Sound Quality Vs. Measurements

Status
Not open for further replies.
Frank,

There are limits to what a reproduction system can do, and indeed I find recording and playback done well is exceptionally realistic; this from perspective of attending very good live performances, and hearing very well done live recordings of them well mixed and then played on my setup.

I've had a number of musicians listen to their own recordings, and pick up stuff they thought had been lost in the mix.
 
There is a school of thought that says that only musically deaf people can hear differences in sound quality; that musicians rarely have good sound systems because to them, all they hear is the performance - they are actually incapable of hearing anything else, whether it's on a 'high end' system or an alarm clock radio. There may be something in it. If so, I would be half way along the scale I think.
I wonder if musicians have more hearing damage than the average music listener They're in rock bands where even the drums and cymbals are amplified 10 to 20dB, or in orchestras just a few feet in front of the trumpets.

Also, musicians hear the music differently just from being so close to their own instruments.
IMO, what is not yet known sufficiently well is how precisely the reproduced waveform needs to match the input waveform so that the illusion of reality can be experienced - the psychoacoustics factor. From personal experience, and other's anecdotes, it seems that certain, boundary conditions need to be reached for this to happen.
It's not just the amplifier and speaker, but the microphone(s) and preamp(s) used (and even here, ignoring the rest of the recording/reproduction chain, pretending the musicians are playing live in an isolated space a room or two away). Yes, the focus of these recent posts is the amplifier, but it doesn't exist in isolation.

If someone gave you a perfect (up to clipping, at say 200 watts) amplifier, could you know it's perfect by listening and doing a/b comparisons with other amplifiers?

I'm thinking it would be a lot easier to detect a perfect speaker, as there's so much more variation between "good" ones.
 
There is a school of thought that says that only musically deaf people can hear differences in sound quality; that musicians rarely have good sound systems because to them, all they hear is the performance - they are actually incapable of hearing anything else, whether it's on a 'high end' system or an alarm clock radio. There may be something in it. If so, I would be half way along the scale I think.

@dvv

What are your musical tastes?

In my experience, inevitably linked to place of residence, musicians rarely have good systems simply because they cannot afford the good stuff. This should not be confused with what they hear - trust me, when developing my own speakers, my friend and I were literally amazed at what musicians can hear, but push aside for the sake of the whole. They do not nit pick in general.

Our greatest surprise was a professor of music from the Musical Academy, about 65 years old - the things that man can hear are amazing. When asked which was the best system he ever heard, he said Levinson.

It is not a matter of hearing or not, it is a matter of priorities - they simply place the whole before the parts. I also do so, probably not as much as they.

As for music, I dislike and avoid most of jazz (or what goes for jazz these days, things like "fusion", etc) and am not too keen pn classical music; in both cases, there are works which I do dearly love (e.g. Dave Bruback's "Take Five").

I collect pop and rock from the 1960-1980 era, I dearly love Irish music (especially by Enya, Loreena McKennit, etc), I like folk music (e.g. Steelye Span), Enigma in particular, and gladly listen to some Country music (Waylon Jennings, Johnny Cash, etc) and I love Gregorians. Plus, of course, the many tastes of music from my own country.
 
Frank,

There are limits to what a reproduction system can do, and indeed I find recording and playback done well is exceptionally realistic; this from perspective of attending very good live performances, and hearing very well done live recordings of them well mixed and then played on my setup.

I've had a number of musicians listen to their own recordings, and pick up stuff they thought had been lost in the mix.

Agreed - quality recordings and productions can indeed sound sublime and VERY near to live music.

That said, we need to realize that live music rarely manages to equal good recordings in far too many cases, obviously implying that many mostly pop and rock groups sound worse live than on recordings. No echo, no effects, andtheir talents shrink.
 
OK, so you can hear the difference between amps that do not measure the same...


...and you prefer the sound of the amps that measure worse.



Why do you find that preposterous? It doesn't contradict your other observations that when there is a difference you can hear, the difference is also measurable.

Godfrey, again - if we knew all there was to be known to be able to make a shall we say credible amplifier, many would be doing it.

Yet, they are not. Most sound generally artifical, implyong but not really conveying the musical message.

Now, why would that be? Poor designer education?
 
The limits of reproduction systems are extremely high, I've had hints many times of what's possible, simply through the use of relatively low cost, nominally lower quality gear yielding far more than one would think possible. PA system levels of volume are possible, which subjectively demonstrate no audible distortion -- as an example, provided the system is sufficiently advanced and refined, it would be possible for a drum kit in full flight to be reproduced, such that a drummer, blindfolded, would be fooled into believing that he was standing right besides a real setup.

With regard to 'perfect' amplifiers, there's no such animal -- but there are sufficient systems, meaning that the number and severity of weaknesses are low enough that an illusion of reality can happen.

Why 'good' speakers all sound different in typical systems is firstly because the power amps misbehave depending upon precisely what load they see, and secondly because the "image" of the musical performance is not presented strongly enough, is not convincing enough, and the distortions and non-perfect behaviours of the speaker can't be easily distingished, and unconsciously, instinctively, discarded, by the ear/brain from the recording's 'message'.
 
Frank,

There are limits to what a reproduction system can do, and indeed I find recording and playback done well is exceptionally realistic; this from perspective of attending very good live performances, and hearing very well done live recordings of them well mixed and then played on my setup.

I've had a number of musicians listen to their own recordings, and pick up stuff they thought had been lost in the mix.

I wanted to congratulate you on your results ( 9227 ) . See Quad 63's for comparison added to my valve amp graphs ( 1 metre on axis ) .

Here is a way of looking at things . Below are the final tests of my no loop feedback SE amp . Up to 5 watts I get my desired spectrum . The amp only has 8.4 watts so to almost maintain spectrum at 8 watts is not bad and took time . Usually the 3rd harmonic rises first . I did get less distortion than this . However not with ideal harmonics . At one watt the amplifier is well behaved and doesn't change right down to the noise floor . In fact so similar as to look as if the scope has frozen . Only at low levels does the hum intrude to say how low it is . The analyzer looks almost identical from one watt down .

The reason I show these is by eye is is hard to tell it is a 1% + distortion amp , near zero looks similar . If it was 1% all second harmonic is would be easier to see . I suggest the eye and the ear are no different . To state the obvious . Regardless of what distortion level you choose keep the harmonics like this . The old Quad 303 did and at about - 80 dB .

Notice my amp is about 0.5 db down at 22 kHz full power and slightly more distortion . 47 kHz - 3dB . 15 Hz - 3dB . Damping factor is low so the bass end is dependent on how the speaker works .

The graphs are my notes so forgive . I take the liberty of saying no harmonic above 1% as my 1% THD goal ruined the amp . I think even Quad used those words or was it Dynaco ? Dynaco , no harmonic above 0.2 % I think ? I haven't bothered to do the sum of the squares , guess less than 1.5 % ?

FJrXtbJ.jpg
 
Hi fas42

My experience is that the distortion that matters

You often say this, but have you any evidence for it? As far as I can tell, you believe that

reduction of distortion = better sound

and therefore

better sound = reduction of distortion

without any reference to measurements of any kind because you believe the the ear to be more sensitive to distortion than measurements and measurements involve digital RFI, so measurements can't validate what the ear hears.

It all seems very circular.

I am happy to believe that distortion changes the sound for the worse, but also that things like noise, phase, frequency response matter. Beyond that, I am open to the suggestion that systems go wrong as more power is demanded, particularly on sudden transients and so on. There has got to be a way to measure these things, but no one ever seems to make any suggestions for new types of measurement that might 'zoom in' on these things..
 
Expanding on Frank's classification above, I'd divide amplifiers into several classes:

1. Poor by any account,
2. Not very good, but not terrible either,
3. On the limits of credibility,
4. Credible, solid choices, and
5. The next thing after the real thing, i.e. live music.

I don't much use this classification because what an amp will do in the end depends not just on how the amp's made, but also how the source was made and how the speakers were made (not to even mention whether the room had been properly sound treated).

Before I file away my view on any amp I am interested enough in to go the whole 9 yards, it has to sound similar to me on all three systems at home, each being an entity all of its own, and each in a different room. That includes a variety of devices, but I vouch that as used at this time, they all represent very credible systems, generally in my own class 4, while mine with the Karan is on the borderline between classes 4 and 5. The room is holding it back, otherwise it would be a clean class 5.

2 out of 3 amps are low global NFB designs at any one time, and overll, of 6 amps I own 3 are lo and 3 are high global NFB types (but not equally so, Karan and Sansui use more global NFB than the Marantz 170DC).

I wonder how many people in general test audio in a similar way?
 
CopperTop, you keep asking for proof, presumably what you consider as scientifically acceptable proof.

I do not see this as an explorers' forum, rather as a forum where people say what they think with no obligatio to anyone else.

In other words, I do not need to prove anything to anyone, I'm not here for the scientific bit, I am here because I like to hear what other people think. Perhaps grab an idea I deem interesting along the way.

Then again, I could always turn this around and ask you to prove that I do not hear a difference. A relatively simple behind the curtain switch test would quite eaisly prove my point, but we can't do it, for annoying practical reasons.

I do NOT deny the value of measurements, heck, I use them myself. Aside from the most obvious, such as THD and IM, which I am generally happy with so long as they are below say 0.1% when operating into adverse conditions at near full power level, the only measurement dear to my heart is looking at the rate of harmonic decay. Experience has taught me that if it satisfies a certain decay pattern, I will like it; I found this out working backwards, measuring an amp which I did like, and comparing it with those I didn't like.

And in most cases (say, 8 out of 10 at least), the ones I did like generally measured poorer in absolute terms than those I didn't like. But there were exceptions . for example, my little Toshiba SB-45. It's a cheap'n'cheerful amp by any measure, uses lots of global NFB, yet manages to sound far better than one would expect it to sound, given its low, low price (in 1983, when it was new).

Point is, there are NO absolutes, just probabilities. Chances are vastly on the side of me liking a low global NFB amp better, but are by no means guaranteed or final. Or the old German made LAS power amp - a Lin topology, whith a few twists, open loop PB no better than 4 kHz, but even so, it's like 10 times more than the usual in its day, and it sounded so good that I will make a sample no matter what, for the sheer pleasure of owning it.
 
You often say this, but have you any evidence for it? As far as I can tell, you believe that

reduction of distortion = better sound

and therefore

better sound = reduction of distortion

without any reference to measurements of any kind because you believe the the ear to be more sensitive to distortion than measurements and measurements involve digital RFI, so measurements can't validate what the ear hears.
My 'evidence' is firstly my personal experiences, and then when the Internet became such an effective way for appreciating other people's activities, coming across similar stories. Very few are to be found, but enough to register, at least to me, as being significant. The key elements in the stories of these people repeat, again and again; the pattern is there to be seen if one's looking for it.

The 'observer effect' is a problem, as in measuring procedures creating disturbing interference; it can be overcome if enough effort is made, but this takes energy and resources, both of which I'm in short supply of, :) ...

I am happy to believe that distortion changes the sound for the worse, but also that things like noise, phase, frequency response matter. Beyond that, I am open to the suggestion that systems go wrong as more power is demanded, particularly on sudden transients and so on. There has got to be a way to measure these things, but no one ever seems to make any suggestions for new types of measurement that might 'zoom in' on these things..
Something like a highly developed, more advanced version of Diffmaker would probably do the job - I note that some people have found the current version has easily reachable limits, which make it not all that useful.
 
Let's be practical here.
There are several of you here who believe that measurements are everything, so when two amps measure the same, they must sound the same. I think that's a preposteruous claim, but that's my problem.
it's a problem because you are not even trying to consider some else's point of view. amps measuring the same is something that simply doesn't exist. we all know that differences come from the tiny stuff and most of the measurements differ in ways that can be described as anything but tiny.

MOST of them measured well
well = ?

If you gents can't hear any differences, that's your problem.
it's only you who thinks that we can't, I repeatedly said that I for one can but you are ignoring it.

You think I'm imagining things, I think you are musically deaf;
dvv, you seem to have a problem in understanding people's POVs. you have done that in the past and I tried it to explain it nicely. I never said you are imagining things, that's what you think because you seem to be incapable of changing perspective for one second. I know I am capable of doing that.
and now you tell me I am musically deaf, which is an insult and I could take it as such. you don't know me and have no idea about how I listen to music, but you are convinced you do, like you are convinced about a lot of incorrect things.

Let's leave it at that we agree not to agree. :p
dvv, your posts contain some conjectures and assertions that IMO are 100% wrong. like that "4 degree phase shift" thing.
and yes, a forum is a place where people say what they think but it's also a place where people will call you up on it, sooner or later. the moment you start conjecturing and don't accept any criticism, it all goes down the toilet and a forum becomes nothing more than a chit-chat place (i.e. useless).

and since you took the liberty to insult me, I'll tell you one thing: you have practically taken over this thread. you're a Juggernaut, you just go on and on and on and the moment someone disagrees you just tell them "let's agree to disagree" or some other politically correct line that actually means "I made this thread mine, go away".

Experience has taught me that if it satisfies a certain decay pattern, I will like it.
the decay pattern tells you where most of non-linearity lies (at the peaks or near zero, xover distortion etc).

I do not see this as an explorers' forum, rather as a forum where people say what they think with no obligatio to anyone else.

In other words, I do not need to prove anything to anyone, I'm not here for the scientific bit
yes, but this is not your thread or forum. some other people have different views which are just as correct. maybe some people are explorers. I don't recall any forum manifest that says this forum is what you think it is.
 
Last edited:
CopperTop, you keep asking for proof, presumably what you consider as scientifically acceptable proof.

Well the basic topic of this thread is how to measure sound quality without accounting for fallible experimenters' prejudices, moods, foibles, earwax. You may be unique in that you have none of those idiosyncrasies, but most of the time I have to take the word of mortals who, to be honest, I think are often just making it up as they go along. I love the idea of better audio quality, but the nuggets of information and research that might actually make it possible are buried beneath tons of subjective made-up opinion.

Do you feel no curiosity why you might prefer the sound of certain amps over others (low NFB vs. high NFB)? Wouldn't it be nice to actually see the difference on a screen and confirm it? And from this evidence modify the existing design to make it even better? Without measurements you would be allowing what could be a 1-dimensional problem to remain a 100-dimensional problem i.e. one you will never solve except by chance.
 
Something like a highly developed, more advanced version of Diffmaker would probably do the job

This is what I was talking into an apparent vacuum about yesterday. To make it practical the system has to be constantly breaking off to measure the amp's frequency and phase response as they vary with temperature etc., so that these can be removed from the measurements. Then the system can concentrate on isolating the short term wobbles that might occur on real music signals. It can still do averaging over many repetitions of the same signal, just like a more conventional distortion test.
 
@mr_push_pull

If you feel insulted by me, I apologize and assure you that was never my objective. May I suggest you look at the same thing, but in the opposite direction?

But never mind that, let me make it as clear as I can - this is by no means MY forum, and suggesting that I think so because essentially I have a different point of view from which I see no reason to budge will not win you any merits for tolerance.

Rest assured, i will not respond to your posts any more, I'm really not into "you did that" and "I did that".

@CopperTop

I assure you I am anything but unique. UIt's just that I've invested 40+ yeras' free time (and sometimes not free) into this hobby (to me) and have drawn my own conclusions and formed my own habits. Neither is guaranteed to be THE truth, nor am I serching for THE truth.

Of course I feel curiosity why some amps sound good to me, while others dont, and, to add insult to ijury, many exalted names and models don't sit well with me. Some I feel are a con game. Most I find grossly overrated. While some humble products I feel are unjustly underrated and sometimes even overlooked.

There was a time, many years ago, when I too felt that all this HAD to show up in numbers, somewhere, somehow. So I measured ad nauseum, only to find that there was no rule I could even remotely rely on. But some things did crop up statistically. Mostly two.

One was that a certain decay rate of harmonics seemed very desirable to me, and of all the amps I liked, at least like 90% had some variant of just that decay rate. Oddly enough, it doesn't seem to make any difference whether the absolute THD was 0.1% or 0.001%. On the other hand, the group I did like consists say 90% of amps using low global NFB rates. Leading manufacturer - Harman/Kardon. Consequently, I own three of their amps, 2 integratedsd and one power amp and preamp system (though frankly, the power amp is the better half of that duo and works better when driven by my Luxman C-03 preamp rather than its series companion, Citation C22 (?)).

The second is that you can change one single part with another from another source, remeasure and find no measurable difference, yet still detect mostly a minute difference in sound, and in very rare cases, a literally momentous difference, like it was now something costing 3 times the price.

A typical example: change the make and model of your filter caps. It will likely measure exactly the same (assuming both sets of caps are of similar quality), but it will not sound the same. Or, change your perfectly good metal film resistors, really down to 1% tolerance, with the same value from say Caddock or Dale, and you will hear a diffrence for the better. Small, but detectable. You may care, or you may not, your business.

Also, as a general observation, I tend to find fast, even insanely fast, amps to sound more open and giving than slow, bandwidth limited amps. Not always the case, but in a convincing majority it is. But one can also overdo the wide bandwidth theme and end up with a shrill amp.

Just MY opinion, no rules carved in rock.
 
I've mentioned variations on the following general testing methodology many times, but it's always ignored or discounted as being too similar to current techniques. Input a mix of several high frequency audio sine waves at very low levels, say -40db from max - do a spectrum measurement of the output. Then, stress the whole system in various ways so that theoretically the distortion components resulting from that original mix are not directly impacted, look at the spectrum content again; my belief is that a 'good sounding' system will show much less variation of levels of the original spectral frequencies than a 'bad' one.
 
My ideal amp would be > 100 W 8 R ( 1000 W 1 R transient ) . < 0.1 % THD with suppression bridge reducing harmonics . Output impedance < 0.5R including cables . No dramatic change in distortion spectrum with level . 5 Hz to 20 kHz -0.1 dB . Noise > - 100 dB . Stable into real loads .

My valve amp sounds more detailed and open than most . If the bass was tight I would not bother with better .

Global feedback is the best there is if it suits the amplifier . That's the problem . In most cases it doesn't . With my valve amp feedback could be made to work . If I did it would be nothing to do with distortion . Trends I saw with the feedback version valve amp also are seen with all feedback amps I have played with . I suspect an amp has to be very fast indeed to be an " ideal " feedback amp ? I was shocked how unhappy valve amps are with feedback . Logically that would be true . It can be made to work . In the process one sees the whole problem . If I built an OTL valve amp the first thing to try is gallons of feedback . With the right valve it has to be fast enough . I have heard OTL .They are so dam fast and open . A tad bright I would say . Without an analyzer on hand I suspect a lot of third harmonic and perhaps some fifth ? To be honest I have never heard anything like them . So dam fast .

Some will say I would prefer my 0.1% distortion ideal amp over one that is " more perfect " because I like distortion . Nonsense , what speakers could prove that ? If it is true it will be mechanisms in the machine the feedback loop is hiding that are different . As Dvv says , what is open loop like . Dvv your amp is almost ridiculously good open loop . If you added just enough feedback to meet my output impedance ideal and pot down the input I bet your amp would meet a good spec ? Try it sometime . I think you might be shocked how you like it . It will sound very open and fast is my guess ? Perhaps a bit grainy at times like a good photograph .

One thing is . An amp might have no virtue except the ability to accept feedback . Like an inherently unstable aircraft with the right mechanisms it might be OK . One designer told me that linearity in amps is totally unimportant if there are no dramatic jumps in gain at any level ( open loop ) . Given enough speed that's all that mattered to get the feedback to work .
 
Rest assured, i will not respond to your posts any more, I'm really not into "you did that" and "I did that".
I'd be very happy if you did that. maybe you don't think it's your thread or forum but you sure behave that way. I absolutely think you should take a break from this thread. it's my impression that people stopped posting here because sooner or later you tell them what's right or wrong. and there's simply too much of you here, the last hundred or so of pages are nothing but conjecture from you.

Just MY opinion, no rules carved in rock.
using polite words or sentences does not make you polite. misrepresenting POVs or playing the "I didn't mean to sound it that way" or throwing assertions around the place is just as bad as swear words. but rest assured too, I'll unsubscribe never to return, or at least until I start seeing an actual discussion and not a monologue led by you.
 
Last edited:
Status
Not open for further replies.