[darrenyeats]

Ah, yes you put it so brutally. Of course it is only your opinion. My opinion is that there are some audio phenomena  that engineers have trouble measuring such as perceived space around the instruments and holographic presentation. Now you said that there was just one parameter that mattered and then you proceeded to list two: low noise and distortion. Could your linear mind grasp that there could be other types of measurement that science has yet to discover or are you of the opinion that there is nothing else to learn? Certainly that is not being open minded which I think is a prerequisite for advancement in any scientific endeavor.

-Roy

Thanks for your post.

Well transparency is just one thing and it means lack of distortion. There are difference kinds of distortion like harmonic distortion (signal-correlated) and noise (signal-uncorrelated). The less distortion of any kind the better. It's a one-dimensional problem in the sense that tackling one kind of distortion doesn't, in principle, cause another kind of distortion. Reducing distortion of any kind is a good thing, period. Compare this to an automobile which has multiple dimensions of performance some of which are, in principle, at loggerheads e.g. speed vs economy. That is what I meant by one thing. Hope that clarifies.

Ask ten race car drivers how they like their car set up and they will all
tell you something different as well, so if it just came down to data, race
engineers wouldn't have much to do.

Ah but I already said that audio sources and interconnects are much simpler than cars. It's just a signal and getting as close to transparent as possible is where it's at. All the flavour of the music is contained in the signal. Most of the argument is getting people to acknowledge that all of the music (including holographic soundstage, PRAT, liquidity, chocolateyness etc) is in the signal, there isn't some bit of soul that somehow lives outside of the signal. Get the signal as accurate as possible and the job is done.

Of course some people don't like how accurate sounds, like some people don't like the colour black. However the meaning of black doesn't change just because some folks don't like it so much. Same with transparency, it means lack of distortion.
Darren

[Steve]

"There is another view - and I shall put it brutally - that a lot of the perceived differences between electronics live in the mind of the listener. This is the principle reason people perceive that "the music" doesn't get reproduced despite measurements showing that the signals are being accurately produced."

>Can you prove that for any, some, even most of the cases? Of course not. You read it somewhere.

"In pro audio signal=music. In audiophile land, somehow music is different to signal. You can have an accurate, clean signal, but somehow it doesn't capture the music."

>Pro audio is about the last place I would look. Akin to amateur hour. All one has to do is listen to the poor quality CDs that are coming out. And I am not just talking about compression.
>And yes, static conditions, such as a signal is quite different than music.

"As I said audiophilia is a philosophy and more of a sociological phenomenon than anything to do with a discussion of physics or engineering principles."

>Can you tell us what your credentials are regarding those areas?
>Speaking of engineering principles, would you explain why a DC coupled gainstage in a typical component is not accurate to the music? Under what conditions would it be?

Any "reviews" that aren't verified blind don't mean a hill of beans to me. Would you like it any clearer?

>Are you talking AB and ABX testing? If so, you are assuming that subjective AB or ABX testing is accurate. Can you prove it is? Can you provide a link to such a test?

"JUST ONE - and that is low noise and distortion (also referred to as transparency)."

>That is two.   Want to add anymore?


"YMMV. Please take it in the spirit intended, which is good-natured.
Darren"

>Will do. Hope you do from me as well. Take care my friend.

Cheers.

[darrenyeats]

>That is two.  Want to add anymore?

Hehe, see previous post for clarification.

As for proof, that's a strong word. I can't prove a negative...I'm the skeptic here, remember? Convince me, you'll find I am quite open-minded despite the bluster.

I'm just putting my opinions out there to see if anyone agrees. (Apparently not, but that's fine.  ) Opinions, including mine and yours too, can change but these are mine.
Darren

[Steve]

"Well transparency is just one thing and it means lack of distortion. There are difference kinds of distortion like harmonic distortion (signal-correlated) and noise (signal-uncorrelated). The less distortion of any kind the better."

>How about hysteresia distortion, DA and ESR type of distortions in capacitors?

"Reducing distortion of any kind is a good thing, period."

>I agree Darren, I like as accurate of sound as possible.

"Ah but I already said that audio sources and interconnects are much simpler than cars. It's just a signal and getting as close to transparent as possible is where it's at. All the flavour of the music is contained in the signal. Most of the argument is getting people to acknowledge that all of the music (including holographic soundstage, PRAT, liquidity, chocolateyness etc) is in the signal, there isn't some bit of soul that somehow lives outside of the signal. Get the signal as accurate as possible and the job is done."

>That is the challenge, getting the signal right, 10db, 30db, 40db or more down. ICs are a little more complex. For instance one IC may "have the music" in one's face, at the speakers, while another has the music layered back well behind the wall (let alone the speakers). Measure the capacitance etc, no difference, but the difference in sound is quite pronounced.

"Of course some people don't like how accurate sounds, like some people don't like the colour black. However the meaning of black doesn't change just because some folks don't like it so much. Same with transparency, it means lack of distortion. "

I agree Darren. Some prefer different but that is there choice. I, like you, prefer accuracy and naturalness.

Take care Darren.

[Steve]

>That is two.  Want to add anymore?

Hehe, see previous post for clarification.

As for proof, that's a strong word. I can't prove a negative...I'm the skeptic here, remember? Convince me, you'll find I am quite open-minded despite the bluster.

I'm just putting my opinions out there to see if anyone agrees. (Apparently not, but that's fine.  ) Opinions, including mine and yours too, can change but these are mine.
Darren

No problem Darren. Taken in good spirit. No harm meant on my end.
Take care.
Steve

[Daygloworange]

I'm just putting my opinions out there to see if anyone agrees. (Apparently not, but that's fine.  ) Opinions, including mine and yours too, can change but these are mine.
Darren

I'm of very similar opinions to yours Darren. My quest is for transparency. I want as pure, and distortion free, a signal as I can get.

Are you talking AB and ABX testing? If so, you are assuming that subjective AB or ABX testing is accurate. Can you prove it is? Can you provide a link to such a test?

Here a link where Arny himself debates it with John Atkinson from Stereophile:

http://www.stereophile.com/news/050905debate/

The ABX protocol is a pretty darn good one IMO. I also agree that blind testing (even single blind), has tremendous merit.

If you were blindfolded and were to listen to a live instrument vs a recording of that instrument in the same room, you would be able to tell which was the live instrument vs the recording.

Your ears can be quite telling. 

Of course, I have the benefit of hearing things I recorded, and have the advantage of knowing what they sound like in real life.

But there are good recordings out there (particularly minimalist recorded classical solo instrument pieces) that one can use as a benchmark, if they are quite familiar with how they sound naturally.



Cheers

[*Scotty*]

Darren, Do you chose your audio equipment on the basis of how it measures or how it sounds?
Scotty

[Daygloworange]

I choose it based on how it sounds, but curiously, the specs and measurents would bear that linearity is what sounds more transparent.

Cheers

[darrenyeats]

Darren, Do you chose your audio equipment on the basis of how it measures or how it sounds?
Scotty

I changed transports in the last year based on a double blind listening test. The old gear was much more expensive but, looking back on it now, quite old. A lot of the new sources perform to a very high measured standard compared to old digital, but anyway I found the old transport sounded better until I did the blind test. Blind I preferred the SB3 That's why I'm quite hot on blind listening now.

As you can see from the footer I'm using the SB3 as a player now, but there are other reasons for that.
Darren

[opaqueice]

Why does amplifier A do a fantastic job with reproducing all the tonal
and spatial cues of an acoustic intrument, while amplifier B does not.

Not to speak for Darren, but I think you might be missing the point just a little.  You're doing precisely what the Porscheophile in his example did - claim that something is true despite all the objective evidence against it. 

There are effective ways to argue against his position, but yours is not one of them.

[*Scotty*]

opaqueice,I think you saying that Jeff didn't really hear the tonal differences between Amplfier A and Amplfier B that he said he did. In as much as I have also heard differences between amplifiers tonal and spacial presentations I am inclined to believe his statements on this subject. As far as Darrens' position is concerned,he hasn't been
particularly consistent or concise in stating his position and I for one am not entirely sure what he advocates. If one is inclined to believe that all amps sound the same there is little to discuss and we are firmly in Julian Hirsch territory.
Scotty

[*Scotty*]

I will say that I have a problem with Jeffs' car analogy. While different cars do have a different feel to them this is directly related to engineered performance parameters that are entirely under the control of the cars' builder. The car is defined by its' design specs. If you want it to do something different you know what to change to accomplish your objective. There are no subjective performance characteristics.  You can reduce a cars objective performance to a laptime on a track and a G-force plot of the lap from the onboard accelerometers. 
   The same cannot be said of audio. It exists as a subjective entity with objective factors that influence its' performance , at the end of the day you have to listen to it to experience it.
Scotty

[TONEPUB]

Why does amplifier A do a fantastic job with reproducing all the tonal
and spatial cues of an acoustic intrument, while amplifier B does not.

Not to speak for Darren, but I think you might be missing the point just a little.  You're doing precisely what the Porscheophile in his example did - claim that something is true despite all the objective evidence against it. 

There are effective ways to argue against his position, but yours is not one of them.

I don't think I'm missing the point at all.  I think I'm making the point exactly.  I spend about 12 hours
a day listening to some of the best (and not so best) gear the industry has to offer.  If there was a simple
set of measurements, why wouldn't everyone just use them and design their gear accordingly?

I also spend a pretty fair amount of time listening to live music, amplified and acoustic along with
keeping a drum set and a Steinway around the house.  (not that I have any major musical talent,
but it does help to have a few acoustic instruments for an absolute reference)

I hate to be so argumentative about this, but if there really was a "right" way to do this, don't you
think after about 40 years now, someone would have figured it out and everyone else would have
copied it?

Every major high end designer I've ever talked to thinks that they have done it right.

That's why CJ sounds like it does, which is different from ARC, and Levinson, and Krell
and ModWright, etc, etc.  I've heard some gear sound more like others, but they pretty
much all have their own signature sounds.  Same with speakers.

Mix that in with the fact that most people perceive sound differently and have their own
set of preferences and the measurement thing becomes pretty worthless.  If all recordings
were done equal and everyone had exactly the same hearing, there might be a place
to start.

What I don't understand is why everyone gets so freaked out about this.

There is so much great stuff out there, find a flavor you like and enjoy it. 
Personally, I've got three great systems and I could care less if anyone likes what I do.

Again, it's really like cars.  Even though I drive BMW's, I can have a ton of fun driving
a Porsche or a Hemi Cuda or a go kart for that matter.  I just enjoy cars.  Same thing
with audio.  I can have fun listening to my system or yours.  And if you enjoy your
system, that's all that matters to me.  I don't care about measurements.  And I have
a tech editor with a PhD in Physics and an EEE.  We've measured more gear than
we've cared to over the years.  For me it's really about the music.

[darrenyeats]

As far as Darrens' position is concerned,he hasn't been
particularly consistent or concise in stating his position and I for one am not entirely sure what he advocates. If one is inclined to believe that all amps sound the same there is little to discuss and we are firmly in Julian Hirsch territory.

Well my point isn't about speakers, that is a different discussion due to acoustic spaces. My point is about sources and interconnects. Amps are similar but different because they have to face unpredictable, complex and sometimes difficult loads especially with passive crossover speakers. When you use amps in an active set up a lot of considerations become relaxed and you don't need monster power or current. However, into any given load, the aim of an amp is just as clear as for a source or interconnect i.e. zero distortion (transparency). It's just that doing it is more difficult and might require more compromises from amp designers. That's one benefit of active speakers (to amp designers).

What I'm advocating is avoiding phrases like "this source or that amp produces an amazing holographic soundstage". The next statement may sound like nit-picking, but bear with me because the difference is subtle but profound. It is the signal that contains the soundstage, and the amp can only reproduce what is there.

Now, I'm sure that's what you guys meant, you would agree with that. But it's a little more than semantics, because when you place the signal at the heart of your philosophy you get a subtle shift in thinking. The subtle shift is realising that all a component needs to do is maintain the signal undistorted, because it's all about the signal. There is no magic powder in a source, IC or amp that you can dust onto the sound to create magic. There is no aura in a box of electronics that can impute grandeur or liquidity into the music. This may sound undramatic, but all the human emotion, sweat, tears and genius of the original musical event is being borne on an electrical signal. Undramatic but, it turns out, just the truth, the whole truth and nothing but the truth. So it takes the mystique away from all this audiophilia and affirms the reality, which is that if a component has low distortion it is by definition better and it's doing all a component can usefully do. If we are asking for more than low distortion from a source, interconnect or amp (provisos stated) we're asking for something that is nonsensical.

So my position is about mainly sources and interconnects (and amps with provisos above). First point, my position is the principle that low distortion to the signal is the only performance parameter that matters, because the signal defines the listening experience and nothing can be reproduced faithfully that is greater than the signal itself.

Second point.

opaqueice,I think you saying that Jeff didn't really hear the tonal differences between Amplfier A and Amplfier B that he said he did. In as much as I have also heard differences between amplifiers tonal and spacial presentations I am inclined to believe his statements on this subject.

My point has nothing to do with whether currently available measurements are "all there is to it". There may be new ways of measuring distortion found in the future that explain the differences that people hear. My point is, if we are ever to progress, the only way to do that is through blind listening judgments. (And my first point was if anything new and useful is to be discovered it can only be in the sense of reducing distortion.) To be clear I believe anyone when they say they hear differences. I hear differences all the time sighted, so the only question is why things are heard. Listening blind is the only way to know if something is heard due to a real audibile difference. Only blind listening can validate distortion measurements, old or new, that correlate to audible qualities. Sighted listening can guide, but only blind listening can validate.

All IMHO. 
Darren

[doug s.]

...What I'm advocating is avoiding phrases like "this source or that amp produces an amazing holographic soundstage". The next statement may sound like nit-picking, but bear with me because the difference is subtle but profound. It is the signal that contains the soundstage, and the amp can only reproduce what is there.

Now, I'm sure that's what you guys meant, you would agree with that. But it's a little more than semantics, because when you place the signal at the heart of your philosophy you get a subtle shift in thinking. The subtle shift is realising that all a component needs to do is maintain the signal undistorted, because it's all about the signal. There is no magic powder in a source, IC or amp that you can dust onto the sound to create magic. There is no aura in a box of electronics that can impute grandeur or liquidity into the music. This may sound undramatic, but all the human emotion, sweat, tears and genius of the original musical event is being borne on an electrical signal. Undramatic but, it turns out, just the truth, the whole truth and nothing but the truth. So it takes the mystique away from all this audiophilia and affirms the reality, which is that if a component has low distortion it is by definition better and it's doing all a component can usefully do. If we are asking for more than low distortion from a source, interconnect or amp (provisos stated) we asking for something that is nonsensical.

all this is great, except i think you are wrong!     why?  because if everything is 100% perfect, undistorted, transparent, then what you will get is a perfect reproduction of A RECORDING.  which may NOT sound the same as the live musical ewent itself.     which is why, imo, electrical components that in fact do measure extremely closely, w/near-identical specs of all known engineering parameters, may in fact sometimes sound a lot different.

ymmv,

doug s.

[doug s.]

I choose it based on how it sounds, but curiously, the specs and measurents would bear that linearity is what sounds more transparent.

Cheers

except that this doesn't explain components whose sound you don't like, tho their specs & measurements are equal or better to what you do like...   

doug s.

[darrenyeats]

i think you are wrong!

I have been wrong before - don't tell my wife. 
Darren

[chadh]

all this is great, except i think you are wrong!     why?  because if everything is 100% perfect, undistorted, transparent, then what you will get is a perfect reproduction of A RECORDING.  which may NOT sound the same as the live musical ewent itself.     which is why, imo, electrical components that in fact do measure extremely closely, w/near-identical specs of all known engineering parameters, may in fact sometimes sound a lot different.

ymmv,

doug s.

Don't you mean "liwe musical ewent"?  And "ymmw"?

By the way, I don't understand your argument at all.  It may be true that someone does not like the sound of a perfectly reproduced recording, and prefers to hear the recording reproduced with various distortions.  But I don't see why that should imply some discontinuity in the mapping from distortion to perceived sound quality.  It would still seem perfectly plausible that two sources that delivered signals that were "close" would sound similar.  If this were the case, then Darren's suggestion of blind testing would allow you to determine which particular types of signal imperfections subjectively sound good to you.

Now, it may or may not be true that that two sources delivering similar measured signals will produce similar sounds.  If they don't, there are two potential explanations:
1) our ability to measure the signal is compromised (measurements may be inaccurate, inappropriate or incomplete); or
2) the mapping from signal characteristics to perceived sound quality is discontinuous

In order for us to infer very much about how a component sounds from measured specifications, it would need to be the case that neither of these failures occurs.  I have no idea how one would establish either of these propositions.  Then again, I'm not sure that one is at very much of a disadvantage if one proceeds assuming these failures do not occur. 

Chad

[Daygloworange]

I choose it based on how it sounds, but curiously, the specs and measurents would bear that linearity is what sounds more transparent.

Cheers

except that this doesn't explain components whose sound you don't like, tho their specs & measurements are equal or better to what you do like...   

doug s.

It does explain it to me Doug, because I listen to recordings I made, and have the knowledge of the what the source sounds like sitting right next to me. So when I change out gear, and listen to my own recording, I know which piece of gear sounds truer to life.

Based on my experiences, I have found a direct correlation to equipment specs playing a very important role in how transparent something sounds.

all this is great, except i think you are wrong!     why?  because if everything is 100% perfect, undistorted, transparent, then what you will get is a perfect reproduction of A RECORDING.  which may NOT sound the same as the live musical ewent itself.     which is why, imo, electrical components that in fact do measure extremely closely, w/near-identical specs of all known engineering parameters, may in fact sometimes sound a lot different.

ymmv,

doug s.

I think your assuming only a 100% transparent playback of current (ie: crappy recorded) recordings.

Theoretically, if the "ewent"  was  recorded and/then played back with 100% accuracy with electronic equipment. It would in fact be a perfect duplicate.

Cheers

[chadh]

I find the analogies between audio gear and cars disturbing.  To see why, let me pose a question for Darren.

Do you take blind driving tests before purchasing a car?

For some reason, I suspect that no blind driving tests take place.  I suspect, instead, that something like the following takes place: 

1.  He compares published technical information about the cars in order to narrow down options;
2.  He conducts sighted tests of cars on some shortlist. 
3.  With relatively high probability, the sighted test confirms the rankings of the cars that he formed through his comparison of technical specifications - although, with non-negligible probability he revises his rankings of the cars due to some perceived but wholly unsubstantiated difference between the vehicles.

If this process is good enough for making those vital choices between Porsche and Ferrari (or BMW and Honda Civic), why does it become so despicable in audio?

Here's another thing that seems weird to me. 

If you're prepared to do blind listening tests, then your listening test must be the only thing that matters.  If you prefer component A over component B, without the influence of anything that would generate biases for you, then it seems you truly prefer component A. 

On the other hand, when you take only sighted tests, you need to acknowledge the possibility that biases exist in your personal assessment that have nothing to do with the component's ability to satisfy your needs or desires. It is in this case that other information should be relevant to you - for example, measurements and technical specifications.  In effect, measurements and technical specifications are really only valuable when you conduct sighted test.  So why is it that "objectivist" types always crusade for both blind testing and a vast array of testing results?  The test results seem redundant for the consumer if the consumer is prepared to undertake blind tests.

Chad