Objectivist or Subjectivist? Give Me a Break

[mep]

I tried to bring some common sense to this thread that was sorely lacking. People need to quit saying that two amplifiers from two or more different companies measure exactly the same. They don’t and never will. Two different amplifiers from the same company will not measure exactly the same due to component tolerance variations that I addressed in my above post. They will be in the manufacturing tolerance of the company, but they will never measure exactly the same. It’s not uncommon for two channels in the same amplifier not to measure exactly the same due to those component variations. ‘Close” is a much better descriptive term. The word “exact” implies perfection, and nothing made by man is perfect.

We live in a world of manufacturing tolerances, and tolerances imply differences. How audible those differences are is certainly up for discussion. But please, let’s stop talking about amplifiers being exactly the same in measurements. They simply aren’t.

 

[Kal Rubinson]

per·cept (pûrspt)
n.
1. The object of perception.
2. A mental impression of something perceived by the senses, viewed as the basic component in the formation of concepts; a sense datum.
Pain

a : a state of physical, emotional, or mental lack of well-being or physical, emotional, or mental uneasiness that ranges from mild discomfort or dull distress to acute often unbearable agony, may be generalized or localized, and is the consequence of being injured or hurt physically or mentally or of some derangement of or lack of equilibrium in the physical or mental functions (as through disease), and that usually produces a reaction of wanting to avoid, escape, or destroy the causative factor and its effects <was in constant pain> b : a basic bodily sensation that is induced by a noxious stimulus, is received by naked nerve endings, is characterized by physical discomfort (as pricking, throbbing, or aching), and typically leads to evasive action

Not bad but the "neurological" portion is clumsy and simplistic. Note, however, the use of the word "induced." In other words, it does not exist as a stimulus but is a percept induced by a range of stimuli. The induced pain is real regardless of the stimulus.

 

[Phelonious Ponk]

I tried to bring some common sense to this thread that was sorely lacking. People need to quit saying that two amplifiers from two or more different companies measure exactly the same. They don’t and never will. Two different amplifiers from the same company will not measure exactly the same due to component tolerance variations that I addressed in my above post. They will be in the manufacturing tolerance of the company, but they will never measure exactly the same. It’s not uncommon for two channels in the same amplifier not to measure exactly the same due to those component variations. ‘Close” is a much better descriptive term. The word “exact” implies perfection, and nothing made by man is perfect.

We live in a world of manufacturing tolerances, and tolerances imply differences. How audible those differences are is certainly up for discussion. But please, let’s stop talking about amplifiers being exactly the same in measurements. They simply aren’t.

Of course. Both sides of the sentence are exaggerations, or broad assumptions, at least. No two amplifiers will measure exactly the same, and no two amplifiers within good manufacturing specifications will sound completely different. We're operating In a world of differences so subtle that they border on in audible, even if thati is arguable. Well, except when we're operating in the world of Audiophile hyperbole. Thanks for clearing that up, Mark.

Tim

 

[FrantzM]

Hi

OT

Tomelex, I listen exclusively with headphones and I would like to think they are very good headphones and some very good amps and DAC... I listened to the cut you provided some time ago.. Well.. They are interesting in the sense that I didn't have music in my head but , nor did it blow my mind the way a good speaker set-up will do? Over expectation?
End of OT

Whatever you were discussing about amps that measures the same but doesn't sound the same ... Please do go back to it

 

[treitz3]

Hello, gentlemen. Please allow me to chime in here. I have heard the Silver 9's, Silver 7's and what some would consider the Silver 6's. I have heard them hooked up to SS pre's as well as tubed pre's, including a 1 of 2 made [22 tube] pre made by Bob himself [referred to as the Silver 1]. I have also been privy to every single TFM amp made to emulate the famed Silver 7 and I can attest that IME, none of the amps sound the same. Not one.

How the Stereophile guys failed is beyond me but that's just me. I'll have to pull up an interview we did with Bob a while back and let him explain it in his own words. What I'm talking about is the actual null test during the CC, along with the percentage of "what makes an SS amp sound like a tube amp".

Tom

 

[Robh3606]

I recall John Atkinson said that the Carver challenge did not include bass.

It's getting obvious knowone reads whats posted in threads here is a partial on the program source material:


Program sources were as follows, for the
following specific sonic attributes: “The
Portrait” and “Peter the Hermit,” from
Growing Up in Hollywood Town (Sheffield
CD-13 and Lab 13) for depth and perspective,
HF maturalness, bass heft and tightness;
Respighi’s Church Windows (Reference
Recordings RR-15) for breadth, depth, bass
range and control, and massed string tone;
Beethoven & Enesco Violin & Piano
Sonatas (Wilson Audio Specialties W-8315)
for tonal accuracy, depth , and imaging
specificity and stability; “Improvisations”
by Jim Keltner, from The Drum Record
(Sheffield CD-14/20) for high-end openness
& timbre and low-end attack, control and
range; and McBride’s “Mexican Rhapsody,”
from a badly worn copy of Fiesta In Hi-Fi
(Mercury Living Presence SR90134) for
treatment of HF stridency and mistracking.

Rob

 

[treitz3]

Tom, there is also good stuff in the audio critic magazine from Carver and some other heavy hitters in audio. But please do post what you can, I for one will enjoy it.

http://www.youtube.com/watch?v=NAmtwNipZoM&feature=player_embedded

Good evening, Tom. Go ahead and take a break and let the video load. When you come back, fast forward to 2 minutes and 30 seconds into the video. Bob will explain it in his own words. I'll be happy to post my own impressions on each amplifier I have heard but I think most folks here are beyond the TFM series. The Silver 6's, 7's and 9's I think speak for themselves and as far as I know are still highly regarded on any audiophile standard.

Tom

 

[jkeny]

Tom, how do different DACs give different sound stage perspectives & can this be correlated to the measurements of the analogue output to the speakers.

Yes, the soundstage depth is generated by the recording and mix engineer given normal quality speakers for playback, actually, speakers that are purposely designed off of a flat group delay can, given certain recordings, increase the depth, but they are then just being an extenison of the knobs and tools of the recording engineer, the engineer does the work on the recording (in daw etc) or the orignal simple mic recording can also deliver this info.

Headphones, playing binaural, will blow your mind, they bring you to the recording event vs bringing the recording event to you as in convetional stereo. Folks dont know what depth or height is until they hear real binaural over headphones.

 

[KeithR]

Of course. Both sides of the sentence are exaggerations, or broad assumptions, at least. No two amplifiers will measure exactly the same, and no two amplifiers within good manufacturing specifications will sound completely different. We're operating In a world of differences so subtle that they border on in audible, even if thati is arguable. Well, except when we're operating in the world of Audiophile hyperbole. Thanks for clearing that up, Mark.

Tim

To keep to my example, I don't believe the difference between a Krell, Levinson, and Spectral amplifier is sonically "subtle"

 

[ack]

To keep to my example, I don't believe the difference between a Krell, Levinson, and Spectral amplifier is sonically "subtle"

Correct. Science goes so much, and beyond measurements, manufacturers simply voice their equipment as they wish.

 

[Phelonious Ponk]

To keep to my example, I don't believe the difference between a Krell, Levinson, and Spectral amplifier is sonically "subtle"

Then I must believe that there is a measured manufacturing specification that defines the differences. It is not reasonable to believe that Krell, Levinson and Spectral voice each amp by hand. And of course this raises more questions: what is subjective/what is high fidelity? If three top amps sound (and measure) very noticably different, unless they're just off by exactly the same % in different directions, by definition there is going to be a hierarchy of "right." This, frankly, is probably why manufacturers are reluctant to publish specifications for the things that distinguish them, because they are distinguishing themselves by deviating from accuracy. Not that there's anything wrong with that if it pleases listeners. Using FR as a simple example, if Spectral (I'm making this up, Spectral fans) has a bit of extra juice in the upper mids, they may have found their market with a sound that sets them apart. That doesn't mean they want their competitors looking more "accurate."

What I still struggle with is my simple example above. Assuming that Krell, Levinson and Spectral all have more than enough power and the appropriate output impedance to drive the load -- there's good driver control, good dynamics, plenty of headroom, no clipping -- most "unsubtle" differences should be in FR. And of course there is no "juice" as referred to above, in a Spectral, there is only volume across the frequency response range. These differences should show up. Not in narrow, manufacturer-serving marketing specs, but in good measurements. Surely some pub has measured well? We should be able to point to what you hear. And if we can't, we should, frankly, question it.

Correct. Science goes so much, and beyond measurements, manufacturers simply voice their equipment as they wish.

The engineeering this far without breaking a sweat. These companies are voicing their products based on measured, defined specfications within an acceptable range of variation, or they would never get them to all sound alike, never establish a "voice."

Tim

 

[microstrip]

Then I must believe that there is a measured manufacturing specification that defines the differences. It is not reasonable to believe that Krell, Levinson and Spectral voice each amp by hand.

No, it is not reasonable.

And of course this raises more questions: what is subjective/what is high fidelity?

I do not see any connection with the previous sentence.

If three top amps sound (and measure) very noticably different, unless they're just off by exactly the same % in different directions, by definition there is going to be a hierarchy of "right."

No, because you can not quantify the % and you do not know which is the measurement that correlates with the difference.

This, frankly, is probably why manufacturers are reluctant to publish specifications for the things that distinguish them, because they are distinguishing themselves by deviating from accuracy.

No. They all know they can not be 100% accurate, and focus on being more accurate each in a set of aspects that defines their sound balance. I know what I get and do not get using a Krell, a Dartzeel or an Audio Research. It would be nice to have it all, but I am not allowed to get your favorite AV receiver.

Not that there's anything wrong with that if it pleases listeners.

No, there is not.

Using FR as a simple example, if Spectral (I'm making this up, Spectral fans) has a bit of extra juice in the upper mids, they may have found their market with a sound that sets them apart. That doesn't mean they want their competitors looking more "accurate."

I think you are joking, but forgot the smile. Going in your line I am tempted to ask if your favorite AV receiver with a bit of extra juice in the upper mids sounds like a Spectral.

What I still struggle with is my simple example above. Assuming that Krell, Levinson and Spectral all have more than enough power and the appropriate output impedance to drive the load -- there's good driver control, good dynamics, plenty of headroom, no clipping -- most "unsubtle" differences should be in FR.

Wrong. Read the reviews with the measurements. They are all flat.

And of course there is no "juice" as referred to above, in a Spectral, there is only volume across the frequency response range. These differences should show up. Not in narrow, manufacturer-serving marketing specs, but in good measurements. Surely some pub has measured well? We should be able to point to what you hear. And if we can't, we should, frankly, question it.

It is what you are doing, although you know there is no answer at this moment. I can live with it while listening to high-end amplifiers. May be you you should consider subsidizing a fund to research on it.

The engineeering this far without breaking a sweat. These companies are voicing their products based on measured, defined specfications within an acceptable range of variation, or they would never get them to all sound alike, never establish a "voice."

I repeat myself – unless you understand all the phases of the creation and manufacturing an high-end electronic device – starting with the manufacture and specification of components, understanding topologies, creation of a model, implementation and many other aspects you will never be able to see the truth. There is currently no universal measurement for all of them – a measurement that has some correlation with a sound parameter in one architecture can be meaningless, inappropriate and even erroneous with other designs.

 

[FrantzM]

Hi

Give the same blueprints to two different engineers or two different Project Managers and you are bound to see differences in the final products. The Leaders personality , quirks and philosophy will show up in the results. Same with Audio equipment. The sets or measurements and philosophies is wide and so are the way the leaders/companies/engineers/designers interprets them and steer their product or results. I don't believe it is a mater of always "voicing" their equipment that suggests that they sound the same when frst produced but the voicing really but the way the designers see the reproduction to be. Often it can become quite cynical or market-wise (depending on the point of view) with designers voluntarily favoring a type of sound which often is far from the signal .. The paradox is that we see this mostly in High End Audio , the area that should have been the epitome of High Fidelity.

I should add that despite shouts of the contrary our tastes as audiophiles are more similar than we want to admit or that the Industry would admit to. Their are aspects, I believe most audiophiles favor, because of the education brought in by listening intently to the sound of music (often opposed to listening to music itself). We have come to be educated, cosnciously or not to recognize some qualities of good reproduction, I have seen people move from their Bose WaveRadio to full fledge Audiophiles gears in a few years. The experience of companies such as Harman and others do bring his to the forefront however hard we push against such.. We favor smooth FR, regularity of modes, the notion of spaciousness, extension and in particular verisimiltude of midrange reproduction. We all do and that make us appreciate if not necessarily love a given equipment that acquit itself of these. We add our preferred spices later.

On the subject of what we call preferences I tend to think that there is a lot of psychology at play. Some brands we like and some we don't .. and once we know the brand our mental/psychological filters kick in. I know some brands I don't like and don't even know why .. I am not sure I like ML, Krell or Mac Intosh and I will be hardpressed to tell someone why. It may come from experiences I have had with poorly set-up systems back 20 years ago or simply of the music played or .. the gear themslves... I do respect some of these krell but in I have not found myself ever auditioning save for My Krell Home THeater which I got because the price was too good to pass on , it was after all Home Theater (yeah there you can feel the condescending tone ) I have never even when possible ever auditioned anything from these companies, mea culpa. It may come from experiences from almost 20 years ago. have I heard a an ML, Krell or Mac Intosh in my 2-Channel system ???? never!!! I have heard them in people systems but have never taken a Krell and audition it in my own system and my own room .. I know some here at WBF simply don't like Magico even without ever having heard one pair in their entire existence or in their dreams . Prejudice? You bet and we audiophiles have them in industrial quantities ...

And from that we start claiming we believe our ears and we start deviating from basic scientific tenets in our posts. We go to logical fallacies such as " it can be reproduced repeatedly but it can't be measured" .. and the whole line of interesting and brain-wrecking affirmations ... meanwhile the manufacturers do their things, measure them at infinitum ... charge us what we want to pay to feel satisfied while we're making of differences often (not always, I maintain there are diferences and some are reliably audible even between electronics) we can't perceive reliably the "Night and Day" and "Not Even Close" that populate audiophile conversations...

 

[Mark Seaton]

I haven't had time to read the whole thread, but saw these last few points wanted to add a few comments...

Then I must believe that there is a measured manufacturing specification that defines the differences. It is not reasonable to believe that Krell, Levinson and Spectral voice each amp by hand. And of course this raises more questions: what is subjective/what is high fidelity? If three top amps sound (and measure) very noticably different, unless they're just off by exactly the same % in different directions, by definition there is going to be a hierarchy of "right." This, frankly, is probably why manufacturers are reluctant to publish specifications for the things that distinguish them, because they are distinguishing themselves by deviating from accuracy. Not that there's anything wrong with that if it pleases listeners. Using FR as a simple example, if Spectral (I'm making this up, Spectral fans) has a bit of extra juice in the upper mids, they may have found their market with a sound that sets them apart. That doesn't mean they want their competitors looking more "accurate."

What I still struggle with is my simple example above. Assuming that Krell, Levinson and Spectral all have more than enough power and the appropriate output impedance to drive the load -- there's good driver control, good dynamics, plenty of headroom, no clipping -- most "unsubtle" differences should be in FR. And of course there is no "juice" as referred to above, in a Spectral, there is only volume across the frequency response range. These differences should show up. Not in narrow, manufacturer-serving marketing specs, but in good measurements. Surely some pub has measured well? We should be able to point to what you hear. And if we can't, we should, frankly, question it.

Tim

Wrong. Read the reviews with the measurements. They are all flat.

Most of the measurements posted for amplifiers are into ideal loads, and many like to dismiss small variations when actually measured into a real or simulated loudspeaker load. Now remember that the complex impedance of a loudspeaker has a small to not-so-small degree of dynamic change. My curiosity was piqued when I was calibrating a system and we swapped in a new amp for the center channel which arrived later in the day of the calibration. The original was a Krell KAV series amp bridged. I had a measurement right before we swapped it out for a Bryston 28BSST. The TEF-25 measurement is extremely consistent when you repeat measurements. I measured the center channel with the Krell twice. We then swapped in the Bryston, not moving the mic or speaker. Took a measurement and adjusted the gain. The resulting curves differed in a small, but potentially audible amount. A broad band HF difference of 0.3dB over at least 1.5 octaves was clearly visible.

I can't know if a very long balanced cable run and the interaction with upstream components played any role, say above 6-10kHz, but the measured response was most certainly different after the swap. While the 0.3dB difference is not huge by any means, the wider the bandwidth, the more this can be audible. This was also just a comparison with one speaker. Some "hi-fi" speakers are quite cavalier in what they assume amplifiers to be able to drive, and I expect the differences would increase accordingly.

Having met a handful of electronics designers, I suspect more of the "family voicing" comes from similar design approaches and priorities which lead to similar interactions and priorities. I know that to be quite true in loudspeaker design where we have the opposite problem of easily finding a plethora of differing measurements, yet difficulty in straight forward correlation to what is heard.

 

[jkeny]

Mark, good to see a real designer who can give real experiences here rather than armchair experts who seem to think they know what real designers do!
Thanks for setting the record straight! The real world does not operate along the absolutes that are being talked about in this thread!

 

[rbbert]

As I posted previously, it would be good to see some comparison testing of amps playing complex signals (like music, at least) into real world loads (such as Mark describes) to get a better idea of what measurements might correlate better with amplifier sound.

Despite often posted (and printed) comments about a specific amplifier's "sound", it's much more likely that it is a speaker/cable/amplifier combination that has a specific "sound", and changing any of the three will change that "sound". In fact, conscientious reviewers are usually aware of that and make specific comments about their findings when making changes of that type.

 

[Gregadd]

Not bad but the "neurological" portion is clumsy and simplistic. Note, however, the use of the word "induced." In other words, it does not exist as a stimulus but is a percept induced by a range of stimuli. The induced pain is real regardless of the stimulus.

No offense Kal but this subject is quickly becoming a "noxious stimulus." Smile

 

[Phelonious Ponk]

Mark, good to see a real designer who can give real experiences here rather than armchair experts who seem to think they know what real designers do!
Thanks for setting the record straight! The real world does not operate along the absolutes that are being talked about in this thread!

I don't think Mark set straight the record you have in mind, John. The absolutes in the later half of this thread are between those who say two components can measure exactly (apologies to mep) the same and sound completely different, and those who say that is exaggeration and would make it all but impossible to produce a consistent product.

In Mark's example, the difference -- 0.3 dB -- is extremely subtle, and he did, in fact, measure it. So rather than "two components can measure the same and sound different," a more accurate Audiophile core value, based on John's anecdote, would be two components can measure slightly differently and sound slightly different.

I have no argument with that; my argument is, and always has been with the overblown, hyperbolic absolutes.

Tim

 

[Robh3606]

In fact, conscientious reviewers are usually aware of that and make specific comments about their findings when making changes of that type.

Many do list a comprehensive list of review equipment.

In Mark's example, the difference -- 0.3 dB -- is extremely subtle, and he did, in fact, measure it.

That does make sense would like to see an impedance curve on the speaker in question.

Rob

 

[jkeny]

OK, Tim, you have used the last 5 or 6 pages to complain about this hyberbole - that's hyberbole in itself, don't you think?