[Karsten]

I have been thinking a lot about this issue as well, and I think that at least part of the explanation can be found here: http://www.audiocircle.com/index.php?topic=39346.0

There are of course several other scenarios which may cause components to match or not, and also poorly designed and inferior components which can course things to go wrong.

Karsten

[miklorsmith]

I think the main problem is that there are some offended that things are measurable.   

Interesting take.  I can only explain my own views, but I assure you I am not offended by anything said here.  Even when I get indignant it's not personal.

I don't think anyone doubts that parameters of audio are measurable.  I think the operative differences focus on how many parameters there are that cannot be measured.  Some folks think there are such dimensions and others do not.  

One of my big difficulties with the "we can measure it" camp is what value that might have.  The end-user asks the question "what will this sound like" (predictive) and the response is "let's hear it first, then try to quantify reasons why it sounds that way" (descriptive).  Whether it is possible to describe sound after the fact is useless to the person asking the question.  Yes, in the long run the descriptive can help an engineer, but not the prospective buyer.

There are a couple of questions wrapped together which are being treated as one.  This question has come up many times before, with the usual suspects taking expected arguments.  I've never seen it spun under the "synergy" label but it seems little different than any of the previous discussions.

Maybe the phrasing should be a little different "can measurements predict synergy" instead of the original thread post, spinning the age-old question into something that consumers might be able to use.

[Kevin Haskins]

I think the main problem is that there are some offended that things are measurable.   

Interesting take.  I can only explain my own views, but I assure you I am not offended by anything said here.  Even when I get indignant it's not personal.

I don't think anyone doubts that parameters of audio are measurable.  I think the operative differences focus on how many parameters there are that cannot be measured.  Some folks think there are such dimensions and others do not. 

One of my big difficulties with the "we can measure it" camp is what value that might have.  The end-user asks the question "what will this sound like" (predictive) and the response is "let's hear it first, then try to quantify reasons why it sounds that way" (descriptive).  Whether it is possible to describe sound after the fact is useless to the person asking the question.  Yes, in the long run the descriptive can help an engineer, but not the prospective buyer.

There are a couple of questions wrapped together which are being treated as one.  This question has come up many times before, with the usual suspects taking expected arguments.  I've never seen it spun under the "synergy" label but it seems little different than any of the previous discussions.

Maybe the phrasing should be a little different "can measurements predict synergy" instead of the original thread post, spinning the age-old question into something that consumers might be able to use.

I think the measurements are something that are useful for engineering.   They will never be useful for consumers simply because there is no way to educate consumers to the level necessary for proper understanding of the subject.    I don't think its an issue of consumers being stupid, usually just the opposite.   They just don't have the specialized background to understand the entire subject.

[WEEZ]

I agree, Kevin.

(reminds me of a song by Tower of Power: The More You Know...The More You Know You Don't Know)

 

WEEZ

[Dan Banquer]

"I think the main problem is that there are some offended that things are measurable.   I'm not sure why that is an issue.   It doesn't hurt my feelings either way. "

That is an interesting observation Kevin and one that I would agree with, but I also readily admit that I do not understand why people get offended.
                d.b.   

[rajacat]

What wasn't measurable 50 years ago is measurable now or is measured more accurately. It follows that 50 years from now there will be more accurate and comprehensive measuring tools than are available now. Science keeps progressing so technicians might think that their instruments are the cat's meow now but they will probably be looked upon as being rather crude in the not too distant future.

Raj

[Kevin Haskins]

What wasn't measurable 50 years ago is measurable now or is measured more accurately. It follows that 50 years from now there will be more accurate and comprehensive measuring tools than are available now. Science keeps progressing so technicians might think that their instruments are the cat's meow now but they will probably be looked upon as being rather crude in the not too distant future.

Raj

That doesn't invalidate the methods we have now.    I often hear people dismiss technology or science based upon the fact that our knowledge of things is fluid and changing.    Just think, Newton's Principia is still valid today.   The methods and calculations are the same and newer levels of understanding will not invalidate the context in which Newtonian methods are useful.   They simply extend our understanding to situations that Newton's methods could not describe.   

The same with our test measurements.   We may get better levels of resolution but that doesn't mean it will be meaningful.   We have electron microscopes that allow us to take meaningful measurements down to microns. That doesn't have ANY effect on bridge building because we had the necessary measurement capability to design good bridges without the need for the electron microscope. 

I'd say the same is true with audio.   The tools we have now are more than up to the task.

[rajacat]

They might be up to the task but still the ultimate testing instrument is the ear. If it doesn't sound good then it really doesn't matter what the instruments say. Correct me if I'm wrong but I don't believe that the testing gear available takes into account quantum mechanics and the influence of subatomic particles on sound. Perhaps such phenomena as the sense of air and space that many hear when using tube gear but not feel/hear when using solid state amps could be explained with more sophisticated lab gear.

Raj

[opnly bafld]

"I think the main problem is that there are some offended that things are measurable.   I'm not sure why that is an issue.   It doesn't hurt my feelings either way. "

That is an interesting observation Kevin and one that I would agree with, but I also readily admit that I do not understand why people get offended.
                d.b.   

I am not sure that it is so much being offended.
How many people do you know that have access to or have a lab with all of the necessary equipment to perform said tests (assuming one even has the knowledge to use the data)?
Therefore, when trying to purchase an amp to go with speakers, as an example, going on what little data is available from the manufacturer the majority would have to use their ears as a reliable source for what works and what does not.
In cases such as this the numbers(specs) the average consumer has to make descisions on are basically worthless  IMO.
Hence we have comments such as "2 amps that measure the same sound different".

Lin

[Dan Banquer]

They might be up to the task but still the ultimate testing instrument is the ear. If it doesn't sound good then it really doesn't matter what the instruments say. Correct me if I'mquantum mech wrong but I don't believe that the testing gear available takes into account anics and the influence of subatomic particles on sound. Perhaps such phenomena as the sense of air and space that many hear when using tube gear but not feel/hear when using solid state amps could be explained with more sophisticated lab gear.

Raj

Just a couple of points I need to make here. Both Kevin and myself covered the distortion spectra as one of the more important reasons why some amps don't sound the same/similar. The distortion spectra applies to tube amps as well. As far as what you percieve? or what you think you percieive? and the reason you give for that perception?
I'm not going there, as it so far removed from anything I've ever experienced, learned, or read.
To the best of my knowledge this has really not much to do with subatomic particles and quantum mechanics, However if it that is what you feel is important to you for your enjoyment of music, then I guess it's important to you.
It's the best I can do on short notice.
               d.b.

[chadh]

I'm not sure whether this is off topic or not...but I thought I'd ask, and people could ignore it if they wanted.

I was sitting around thinking about how we know how much measurement can tell us about how something sounds.  And I wondered whether anybody has ever run the following kind of experiment, and whether it would reveal anything.

Suppose I played a a real piano, or guitar, or violin.  And I took a comprehensive set of measurements of those sounds.  The measurements are recorded, and the sounds themselves are also recorded.

Then I go to my computer, and somehow convert those measurements into a simulation of the sound.  Now, if the measurements are truly capturing EVERYTHING I care about in the music, then the measurements alone should enable me to recreate a perfect facsimile of the sound.  And that should mean that if I play back the recorded sounds and play back the facsimile that I have created, I should not be able to tell the difference between the two.  If I can distinguish between the facsimile and the recoding of the actual sounds, it would suggest that something other than what was measured is relevant to the sound.  Does this comparison seem valid?  Or are there too many variables for which I cannot control (like failures in the recording process of the original sounds)?

Chad

[Dan Banquer]

"I think the main problem is that there are some offended that things are measurable.   I'm not sure why that is an issue.   It doesn't hurt my feelings either way. "

That is an interesting observation Kevin and one that I would agree with, but I also readily admit that I do not understand why people get offended.
                d.b.   

I am not sure that it is so much being offended.
How many people do you know that have access to or have a lab with all of the necessary equipment to perform said tests (assuming one even has the knowledge to use the data)?
Therefore, when trying to purchase an amp to go with speakers, as an example, going on what little data is available from the manufacturer the majority would have to use their ears as a reliable source for what works and what does not.
In cases such as this the numbers(specs) the average consumer has to make descisions on are basically worthless  IMO.
Hence we have comments such as "2 amps that measure the same sound different".

Lin

Many of the magazines you may subscribe have the equipment, and the personnel to take the measurements needed. They also could take the time to explain them.
However, the great majority of audiophiles, music lovers, etc.etc. are simply not interested and would prefer not to be bothered.
It's the way it is, and to tell you the truth, I find it as frustrating as you coming from the other side of the coin so to speak, because I have posted here on so many subjects, and still I get the same questions over and over again.
Please feel free to do a search, I've posted a lot of stuff on grounding, stereo separation, and distortion. I've also referenced that at the beginning of this thread.

So if you think you are the only one who is frustrated...............guess again
                 d.b.

[rajacat]

Kevin,

Do you have enough confidence in your numbers that you could design a speaker to the best measurements you could get and then put it on the market without ever listening to it? If the numbers are that good then the human ear test is just superfluous. Of course I doubt that you would ever do that, it's just a discussion point.

Raj

[rollo]

Chad,
          Not being a scientific type, sounds reasonable to me.

  Guys, this is same old stuff all over again. It always turns out to be a battle between the measurees and the anti-measurees when someone asks the measurement question. No matter what format the battle begins.
      This one so far is not heated, but its getting there. ENOUGH ALREADY! Get a grip and realize it is fruitless.
        I mean a valid question would be, Can you measure what we hear? Can we hear what you measure? Ha,ha,ha

  rollo

[rajacat]

rollo,

Relax! I enjoy these musings and sometimes I actually learn something.

Raj

[konut]

Synergy is definately measurable. But only the owner of a system can measure it. The goesintos and goesouttas might be all up to spec, but if the combo is not satisfying to YOU then you got no joy. You can only go so far to ensure that components are matched and, for the most part, play nice together but the proof of the pudding is in the eating. To use the analogy previously used in this thread, a fart my be able to be measured, but for the most part, you are only satisfied by your own, and no one elses no matter how it measures.  

[Kevin Haskins]

Kevin,

Do you have enough confidence in your numbers that you could design a speaker to the best measurements you could get and then put it on the market without ever listening to it? If the numbers are that good then the human ear test is just superfluous. Of course I doubt that you would ever do that, it's just a discussion point.

Raj

I could do it and be relatively certain I'd have something that people would like.   I don't do it though because loudspeaker voicing also has a personal preference component.   

Most of this discussion I've been talking about amplifier measurements.   With loudspeakers you have the additional complexity of designing without one of the largest variables, the room.  I like to take a loudspeaker around to several different listening environments when I do a design because the room has such a profound effect on the final results.

[Daryl]

They might be up to the task but still the ultimate testing instrument is the ear. If it doesn't sound good then it really doesn't matter what the instruments say. Correct me if I'm wrong but I don't believe that the testing gear available takes into account quantum mechanics and the influence of subatomic particles on sound. Perhaps such phenomena as the sense of air and space that many hear when using tube gear but not feel/hear when using solid state amps could be explained with more sophisticated lab gear.

Raj

Hi Rajacat!

In fact the ear is not at all suitable as a test instrument.

There is not a single aspect of sound where the ear has any sort accuracy at all (the most remarkable thing about your ears is their considerable dynamic range).

The sounding good aspect doesn't hold water either.

Often you measure a system which is subpar by far but it sounds great, often better than the original.

A lot of equipment is designed not for accuracy but to spruce up the sound.

Distortion doesn't neccesarily sound bad.

The best and worst thing about the ear is how the brain processes signals.

On one hand your brain uses your ears which are not an exceptionally accurate instrument and through incredible OSP (organic signal processing) techniques are capable of gleening incredible details about what you are hearing which you might not think possible given how the information is aquired.

On the other hand your brain interpeting the information from your ears is also very impressionistic and causes people to imagine they are hearing things that aren't real and this is the basis for most audiophile talk.

Test instruments can easily measure acoustics and anything else.

An audio signal is a one dimensional data set and all pertinent information is contained within it.

What was involved in how the signal arrived their is not related in any way to being able to measure the quality of the signal.

Test equipment will get more accurate though it does not need to for the purpose of audio.

Where there is room and where advancement (in the area of measuring reproduction accuracy) will be made in the future is better understanding of how the brain processes data aquired by the ears.

While any type of reproduction error can easily be measured the signifigance of said errors is not so easy to predict and better understanding of how the brain processes sound information will shed new light in this area.

[Daryl]

I'm not sure whether this is off topic or not...but I thought I'd ask, and people could ignore it if they wanted.

I was sitting around thinking about how we know how much measurement can tell us about how something sounds.  And I wondered whether anybody has ever run the following kind of experiment, and whether it would reveal anything.

Suppose I played a a real piano, or guitar, or violin.  And I took a comprehensive set of measurements of those sounds.  The measurements are recorded, and the sounds themselves are also recorded.

Then I go to my computer, and somehow convert those measurements into a simulation of the sound.  Now, if the measurements are truly capturing EVERYTHING I care about in the music, then the measurements alone should enable me to recreate a perfect facsimile of the sound.  And that should mean that if I play back the recorded sounds and play back the facsimile that I have created, I should not be able to tell the difference between the two.  If I can distinguish between the facsimile and the recoding of the actual sounds, it would suggest that something other than what was measured is relevant to the sound.  Does this comparison seem valid?  Or are there too many variables for which I cannot control (like failures in the recording process of the original sounds)?

Chad

That is somewhat possible.

The most limiting componet is loudspeakers.

You have to reallize that when listening to a piano (or anything else) in a room that you have sound reaching you from every direction and the result is a complex soundfield that a single loudspeaker cannot create.

So to perform such an experiment you would want to use an anechoic environment and I think you would be suprised by the outcome.

There have been experiments done creating complex soundfields with speaker systems called 'linear arrays'.

This is a row of many loudspeakers each with their own record/plaback channel placed very close to each other that can reproduce a complex soundfield.

So far these have only been one dimensional arrays which are still not capable of completely replicating an acoustic field but even still the results are said to be uncanny.

[chadh]

I'm not sure whether this is off topic or not...but I thought I'd ask, and people could ignore it if they wanted.

I was sitting around thinking about how we know how much measurement can tell us about how something sounds.  And I wondered whether anybody has ever run the following kind of experiment, and whether it would reveal anything.

Suppose I played a a real piano, or guitar, or violin.  And I took a comprehensive set of measurements of those sounds.  The measurements are recorded, and the sounds themselves are also recorded.

Then I go to my computer, and somehow convert those measurements into a simulation of the sound.  Now, if the measurements are truly capturing EVERYTHING I care about in the music, then the measurements alone should enable me to recreate a perfect facsimile of the sound.  And that should mean that if I play back the recorded sounds and play back the facsimile that I have created, I should not be able to tell the difference between the two.  If I can distinguish between the facsimile and the recoding of the actual sounds, it would suggest that something other than what was measured is relevant to the sound.  Does this comparison seem valid?  Or are there too many variables for which I cannot control (like failures in the recording process of the original sounds)?

Chad

That is somewhat possible.

The most limiting componet is loudspeakers.

You have to reallize that when listening to a piano (or anything else) in a room that you have sound reaching you from every direction and the result is a complex soundfield that a single loudspeaker cannot create.

So to perform such an experiment you would want to use an anechoic environment and I think you would be suprised by the outcome.

There have been experiments done creating complex soundfields with speaker systems called 'linear arrays'.

This is a row of loudspeakers placed very close to each other that can reproduce a complex soundfield.

So far these have only been one dimensional arrays which are still not capable of completely replicating an acoustic field but even still the results are said to be uncanny.

Thanks for your reply.

I was imagining that the loudspeaker issue would be mostly irrelevant, though.  I envisaged a comparison between the recording of the music (played through the loudspeakers) to the synthesized sound (played through the loudspeakers) - so both sounds compared would face the same constraints of the loudspeakers.  Maybe this creates other difficulties though.

Chad