subjective - objective clash
Sensitive issue, small wonder this thread goes lively along, for it touches perhaps the most controversial (and economicaly significant) side. Witness the prices commanded by "premium directional" loudspeaker cables to name one.
Yet I feel there is the technology - available even to dedicated and modestly equiped individuals - to make some inroads at least to grasp a better understanding. Probably what I am proposing below has been already done.
Imagine you connect a good quality A/D converter capable of 24 bit resolution and perhaps up to 96 kHz sampling rate or beyond, to the input of an amplifier under characterization. Connect another one to the amplifier output, and record both signals simultaneously. This could be done with different program sources including everything you could possibly be interested in, with regards to testing the selected amplifier.
Now scale and substract the signals and keep the difference. With this data set, do as your imagination conceivably brews like spectral / time plots, edge detection, whatever.
I guess this "signature" will be broadly similar for units of the same design and brand, and different from others.
Now comes the good part, how does this relate to the listening experience?
Is a "perfect" i.e. an almost defect free in terms of distortion, slew rate and transient response etc. unit nicer to the ears?
Is a "nice to listen" (at least for some source material) amplifier signed by a peculiar aberration from perfection, can it be characterized?
I guess even the most hard core supporter of music listening as a whole experience rather than the addition of objectively perfect parts, will admit a system that essentially scales a signal without introducing any artifact, is as far as can be done to bring the stage to the living room. Here is a way to test it.
The test may of course be extended to include the drivers, but this requires not only a well characterized quality pickup, but certainly special facilities like an anechoic room etc.
Rodolfo
Sensitive issue, small wonder this thread goes lively along, for it touches perhaps the most controversial (and economicaly significant) side. Witness the prices commanded by "premium directional" loudspeaker cables to name one.
Yet I feel there is the technology - available even to dedicated and modestly equiped individuals - to make some inroads at least to grasp a better understanding. Probably what I am proposing below has been already done.
Imagine you connect a good quality A/D converter capable of 24 bit resolution and perhaps up to 96 kHz sampling rate or beyond, to the input of an amplifier under characterization. Connect another one to the amplifier output, and record both signals simultaneously. This could be done with different program sources including everything you could possibly be interested in, with regards to testing the selected amplifier.
Now scale and substract the signals and keep the difference. With this data set, do as your imagination conceivably brews like spectral / time plots, edge detection, whatever.
I guess this "signature" will be broadly similar for units of the same design and brand, and different from others.
Now comes the good part, how does this relate to the listening experience?
Is a "perfect" i.e. an almost defect free in terms of distortion, slew rate and transient response etc. unit nicer to the ears?
Is a "nice to listen" (at least for some source material) amplifier signed by a peculiar aberration from perfection, can it be characterized?
I guess even the most hard core supporter of music listening as a whole experience rather than the addition of objectively perfect parts, will admit a system that essentially scales a signal without introducing any artifact, is as far as can be done to bring the stage to the living room. Here is a way to test it.
The test may of course be extended to include the drivers, but this requires not only a well characterized quality pickup, but certainly special facilities like an anechoic room etc.
Rodolfo
john curl said:
But is that perception in fact due to "info" being removed, or the lack of "distortion" being added?
It's been fairly well demonstrated that various "distortions" can be subjectively perceived as fuller, richer, more detailed, more dynamic, more "real," etc.
If you look at the SET/horn contingency, you'll find that the most common reason given for their prefering such sustems over other, more linear systems is that to them these systems "sound more like real music."
I can't help thinking of Gordon's second definition of "accuracy" in his The Audio Glossary:
The ultimate objective of an ideal sound system, which everyone claims to want but nobody likes when he has it.
I think we need to get out of this hard-core objectivist rut that assumes that if something sounds better, it must be due to the improvement of some objective performance parameter.
While most audiophiles consider themselves subjectivists, I think that deep down, many are really objectivists. I often see it said that such things as SET tube amps sound good not because of their distortions, but in spite of them, which implies that there's something going on so much better objectively than everything else that it trumps what's going on which is objectively worse.
In a society where the quantitiative dominates the qualitiative, I can see where the self-proclaimed subjectivists would want to cling to their "inner objectivist."
se
Yes indeed, John.
This is my experience too. I was there in 1983(?) in the University Audio Society - a complete bunch of turntable junkies - when we got a sneak preview of the first Philips CD player. All of us, fed up with the idiosyncracies of turntables and phono amps and dusty vinyl waiting with bated breath for the new digital, distortion-free, perfect reproduction. Expectations were high and attitudes open. Of course the player sounded worse than a Dual 505 on a very bad day. It took another 10 years or so for the so-called "perfect reproduction" to start rivaling vinyl and even today it is controversial. The reasons are widely known: phase jitter (a little like PIM, eh?), non-linear DACs, imperfect brick wall filters, digital noise correlated to the music signal, crummy analogue sections and so on. And some of the life-killing problems in CD are similar in effect to those in amplifiers, particularly high NFB amps. It took a lot of time for me to find a CD I could enjoy and had to pay a pretty penny for a ML No39. I don't doubt it could be beaten in an "involvement" contest by a Linn LP12.
john curl said:
Accidentaly just yesterday night we were listening to SONY 777 player and quite clearly hearing these differences. With IC based preamp. No standard cable used, but http://www.pha.inecnet.cz/macura/buffer_en.html
This one replacing CD output cable, and the end of cable terminated by 50 Ohm. This is very interesting how dramatic sound improvement this brings and removes mentioned CD problems. Nobody who did not try believes and everybody who heard comparison between usual cable and buffer+terminated cable is very very surprised. The same for preamp output, this means that it is a pretty high-power link signal transmission. Don't ask me for exact explanation why it works (I do not know why) but try it. The preamp used can also be seen on my page.
Well folks, what do I know? ;-) I do know this: I have all discrete high speed electronics of my own design following the Sony player, up to the Wilson WATT speakers. Maybe I'm just deaf or crazy, but I can't easily tell the 'signature' of DVD vs SACD on my player. CD, I can immediately hear as inferior to either DVD or SACD. I still think it is the cheap IC in the line stage, that is the weak link in the chain. My associate and I have had the same problem before, when we added analog IC based products with our discrete products at CES shows.
MikeB said:
This is almost always true, given that loud speaker-room interaction is bound to make a mess of your systems response...
Moreover, even if the later were not an issue, human ears do NOT have a flat frequency response....
Thus even assuming an ideal system with zero distortion of whatever complexion, your system is bound to sound different in different rooms....and in many instances, it may sound totally crapulous...
Solution? At the very least, use tone controls....DSP solutions prefered though...
http://www.harman.com/wp/pdf/Loudspeakers and Rooms - Working Together.pdf
http://www.harman.com/wp/pdf/MaximizingLoudspeaker.pdf
http://www.harman.com/wp/pdf/AudioScience.pdf
http://www.harman.com/wp/pdf/Loudspeakers&RoomsPt1.pdf
http://www.harman.com/wp/pdf/Loudspeakers&RoomsPt2.pdf
http://www.harman.com/wp/pdf/Loudspeakers&RoomsPt3.pdf
http://www.harman.com/wp/pdf/HowManyChannels.pdf
http://www.harman.com/wp/pdf/multsubs.pdf
The Flat Response
Mike,
Whilst not for one second decrying thre goal of acheiving a perfect in-room response, my experience is such that even the often large tonal anomalies that one gets in real rooms, with real speakers have little, if any, effect on the subject of 'realism' as I percieve it to be understood by Traderbam and others.
The same problems arise at live events, and in many cases the response aberrations would be many times greater, yet this does nothing to the fact that when I close my eyes I know it's the real thing I'm listening too. Theres way more important things to musical communication than the odd response error.
i do not believe for one moment that the solution to the suspension of disbelief problem is to feed the signal through ever more complex DSP systems, in order to preserve whatever it is that gets lost in all HiFi systems - do you really believe that?
Andy
Mike,
Whilst not for one second decrying thre goal of acheiving a perfect in-room response, my experience is such that even the often large tonal anomalies that one gets in real rooms, with real speakers have little, if any, effect on the subject of 'realism' as I percieve it to be understood by Traderbam and others.
The same problems arise at live events, and in many cases the response aberrations would be many times greater, yet this does nothing to the fact that when I close my eyes I know it's the real thing I'm listening too. Theres way more important things to musical communication than the odd response error.
i do not believe for one moment that the solution to the suspension of disbelief problem is to feed the signal through ever more complex DSP systems, in order to preserve whatever it is that gets lost in all HiFi systems - do you really believe that?
Andy
quote:
Originally posted by FrankWW
The listeners sorted out NFB taste on the lines of preference for musical genres which are quite different in their demands on the amplifier. I think that's interesting, also.
Statistically this said much about the listeners taste in music, and hardly aything about the amps. per se...
__________________
The artist formerly known as mikek
But it does tell us something besides that, doesn't it? Playing classical music, especially orchestral, requires the amplifier provide a greater dynamic range and deliver us more information than do jazz or folk music. Classical music gives the amp more opportunity to produce nonlinear products. The NFB regime has a great deal to do with the quality and even the amount of information listeners receive.
For instance, classical singers' formants and solo violin harmonics, which intentionally often are not masked by orchestral sound, are sitting in the area in which our hearing is most sensitive. And when these singers and instruments start to sound like they're on steroids, you can know they're getting nonlinear help.
Re: The Flat Response
Of course a live event will invariably sound different from two channel stereo...if only because said event has no resemblance to the traditional stereo system model...
Indeed.....dynamic range is a significant one...
If you are going to reproduce significantly more than two discrete channels with negligible deterioration in quality, then analogue processing is sub-optimal....
'Complex' DSP systems need not compromise quality to any significant degree...the same cannot be said of several 100 op amp. stages in series....
Which is why this...:
http://www.meridian-audio.com/m_800_bro_intro.htm
http://www.meridian-audio.com/m_Instal_bro_ultimate.htm
...is the best all-round audio system in the world.
ALW said:
Of course a live event will invariably sound different from two channel stereo...if only because said event has no resemblance to the traditional stereo system model...
ALW said:
Indeed.....dynamic range is a significant one...
ALW said:
If you are going to reproduce significantly more than two discrete channels with negligible deterioration in quality, then analogue processing is sub-optimal....
'Complex' DSP systems need not compromise quality to any significant degree...the same cannot be said of several 100 op amp. stages in series....
Which is why this...:
http://www.meridian-audio.com/m_800_bro_intro.htm
http://www.meridian-audio.com/m_Instal_bro_ultimate.htm
...is the best all-round audio system in the world.
Mr.Curl has gone through most of the CCT acrobatics (for over 45years), successfull with his products, have discussion with world's best audio designers, engineers, Studying TIM-PIM, Scoping audio traces to FM frequencies, Exploring "Dynamic Phase Shift" and still cannot tell why quality Phono sounds better than quality CD player.
I dont think the engineers at Sony or Denon have answer of this also, we dont have any digital product from them that has the same performance as quality phono.
If we cannot see or measure what cause the difference (dont talk about replicating) that cause could be something "extraterrestial", cannot be caught by human's measurement equipement.
Hi folks.... hello Mikek... do not let the thread die
Let's go friends...came in Mikeks and say.
"well, i am completely agreeing with the disagreements related you one and all guys.
I do not know if you understand what i mean,or if what i mean was the thing you understood or not, because the meaning maybe was that one you get, but can be the other one you did not get!...got it?
hehe
Life is Beatifull, big brains as Mikeks too.
Carlos
Let's go friends...came in Mikeks and say.
"well, i am completely agreeing with the disagreements related you one and all guys.
I do not know if you understand what i mean,or if what i mean was the thing you understood or not, because the meaning maybe was that one you get, but can be the other one you did not get!...got it?
hehe
Life is Beatifull, big brains as Mikeks too.
Carlos
Corn circles to you!
History tells us that even those who have studied a subject for the longest time don't discover everything. It takes new approaches, new knowledge and new techniques to push progress on and on. It is nonsense to assume something cannot be understood just because it isn't. It is an unfortunate element of human nature, which we all fall foul of from time to time, to invent and adhere to arbitrary, often unprovable, explanations where no proven explanation is available. Corn circles.
Resolving the NFB problem does not require aliens. Unless I'm an alien. It requires a certain cunning, a broad understanding of science, solid maths and, probably most importantly, an eagle eye for mis-assumptions.
Folks, Walt Jung is working to putting Barrie Gilbert's article: 'Are Op Amps Really Linear?' on Walt's Website. By working together, Walt and I are reconstructing the entire article for people to read for themselves.
Walt and I discussed its importance, just yesterday, and our conclusion is that negative feedback is problematic, and Barrie Gilbert has taken one of the potential problems into the public view. Matti Otala originally gave an even more complete 'qualitative' (just equations) paper years before, but Barrie has put forth a few 'quantitative' (with numerical calculations) examples into the scene.
The paper is now 8 pages of difficult reading for non-engineers. A challenging read even for engineers, as much has been glossed over for brevity.
I don't know what Mikek's problem is: Why ignore the obvious?
Walt and I discussed its importance, just yesterday, and our conclusion is that negative feedback is problematic, and Barrie Gilbert has taken one of the potential problems into the public view. Matti Otala originally gave an even more complete 'qualitative' (just equations) paper years before, but Barrie has put forth a few 'quantitative' (with numerical calculations) examples into the scene.
The paper is now 8 pages of difficult reading for non-engineers. A challenging read even for engineers, as much has been glossed over for brevity.
I don't know what Mikek's problem is: Why ignore the obvious?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.