 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
216.53.159.98
In Reply to: Re: Why Not Write Your Own Thesis To Disprove It? posted by Soundmind on June 13, 2006 at 09:18:25:
Soundmind, I don't have a BS in anything so my hat is off to you.I'm a bit surprised you responded to my statement "we should try to disover how something can measure poorly and sound great." with this comment "...I suggested this a year or two ago to a famous amplifier designer who posts at the Asylum and freely admits he can't find any correlation between what he measures and how his equipment sounds to him" That shows me you are a lot more open minded than I took you for.
I've stated numerous times I believe both Objectivists & Subjectivists have it partly right & partly wrong. The fact that at this time we cannot find any correlation between how audio equipment measures and how audio equipment sounds, seems to dictate that the measurements being used today are NOT the correct measurements required for determining why individuals perceive audio components as either being realistic or not.
That alone, it would seem to me, should open the mind of Objectivists that it's possible Subjectivists might POSSIBLY be right in their beliefs...
Thetubeguy1954
Follow Ups:
As I have said, ultimately accurate sound reproduction like any other sensory perception is subjective. But it is testable and designs are verifiable according to scientific principles. This means that the limits of perception are tested and determined for human beings, the salient aspects of sound fields which are perceptable are determined, and the designs are measured against their ability to meet the demands of duplicating those salient aspects of sound fields. Almost none of this work has been done. The art and science of sound recording and reproduction insofar as the state of the art technology's ability to record and reproduce musical instruments convincingly is to a large degree so far a failure. This is verified in the TAS article and is hardly surprising in light of the fact that critical scientific principles have yet to be discovered and serious engineering solutions have yet to be applied. The technology of the SET non feedback amplifier may have been state of the art in 1936 but in 2006 it is a pathetic joke, inferior to many more advanced technologies in EVERY conceivable way. That it would be offered as state of the art selling for absurdly high prices is outrageous. Of all the limitations in the sound recording/playback process, audio power amplifiers have reached near theoretical perfection. If the people who sell this reprise of 70 year old technology want to be taken seriously by engineers, not just hobbyists, then it is up to them to prove their superiority with scientifically respectable demonstrations which their collegues find credible, not mere testimonials of hobbyists and people looking to write a thesis for a master's degree. And you can be sure, the onus in on them, not on the other 99.9999% of electrical engineers and scientists who rely on the established system of measurement to make critical decisions.
So if an amplifier sounds bad but measures good, it's good?
I don't get that. I really thought that it was well established that upper order harmonic distortions can be heard, as sounding bad, even when very low in percentage (compared to lower order distortions). Thus, without weighting the upper order distortions, THD is meaningless.
Are you saying that this is not correct?
Thanks....Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
"So if an amplifier sounds bad but measures good, it's good?"Amplifiers don't make any sound, they perform an electrical function in a sound system. Sound recording/reproduction systems sound accurate or inaccurate and the nature, degree, and source of their inaccuracy needs to be studied and determined. But if a sound system sounds inaccurate, it is foolish to look first at the elements which by every test devised perform their functional roles nearly perfectly when others don't even come close. Unless you can find a flaw in the test procedure or rationale behind the test, look for the source of the error elsewhere, such as in the recording or in the loudspeakers where it most likely is. There is much agreement that loudspeakers are the weakest link in the chain so if the object is to find an amplifier whose flaws are complimentary to a particular loudspeaker, you can spend your entire life looking for one and then when you buy another loudspeaker to connect it to, start looking all over again because the one you bought before has the wrong flaws. As for recordings, the variables are so great that when the editor of a famous consumer audio magazine who posts here said he equalizes his recording so precisely that they must be correct to within 0.1 db but when listening to recordings doesn't bother to try to equalize them because it is hopeless. Why then would you look to solve an accuracy problem at an amplifier which is doing its job? If this is just a matter of personal preference, then any amplifier is as good as any other because there is no independent standard or reference by which to judge any of them. And that is what the industry which makes them would like you as a consumer to believe. If you take that attitude, they will be able to constantly sell you an endless parade of them each one claiming to be more likeable than all of the others including their own. If on the other hand, you decide that there is an objective limit beyond which further improvement or change is of no subjective value, then once you have an amplifier which meets those criteria, there will never be a need to buy another one unless it fails and you decide not to repair it. That is this industry's worst nightmare, that everyone would see it that way. If it were so, their market and profits would evaporate.
> So if an amplifier sounds bad but measures good, it's good?If an amplifier has inaudible distortion components then it has achieved what most would consider to be ideal in an audio amplifier.
If another amplifier with audible distortion sounds more pleasant then one needs to examine the purpose of the amplifier. I would suggest that most rational people would prefer to split the functions and use an amplifier with inaudible distortion plus a separate processor to introduce the distortion in a controlled manner.
> Thus, without weighting the upper order distortions, THD is
> meaningless.THD has a very precise and well understood meaning. However, as with all integrated "figures of merit" it only conveys what it was intended to convey and that is the total harmonic distortion.
If you want an integrated "figure of merit" for another purpose then it is perfectly reasonable to derive one. What is not reasonable is to disparage a particular integrated "figure of merit" for not representing something that no reasonable person would expect it to represent in the first place.
"I would suggest that most rational people would prefer to split the functions and use an amplifier with inaudible distortion plus a separate processor to introduce the distortion in a controlled manner.
"http://stereophile.com/reference/406howard/
Keith Howard has found that adding distortion to a recording does not result in any euphonic enhancement. This is a falacy that people who have a hard time believing tube gear can sound good use as a justification for why it subjectively sounds better than their SS amp. The truth is that ALL distortion degrades the sound. Some does more damage than others. You draw the conclusion.
> The truth is that ALL distortion degrades the sound.How do you square this with significant numbers of people prefering the sound of a heavily distorting valve amplifiers to an inaudibly distorting solid state amplifier?
I suggest this is a foolish statement to make based on one article from a non-technical journal when there is so much contradictory information including, I suspect, your own experiences with the sound of valve and solid state amplifiers.
"How do you square this with significant numbers of people prefering the sound of a heavily distorting valve amplifiers to an inaudibly distorting solid state amplifier?I suggest this is a foolish statement to make based on one article from a non-technical journal when there is so much contradictory information including, I suspect, your own experiences with the sound of valve and solid state amplifiers. "
And I would suggest that you don't understand my statement because the answer is obvious if you think about it. 1) It is an assumption on your part that valve amps are "heavily distorting" compared to SS amps. At what power (it matters for amps with no feedback you know)? In which frequency range? 2)It is an assumption on your part that SS amps are inaudibly distorting. Not many people even here in the PHP would agree with this statement (I know John Curl, JJ, and hell even Dan Banquer would not agree with you that SS amps are not audibly distorting).
One argument is that while both are distorting and it is audible that one is more audibly damaging to the sound than the other. Clear enough for you? Howard made it clear that while none of the distorted tracks were "euphonic" compared to the original, he also made it clear that some distortion types he added were less pleasant than others. The ones that were the least pleasant contained high order and/or odd order harmonics. Not such a great surprise unless you think those levels to be inaudible.
> > The truth is that ALL distortion degrades the sound.
> And I would suggest that you don't understand my statementCan you please explain since it seems both unambiguous and incorrect and was not really addressed in your reply. It is incorrect because if the distortion is too small to be audible then the perception of the sound is not degraded.
> because the answer is obvious if you think about it. 1) It is an
> assumption on your part that valve amps are "heavily distorting"
> compared to SS amps.You are being dishonest. I asked: "How do you square this with significant numbers of people prefering the sound of a heavily distorting valve amplifiers to an inaudibly distorting solid state amplifier?" The assumption here is that heavily distorting valve amplifiers exist (e.g. SET amplifiers) and people exist who prefer their sound to undistorting solid state amplifiers (e.g. those who have paid a lot of money for SET amplifiers). Do you deny this?
> 2)It is an assumption on your part that SS amps are inaudibly
> distorting.I asked: "How do you square this with significant numbers of people prefering the sound of a heavily distorting valve amplifiers to an inaudibly distorting solid state amplifier?" The assumption here is that inaudibly distorting solid state amplifiers exist in significant numbers but not that all solid state amplifiers are inaudibly distorting which is obviously nonsense. Do you deny this?
> Not many people even here in the PHP would agree with this statement
I would hope not since it would appear to be wrong although it could stand a bit of clarification as to whether the words you are putting in my mouth refer to all solid state amplifiers under all conditions or competently designed ones operating under conditions which allow inaudible levels of distortion (and which you, presumably, dispute are inaudible).
> One argument is that while both are distorting and it is audible
> that one is more audibly damaging to the sound than the other. Clear
> enough for you?It could stand some clarification. It would appear to follow from this that you are stating that a competently designed solid state amplifier operating well away from clipping and into a relative benign load so that it can maintain inaudile levels of distortion is audibly damaging the sound? If so, how?
Why if the sound is degraded by low levels of distortion do all amplifiers, whether valve or solid state, that achieve these low levels of distortion sound indistinguishable from each other? That is, why are they all degrading the sound in the same way?
Or do you perhaps believe that all levels of distortion in all amplifiers are audible?
"It is incorrect because if the distortion is too small to be audible then the perception of the sound is not degraded."Please tell me where that level lies. See my post above for the whole discussion.
"You are being dishonest. I asked: "How do you square this with significant numbers of people prefering the sound of a heavily distorting valve amplifiers to an inaudibly distorting solid state amplifier?" The assumption here is that heavily distorting valve amplifiers exist (e.g. SET amplifiers) and people exist who prefer their sound to undistorting solid state amplifiers (e.g. those who have paid a lot of money for SET amplifiers). Do you deny this?"
I am never dishonest. You are the one using the word heavily distorting not me. IMO SET amps run within their true power limits are not heavily distorting. That is my point. Of course there are many people who prefer the sound SET amps...and with good reason.
"I asked: "How do you square this with significant numbers of people prefering the sound of a heavily distorting valve amplifiers to an inaudibly distorting solid state amplifier?" The assumption here is that inaudibly distorting solid state amplifiers exist in significant numbers but not that all solid state amplifiers are inaudibly distorting which is obviously nonsense. Do you deny this"
Do I deny what? Sorry I am not understanding the point you are trying to make here. I do deny that there exists any amp on the market that is inaudibly distorting. Is that what you mean?
"It would appear to follow from this that you are stating that a competently designed solid state amplifier operating well away from clipping and into a relative benign load so that it can maintain inaudile levels of distortion is audibly damaging the sound? If so, how"There you go again assuming that the distortion it is generating is inaudible. I dispute this. It will make distortion and I would argue that it is audible. Therefore it is damaging to the sound.
"Why if the sound is degraded by low levels of distortion do all amplifiers, whether valve or solid state, that achieve these low levels of distortion sound indistinguishable from each other"They do? Again, most here, including myself would disagree with this statement. You seem to think I accept your basic premise that all low distortion amps sound the same but I do not.
"Or do you perhaps believe that all levels of distortion in all amplifiers are audible"All levels? Probably not. Much lower than is generally given credit, yes.
> I do deny that there exists any amp on the market that is inaudibly
> distorting.The penny drops and I think I now understand your beliefs about amplifiers and distortion.
The obvious question is how do you square this belief with the results of experiments into the audibility of distortion? For example, how have experimenters managed to find the thresholds if their amplifiers (almost certainly quite ordinary solid state amplifiers) possess audible levels of distortion? Are all the researchers, referees and readers of journals in the area simply incompetent and the scientific method has broken down in this case?
"What is not reasonable is to disparage a particular integrated "figure of merit" for not representing something that no reasonable person would expect it to represent in the first place."OK, sorry. THD tells us how much THD an amplifier has but tells us very little about how that amplifier will sound. That's what I meant by "meaningless". Nothing more. I don't think that's disparaging.
On the other hand if THD does not tell us much about how an amplifier will sound, what "merit" does it have?"If an amplifier has inaudible distortion components then it has achieved what most would consider to be ideal in an audio amplifier."
Please lead me to this amplifier, I want to buy it.
"If another amplifier with audible distortion sounds more pleasant..."
How would an amplifier with audible distortion sound more pleasant to a person with a good knowledge of the sound of real instruments? Are the distortions making up for something missing in the recording process or a shortcoming in the speakers or what? Just exactly what are you saying?
Thanks....Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
> On the other hand if THD does not tell us much about how an amplifier
> will sound, what "merit" does it have?For most people an ideal amplifier is not supposed to have a "sound" whether good or bad. It supposed to be neutral and this will be achieved when the distortion is inaudible.
Now many audiophiles do not seek such amplifiers and prefer ones with a more pleasing sound. However, they do not describe such amplifiers as sounding more pleasing but use words like natural, life like, or similar positive sounding adjectives. There is not much wrong with this until they use words like accurate which have a well defined meaning when applied to amplifiers and tends to cause a few people to object.
The merit of THD is principally that it is simple to measure and if well below the threshold of audibility then this form of distortion will be inaudible for the conditions of the test. Unfortunately, these conditions are not always realistic/representative and extrapolating to real conditions is not always straightforward. Hence, for example, it is unwise to simply take a quoted THD figure of below 0.01% (comfortably inaudible in all experiments) and assume that this will be achieved under all conditions when driving a real loudspeaker.
If the THD is at or above the audibility threshold then one would need to look at the individual components to determine the likelihood of audibility or whether the colouration may be perceived as pleasant or not. At this level it is not particularly useful as a single integrated number but a map of the individual components against frequency and power can be useful diagnostic information for people familiar with interpretting it.
> Please lead me to this amplifier, I want to buy it.
Most competently designed solid state amplifiers fall into this category when driving speakers with reasonable loads and operating well away from clipping. Perhaps a more interesting question is how cheap is good enough and, I am afraid, this I cannot answer with any authority since I have not tested or listened under controlled circumstances to the current crop of amplifiers. Nor, obviously, am I aware of a trustworthy source for this information. Nonetheless, sticking my neck out and having a punt it may be around 231 Euros for a 2x150W Behringer A500 amplifier. Anyone else?
> How would an amplifier with audible distortion sound more pleasant
> to a person with a good knowledge of the sound of real instruments?
> Are the distortions making up for something missing in the recording
> process or a shortcoming in the speakers or what? Just exactly what
> are you saying?For example, most people are familiar with the sound of valve amplifiers and will often prefer the sound of a voice recorded through such amplifiers even though they have audible levels of distortion compared to that of a solid state amplifier. This can occur for reasons such as modifying the frequency response and enhancing the harmonic content of the voice/instrument. The departure from neutral is not making up for something missing but adding to what is present.
"For most people an ideal amplifier is not supposed to have a "sound" whether good or bad. It supposed to be neutral and this will be achieved when the distortion is inaudible."Tre' asked you which amp this was. I would like to know too so I can also get one.
"Now many audiophiles do not seek such amplifiers and prefer ones with a more pleasing sound. However, they do not describe such amplifiers as sounding more pleasing but use words like natural, life like, or similar positive sounding adjectives. There is not much wrong with this until they use words like accurate which have a well defined meaning when applied to amplifiers and tends to cause a few people to object."
Isn't it reasonable to assume that more natural would also be more accurate unless you think real instruments playing music isn't natural? People with a good knowledge of live unamplified instruments respond to natural reproduced sound. Why do YOU assume that this is in some way "euphonic" or "enhanced".
The reason you assume that it can't be what is really on the recording is from your past experiences with recorded music...likely from inexpensive SS electronics and inferior speakers (we all started there...at least those of us born in the last 30 to 40 years). If your experience with live unamplified music is not extensive you don't even realize what is missing from the playback. Then as you developed as an audiophile (hopefully) you heard more live music and you heard gear that suddenly made the music more vibrant dynamically and colorful (not the same as colored). THis is what is meant by natural.
The other part of experience is the measurements. Now you have seen the measurements and this makes certain expectations as a result. You EXPECT the better measuring gear to sound more accurate (ie. more like what was on the recording). THat is not what many people's ears are telling them; however, and it causes a lot of dissonance in the industry and among audiophiles. To many listeners if you swapped tube amp measurements with SS measurements many people would say the measurements have a pretty good correlation with the listening tests!
"Most competently designed solid state amplifiers fall into this category when driving speakers with reasonable loads and operating well away from clipping."
What does this have to do with audibility of distortions? WHat if the distortion is audible at all power levels below clipping? No one knows how low of distortion is low enough and it depends strongly on the harmonic content of that particular distortion.
"For example, most people are familiar with the sound of valve amplifiers and will often prefer the sound of a voice recorded through such amplifiers even though they have audible levels of distortion compared to that of a solid state amplifier"
This should make you wonder very much about why? Mr. Howard (see stereophile article of his about added distortion to recordings and "euphonic" results) has found that ALL added distortion is a degradation. So if voice sounds better its because the tube amp is doing LESS to degrade the sound of that recorded voice. The other thing you have to realize is that at low power, many tube amps with little or no feedback have LOWER distortion, especially in higher harmonics, than SS amps. This is because noise and distortion decrease rapidly with reduction of power in SET amps, for example, and do not decrease at all (in fact often they increase, relatively) at lower power.
> Tre' asked you which amp this was. I would like to know too so I can also
> get one.??? I answered Tre' question.
> Isn't it reasonable to assume that more natural would also be more accurate
> unless you think real instruments playing music isn't natural?Accurate has a precise technical meaning when applied to an amplifier and relates to the accuracy of the amplified signal relative to the input signal. It cannot be validly used to describe a significantly distorting amplifier compared to one distorting less.
Natural does not have a precise technical meaning and so can be used to describe a significantly distorting amplifier without being an absolutely false statement. One can object on qualitative grounds if inclined but things are not black and white.
> People with a good knowledge of live unamplified instruments respond to
> natural reproduced sound. Why do YOU assume that this is in some way
> "euphonic" or "enhanced".If the distorted version is preferred to the undistorted version then how would you describe it?
> The reason you assume that it can't be what is really on the recording is
> from your past experiences with recorded music...What leads you to believe that?
> ...Then as you developed as an audiophile
I do not hold audiophile beliefs and would be offended to be called an audiophile by a person outside the audiophile world although I do appreciate you do not mean it in a derogatory way here.
> ...THis is what is meant by natural.
No problems so long as you avoid terms with precise technical meanings you should be on safe(ish) ground.
> You EXPECT the better measuring gear to sound more accurate (ie. more like
> what was on the recording).Expectation is irrelevant. If the distortion is less than it is more accurate (subject to all forms of distortion being less). This is what accurate means. However, it does not necessarily follow that the more accurate device will be preferred.
> THat is not what many people's ears are telling them; however, and it
> causes a lot of dissonance in the industryI am not aware of any dissonance in the technically educated part of the audio/acoustical industry or in the professional or consumer audio industry but then such people do not hold audiophile beliefs about what is going on.
> and among audiophiles.
Possibly. From the evidence in these pages (my only real exposure to audiophiles) there certainly are audiophiles who appear uncomfortable with their beliefs and seem to be threatened by a more scientific view. Nonetheless, I suspect they are probably in the minority and, as far as I can judge, most have few problems dismissing contradictory information when it intrudes from the world outside.
> "Most competently designed solid state amplifiers fall into this category
> when driving speakers with reasonable loads and operating well away from
> clipping."
> What does this have to do with audibility of distortions?I am struggling to follow your train of thought here. If you clip an amplifier, trigger protection circuitry, ask for too much current, etc... an amplifier will distort.
> WHat if the distortion is audible at all power levels below clipping?
??? Then it will be audible?
> No one knows how low of distortion is low enough and it depends strongly on
> the harmonic content of that particular distortion.Nonsense. There are plenty of studies in the technical journals concerning the audibility of harmonics.
> This should make you wonder very much about why? Mr. Howard (see
> stereophile article of his about added distortion to recordings and
> "euphonic" results) has found that ALL added distortion is a degradation.
> So if voice sounds better its because the tube amp is doing LESS to degrade
> the sound of that recorded voice.I think you might want to slow down a bit. You are making wild and unjustifiable extrapolations from one limited source of information. Do you think this is wise?
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
