In many thread many members argue about the relation of THD and sound quality. I've got heavier case 
I've got friends who loves audio. One of them wanted me to upgrade his power amp. In the process, he come to my house and he listens to the "not finished" modification. He said "Hey, that's it, I like it..."
He said this when the amp is oscilating (not in all level, it only oscilates at full power with square wave. With moderate signal it doesn't oscilate).
I said, "no, it is not finished yet". He said, "OK, the final result must be something......"
When I finished modification (I eliminate the oscilation fully, thanks to Mr. Pass and Jocko Homo), I return the amp to him.
The next day, he return the amp to me and said, "I don't like the sound, it is different than when I heard in your house..."
HAAAAH.....I tough myself. He likes the sound of oscilating amp??? To test it, I change back only the oscilating modification, putting it back to where it oscilates. When he hear it, he said "Yes, you are good. I like this very much". I'm not afraid of the oscilation, because it is not "heavy oscilation" just "light oscilation" so I keep the setting of the oscilating one.
Another case. I've got another friend who also loves audio very much. They all don't know electronics, just listens.
He used to have very-very hot Gryphone. The heatsink is massive, still very hot. It must be biased at very big ampere.
Now he change his amp to FM Accoustic, but what surprises me, when I touch the heatsink, it is only slightly warm. I think FM Accoustic is low biased class AB. Won't reach 10W class A in the AB range. And yet this friend of mine said "This is how an amp supposed to sound. This is the ultimate sound"
Low biased classAB will contain Xover distortion artifact, compared to full classA. And he likes the sound of Xover distortion?
Another case. ClassD is perfect from the point of view of Xover distortion. ClassD won't have Xover distortion. Most ClassD will likely cancel most harmonics (leaving something in the form of "noise floor"). But some who has listened to it said the sound of classD is "sterile" "without emotion". I really suspect that the "not sterile" of "with emotion" sound is actually the distorted sound.
This is my analisis. The music played various, from mellow sweet jazz to heavy full classic concerta (like PMA used) with fast high tone violin and full impact big drums.
Both of my friend likes when there is something "added". With the oscilating amp, the details is like "having turbo", revealing every detail, and when the amp is not oscialting, those details gone. The trebles have more edges. When the circuit is intentionally made unbalanced (to have low order harmonic), they like it, said "the sound is warmer, have more soul".
The same with Gryphone VS FM Accoustic. ClassA cannot have details (Is it details?......) like low biased classAB+Xover distortion; it has "sst....ssst" on the edges of the trebles. I know that this is not supposed to sound right, but they like it. Too much of this will be fatiguing. But when getting the right "balance" the sound is considered "superb"
Some time ago, PRR said "Why do you afraid of TIM? Slight TIM is OK". Now I know what he means.....
The question is : I'm confused. If making amp that is perfect from theories, from scopes, the ones who listens to it don't like the sound. But when making amp that has slight 2nd harmonic (intentionally make unbalanced like PLH), has Xover distortion, or worse, the amp that oscilates (slightly), the consumer likes it better.
So what is the point of learning how to make very good (technically) amp? If ones who listens to it prefers the "WRONG" design?
They are not ordinary people. They have years of Hi-End stuff. One of them even have his own recording studio. They both amateur musician. So they MUST know that the sound they are hearing is not "clone" of the real musical instrument, but they like the modded sound. I cannot argue with them.
Is it my knowledge or their ears is wrong? Is this happening elsewhere or just in my place?
I've got friends who loves audio. One of them wanted me to upgrade his power amp. In the process, he come to my house and he listens to the "not finished" modification. He said "Hey, that's it, I like it..."
He said this when the amp is oscilating (not in all level, it only oscilates at full power with square wave. With moderate signal it doesn't oscilate).
I said, "no, it is not finished yet". He said, "OK, the final result must be something......"
When I finished modification (I eliminate the oscilation fully, thanks to Mr. Pass and Jocko Homo), I return the amp to him.
The next day, he return the amp to me and said, "I don't like the sound, it is different than when I heard in your house..."
HAAAAH.....I tough myself.
He likes the sound of oscilating amp??? To test it, I change back only the oscilating modification, putting it back to where it oscilates. When he hear it, he said "Yes, you are good. I like this very much". I'm not afraid of the oscilation, because it is not "heavy oscilation" just "light oscilation" so I keep the setting of the oscilating one. Another case. I've got another friend who also loves audio very much. They all don't know electronics, just listens.
He used to have very-very hot Gryphone. The heatsink is massive, still very hot. It must be biased at very big ampere.
Now he change his amp to FM Accoustic, but what surprises me, when I touch the heatsink, it is only slightly warm. I think FM Accoustic is low biased class AB. Won't reach 10W class A in the AB range. And yet this friend of mine said "This is how an amp supposed to sound. This is the ultimate sound"
Low biased classAB will contain Xover distortion artifact, compared to full classA. And he likes the sound of Xover distortion?
Another case. ClassD is perfect from the point of view of Xover distortion. ClassD won't have Xover distortion. Most ClassD will likely cancel most harmonics (leaving something in the form of "noise floor"). But some who has listened to it said the sound of classD is "sterile" "without emotion". I really suspect that the "not sterile" of "with emotion" sound is actually the distorted sound.
This is my analisis. The music played various, from mellow sweet jazz to heavy full classic concerta (like PMA used) with fast high tone violin and full impact big drums.
Both of my friend likes when there is something "added". With the oscilating amp, the details is like "having turbo", revealing every detail, and when the amp is not oscialting, those details gone. The trebles have more edges. When the circuit is intentionally made unbalanced (to have low order harmonic), they like it, said "the sound is warmer, have more soul".
The same with Gryphone VS FM Accoustic. ClassA cannot have details (Is it details?......
) like low biased classAB+Xover distortion; it has "sst....ssst" on the edges of the trebles. I know that this is not supposed to sound right, but they like it. Too much of this will be fatiguing. But when getting the right "balance" the sound is considered "superb"Some time ago, PRR said "Why do you afraid of TIM? Slight TIM is OK". Now I know what he means.....
The question is : I'm confused. If making amp that is perfect from theories, from scopes, the ones who listens to it don't like the sound. But when making amp that has slight 2nd harmonic (intentionally make unbalanced like PLH), has Xover distortion, or worse, the amp that oscilates (slightly), the consumer likes it better.
So what is the point of learning how to make very good (technically) amp? If ones who listens to it prefers the "WRONG" design?
They are not ordinary people. They have years of Hi-End stuff. One of them even have his own recording studio. They both amateur musician. So they MUST know that the sound they are hearing is not "clone" of the real musical instrument, but they like the modded sound. I cannot argue with them.
Is it my knowledge or their ears is wrong? Is this happening elsewhere or just in my place?
Hi lumanauw,
Nothing is wrong. A bit of distorsion is very often prefered to no distorsion at all. A long time ago, I had the following experience.
I had two lo-fi recorders whose signal to noise was probably no better than 35 dB. Despite of this noise, all material recorded on them seemed very detailed. I have been puzzled by this strange behaviour for long until I conluded that the detail was only due to a nice and important distorsion which makes the sound more alive.
Such a reasoning could explain the love affair with tubes and particularly single ended amps
If you take thirty amps with impeccable specs but not said to be appeally sounding and connect them in series with an attenuator equal to their gain betwwen each and then compare the result with thirty other amplifiers said to be musical, connected in the same manner, I bet that the first series would stay audible and the second one not.
I suggest you build a box with adjustable distorsion and include it ahead of an amplifier having a very low distorsion. Test how react your friends without telling them what electronics they are listening to. Tell us what are the results, it will be really interesting.
I think once you have a system with a three way active crossover using steep slopes, the claimed difference between amps if of little relevance. There is a real gain in clarity and loudpseakers driven by "blameless" amps and the like are really full of life.
~~~~~ Forr
§§§
Nothing is wrong. A bit of distorsion is very often prefered to no distorsion at all. A long time ago, I had the following experience.
I had two lo-fi recorders whose signal to noise was probably no better than 35 dB. Despite of this noise, all material recorded on them seemed very detailed. I have been puzzled by this strange behaviour for long until I conluded that the detail was only due to a nice and important distorsion which makes the sound more alive.
Such a reasoning could explain the love affair with tubes and particularly single ended amps
If you take thirty amps with impeccable specs but not said to be appeally sounding and connect them in series with an attenuator equal to their gain betwwen each and then compare the result with thirty other amplifiers said to be musical, connected in the same manner, I bet that the first series would stay audible and the second one not.
I suggest you build a box with adjustable distorsion and include it ahead of an amplifier having a very low distorsion. Test how react your friends without telling them what electronics they are listening to. Tell us what are the results, it will be really interesting.
I think once you have a system with a three way active crossover using steep slopes, the claimed difference between amps if of little relevance. There is a real gain in clarity and loudpseakers driven by "blameless" amps and the like are really full of life.
~~~~~ Forr
§§§
nothing is wrong.
people have different tastes and it is perfectly legit for you or your friends to love things that others hate, and vice versa.
However, when others design amp, they don't know precisely what your tastes are and it is hard for them to design to your tastes - your are the only ones who can do that, unless of course you have gobs of money to hire someone to do it for you.
so an otherwise perfect amp can sound ****** to you, and an otherwise ****** amp can sound perfect to you.
That is why it is so important to have a standard for all of us to compare amps: a distortion-less amp. otherwise, we are essentially comparing the superiority of our "tastes" or preference.
and we all know how futile that is.
that is why it is so laughable to "guaranttee" that an amp will sound perfect to others based on our personal experience with the amp.
people have different tastes and it is perfectly legit for you or your friends to love things that others hate, and vice versa.
However, when others design amp, they don't know precisely what your tastes are and it is hard for them to design to your tastes - your are the only ones who can do that, unless of course you have gobs of money to hire someone to do it for you.
so an otherwise perfect amp can sound ****** to you, and an otherwise ****** amp can sound perfect to you.
That is why it is so important to have a standard for all of us to compare amps: a distortion-less amp. otherwise, we are essentially comparing the superiority of our "tastes" or preference.
and we all know how futile that is.
that is why it is so laughable to "guaranttee" that an amp will sound perfect to others based on our personal experience with the amp.
forr said:
Hi forr!
That is well said.
That is the reason that make Harold Black invent the negative feedback, because he remarks that , after the signal have passed for a string of many amplifier repeaters , the signal was barely understandable .
But ...as always... many people don't want to find " the wire with gain"
Salut!
Hi, Forr,
The "distorted" output stage behavior=more detailed sound+live (to my friend). The hard part is knowing until how far this "wrong-ness" has to be adjusted. Too little of it will be considered sterile, too much of it will be fatiguing and muddy sound.
Maybe this thread is a joke to some. But if you make power amp for living, you will have to "listen" to what the customers want. Even if what they wanted is actually wrong from your technical point of view. You cannot push them to "Hey....this is how a correct amp should sound like.....you should buy this kind of amp, not the kind you used to think is good........". An amplifier maker make money from the amps that he sell. The only way to make it sellable is to fit to customer's desire, regardless whether it is right or wrong in the technical point of view of the amp maker. If you cannot follow what your customer's ear want, you won't make money for your family Technically ideal sound cannot be pushed in 1 day to fit to the customer's ear. Their ears are years and years used to listen to distorted amps. You can introduce the sound, but maybe will take years and years also until they used to hear that this is the "right" amp supposed to sound.....naaa...maybe they will never ever accept the "technically perfect" sound.....
Yes, it will be interesting. But what happened is not like adding another "niceness button" unit infront of the power amp, it have to do with the internal setting of how the output stage acts. In my case, the output is intentionally made oscilating (in full power), and in PLH, the output stage is intentionally made unbalanced between positive and negative section.
The "distorted" output stage behavior=more detailed sound+live (to my friend). The hard part is knowing until how far this "wrong-ness" has to be adjusted. Too little of it will be considered sterile, too much of it will be fatiguing and muddy sound.
Maybe this thread is a joke to some. But if you make power amp for living, you will have to "listen" to what the customers want. Even if what they wanted is actually wrong from your technical point of view. You cannot push them to "Hey....this is how a correct amp should sound like.....you should buy this kind of amp, not the kind you used to think is good........". An amplifier maker make money from the amps that he sell. The only way to make it sellable is to fit to customer's desire, regardless whether it is right or wrong in the technical point of view of the amp maker. If you cannot follow what your customer's ear want, you won't make money for your family
Technically ideal sound cannot be pushed in 1 day to fit to the customer's ear. Their ears are years and years used to listen to distorted amps. You can introduce the sound, but maybe will take years and years also until they used to hear that this is the "right" amp supposed to sound.....naaa...maybe they will never ever accept the "technically perfect" sound.....You might want to review some of the posts by, say, John Curl, where he talks about negative feedback generating distortion. Lower order harmonics are traded for somewhat lower levels of higher order harmonics. The overall amount of distortion decreases, so the numbers look good and those who read spec sheets are happy.
However, it's not quite that simple...
There are people who listen to these harmonics piling up in the upper midrange and think they are hearing 'detail' or 'more information.' They have a strong tendency to use the buzzword 'accuracy.'
Only, that's not the way real intstruments sound.
One of the things that sends shivers up the spine of a person like this is the sound of a musician's fingers rubbing over round wound strings as he positions his hand for the next note. The problem is that if you play an instrument, or sit very near someone playing, you'll hear that sound, but much, much more muted.
The fact that some musicians fall for this doesn't mean a thing. There's no law that says that musicians have good hearing. (Actually, there's no law that says that musicians are even very good at being musicians, but that's another matter entirely.)
Partly, it's a matter of taste. Partly, it's a matter of education, like learning to taste wine, rather than simply drinking it.
There's a simpler solution than generating distortion...give them tone controls and see if they turn the treble up.
I'll bet you they do...
Grey
However, it's not quite that simple...
There are people who listen to these harmonics piling up in the upper midrange and think they are hearing 'detail' or 'more information.' They have a strong tendency to use the buzzword 'accuracy.'
Only, that's not the way real intstruments sound.
One of the things that sends shivers up the spine of a person like this is the sound of a musician's fingers rubbing over round wound strings as he positions his hand for the next note. The problem is that if you play an instrument, or sit very near someone playing, you'll hear that sound, but much, much more muted.
The fact that some musicians fall for this doesn't mean a thing. There's no law that says that musicians have good hearing. (Actually, there's no law that says that musicians are even very good at being musicians, but that's another matter entirely.)
Partly, it's a matter of taste. Partly, it's a matter of education, like learning to taste wine, rather than simply drinking it.
There's a simpler solution than generating distortion...give them tone controls and see if they turn the treble up.
I'll bet you they do...
Grey
This is a very limited observation of mine, but there has also been mention on this forum and others more or less that:
"When you are very close to a live horn section you will be surprised how harsh, spitty, and bright the sound is."
My son plays alto and tenor sax in a jazz combo, and in a big band, and has various teachers that are very good players. I have heard them close-up, and what surprised me most was how GREAT they sound close up, The players work very hard at their tone, and even with trumpets, the sound is not harsh and distorted.
I think a lot of people hear the same effect that you mention with the guitar strings, and ASSUME that they are hearing what those instruments sound like up close because it seems to have more detail
"When you are very close to a live horn section you will be surprised how harsh, spitty, and bright the sound is."
My son plays alto and tenor sax in a jazz combo, and in a big band, and has various teachers that are very good players. I have heard them close-up, and what surprised me most was how GREAT they sound close up, The players work very hard at their tone, and even with trumpets, the sound is not harsh and distorted.
I think a lot of people hear the same effect that you mention with the guitar strings, and ASSUME that they are hearing what those instruments sound like up close because it seems to have more detail
Now you're discussing psychoacoustics, and there's precious little that's known about that. Even as long ago as the early 1960s it was demonstrated that people could easily be "trained" to prefer the sound of deliberately horrible amps, and to judge that one "better" than a really good one. Now, you take that guy who's been listening to a SET for the past ten years, give him that theoretical "wire with gain", and you know he's going to wonder why his music doesn't sound "right".
Furthermore, there is evidence that factors long overlooked are important, including what were once considered "ultrasonics", and subtle information "lost" under the noise floor. Nor is all distortion created "equal", and it's not necessarily a bad thing. After all, if it was, then you wouldn't include tone controls. Turn up the bass (or treble) and what comes out is no longer the same as what went in: i.e. it's distorted. Now Doug Self published a long polemic against B. Ollsen and the quasi-comp topology. Self had all the numbers to back up his claims against Ollsen. However, I've built them both ways, and Ollsen was right: quasi-comp sounds better AFAIC.
This whole business is going to be more art than science for the foreseeable future.
And that quest for audio nirvana is going to make a lot of people a lot of $$$$$, both from legitimate endeavours, and from $500 Volume Control Knobs.
Furthermore, there is evidence that factors long overlooked are important, including what were once considered "ultrasonics", and subtle information "lost" under the noise floor. Nor is all distortion created "equal", and it's not necessarily a bad thing. After all, if it was, then you wouldn't include tone controls. Turn up the bass (or treble) and what comes out is no longer the same as what went in: i.e. it's distorted. Now Doug Self published a long polemic against B. Ollsen and the quasi-comp topology. Self had all the numbers to back up his claims against Ollsen. However, I've built them both ways, and Ollsen was right: quasi-comp sounds better AFAIC.
This whole business is going to be more art than science for the foreseeable future.
And that quest for audio nirvana is going to make a lot of people a lot of $$$$$, both from legitimate endeavours, and from $500 Volume Control Knobs.
AKSA said:
The reason is not low distortion, but different design flaws. Low distortion does not necessarilly mean grainy sound, though it may happen.
We should find a reliable method to evaluate dynamic spectrum rather than FFT + averaging above steady state signals. This does not say much.
Every time when speaking about low distortion let's not forget we must not allow high order distortion even at very low amplitude levels and newly created high order distortion components. On the other hand, there is no reason to try to reduce H2 to -140dB
, and having H11 = -100dB.
Music is harmony, produced by humans to humans apreciation.
When one person evaluates...feeling the music with his sensors...ears...electrical to chemical transducers and brain recognizing music.... subjective evaluation can be translated in some human noises called speach.
Of course, there are objective values human created...and those values, Mathematics values and Mathematics expressions are another good formula of communication....signs in place of music..to evaluate music...interesting...not a direct formula...but can produce some results too.
The result we use to see, is that objective evaluations are matter of ethernal discussions, and i could not find conclusion nowhere....if some conclusion was found...show me the link to read...and thanks in advance by your kindness to show me this.
I am talking about a forum thread...that had more than 12 visitors, and unless 12 giving their opinnions...show me when majority agreed over numbers...i could see agreement related sonics subjective evaluation only.
But maybe i have searched in the wrong places, of course this is possible.
But, there are many times, that, despite all human differences in musical tastes....preferences of instruments, culture differences (Culture here as people characteristic...not knowledge quantity)... i could see that there are many moments that humans agree.
Maybe because direct communication is more clear...despite the Mathematics precision, the ones use those formulas are always discussing, disagreeing...so...nothing positive result...only doubts.
there are many that agree that JLH is good, despite distortions, despite this, that or those...so.... subjective evaluations, unless result in more positive communication, and could produce for us one reference of quality...one standard that allows our communication....better than...equal them...worst them.
When IMD...transient...clipping, harmonics are beeing already discussed...and now there are difficulties to find some point of agreement.....another metric can be a good idea.
By example, the study of human perception to increase the subjective evaluation...more direct....sound to humans.....in the place of sound to numbers and numbers to humans.
regards,
Carlos
When one person evaluates...feeling the music with his sensors...ears...electrical to chemical transducers and brain recognizing music.... subjective evaluation can be translated in some human noises called speach.
Of course, there are objective values human created...and those values, Mathematics values and Mathematics expressions are another good formula of communication....signs in place of music..to evaluate music...interesting...not a direct formula...but can produce some results too.
The result we use to see, is that objective evaluations are matter of ethernal discussions, and i could not find conclusion nowhere....if some conclusion was found...show me the link to read...and thanks in advance by your kindness to show me this.
I am talking about a forum thread...that had more than 12 visitors, and unless 12 giving their opinnions...show me when majority agreed over numbers...i could see agreement related sonics subjective evaluation only.
But maybe i have searched in the wrong places, of course this is possible.
But, there are many times, that, despite all human differences in musical tastes....preferences of instruments, culture differences (Culture here as people characteristic...not knowledge quantity)... i could see that there are many moments that humans agree.
Maybe because direct communication is more clear...despite the Mathematics precision, the ones use those formulas are always discussing, disagreeing...so...nothing positive result...only doubts.
there are many that agree that JLH is good, despite distortions, despite this, that or those...so.... subjective evaluations, unless result in more positive communication, and could produce for us one reference of quality...one standard that allows our communication....better than...equal them...worst them.
When IMD...transient...clipping, harmonics are beeing already discussed...and now there are difficulties to find some point of agreement.....another metric can be a good idea.
By example, the study of human perception to increase the subjective evaluation...more direct....sound to humans.....in the place of sound to numbers and numbers to humans.
regards,
Carlos
Grey, that is exactly what happens
With oscilating amp, the finger movement in the classic guitar's string becomes sooo obvious, when the amp is made to be stable/no oscilation at all, that sound become very faint, hardly noticeable. I see in scope, the oscilation is about 15Mhz, the guitar's sound is certainly below 20khz. The question is, why the 15Mhz oscilation can make signals below 20khz tobe "super detailed"? What is the explenation?This fact of having a "background" noisefloor can "boost" the details, is this the same case with Phono player system?
Audio maniacs hear the phono gives more detail, more "right" music compared to CD. But when there is "silince" between tracks, the phono is not quiet at all, it has low frequency hiss.
Is that the "more details" of phono plate comes because it has that very big noise-floor / all the time low frequency hiss?
Anyone has tried an experiment that if we "paste" the music program with a certain frequency test tone, will it reveal more detail to the music?
Hi Darkfenriz,
Some also enjoy something in between, not the hard ones, not the plain ones , but the "semi" ones
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.