being revealed or it's original specs. Sureluy somebody knows.
Greg

 

[jkeny]

Don, I believe this paper bears more attention than a quick skim as it makes a number of fundamental & interesting points:
- In signal processing, in band-limited systems there is a limit to how precisely we can simultaneously measure both time & frequency i.e there is an uncertainty! These experiments show that in a number of people tested that they could beat this uncertainty, some by 10 fold "We study human ability to simultaneously judge the frequency and the timing of a sound. Our subjects often exceeded the uncertainty limit, sometimes by more than tenfold, mostly through remarkable timing acuity."
- So what? Well "When applied to filters, the result is that one cannot achieve high temporal resolution and frequency resolution at the same time; a concrete example are the resolution issues of the short-time Fourier transform – if one uses a wide window, one achieves good frequency resolution at the cost of temporal resolution, while a narrow window has the opposite trade-off."
- OK, so what again? "Our results establish a lower bound for the nonlinearity and complexity of the algorithms employed by our brains in parsing transient sounds, rule out simple "linear filter" models of early auditory processing, and highlight timing acuity as a central feature in auditory object processing."
- Eh? "In many applications such as speech recognition or audio compression (e.g. MP3 [18]), the first computational stage consists of generating from the source sound small sonogram snippets, which become the input to latter stages. Our data suggest this is not a faithful description of early steps in auditory transduction and processing, which appear to preserve much more accurate information about the timing and phase of sound components than about their intensity"
- so the suggestion is that being very accurate in recreating the frequency/amplitude of a signal is actually the wrong focus if this model of the perception of hearing is correct. I've highlighted this as it is fundamentally counter to what is still the current focus in audio development & as Jack says & I say myself, without an accurate understanding of how we hear what we hear, we are perhaps chasing accuracy where it's not needed & ignoring the real issues.

I barely skimmed the paper. It appears to target time delay, and it has been shown that our ability to locate objects uses pretty small time delays (on the order of a few us IIRC). For components, the only place I see it being a significant issue is speakers, and that is part of what led to the effort to time-align the wavefroonts from speakers (and my love of planar speakers, but that's another story). Phase shift and group delay is easily measured over frequency for electronics, a little harder for speakers, but should certainly be measurable. I think it's another of those things we can measure but usually don't. Consumers seem to have a hard time undestanding some of the basic specs that are provided, and there seems to be little incentive to provide more (and a lot of reasons to not, including cost and time of testing, educating customers about why it matters, etc.) Seems to me most specs are there because they are required, either by regulations or because "everybody does them". The few added specs usually crop up when the company feels there is a marketing advantage (which may have nothing to do with actual audible advantage).

I do not know how much delay we can sense in say a musical selection if the highs and lows are delayed differently in time by a small amount. I would be curious if the paper (or any paper) has addressed that. I suspect it would alter the image somewhat.

My 0.000001 cents (microcent) - Don