Do all amps sound the same?
What are people's general thoughts on this article by Dave Clark that seems to prove that a widely differing amps actually sound the same under controlled conditions:
http://bruce.coppola.name/audio/Amp_Sound.pdf
There is even an amp challenge that enables one to demonstrate that they can actually hear the differences between amps that it turns out nobody has passed to-date. Certainly worrying if you have just spent $3000 on a monster amp:
http://www.tom-morrow-land.com/tests/ampchall/index.htm
Want to weigh up the arguements before making any accurate judgement.
Kevin
Discussion started by Acoustic Insight , on 852 days ago
Also, please re-read Josh's comment:
"It's sort of akin to a passionate art collector deciding that a Manet hanging on the wall is no longer worthy of admiration because someone presented evidence that the artist died of syphilis."
He is not comparing a painting to an amp. Rather, he is saying (my interpretation--correct me if I am wrong, Josh) that why would anyone stop loving a Manet that they have always love, just because they find out that artist that painted it died of a disease that is considered distasteful?
Or, in my words, if one hears a difference, and is pleased, why would they give a rats ass what others think?
I, for one, have found over the last 25 plus years, and hundreds of thousands of dollars spent, to trust my ears, and if someone else does not agree with my assessment, oh well, we amicably agree to disagree.
Lee
"It's sort of akin to a passionate art collector deciding that a Manet hanging on the wall is no longer worthy of admiration because someone presented evidence that the artist died of syphilis."
He is not comparing a painting to an amp. Rather, he is saying (my interpretation--correct me if I am wrong, Josh) that why would anyone stop loving a Manet that they have always love, just because they find out that artist that painted it died of a disease that is considered distasteful?
Or, in my words, if one hears a difference, and is pleased, why would they give a rats ass what others think?
I, for one, have found over the last 25 plus years, and hundreds of thousands of dollars spent, to trust my ears, and if someone else does not agree with my assessment, oh well, we amicably agree to disagree.
Lee
@Lee - you distilled the gist of my “painting” comment perfectly. It is difficult to imagine that it would need further elucidation.
Kevin, I agree with many of your well-articulated comments about the purpose and methodology of ABX testing. Perhaps the emotional responses and flat denials here can be accounted for because many in the audio community (including myself) feel a dizzying sense of accumulated fatigue every time this old warhorse is trotted out in forums, like a cockroach that refuses to die despite being stepped on repeatedly. The “tubes versus solid state” and “analog versus digital” debates are equally pointless. To wit: I can’t see how ABX testing, or the issue of amps (or any other componentry for that matter) sounding potentially the same, has any practical value in common audio practice, at least as I have observed it and engaged in it on the consumer side of things, not necessarily as a designer or manufacturer. If one has the goal of presenting a scientific paper to the AES, of conducting studies in psychoacoustics or audiology, of rigorous product development where only one variable at a time is manipulated, or of engaging in endlessly fascinating discussions about epistemology, then yes, I can see that ABX testing is a useful tool. That being said, I have four problems with ABX testing and the sorts of audio-related questions to which it is commonly (mis)appropriated:
1.) It is impractical. The key to good and meaningful and useful science is asking the right questions; trying to establish something as broad as whether “all” amps sound the same is absurdly impractical. If you want to find out whether a small pool of three or four amps sound the same, that’s spiffy; ABX away. But to attempt to extrapolate those results across all amplifiers, with their endless architectural and upstream / downstream component permutations, would violate the tenets of the very sort of scientific procedure that was presumably a sacrosanct component of the original experiment, would it not?
2.) It is irrelevant. For the sake of discussion, let’s say all of the practical issues have been somehow overcome: working samples of every amp that is or ever was on the market have been obtained, and a panel of judges has been assembled who, by some miracle, are capable of enduring the Sisyphean task of thousands of hours of comparative listening while maintaining sufficient interest and sincerity to qualify their answers and efforts as genuine. Say these tests were conducted in an underground facility, shielded from EMI or other spurious radiation, with controlled temperature and humidity, using an independent, tightly-regulated, battery-derived power supply. Say one found a way to control a hundred other variables I haven’t named. Say this test yielded a statistically significant result, the conclusion of which was that all amps indeed sound the same under controlled conditions. What would be the point of all that? Should we all pitch our Levinsons and Krells and Wavacs, and start listening to $299 commodity amps? Would the enjoyment brought by those unmasked impostor products be instantly nullified and whisked away by our newfound knowledge that a panel of listeners was unable to distinguish them from some lesser amp?
3.) It is misapplied. The point of any scientific method is to obtain a result that is uncolored by human bias. This is terrific if our goal is to arrive at some objective evaluation of, say, gas mileage. It is entirely possible and practical to design an experiment that produces and measurable and repeatable result, namely: did car A consume as much fuel when driven under identical conditions as car B? Audio is not like that; it doesn’t produce an objective result. Audio isn’t audio unless it is listened to - otherwise it’s just a signal. Given that the end goal of audio practice (as discussed in this forum, at least) is to present an analogue of a signal to a user’s ear-brain mechanism, that result can never be measurable or repeatable because the conditions are always dynamic, and given the absence of a device that somehow records human consciousness, parses out the portion relating to auditory experience, and interprets it in any meaningful way, it isn’t measurable either. The subjective result of an input to the ear-brain mechanism can be experienced by and is relevant for one individual only: its owner. Even if an experiment were devised in such a way that individual tester influence were compensated for (that is, my “sounds the same” or “sounds different” result is truly the same as yours), I refer again to point #2 above - so what? The results might be interesting to a cognitive neuropsychologist, but they certainly wouldn’t influence my audio experiences or equipment choices.
4.) It ignores or dismisses many of the perfectly valid reasons why people purchase high-end amplifiers. It has always seemed the case to me that proponents of ABX testing consider the test results themselves as the only valid criteria for an amplifier purchase or comparison, and that any other factors are tantamount to deception, fraud, or ignorance. In reality, the reasons someone settles on a particular equipment purchase are some inscrutably complex mixture of peer influence, preconceived ideas, how it makes them feel, what they ate for dinner that night, the skill of the salesperson, the product’s size, build and heft; its sound quality, aesthetics, packaging, magazine reviews, quoted specifications, price, whether it was on sale, and perhaps dozens of other influences. If the result of the myriad factors in this equation is ultimately that a lightbulb blinks on in the buyer’s head that signals “this amp is worth $10k to me; I’m buying it,” would a study demonstrating that it sounds the same as a $500 receiver under specific conditions really have any bearing? Should it?
To summarize: I understand and respect the point of ABX testing in a number of particular scenarios, but I don’t think it is capable of revealing any universal truth about whether all amplifiers sound the same, and even if it did, I just don’t see how it matters or why it is worth discussing (again). We all have our reasons for being involved in audio. If you enjoy audio because you find it compelling to apply scientific methods to deduce whether there are statistically significant differences between audio components, more power to you; that’s perfectly legitimate. Although I love and appreciate science, and use its methods regularly in the course of designing my products, at the end of the day I’m involved in the audio field because I love music and I enjoy the subjective experiences I gain from reproducing it in my environment; those experiences exist, as they should, independent of anyone else’s opinions or evaluations.
Kevin, I agree with many of your well-articulated comments about the purpose and methodology of ABX testing. Perhaps the emotional responses and flat denials here can be accounted for because many in the audio community (including myself) feel a dizzying sense of accumulated fatigue every time this old warhorse is trotted out in forums, like a cockroach that refuses to die despite being stepped on repeatedly. The “tubes versus solid state” and “analog versus digital” debates are equally pointless. To wit: I can’t see how ABX testing, or the issue of amps (or any other componentry for that matter) sounding potentially the same, has any practical value in common audio practice, at least as I have observed it and engaged in it on the consumer side of things, not necessarily as a designer or manufacturer. If one has the goal of presenting a scientific paper to the AES, of conducting studies in psychoacoustics or audiology, of rigorous product development where only one variable at a time is manipulated, or of engaging in endlessly fascinating discussions about epistemology, then yes, I can see that ABX testing is a useful tool. That being said, I have four problems with ABX testing and the sorts of audio-related questions to which it is commonly (mis)appropriated:
1.) It is impractical. The key to good and meaningful and useful science is asking the right questions; trying to establish something as broad as whether “all” amps sound the same is absurdly impractical. If you want to find out whether a small pool of three or four amps sound the same, that’s spiffy; ABX away. But to attempt to extrapolate those results across all amplifiers, with their endless architectural and upstream / downstream component permutations, would violate the tenets of the very sort of scientific procedure that was presumably a sacrosanct component of the original experiment, would it not?
2.) It is irrelevant. For the sake of discussion, let’s say all of the practical issues have been somehow overcome: working samples of every amp that is or ever was on the market have been obtained, and a panel of judges has been assembled who, by some miracle, are capable of enduring the Sisyphean task of thousands of hours of comparative listening while maintaining sufficient interest and sincerity to qualify their answers and efforts as genuine. Say these tests were conducted in an underground facility, shielded from EMI or other spurious radiation, with controlled temperature and humidity, using an independent, tightly-regulated, battery-derived power supply. Say one found a way to control a hundred other variables I haven’t named. Say this test yielded a statistically significant result, the conclusion of which was that all amps indeed sound the same under controlled conditions. What would be the point of all that? Should we all pitch our Levinsons and Krells and Wavacs, and start listening to $299 commodity amps? Would the enjoyment brought by those unmasked impostor products be instantly nullified and whisked away by our newfound knowledge that a panel of listeners was unable to distinguish them from some lesser amp?
3.) It is misapplied. The point of any scientific method is to obtain a result that is uncolored by human bias. This is terrific if our goal is to arrive at some objective evaluation of, say, gas mileage. It is entirely possible and practical to design an experiment that produces and measurable and repeatable result, namely: did car A consume as much fuel when driven under identical conditions as car B? Audio is not like that; it doesn’t produce an objective result. Audio isn’t audio unless it is listened to - otherwise it’s just a signal. Given that the end goal of audio practice (as discussed in this forum, at least) is to present an analogue of a signal to a user’s ear-brain mechanism, that result can never be measurable or repeatable because the conditions are always dynamic, and given the absence of a device that somehow records human consciousness, parses out the portion relating to auditory experience, and interprets it in any meaningful way, it isn’t measurable either. The subjective result of an input to the ear-brain mechanism can be experienced by and is relevant for one individual only: its owner. Even if an experiment were devised in such a way that individual tester influence were compensated for (that is, my “sounds the same” or “sounds different” result is truly the same as yours), I refer again to point #2 above - so what? The results might be interesting to a cognitive neuropsychologist, but they certainly wouldn’t influence my audio experiences or equipment choices.
4.) It ignores or dismisses many of the perfectly valid reasons why people purchase high-end amplifiers. It has always seemed the case to me that proponents of ABX testing consider the test results themselves as the only valid criteria for an amplifier purchase or comparison, and that any other factors are tantamount to deception, fraud, or ignorance. In reality, the reasons someone settles on a particular equipment purchase are some inscrutably complex mixture of peer influence, preconceived ideas, how it makes them feel, what they ate for dinner that night, the skill of the salesperson, the product’s size, build and heft; its sound quality, aesthetics, packaging, magazine reviews, quoted specifications, price, whether it was on sale, and perhaps dozens of other influences. If the result of the myriad factors in this equation is ultimately that a lightbulb blinks on in the buyer’s head that signals “this amp is worth $10k to me; I’m buying it,” would a study demonstrating that it sounds the same as a $500 receiver under specific conditions really have any bearing? Should it?
To summarize: I understand and respect the point of ABX testing in a number of particular scenarios, but I don’t think it is capable of revealing any universal truth about whether all amplifiers sound the same, and even if it did, I just don’t see how it matters or why it is worth discussing (again). We all have our reasons for being involved in audio. If you enjoy audio because you find it compelling to apply scientific methods to deduce whether there are statistically significant differences between audio components, more power to you; that’s perfectly legitimate. Although I love and appreciate science, and use its methods regularly in the course of designing my products, at the end of the day I’m involved in the audio field because I love music and I enjoy the subjective experiences I gain from reproducing it in my environment; those experiences exist, as they should, independent of anyone else’s opinions or evaluations.
The perfect example in car world would be to compare a Z06 Corvette, a Porsche GT3 and an Aston Martin V-12 Vantage.
On paper, all three of these cars have very similar "specs". They all will go about 190 mph, all accelerate from 0-60 within
about .2 sec of each other and stop from 60-0 within about 10 feet of each other, so you would think from looking at the
"measurements" that these cars drive about the same.
However, nothing could be further from the truth when you get behind the wheel. And it doesn't have to be megabuck sports
cars. Compare the Audi A4, BMW 330i, the C-Class MB and the Lexus IS250. Same thing - nearly identical specs, COMPLETELY
different performance.
And, just like the amps, chances are you will respond strongly to one and not so much to the others.
For all the logical arguments I've heard about measurements and testing, I still think amplifiers sound different, sometimes
very different. And no set of measurements have convinced me otherwise so far.
On paper, all three of these cars have very similar "specs". They all will go about 190 mph, all accelerate from 0-60 within
about .2 sec of each other and stop from 60-0 within about 10 feet of each other, so you would think from looking at the
"measurements" that these cars drive about the same.
However, nothing could be further from the truth when you get behind the wheel. And it doesn't have to be megabuck sports
cars. Compare the Audi A4, BMW 330i, the C-Class MB and the Lexus IS250. Same thing - nearly identical specs, COMPLETELY
different performance.
And, just like the amps, chances are you will respond strongly to one and not so much to the others.
For all the logical arguments I've heard about measurements and testing, I still think amplifiers sound different, sometimes
very different. And no set of measurements have convinced me otherwise so far.
Locus
‘Or, in my words, if one hears a difference, and is pleased, why would they give a rats ass what others think?’
I am really not saying you have to give a damn if you are quite happy. You can take it further and ask why bother joining an on-line forum at all if you are quite happy living in your own world and are quite happy with the premise ‘What you hear is what you get’.
Jeff
Yes, some intelligent and valid comments – I like the analogy with cars. Dynamics is one issue I am investigating myself.
Josh
Why indeed bother with such studies at all. It is purely up to you whether you accept them or not – I am certainly not forcing anything here.
All I would say is if you are (1) dead against the ABX test, (2) don’t see any value in testing at all, and are (3) actually not interested at all in the results either way, then just ignore this post. You are not telling me anything new either.
I am interested in if there are any better/alternative tests or ones that display the opposite result, but thanks for you personal opinions. I agree that science is not for everyone and 'the man-in-the-street just don’t give a damn’ so to speak.
Yes I am interested in music and it is my passion. I play a wide variety of instruments, compose music. I am also into Hi-End audio design as you know. I don’t question at all that properly designed trials have their place, however. Audio would not progress at all without them. I could write a few pages on this but we all know it anyway – I would not be saying anything new and just be boring you some more .
Sorry to bore you with this post. I think that has just about killed the cockroach!
‘Or, in my words, if one hears a difference, and is pleased, why would they give a rats ass what others think?’
I am really not saying you have to give a damn if you are quite happy. You can take it further and ask why bother joining an on-line forum at all if you are quite happy living in your own world and are quite happy with the premise ‘What you hear is what you get’.
Jeff
Yes, some intelligent and valid comments – I like the analogy with cars. Dynamics is one issue I am investigating myself.
Josh
Why indeed bother with such studies at all. It is purely up to you whether you accept them or not – I am certainly not forcing anything here.
All I would say is if you are (1) dead against the ABX test, (2) don’t see any value in testing at all, and are (3) actually not interested at all in the results either way, then just ignore this post. You are not telling me anything new either.
I am interested in if there are any better/alternative tests or ones that display the opposite result, but thanks for you personal opinions. I agree that science is not for everyone and 'the man-in-the-street just don’t give a damn’ so to speak.
Yes I am interested in music and it is my passion. I play a wide variety of instruments, compose music. I am also into Hi-End audio design as you know. I don’t question at all that properly designed trials have their place, however. Audio would not progress at all without them. I could write a few pages on this but we all know it anyway – I would not be saying anything new and just be boring you some more .
Sorry to bore you with this post. I think that has just about killed the cockroach!
Hi Kevin(Acoustic Insight). :) No scientific anything will ever be produced that can/will prove scientifically that amp A sounds better than amp B and so on. Human interpretation of music along with a consumers pocket book will always make that determination, which of course is not very scientific. :) My 2 amps for each one of my systems capture the essence of the performance , not a exact replecation of the actual musical performance. I'm sorry that your experience you mentioned earlier didn't meet this criteria. With regards to all things being equal, the only thing that would change is the source component, at least that's what I meant by saying that earlier. You lost me when you said that the source and the speakers changed.
By changing just the source components I was able to form a opinion about digital vs analog for source component playback & tube amps vs solid state. I did this at least 15 times with a variety of amps/brands. While each amp had its own sound, digital consistantly(to me) sounded better with tube amps and analog with solid state.
By changing just the source components I was able to form a opinion about digital vs analog for source component playback & tube amps vs solid state. I did this at least 15 times with a variety of amps/brands. While each amp had its own sound, digital consistantly(to me) sounded better with tube amps and analog with solid state.
"ask why bother joining an on-line forum at all"
Meet with friends, entertainment, laughs. You know, have some fun.
"Sorry to bore you with this post. I think that has just about killed the cockroach!"
Not sure what that means...
I'm out of this nonsense thread. Best of luck with your speaker company.
Peace,
Lee
Meet with friends, entertainment, laughs. You know, have some fun.
"Sorry to bore you with this post. I think that has just about killed the cockroach!"
Not sure what that means...
I'm out of this nonsense thread. Best of luck with your speaker company.
Peace,
Lee
No Kevin, I did not say I was dead against the ABX test in general; it is a good tool for answering a particular kind of question. The question of whether all amps sound the same is not the right question for an ABX test, or for any other test - it is essentially unknowable because it is undefined and impractical to answer. The word "all" is the killer here. If you were asking a much more specific question, then the ABX test may indeed have value. I also did not say or imply that I didn't see any value in testing. I thought I took pains to be quite explicit to the contrary, in fact, but just to clarify again: I own many tens of thousands of dollars worth of test gear, use it regularly, enjoy doing so because I often find the results illuminating, and I believe the results. Finally, I don't think talking about testing is boring - as a designer, quite the opposite really. But I do agree that this particular question of all amps sounding the same is boring because it has been discussed to death in forums ad nauseam, and the results are always the same: people's feelings are hurt, divisive and unnecessary lines in the sand are drawn between "objectivists" and "subjectivists," and no real progress is made. I hang out at AudioEvo and not the other audio forums because I’ve seen far too much of that sort of thing and have no patience for it anymore.
The thing about your original topic that rubbed me the wrong way (and provoked my original, much shorter response) was the idea that the references you cited should cause anyone who had purchased an expensive amp to "worry." When I started in this business almost twenty years ago, I was much more dogmatic and unyielding about my audio-related beliefs. I am not any less passionate or opinionated today about the way I think things should be done, but I have come to realize and deeply respect the idea that there are many ways to skin the audio cat, and that people who espouse beliefs and practices radically different from my own are also able to achieve results that I must admit are excellent by any standard. You claim below not to be forcing anything on anyone, but the tone of your post (and others you’ve made) seems to suggest otherwise, at least to me. You and I have more in common than you might think at this point, but live and let live, brother - if someone wants to spend $3k on an amp or even ten times that amount, who are we to tell them they shouldn’t or suggest they’re ignorant for doing so? I wholeheartedly agree that test results are essential for progress in audio, but using them to convince others they’re wrong is futile at best, and only serves to introduce schisms into our avocation.
The thing about your original topic that rubbed me the wrong way (and provoked my original, much shorter response) was the idea that the references you cited should cause anyone who had purchased an expensive amp to "worry." When I started in this business almost twenty years ago, I was much more dogmatic and unyielding about my audio-related beliefs. I am not any less passionate or opinionated today about the way I think things should be done, but I have come to realize and deeply respect the idea that there are many ways to skin the audio cat, and that people who espouse beliefs and practices radically different from my own are also able to achieve results that I must admit are excellent by any standard. You claim below not to be forcing anything on anyone, but the tone of your post (and others you’ve made) seems to suggest otherwise, at least to me. You and I have more in common than you might think at this point, but live and let live, brother - if someone wants to spend $3k on an amp or even ten times that amount, who are we to tell them they shouldn’t or suggest they’re ignorant for doing so? I wholeheartedly agree that test results are essential for progress in audio, but using them to convince others they’re wrong is futile at best, and only serves to introduce schisms into our avocation.
.. OK the cockroach still lives (somebody mentioning this thread being akin to this insect that keeps coming back if you don’t know what I am talking about here).
Here’s a more in-depth a look at professional testing for those still interested:
It’s very simple and there is no voodoo involved. If I was trying to assess whether a modification to a system (say B) sounded any different to the original (A), then I would set up a system where I can readily switch between the each system. By continual switching and with no indication at all whether one system is switched in at one time I would then try and determine if I could identify first
1.If I could tell the difference in switching
2.If I could eventually identify which system is switched in
The next question is ‘is B actually moving in the right direction? – is it a modification that is doing the right thing.’
It is then necessary to perform the test with an audience. I would hide the systems from view and invite a number of professionals to say if they actually prefer A or B. We have already established that there is an observable difference. It is now time to see if the observable difference is detectable by a general audience. If it can’t be detected by a group of professionals then there is probably no point in building in the modification at it would probably involve some expense.
None of this testing is rocket science – it’s simply an accepted approach while developing anything that’s based on science which impacts on an observer. With audio you are already half-way there as measurements can tell you a great deal of what is going on and does tend to point to the result (but I agree this is not always the case and there are areas where measurements can’t be taken any further).
Now to say ‘I just trust my own ears’ works for some people and that’s all very well – what they have to realise, however that just about every system out there has been tested for many hours using the very kind of test that some people ridicule here. In fact exactly the same test that I first brought to attention – an ABX trial.. I graduated in science and in my higher degree + research post I used such tests extensively in any circumstance surrounding an ‘objective’ area. Audio is a typical such area – it’s not some special case. An audio system does not have some magical ‘ethereal’ property that is beyond complete evaluation.
Sorry I don’t agree with the approach that there is no proper test to ‘measure’ something as subjective as audio. This is a complete myth and just about any scientist in the world who has not been brainwashed by this purely ‘audiophilic misconception’ will tell you otherwise. Even some non scientists support it. The ABX test has been designed for this very purpose and is a very powerful tool indeed.
Now some people will say ‘Audio is very personal and so we just can’t trust anyone else’s subjective opinion. Just listen and make your own mind up.’ This argument is flawed for the following reasons:
1.Anything you hear has already been developed using subjective testing (personal and group)
2.By the very admission subjective listening by others is not reliable
Controlled trials (ABX, single blind, double-blind etc) are in fact the ONLY reliable way of testing human response in this manner as it is designed to cancel out anything that is not relevant to the test – the ‘circumstantial and personal’ factors that are inherent in a single person’s assessment. Actually an audio review is a good example of a single person’s subjective test and is very useful, but has the major flaw that you can’t rely on the ‘circumstance and personal factors’ being the same as your own.
The development of audio systems is a science – pure and simple. I simply don’t agree with the premise that it a system can’t be fully evaluated beyond question from a human perspective, any less than an overhead projector, TV or digital weighing machine can. The purpose is well defined – it’s a bunch of scientifically designed components that does a very clearly defined thing to an audio signal, and degrading it as little as possible. It does what it says on the tin, so to speak.
Now if you have an ‘emotional attachment’ to you amp, see it something more than this and even give it a name ‘Sam’, for that matter; good luck to you. Just grow up a bit and try and see things from a scientific viewpoint – science FULLY describes it and is indeed the ONLY thing that describes it. We are not dealing with quantum mechanics here – something that is really beyond the realm of current scientific understanding. Any component of the audio chain is fully measurable by scientific or observational means and the science that went into the design is COMPLETELY understood.
The music it conveys is something entirely different, however and I agree totally here that it goes beyond any reasonable subjective evaluation – it s all too often the case that this is confused with the medium that simply ‘conveys’ it.
The cockroach is dead… long live the cockroach!
Here’s a more in-depth a look at professional testing for those still interested:
It’s very simple and there is no voodoo involved. If I was trying to assess whether a modification to a system (say B) sounded any different to the original (A), then I would set up a system where I can readily switch between the each system. By continual switching and with no indication at all whether one system is switched in at one time I would then try and determine if I could identify first
1.If I could tell the difference in switching
2.If I could eventually identify which system is switched in
The next question is ‘is B actually moving in the right direction? – is it a modification that is doing the right thing.’
It is then necessary to perform the test with an audience. I would hide the systems from view and invite a number of professionals to say if they actually prefer A or B. We have already established that there is an observable difference. It is now time to see if the observable difference is detectable by a general audience. If it can’t be detected by a group of professionals then there is probably no point in building in the modification at it would probably involve some expense.
None of this testing is rocket science – it’s simply an accepted approach while developing anything that’s based on science which impacts on an observer. With audio you are already half-way there as measurements can tell you a great deal of what is going on and does tend to point to the result (but I agree this is not always the case and there are areas where measurements can’t be taken any further).
Now to say ‘I just trust my own ears’ works for some people and that’s all very well – what they have to realise, however that just about every system out there has been tested for many hours using the very kind of test that some people ridicule here. In fact exactly the same test that I first brought to attention – an ABX trial.. I graduated in science and in my higher degree + research post I used such tests extensively in any circumstance surrounding an ‘objective’ area. Audio is a typical such area – it’s not some special case. An audio system does not have some magical ‘ethereal’ property that is beyond complete evaluation.
Sorry I don’t agree with the approach that there is no proper test to ‘measure’ something as subjective as audio. This is a complete myth and just about any scientist in the world who has not been brainwashed by this purely ‘audiophilic misconception’ will tell you otherwise. Even some non scientists support it. The ABX test has been designed for this very purpose and is a very powerful tool indeed.
Now some people will say ‘Audio is very personal and so we just can’t trust anyone else’s subjective opinion. Just listen and make your own mind up.’ This argument is flawed for the following reasons:
1.Anything you hear has already been developed using subjective testing (personal and group)
2.By the very admission subjective listening by others is not reliable
Controlled trials (ABX, single blind, double-blind etc) are in fact the ONLY reliable way of testing human response in this manner as it is designed to cancel out anything that is not relevant to the test – the ‘circumstantial and personal’ factors that are inherent in a single person’s assessment. Actually an audio review is a good example of a single person’s subjective test and is very useful, but has the major flaw that you can’t rely on the ‘circumstance and personal factors’ being the same as your own.
The development of audio systems is a science – pure and simple. I simply don’t agree with the premise that it a system can’t be fully evaluated beyond question from a human perspective, any less than an overhead projector, TV or digital weighing machine can. The purpose is well defined – it’s a bunch of scientifically designed components that does a very clearly defined thing to an audio signal, and degrading it as little as possible. It does what it says on the tin, so to speak.
Now if you have an ‘emotional attachment’ to you amp, see it something more than this and even give it a name ‘Sam’, for that matter; good luck to you. Just grow up a bit and try and see things from a scientific viewpoint – science FULLY describes it and is indeed the ONLY thing that describes it. We are not dealing with quantum mechanics here – something that is really beyond the realm of current scientific understanding. Any component of the audio chain is fully measurable by scientific or observational means and the science that went into the design is COMPLETELY understood.
The music it conveys is something entirely different, however and I agree totally here that it goes beyond any reasonable subjective evaluation – it s all too often the case that this is confused with the medium that simply ‘conveys’ it.
The cockroach is dead… long live the cockroach!
Sorry Greg, I was so carried away I forgot to repond to your observations.
I agree with you here. If we can clearly observe that amps A does things a bit differently to amp B then the next question is 'Is this actually an improvement or not'. This may often come down to a subjective opinion - is this trade-off worth that gain for instance. Its a more difficult question to answer. Trials will have a place in answering this and we can statistically determine if a sample prefer the change or not - it then becomes more of an 'objective' measurement.
This is not at all the case in the amp comparison aforementioned, that is designed to establish simply 'is there an observable difference at all' not 'do you prefer one over the other'. It's a very basic premise/and test. The latter issue is a bit more complex as you have identified.
I am still unconvinced about this amp test either way, by the way. It certainly stands up to scientific scrutiny and I can't smell a 'dead cockroach', but it does come as quite a surprise to me and thought would be a stimulating discussion point.
I do try to stimulate (or bore as the case may be). Take you pick!
Kevin
I agree with you here. If we can clearly observe that amps A does things a bit differently to amp B then the next question is 'Is this actually an improvement or not'. This may often come down to a subjective opinion - is this trade-off worth that gain for instance. Its a more difficult question to answer. Trials will have a place in answering this and we can statistically determine if a sample prefer the change or not - it then becomes more of an 'objective' measurement.
This is not at all the case in the amp comparison aforementioned, that is designed to establish simply 'is there an observable difference at all' not 'do you prefer one over the other'. It's a very basic premise/and test. The latter issue is a bit more complex as you have identified.
I am still unconvinced about this amp test either way, by the way. It certainly stands up to scientific scrutiny and I can't smell a 'dead cockroach', but it does come as quite a surprise to me and thought would be a stimulating discussion point.
I do try to stimulate (or bore as the case may be). Take you pick!
Kevin
No worries Kevin. :) Let us know if you come up with a double blind controlled study that would scientifically prove that a certain amp(s) are better & why. You might get published in Scientific American or Nature Science Journal- maybe even the win the Nobel Prize in science. (?) Who knows! :) BTW, stimulating discussions are a good thing. It's always good to hear others point of view regarding such matters.
:) If only I had the time - I have a full-time job, young son (just hope he doesn't take after me), a speaker company to develop and a recent interest in this site amongst several other things on the agenda. Now if I could live to 200 (or even clone myself) that would solve some issues.
How'd you know I named my amp Sam? My last one was name Florence (or Flo for short; God rest her fried Darlingtons), and I'm thinking of naming my next one Horace.
What, you guys thought I was kidding??? Okay, I was. I thought this thread could use an injection of levity.
You need to be a member of this group before you can participate in this discussion
