Why Do People Come To Computer Audiophile To Display Their Contempt For Audiophiles?

[esldude]

1 hour ago, jabbr said:

 

You describe an ideal amp below as "transparent", ok, let's use that goal for "fidelity". Now how do the Stereophile measurements tell you how "transparent" the amplifier sounds?

 

 

Are you suggesting the J2 isn't transparent? or the M2? Voicing? What voicing? The designs are different topologies with the goal of determining which topology works best in which circumstance. In many cases there is a new transistor technology which is being tested (e.g. Silicon Carbide, SIT, Power JFET), or perhaps positive current feedback ala F7 which is useful in certain difficult speaker loads.

 

 

Yeah ok but use the measurements to tell me how they "sound", or how "less transparent" they sound. Interpret the measurements...

 

 

Nelson Pass tells us how his M2 is not transparent in the manual:

 

Does it have any similarities to no-feedback Class A tube amplifiers

with an output transformer? Yes and no.

 

It does have some of the qualities that SET aficionados look for. The auto-former

does bring its own signature to the sound, although in smaller doses than we

usually experience. The midrange has a lush, warm character that I associate

with tubes and transformers, but subtly so. Part of this is due to the limited

bandwidth of transformers, and part of it is the lack of feedback artifacts.

On the other hand, there is considerably less harmonic distortion than with no-

feedback SETs, and if you really do like that distortion then the M2 might not be

what you want. Also, there is more power than usual and a relatively high

damping factor, so the M2 is different in that regard.

snippage.........

I think you will find the M2 a very pleasant amplifier to listen to. It is relaxed and

lively with a slight softness to both the bottom and top frequencies. I hope that

you will find as I do that it disappears and leaves the music for you to enjoy. I also

hope that it will make you go back through your entire record collection and make

you listen to it all over again.

---------------------------------------------------------------------------------------------

 

So yes, the designer of the M2 doesn't think it is transparent.  He says so.  The

usual measures under the appropriate conditions would show you this without

listening to it.  Much of it if you know the circuitry and components chosen, and

how some measures are done in the spec's even clue you into how it would measure

indicating a lack of transparency.   It is those measures omitted. 

[esldude]

6 hours ago, jabbr said:

 

 

"disappears and leaves the music for you to enjoy" sounds transparent to me

 

Every electronic component: resistors, capacitors, transformers, transistors have "electronic signatures" i.e. measureable non-linearities. Thus each component has a "sound". In his description of the M2 the "sound" is ascribed to the transformer but many amps have transformers so isn't this generally useful info? The goal of ultimate transparency involves tradeoffs. Yes, these tradeoffs can be described as "not transparent" to a degree. So which "non transparencies" are important? What are the best tradeoffs?

 

The FirstWatt line is designed to be low power (as opposed to the main Pass Labs offerings). Thats why I am comparing two of these similar designs. You have suggested that the M2 is better than the J2 because it has more power, but there is also the issue of the transformer and its characteristics, so perhaps the J2 has advantages in other regards. These are competing factors. This makes an absolute judgement of relative transparency difficult.

 

What I am suggesting is that measurements can help but not (alone) answer that question ... and yes, each amplifier in existence has its own sonic signature determined by the non-linearities in its components.

In context of everything he said, no he isn't even claiming transparency.  Especially as sins of omission can be ignored whiles sin of commission in audio gear can't.  He is simply saying despite the limits it could disappear and be musically enjoyable. 

 

While any stage may have non-linearities not all puncture transparency.  Some do, some don't.  Going from .000005% distortion to .00005% is not going to be heard.  Going from flat response to being 3 db down at 20 hz and 20 khz might not be heard in isolation and with some music though it can be heard as lacking compared to flat response. 

 

 

[esldude]

58 minutes ago, Kimo said:

Regarding the transparency of Pass designs, Nelson is also stating that the sonic signature of the M2 is more transparent, not less, at least in some ways.

 

Part of this is due to the limited bandwidth of transformers, and part of it is the lack of feedback artifacts.

 

 

I am still trying to figure out in what world limited bandwidth, described by the designer as part and parcel of a sonic character indicates transparency.  Now low gain, low bandwidth amplifiers similar to these two can have difficulties with feedback causing problems.   So he wisely avoided enough feedback or any feedback (one uses some and one uses none) in a range that causes audible problems that would not sound okay.  The part you excerpted was where he was saying the sound shares characteristics associated with transformers and tubes.  Just after saying it has a lush warm character.  Again, Nelson Pass isn't claiming transparency for these designs.  He is claiming a highly musical enjoyable character.  Given the design and the specs that is exactly what I would expect.   A warm enjoyable sound as he has carefully removed any warts that would disrupt that character.

[esldude]

33 minutes ago, Kimo said:

Since we are focusing on measurements and amps, I am currently listening to a push/pull integrated tube amp that claims a s/n ratio of >95.  Is that even possible for tube amp?

It could be.  Which amp is it?

 

You can make op-amp like circuits with tubes, and I seem to recall one design which basically did that, and combined with a good output stage achieved something near what you are listing here.

[esldude]

8 minutes ago, Kimo said:

Qualiton 50i.  New to the USA.  Doesn't look like a simple design.  They wind their own transformers and utilize a more complex circuit design.  You could understand it all, but it is beyond my humble comprehension.

Finally remembered the one I had in mind.  The Wolcott tube amps.  Instrumentation amp design.  I think it has SNR of 100 db, and distortion like .0001% or some such.  Was pretty powerful too at about 220 wpc.  Used feed-forward as well as feedback. 

 

Reading translated article about the Qualiton 50i it looks to be fairly conventional push-pull tube amp design done to a high quality level.  I've had VTLs which were in the high 70 db range for 1 watt which put them near 90 db SNR for full rated power.  So sure its possible. 

[esldude]

 

Doesn't change what you are saying.  However, the M2 uses an input autoformer type transformer for the 14 db voltage gain in that amp.  It feeds MOSFET followers.  Simple design.  Per the designer not transparent.  Not the designer's aim.

 

 

 

 

 

[esldude]

3 hours ago, jabbr said:

Why am I going on and on about this?

 

Precisely for the topic as stated. I give an example of FirstWatt as exemplary of a designer that "knows his stuff", is unquestionably an expert in his field and is respectful of the goals of his customers i.e. audiophiles

 

I present this to highlight the dichotomy between "blind" measurement and listening. I am hardly anti-measurement, but I think at the end of the day knowledge of the schematic is critical to understanding the measurements.

 

For "subjectivists" here are a series of different designs that can be listened to. For "objectivists", here are different schematics that can be analyzed. I have suggested that it is easier to just listen  That's what I do when going over schematics  

 

We should try to look at things from the "other guy's" perspective (I didn't bring up the schematics initially because that's not where we started but am happy to discuss...). Rather than directing what people should like, we can observe what people do like and try to explain that. There doesn't need to be value judgment.

I don't think anyone will dispute 'just listen' is easier.  It also is the least reliable, least repeatable and least accurate method.  In fact if listening were all that it would be better than knowing the schematics, and measurements.  Yet it is not. 

[esldude]

3 hours ago, christopher3393 said:

 

Don't you think that statement is a sweeping generalization? Are you saying "in all cases" or would you qualify that further?

It's too sweeping.  Sometimes the measurement is easier.  Take an audible minor FR issue. In the time spent listening to one song you could do a very accurate measure finding out more easily than listening how important the FR deficiencies are.

[esldude]

12 minutes ago, mmerrill99 said:

That's fine if you can adjust the frequency issues but what if that adjustment is not in your purview? Is it not enough to know that it doesn't sound 'right'?

If you know it doesn't sound right, and that is all, then where do you go from there?

 

If you measure and also know it doesn't sound right you know where you need to go.  You know this even if you don't have the ability to make that adjustment for what is making the sound different.  You may measure being unsure of the sound and would know where you need to go next.

[esldude]

9 minutes ago, mmerrill99 said:

You find something that sounds 'right'

 

 You make my point so well.  Your solution following your method limits the possibilities to any other piece of gear that might be available.  Not exactly a  big help. My solution will help you zero in on what gear will fill your needs, and allow you verify it as so.

Quote

I'm sure that audio show-rooms will allow you to bring in measuring equipment - I regularly see AP scopes being brought in by potential customers.

Well, I don't know if show rooms at audio salons would allow that or not. You are of course changing the subject at this point.  You can do plenty of testing without an AP if need be.  I can put it all in a backpack and leave room for lunch. 

 

I do remember one audio dealer who used to host those McIntosh clinics where you bring in your gear and see if it measured as well as McIntosh gear.  They also sold Nakamachi and were very happy to show you the real response between a Nak deck and any other.  As they had the gear and people to do repairs they would tell you other info from their little testing station if you knew enough to ask. 

[esldude]

2 hours ago, mmerrill99 said:

 

I was using an AP to show how ridiculous this is. How many audio show-rooms have you unpacked your measuring gear from your backpack & proceeded with measurements (while having your lunch)? 

Yes, you are being ridiculous.

 

I have not measured potential candidates for purchase in a showroom.  I have done so when helping friends with issues or deciding on how to proceed.  Were there any audio dealers closer than 150 miles away, I could and would do such testing if it was related to potential purchases of myself or friends.  Which brings us of course to another advantage of measurements.  When done right you don't have to be present.  In my case, there are no dealers with showrooms reasonably close.  So how easy is listening you can't do?  Not very as you already mentioned. 

 

Maybe magazine reviews need to add to their review process.  They should start with a section that explains how they expect a design to sound and why before listening to it.  Then contrast that with how it sounded.  Even better if they always have two reviewers for each piece of gear in the magazine (HP at TAS did this at one time).  Let one see the measurements before listening and the other not.

[esldude]

5 minutes ago, mmerrill99 said:

I presume you've been to audio showrooms & yet not once have you done any measurements there with your backpacked equipment?

 

I see

 

QED

Well the last showroom closer than 120 miles of me was something like 10 years ago.  So not since then has one been local.  I believe I have only been in showrooms while out of town twice since then.  Neither was a planned trip, so no I didn't have the gear to measure with me.

 

 

[esldude]

1 hour ago, mmerrill99 said:

So you are relying on others measurements or you are measuring the equipment taken home on pot luck (no measurements available or done in the showroom) - pretty much the same as the subjectivist who takes note of other people's listening impressions, listen themselves in the showroom but ultimately listens to the equipment on their own system.

 

Except that after all your measurements, you also listen & then make your decision (or do you?).

 

Whereas the subjectivist makes their decision without the pre-bias of measurements.

Actually, most recent purchases were made with no auditioning of any kind.  Just the specs vs my needs. Yes, those were also measured by someone else so I wasn't just at the mercy of the vendor being honest.

 

The showroom you seem locked in on is a complete non-issue for me as there are none around.  I would venture most people buying gear are in a similar situation today. 

 

You might find understanding my approach easier if you remember, it is my opinion, other than gear designed to have a particular sound, everything between the transducers is not detectable with a slight exception for matching amps to speakers. And exceptions for low quality gear which has sub-standard measured performance.

[esldude]

@mmerrill99  Well we have like 3 pages of post since I last had a chance to post.  You seem intent on attacking the credibility of anyone choosing gear without extensive sighted listening.   You seem intent on twisting things to imply other ways are problematic.

 

I'll try and summarize and simplify for myself.  I found when testing myself unsighted, looking at results of other people unsighted and learning about how things work as much as I had the chance it was simple.  Once you match levels, and remove other bias, frequency response was the most common and largest reason for things sounding really different. Skipping over all the in between steps, if some gear has flat response, low enough noise, low enough distortion, then it is difficult to impossible to hear when you don't know which is which. 

 

So if you wish to say now, that knowing measurements before listening or having tested out the proposition prior has biases for me to believe the measurements.  Then okay, I am fine with that.  Those biases as you call them however are simply knowledge gained to me.  You asked way back if I had ever returned something I purchased after choosing via measurements.  No I haven't.  Once I was comfortable with the idea, which was tested by me, that a good set of measurements would give me a good transparent sound, I can use that and move on.  That approach has been simpler, easier and effective.

 

The other approach is messier, less efficient, and if differences are real I am left with only vague ideas about what they are and which factors are important.  I've made the decision to go with transparent gear and alter sound to my preference in other ways.  With target curves or EQ mostly.  Works very nicely.  When half of real differences were only vaguely understood and the other half imaginary it made getting really good sound a large sized hassle.  Some people turn the hassle into a ritual and make it almost spiritual in nature to their practice of it.  I was something like that at one time.  For me at least, this other way is much better.

[esldude]

7 hours ago, mfsoa said:

But PKane, what do you do if a component that sounds great doesn't measure well? I had a VAC preamp that sounded glorious. Probably a thousand times more distortion, worse freq response etc. etc. compared to a SS pre. But it sounded great. I thought so, everyone that heard it thought so. 

 

So what would you do? Would you get rid of the better sounding component for one that measures better and clearly sound inferior?

Having been down that road, I can say those distortions and other non-linearities are why it sounded that way.  It was the reason you liked it and others too.  Not a thing wrong with that.  If I were happy with that I would not at all get rid of it.

 

The problem imo, is going from there to saying the measurements are wrong, and that better measuring gear is defective in some way we can't measure.  Then proceeding as if the other gear has deficiencies that need fixing.  I am not attributing this to you as you didn't say it.  It is a common next step after such an experience.

 

What you would find is there are pre-amps or amps you can put after the VAC which sound less good than the VAC according to your tastes and preferences on their own.  Perhaps the tastes of most people.  When conveying the VAC output signal they would do so accurately and not mess up the sound of the VAC which sounds glorious to you.  That glorious sound is an additive coloration.  It is the old mistake of if it sounds better it must be because it is of higher accuracy.  Yet we know now for some years that is not always the case.  Some inaccuracies actually sound more pleasing. 

 

In fact a pre-amp is the best place imo to have such a thing in a system.  If frees up sources, and other components so you get the sound you love and all the other gear only needs to be effective and perform accurately for you to have the sound you want.   A pre-amp is the smartest place to have a designer sound.  Even better to me would be DSP to get that sound.  That isn't common among consumer audio, it is somewhat available in pro recording circles.  In principle one could have an emulation plug in for your VAC preamp. 

[esldude]

5 hours ago, mmerrill99 said:

Ah but esldude & Pkane don't use listening without first measurement so this doesn't apply to them - they know what they are going to hear & guess what? - they hear what the measurements tell them - it's nice to always be validated, isn't it?

 

Yes some people may perceive a problem & need help to identify it, agreed - I'm not disputing that & measurements will certainly help - I'm just disputing the use of measurements as the defining aspect of what is/will be heard as was the discussion with esldude & pkane.

 

I'm also sure that differences/improvements in sound have not yet been discovered in the measurements - take the Regen & iso regen for instance!

 

So if we are relying on measurements alone as the yardstick of what's worth auditioning then, as I said before, I find it a limitation 

Actually most often these days I don't do listening in the sense you mean.  The measurements aren't biasing my auditioning.  If the gear is made to do what I need I plug it in and use it.  I am listening to music then.  Not listening to possible deficiencies or glories of the gear. 

[esldude]

6 hours ago, mmerrill99 said:

A DAC may perform well in the FR area with simple test signals but not so well with more complex signals.

FR measurements don't tell all - what I'm saying is that sounding better may not be due to euphonic distortions - there may well be other reasons not measured. So in all cases you are limited by the scope & types of measurements done.

 

If we all agree that measurements don't tell us how something will sound then we can't use measurements as the yardstick by which to judge the sound of something 

While I can imagine a design that would do what you describe, I don't know of any such designs in use with DACs.  There are DACs that have okay FR, but not okay IMD or THD or noise floors.  So yes you need more than just FR.

 

While one is limited by the scope and types of measurements done, they have been chosen with good reason over the years, and the true limits to that is much less than you are implying.  You also are ignoring the limited scope of simple listening.  Much, much more circumscribed than your average basic set of measurements.

[esldude]

31 minutes ago, fas42 said:

 

So, how you determine "low enough noise, low enough distortion"?

Low enough noise you don't hear noise.  So I would place that 80 db down from normal levels.  For most things this isn't at all difficult to manage. 

 

Low enough distortion I would consider THD and IMD below .1% at max level.  As we can usually get most gear another factor of 10 lower than this or even a factor of 100 these days it isn't much of a problem for most gear.

[esldude]

4 minutes ago, fas42 said:

 

So, "simple listening" is insufficient for detecting "not okay IMD or THD or noise floors"?

Simple listening for these issues is not as simple as simple measurement.  So unless I have no access to measurements listening for them would be at a lower level of diagnostic utility.  Like when I was helping straighten out some bad sound at a strip club once.  One didn't need measuring gear to improve upon the result they were getting.  One could have done even more if you had the measuring to work with. 

 

Then there is the issue that some particular types of marginally audible noise or certain THD profiles can sound a little better to us than clean results at times.  So much simpler to have everything plenty clean and you avoid all of those pitfalls.

[esldude]

20 minutes ago, Superdad said:

 

Now those are some serious hardcore audiophile bonafides.  Dude goes to a strip club, and what doese he do?  Helps them improve their sound system!

 

I'm sure the other patrons were very appreciative Dennis, but I hope you were not disappointed when they declined to participate in a "blind" test.  Not what they were there for!  

 

Actually it was the girls that were appreciative.  The sound was so bad they were complaining of headaches.  Someone knowing me told the owner, "I bet that guy could fix it up for you".  So I did. 

 

The girls loved the new sound.  No headaches and the patrons liked it better and spent more money.

 

I had free drinks and the girls gave free dances anytime I was there.  Enough to get a guy in trouble if he isn't careful there. 

[esldude]

18 minutes ago, mmerrill99 said:

What test signal do you use in that measurement?

Signal to noise ratio would be one.

 

Dynamic range would be another. 

 

The latter is probably the better test.

 

Another variant with digital gear would be spurious free dynamic range.

[esldude]

9 hours ago, mmerrill99 said:

I should have been more accurate in my request - how would you fully characterise noise to show that it is below audibility which I believe is what Fas42 was getting at? 

80 db below your normal listening level should be satisfactory.  I said the same thing somewhere in this thread.

 

If you want more detail I would want the quiescent noise level 80 db below normal listening levels.  I would want dynamic range at those levels or less. 

 

 

[esldude]

9 hours ago, mmerrill99 said:

No, this your defensiveness kicking in - just because I point out possible flaws in your method doesn't mean I'm "intent on attacking the credibility of anyone choosing gear without extensive sighted listening." This seems a very biased view If one can't examine a methodology, objectively then it's much more of a belief system than an objective approach 

 

Don't have a problem examining methodology.  If anything you are tiresome in refusing to accept the methodology by using sophistry to act as if you have found some flaw in it. Then no examination satisfies you because you have decided you aren't going to be satisfied with it. 

 

9 hours ago, mmerrill99 said:

You have inadvertently revealed the problem - your assumption is that your measurements have revealed everything that can be heard & therefore you have never heard anything (or returned a product) once you are happy with the measurements. So two questions:

- do you believe that your measurements have revealed ALL that can be heard - please tell us what set of measurements you use to qualify for this accolade

- do you think it's possible that your measurements have so biased you that you are convinced there is noting which you will hear that has not been revealed in the measurements & therefore you don't hear anything - it's self-fulfilling as all biases are!  

Yes, it's messier, less efficient & prone to error - so what - it has the advantage that it isn't approaching the real world with a theory of operation & therefore is open to being surprised & taught by experience.

 

 

Very strange ideas you imagine.

 

The old ridiculous claim about everything that can be heard. 

 

I won't go into the particular measurements as I have done so more than once.  Whether the measurements reveal everything that can be heard I don't know.  They reveal much of what can be heard.  Each measurement properly applied narrows down the area where remaining problems could be.  Our basic good set of measurements has narrowed that way down.  Anything that might get through undetected will not cause major differences in sound quality. They will  necessarily be much smaller.  These measurements already well exceed what can be detected by simple listening while knowing the identity of what you are listening to.  The noise in the measurement results of the sighted listener swamps out whole areas that can be measured very well by other means.  Measurements with measuring instruments don't have to be perfect to handily beat just listening. 

 

The business about measuring biasing me to not hear differences is one that can be tested.  I can do a test of things that are really different without knowing they are and someone can see if they are detected by me.  One inadvertent test of that was last year hearing files posted from what should have been virtually identical signals.  The details certainly made me believe that.  Others insisted they sounded different.  I listened using some ABX software and did indeed detect a difference.  I then looked at the files with more scrutiny and found a level difference and an FR difference (the latter being sizable).  In the end the details omitted information.  Yet I was biased to think they would sound the same and they did not.   As usual the answer isn't to worry about biases so much as it is to test in a way they aren't as likely to effect the results. 

 

Yes, it's messier, less efficient & prone to error - so what - it has the advantage that it isn't approaching the real world with a theory of operation & therefore is open to being surprised & taught by experience.

 

The other approach can be surprised as well.  Plus it makes good use of the experience of many people honing in on the truth beyond a single lifetime.  Approaching things without a theory of operation being a superior approach sounds stupid if you ask me.  And this an approach you approve of after stating the obvious that it is less efficient and prone to error.  Inefficient and prone to error is not a negative folks!  Now that is news.  Let us embrace the inefficient and prone to error approaches in all of life. 

 

[esldude]

9 hours ago, mmerrill99 said:

Again, you reveal the shortcomings of measurements - you don't really know what is the complete set of measurements needed.

 

Example: You suggested on Audioscience review that you would like to see measurements which reveal noise modulation issues - you don't itemise this here?

 

I asked you  a while back for these noise mod measurements of the 3 or 4 DACs you own - you didn't reply so I take it you have not done these measurements.

 

Do you therefore consider that you have fully qualified the noise behaviour of these DACs?

 

So let me get this straight.  A measured approach is bad because it isn't capable of being surprised and people who use it think they know all that there is to know. 

 

Then if they instead (contrary to your mischaractarization) express the possibility of surprise or at least curiosity about whether something could stand more scrutiny it is an indictment of their approach.  You probably even think you win arguments this way don't you?

 

Whether I have fully qualified noise behaviour of DACs, I have well measured the quantity of the noise and put limits upon how much it might matter.   Perhaps noise floor modulation could matter a little or not.  It would be nice to see additional info. 

 

So far noise modulation appears to be a relatively small effect except when the signal has significant jitter (like over HDMI) or when people use leaky filters that let aliasing modulate the noise floor as the level of the 20 khz range changes.  Leaky filters are seen in gear that does so to prevent pre-ringing and in gear which is cheap and they are simply not doing good filtering. 

 

I must have missed where you asked me about the noise previously.  Yet again you seem to think I have to claim perfect knowledge in every aspect for it to have validity or if I admit less than perfect knowledge this throws into disarray the whole idea of measuring.   I neither claim perfect knowledge nor need to. 

[esldude]

1 minute ago, Ralf11 said:

esl, he is essentially saying that one might be biased to look only in certain places for something

 

thus, a bias in the design of an experiment could result in a bias that affects the outcome

 

a bit similar to the classic joke of a guy looking for his car keys under a light pole, a passerby asks about them and helps look only to find out he lost them in a dark spot several yards away ("but it's light over here")

Sounds more like the subjectivist approach to me.  It is easy to simply sit and listen without trying to really understand anything.  And it makes me feel like I can know something much easier.  So we'll just do that instead of all the bother of doing more. (or bother looking over there where it is a bit dark even though the keys are there).

 

Bias can of course influence where you look and what you look for.  Good measured approaches that have found success and testing of those results that as much as possible control for bias have a much better track record than those who simply make no effort.  So sure with no controls anything is possible, but then anything is possible.  Doesn't exactly guide anyone to what is important does it?