[boead]

The point is that lack of distortion may not be all it's cracked up to be.
Another example is most digital amps (the ones I've heard) all leave me feeling emotionless.  They do many things right, but just don't involve me...

Yeah, I hear lots of people saying that. They do lots of things correctly but lack musicality.
My little flea-powered amp is well modified, I likely spent more on it then it’s ever going to be worth but it is really very musical. I can list its faults and complain about them but in the end its musicality is what makes me keep it.

[John Ashman]

Lack of distortion is almost always a plus.  But sometimes it takes one set of colorations to compensate for another.  For instance, speakers that are called "too accurate" aren't too accurate, they just are exposing their own weaknesses in distortion or other areas that might be pleasantly masked by a warm tonal balance.  But if the speaker is truly low in distoriton, the accuracy will simply be a plus.  And vice versa.  A lot of times when I hear that an amp or whatever is too "analytical" or otherwise flawed, it's  usually the speaker, but the amp is getting the blame for not covering for it.

[timbley]

The point is that lack of distortion may not be all it's cracked up to be.
Another example is most digital amps (the ones I've heard) all leave me feeling emotionless.  They do many things right, but just don't involve me...

Yeah, I hear lots of people saying that. They do lots of things correctly but lack musicality.
My little flea-powered amp is well modified, I likely spent more on it then it’s ever going to be worth but it is really very musical. I can list its faults and complain about them but in the end its musicality is what makes me keep it.

The above quote with my name on it isn't actually my quote. I'm shooting for the lowest possible distortion that I can afford in my system. But I do think some types of distortion are a lot less irritating than others. And sometimes correcting one thing too much when something else is way out of whack can be more distracting than if everything is more or less even. For instance, I once auditioned a pair of ribbon speakers with a tuned port bass. That was a bad combination because the ribbon just pointed out how scuffy and boxy that little tuned port box really was. I preferred my Polk speakers because they were nice and evenly murky and boxy, allowing my mind and ears to adjust and get into the music.

[jules]

All respect to the latest turn in this thread and no doubt distortion of one type or another is important but at the same time this quest for low .0000001% or whatever distortion has been going on for some time and I can't see it as the driver of a great deal advance. Apart from implying that an amp might be 99.99999% perfect it also suggests that all that is needed is to get rid of that last pesky bit of distortion and bingo ... the perfect amp.

On the other hand I fully support the point boead made about the sound of a live trumpet v the recorded sound. As I see it the same applies across the board. I have a fair bit of experience with the clarinet. On the face of it this might seem like a fairly un-challenging sound to reproduce but I have never heard this instrument in recorded form come anywhere near the complexity of the real thing. What is lacking [and this concurs with boead observation] is a sort of egdy roughness. This quality makes up a minor part of the whole but that's the point isn't it. A musical instrument isn't a tone generator and any particular note on any particular instrument is branded not by it's purity but by it's variation from same [otherwise a violin would sound the same as a trumpet].

It would be great if, instead of having distortion as a revered measurement we could have a "% of the original sound" measurement. Of course it's a fantasy in that it can't be measured. I didn't really tie together my earlier post in which I suggested that we might be hearing something between 5% and 50% of a musical source [in a recording] but if this is so then "neutral" simply becomes a sampling of the bland from what is available and "transparent" might merely be something that leaves a lot out!

Luckily our heads and our imagination are great at filling in gaps but perhaps if we were a bat  or a dog we might be more critical.

jules

edited

[timbley]

It would be great if, instead of having distortion as a revered measurement we could have a "% of the original sound" measurement. Of course it's a fantasy in that it can't be measured. I didn't really tie together my earlier post in which I suggested that we might be hearing something between 5% and 50% of a musical source [in a recording] but if this is so then "neutral" simply becomes a sampling of the bland from what is available and "transparent" might merely be something that leaves a lot out!  
 
 Luckily our heads and our imagination are great at filling in gaps but perhaps if we were a bat  or a dog we might be more critical.
 

I think distortion in full context is a "% of original sound" measurement. To get the clarinet to sound something like the real thing, we have to measure where distortion is letting us down so we can fix it. I don't see how getting rid of distortion could do anything but help. I don't think we're really anywhere near 99% with our equipment yet, particularly with loudspeakers and lousy listening rooms.  If you were to measure the sound of a real clarinet, you'd still be looking at the microphone's distortion, although that would probably be comparibly small. But If you were to look at that recorded waveform, and then try to play it back through any loudspeaker, amplifier combination and measure it again, my guess is it would be very easy to see, and hear the difference.

[jules]

Well no, that's my point. I don't think it's the distortion that's the issue so much as "insensitivity" if I can put it that way.

The primary tone that an instrument emits is the loudest part of it's output but what makes it distinctive is the subtler overtones, resonances etc etc that can't be measured but can very easily be heard if you happen to have the chance to make a live comparison. Simple as that might sound, a live comparison sometimes isn't easy to come across. At the same time it doesn't mean you need to listen to a virtuoso performance and you don't even need a listening room. A kid practicing a musical instrument is enpough to do a test.

Maybe I should modify my suggestion slightly but I stand by the point that it's easy to reproduce the louder fundamentals while the kaleidescope of subsidery tones that makes up the full picture could easily be down in the range of 5% to 50%.

jules

[timbley]

I would say that insensitivity, the inability to produce subtler overtones, would count as a type of distortion for sure. Maybe this is just a semantics thing, since so often in specs distortion means THD plus noise, which I certainly agree comes no where near telling the whole story.
OTH, compression could be considered distortion. But it can also help to reveal subtler sounds, and I've read that in listening tests a lot of people actually prefer some compression. So maybe there's a case to made that some distortion could help.

[lonewolfny42]

the cleaner the window the more you can see.

That is correct....helps to have a clear view.

I like to hear a lot of detail in my music..so when I discribe a speaker as detailed I don't mean a lot of treble or harsh sounding . I guess I should switch to saying the speaker or sound was very "transparent"...[/list:u]
John Ashman:

"Detail" as it's currently used usually means "bright" and/or "harsh". Most "detailed" speakers are anything *but* detailed and they are often laden with distortion. Some very popular "high-end" salon brands come to mind. If the first thing you notice about a speaker or amp is how "detailed" it is, it is almost certainly just brightness. Real detail is the lack of distortion. It is easier to fake detail

[/list:u]

[boead]

I don't see how getting rid of distortion could do anything but help....

Well techniques like negative feedback used in amps will reduce distortion but will also reduce fidelity. The type of distortion is very relevant here. Some types are perfectly fine to the human ear, they may actually ‘add’ to the musicality because the distortions are somewhat natural. While others clearly aggravate the senses.

I have never been much to pay attention to distortion and after diving into the world of tubes, I sort of abandoned the notion of distortions as a gauge to fidelity. However, when I was shopping for a digital source, I started reading how some DAC designers attributed fidelity to distortion. dCS ring chips boast technology (inherent design) that lowers distortion and increases linearity and according to the makers, greatly improves musicality. After listing to many players, I can attest to the results. I ended up buying an Arcam with the dCS ring chips because I was impressed with its ability to be not just accurate but tremendously musical.

[Steve]

I think Lonewolf hit the nail on the head. It is true transparency and resolution (detail but not bright) that are key. The clearer the window, the more is actually heard.

I don't think artifically adding distortions is the way to go as true nuisances are still missing, but it is certainly better than alot of the bland components I have heard.

I think the main weakness is that the predictable/constant amounts of harmonic distortioins, 2nd, 3rd, 4th and IM distortion, etc that will change the character of the music depending on the instrument. IM, in tubes is typically 3 times that of HD. So if 1%, the IM may be close to 3%.

There are also many other types of distortions besides harmonic or IM.
There are sonic signatures from each individual part. And global feedback, to lower the overall HD, produces even higher order harmonics because the first stage becomes a mixer of two signals.

 As someone nicely pointed out, the designer may be trading one form of distortion for another, and some distortions are easier to live with than others.

I would say under the very best conditions, the very best systems, the very best recordings maybe close to 70% or more true to the music. I think alot of the remaining problems are in the recording studios. Luckily, there are some better studios out there, but they are few and far between.

[timbley]

I'm interested in negative feedback.
Negative feedback trades lower order harmonics for higher order? How does the listener typically perceive this?
I just went from a Panasonic digital receiver, which I read is intrinsically a zero feedback design, to a JVC "Hybrid feedback" receiver, which has a name that pretty much says it: two kinds of feedback at once, digital and analog.
I would say the JVC has a more colored sound, being richer and warmer. The Panasonic seems neutral to me, neither warm or cold that I can tell.
A little warmth is actually kind of nice, and not at all fatiguing. But I'm not sure that I prefer it.
Another post quoted a Phillips engineer who said that the Texas Instruments zero feedback, digitally controlled concept was  dead end due to unfixable high distortion levels. Hmm, it sounded awefully nice to me.

[Steve]

I'm interested in negative feedback.
Negative feedback trades lower order harmonics for higher order? How does the listener typically perceive this?">

Well, if you have two stages, each producing Only 2nd and 3rd harmonics, the feedback signal also contains these harmonics. Now, when combined with the input signal, at the first stage input, one gets 2nd, 3rd, 4th, 6th, and 9th order harmonics thru the 1st stage. Of course, thru the 2nd stage is even worse.

Add a 3rd stage, and look out. Also depends on the phase relationships between each of the feedback harmonics and each of the music harmonics. Some may add, cancel, or in between.

The IM distortion is a real culprit, as it adds and subtracts between each and all the harmonics present, including the main signal. If an amp has 5% HD, the IM is generally around 15%. Ever hear dirty music, lots of garbage? May be IM distortion. That is why some amps don't like complex music, like orchestras, and prefer simple music.

Depends on how much feedback, what kind etc. If not much, maybe very little. It depends on the phase relationships. If alot, the instruments may be perceived as having either a tinney/bright sound as the higher order harmonics of the instrument are artifically accentuated, or a lack of brightness if the higher order cancel out. Or in between. It is difficult to say.

As one can see, the most profound affects occur with more stages.

That is why the output device needs to be carefully chosen. One device may be slightly cleaner, but may need a higher drive signal, more stages, with more distortion (% wise). So the previous stages are producing alot of distortion.  
The second output device may produce slightly more distortion, but needs less drive, fewer stages, less distortion (% wise). The previous stages produce much less distortion.

Passive parts have a tremendous impact on the sound. Most slant the sound either warmer, or cooler. Most also add harshness, bloated bass, lack of focus etc. Just because an ad says "high quality metal film", carbon, or other brand resistors doesn't make the resistor transparent, balanced tonal wise, or "truly clear".

<<"I just went from a Panasonic digital receiver, which I read is intrinsically a zero feedback design, to a JVC "Hybrid feedback" receiver, which has a name that pretty much says it: two kinds of feedback at once, digital and analog.
I would say the JVC has a more colored sound, being richer and warmer. The Panasonic seems neutral to me, neither warm or cold  ...">

You are probably noticing more of the sound of the passive parts used. As mentioned above, most parts add their own sonic signature, many times profoundly affecting the music.

[John Ashman]

As anyone ever measured IM distortion on a speaker?  It's funny that speaker companies can quote .5% distortion on a product that is +/-5dB, not time or phase correct and sounds terrible.

[Steve]

As anyone ever measured IM distortion on a speaker?  It's funny that speaker companies can quote .5% distortion on a product that is +/-5dB, not time or phase correct and sounds terrible.

Nice post John. Seems like along time ago I heard something about it, and article or a spec. But I cannot remember now.
As you mention, +/- 5db (10db) can be pretty bad and also depends on where the "bumps" and "dips" are. Almost like having a room that echos when unfurnished (real resonant problems).

[doug s.]

i've been reading this thread & find it quite interesting.  i think what someone equates to "transparency" & "detail" is whatever they think it is.  i, for one, equate them to be pretty much interchangable terms.  i have never thought "extremely detailed" to mean "overly bright", "harsh" or "edgy".  for that, i use the terms "overly bright", "harsh" or "edgy".    

now, when i bought my stock art di/o & started modding it, i noticed improvements in detail.  as i modded it in stages, i noticed it getting more & more detailed.  then, i went back & switched out some parts - it got even smoother, while not losing the added detail.  never was it harsh or edgy - the added smoothness was noticable only in relation to how it was before.  added smoothness was gained when i swapped out the holco resistors for riken ohms, & when i swapped out the lt1365 opamp for the lt1362.  

one cd i was listening to, that made the added detail apparent was "hawaiian slack key guitar masters, wolume ll".  this cd had what appeared to be noise on some closely miked solo recordings.  as the detail was gained w/mods, the noise became more & more apparent, as actual sounds the musician was making, breathing, moving in his chair, etc.  i also noticed stringed textures becoming more apparent on the jerry douglas/edgar meyer/russ barenberg cd "skip hop wobble".

another thing about detail, is that, after i had finished modding my di/o, auditions of other speakers made me aware of when i was losing detail.  if my main speakers weren't as detailed as they were, i woulda never heard all the additional details i was getting w/my di/o mods...  so, earlier comments that you can hear the detail once it is brought to your attention, yust don't agree w/my experiences.  once i have heard something, i listen for it; sometimes i cannot hear it, if the speakers/electronics aren't as detailed.

re: distortion, i have recently been playing w/a kailin tubed audio enhancer i recently bought.  i have always been fond of my modded ase z-man tubed buffer stages, & i wanted to give this newcomer a listen.  i have presently been using it w/my tunas, i may try it on my cd system.  i am also awaiting some nos 12ax7's to show up, which will most certainly be a step up from the junk chinese 12ax7's supplied w/the thing...  in any ewent, even w/the chinese tubes, i ever so slightly prefer the sound of my s/s tunas (at least the tandberg 3001a & the aiwa at-9700u), w/the kailin in the chain.  and, this was true w/pretty-much every tuna i have tried my ase z-man with, also.  if measurements were taken, i would be *wery* surprised if the distortion figures ding go up when these tube buffer stages were in the audio chain.  but, i find the total presentation ever so slightly more musical, w/o losing any detail.  yust that slight bit of added distortion, uh, i mean warmth & smoothness that i like.

ymmv,

doug s

[WEEZ]

I think that a lot of the audio 'terms' that get thrown around probably mean something different to some than to others, as doug s. said.

If one accepts the definition of transparency as a 'lack of colorations'- then 'warmth' is a detriment to transparency- as it is probably a coloration. Of course, we could try to define "warmth" and we would come up with many different ideas on what constitutes warmth, too.  

Yet most listeners prefer a warmer sound to a cooler sound. I know I do. That's probably a trade-off that some would say is "giving up some transparency". But it does not mean losing resolution. Or the resulting detail.

Too much warmth, however, - and your stereo could end up sounding like a Bose wave radio.  

Even more serious; a lack of resolution could make a saxaphone sound like a kazoo.  

WEEZ

[_scotty_]

I have to agree with other contributors to this thread who think warmth is a coloration.  I have never heard "warmth" present in live music.
  What I have heard are a number of qualities that are associated with
the performance of live music.  These are; unrestricted dynamics, no distortion whatsoever, and the same sense of attack from all instruments regardless of the frequency range that the instrument produces when played.
I also always hear the sound projected towards me with a sense that the air between me and the instrument is being highly energized by the action of playing the instrument.  As most instruments are constructed to be heard by
more than just person playing them this isn't too suprising.  There is also never any doubt that I am experiencing reality, you know it when you hear it.
Another quality that is present when music is experienced live is the ability to hear the softest passages just as easily as the loudest passages, the only caveat being that one is not too distant  from the performer.

 Live music is very hard to ignore even when you don't particularly care for the style being performed. This is in stark contrast to Muzak which almost no
one even realizes is playing because is doubles as musical wallpaper.
    These qualities are what I expect from from reproduced music
in my home. There should be the same attack at all frequencies, the same
sense of unrestricted dynamics, the music should still sound dynamically alive even when soft passages are being played or the playback level is reduced.The lack of perceivable distortion should be the same, and the sense that the energy is being projected out into the room at me by instrument being played should be present. There should  be no sense of frequency response limitations.  
  While I have made a lot of progress towards a realistic sounding system using the aforementioned list of criteria to evaluate equipement, the unmistakable and illusive quality of being real
is not present in my listening room yet. I still have to buy a ticket or pay a cover charge.  I hope my evaluation list is helpful,
Scotty

[Steve]

Couldn't agree with you more Scotty. Nice post.

[timbley]

Thanks for the input on negative feedback Steve. The JVC's implementation of it doesn't seem to be causing much harm, because it strikes me as smooth and sweet, not at all edgy.

When I think of the JVC's sound compared to the Pannys, it's like the difference between an incandescent and a daylight bulb, although not nearly that extreme. Neither seems to give up anything in the way of clarity. Either bulb can show a full pallet of colors, although the relationships get changed slightly. Once you are immersed in incandescent light, it ceases to seem quite so yellow, and a daylight bulb can appear blue.
When I first hooked up the JVC, the warmth was obvious. But now that I've been using it (them) for a few weeks, I don't notice it so much. I'll bet the Panasonic will sound somewhat cool at first if I hook it up again.

[timbley]

Now here's a question:

Will it help a negative feedback amplifier if it is only asked to cover a limited frequency range, such as with an active crossover in front. More specifically, will it reduce the negative effects of negative feedback.