Toward a Theory To Increase Mutual Understanding and Predictability

[amirm]

I'd also say I completely disagree with Harmans findings and call into question the testing and the quality of their test systems. Their findings that reflections with shorter delay times, such as 1st reflections of sidewalls and floor are actually beneficial is simply a reflection (oh no...) of their predisposition and acclimation to that kind of sound.

That research is confirmed by many, many other researchers. Indeed Harman's research only confirms that position. See my article on perceptual effects of room reflections to see how/why this is true both objectively in the way our hearing system works and research by others. I can fill this thread with more references than you can count on value of first side reflections.

Also this research goes back to NRC, way before Harman and studies were done using ordinary listeners so there is no bias for anyone at Harman there.

Personally I don't like rooms that do as you say as the sound then comes out of the point source of the speaker, focusing attention on them rather than a nice soundstage. When was the last time you heard a live band have its sound come out of two points left and right? Side reflections stretch those sources which sharply lessens this effect.

That said, there is a small percentage of population that can hear these reflections (usually recording engineers) by training and for *work* they like to have them masked. Ironically when going home to listen to music, many like the effect. You may be falling in this population but you are definitely not the majority.

 

[Folsom]

Your stereo is broken if it sounds like things are coming only out of the right and left. It should leave wondering if the speakers are "on".

 

[RogerD]

Your stereo is broken if it sounds like things are coming only out of the right and left. It should leave wondering if the speakers are "on".

Lol....since I was a kid and went to the movies, I never heard left and right.

 

[opus112]

http://www.claritycap.co.uk/products/tc.php

They're saying -

The TC capacitor’s equivalent series resistance (ESR) and inductance are a fraction of that of an electrolytic, and therefore instantly deliver all their power precisely when it is needed.

They provide data on ESR and certainly with ESRs in the single digit milliohms they're an order of magnitude or more lower than electrolytics.

However - they're also claiming inductance is a fraction of that of an electrolytic but its odd that they provide zero data on the ESL. Could it be that their claim about ESL is pure BS? This inquiring mind wishes to know - anyone got any pointers on ESL for these caps?

<later>

Having done my own research I'll let Morgan Jones do the talking - this table is from his book on Valve Amplifiers, Clarity caps include the adjacent graph on their site.

 

[Ron Resnick]

So Ron, how do you think this thread is going?

I like the discussion! While I had hoped the replies would track more closely my original thesis rather than debate my premises I am grateful to respondents who ultimately did state the objective to which they subscribe.

When we start new topics on new threads it will be interesting to see if anyone takes up the challenge by stating his objective and his musical preferences in the opening post.

 

[Ron Resnick]

I think original event should be interpreted as "realistic concert like sound". It might not sound exactly like the violin that was played at the concert on that day, but should not sound like a violin that was never ever played anywhere because it is too plastic or too rolled off.

This seems like a helpful amendment to improve the usefulness of Objective 1 and to make it less objectionably theoretical.

 

[Ron Resnick]

When listening to "regular rock and pop", how is it possible to know what the original musical event was?


There is so much studio manipulation that goes into "regular rock and pop" recordings, how can one have any idea what the original musical event sounded like?

I agree! For regular multi-track rock and pop there is no original musical event at all.

 

[Ron Resnick]

Despite our worthy and noble obsession with the "absolute sound" my Objective 1 and Objective 2 are theoretical and aspirational.

The discussion has proven that the only way to achieve Objective 1 is to experience an original musical event in your listening room and to record it on reel-to-reel tape. Then you can play back the recording on your high-end audio system in the same room and evaluate how closely the playback matches what you heard during the original musical event (Objective 1). You can assemble components and employ room treatment and other techniques to make that comparison match as closely as possible.

If you cut a lacquer and produce a vinyl record you can play the record on your system in the same room and compare it to the reel-to-reel tape (Objective 2).

I think I agree with Diapason and Peter A. and others who have commented that despite our noble efforts ostensibly and rhetorically in support of Objective 1 and Objective 2 what we really are doing is achieving Objective 3 by making educated guesses about what real voices and real instruments sound like from our live experiences with them, and then assembling systems which comport with our subjective, composite impressions of what we think recorded voices and recorded instruments should sound like on our systems.

 

[RogerD]

Speaking of movies,and immersion sound, back in the days when Altec,WE,Ampex and others equipment were used in every town and city cinema theater. I remember opening the door to walk down the aisle and was hit by the glorious sound that was totally transparent. Now the speakers were large horns and the electronics were all tubes,but the source was 35mm film...the king of analog. The systems were the opposite of today's high end,very simple. The electronics had no low esr capacitors or special resistors,the transformers were nothing special except now considered vintage. So how was that dynamic,transparent,clear sound accomplished. Btw that level of reproduction has always been in my memory. I'm not saying that nothing has changed,nor today's systems can't be better. I just think for what they had back then as far as sound quality was somewhat remarkable.

Also I've always aspired to a similar sound. I do think I have moved past that,but it's hard to judge in a 300 sq ft room.

 

[Ron Resnick]

Having retreated to the view that Objective 1 and Objective 2 are largely aspirational I continue to believe, nonetheless, in the thesis of the opening post -- that declaring our aspirational objective and our musical preference at the commencement of a discussion will assist considerably in allowing us to understand each other.

 

[RBFC]

Ron, I feel your frustration. A while back, I attempted to gather opinions on which recordings may form a "foundational list" of audiophile-quality sound. These recordings could provide common ground in discussions about many aspects of sound quality, since one could refer to exact time stamps, etc for others to compare. The project was met with rather "interesting" resistance.

Lee

 

[RogerD]

Ron, I feel your frustration. A while back, I attempted to gather opinions on which recordings may form a "foundational list" of audiophile-quality sound. These recordings could provide common ground in discussions about many aspects of sound quality, since one could refer to exact time stamps, etc for others to compare. The project was met with rather "interesting" resistance.

Lee

I listened to this tonight....pretty incredible. Listen to the baritone sax,literally fills the room and the growl is really impressive,Harry Carney was a great one. I wish my room was bigger yet as the power of Ellington's band on some cuts is jaw dropping. The brass just soars.

 

[morricab]

Spinning wheels here. I still haven't seen all these horrible harmonic amps.

And if you don't understand what 3rd harmonic or lower designing means, I don't think you really get the subject matter well. It means people purposely design for dominate harmonics to be 3rd or below. Not sure how else to phrase that. And do you distortion past 3rd, higher, sure, but at what dB? Voltage only divides so there's always limitations.

Why don't manufacturers advertise their harmonic distortion choices? Because it's about as interesting as a plumber designing a house's pipes to not leak.

I eagerly await your LTspice models along with topology that has perfect distortion patterns.

Look, I won't "go fer" the measurements for you...suffice to say they can be found easily in Sterephile and Soundstage...

I have never heard 3rd harmonic or lower designing...the only thing lower than 3rd is 2nd so that is an absurd way to phrase it. There are precious few amps that even come close to the design philosophy of having only 2nd and 3rd harmonic distortion products. It is not really possible with a multi-stage amp. In fact, based on the measurements I have seen out there, I am struggling to think of even one that matches this criteria.

"Why don't manufacturers advertise their harmonic distortion choices? Because it's about as interesting as a plumber designing a house's pipes to not leak. "

This is also an absurd statement. They do publish THD...but it is meaningless. Do they have the FFT data? Most likely , but as a few magazines that measure show, it is often not so pretty and has an impact on the SQ. They are just lucky that no one in the reviewing world is running their distortion data through a hearing prediction model or they might get some unfavorable press. If you would bother to go look you will see that every pattern under the sun is out there, so a lot of leaky plumbing can be found...even in 2017. Go look at the D'agostino Momentum monos distortion measurements in Stereophile...there is even evidence of zero crossing distortion!! So much for that being a thing of the past... It is always there at some level in all Class AB amps...it might be below obvious detection but it will manifest in the harmonic pattern.

I eagerly await your model that shows an amp with only 2nd and 3rd harmonic distortion generated. Boyk and Sussmann did computer modeling of devices and found that only a MOSFET, which has a quadratric transfer function, singled ended and running in Class A or push/pull running in Class A...with no feedback and no other circuitry (so a single stage) could Theoretically produce only low, even order harmonics (or for PP none!). However, that perfect theoretical device does not exist. Adding feedback or running the push/pull in other than Class A gave a myriad of harmonics. A bipolar transistor will make a myriad of harmonics no matter what because of its sharp non-linearity. Tubes will has make harmonics but are far closer to linear and therefore for the same drive level they make far lower levels.

Once you start adding stages like a differential input and a driver, all bets are off. Only 2nd and 3rd order?? I think not. Crowhurst demonstrated that adding negative feedback multiplied the number of high order components while reducing the low order ones in level. This continued indefinitely until you get what is indistinguishable from a noise floor but with the exception that it is not really noise because it is signal correlated and true noise is not signal correlated. An amp without negative feedback will not have this issue and will have less intermodulation possibilties with noise from the power supply becauase the signal only goes through the amp once.

 

[morricab]

Spinning wheels here. I still haven't seen all these horrible harmonic amps.

And if you don't understand what 3rd harmonic or lower designing means, I don't think you really get the subject matter well. It means people purposely design for dominate harmonics to be 3rd or below. Not sure how else to phrase that. And do you distortion past 3rd, higher, sure, but at what dB? Voltage only divides so there's always limitations.

Why don't manufacturers advertise their harmonic distortion choices? Because it's about as interesting as a plumber designing a house's pipes to not leak.

I eagerly await your LTspice models along with topology that has perfect distortion patterns.

https://www.its.caltech.edu/~musiclab/feedback-paper-acrobat.pdf

"The spurious signals generated by feedback in the single-ended FET amplier, and enhanced by feedback in the class-B FET pair, are correlated with the program material, since they are constructed from sums and differences of integer multiples of the program-material frequencies. We do not know whether or not this program-correlated noise is psychoacoustically signicant, but its presence is certainly suggestive."


"So high-order products can be enhanced by feedback in some simple amplier stages. Although we have not investigated whether or not this happens in more complex ampliers, it is tempting to relate this finding to the perception of feedback introducing what can sound like a badly-integrated \super-tweeter". Similarly, the fact that feedback can sometimes increase the relative distortion of very low-level signals makes it tempting to relate this finding to the loss of fine detail and room sounds. Whether or not these are appropriate attributions can be determined only by psychoacoustic experiments."

This paper is from 2003...hardly ancient history.

 

[microstrip]

(...) The discussion has proven that the only way to achieve Objective 1 is to experience an original musical event in your listening room and to record it on reel-to-reel tape. Then you can play back the recording on your high-end audio system in the same room and evaluate how closely the playback matches what you heard during the original musical event (Objective 1). You can assemble components and employ room treatment and other techniques to make that comparison match as closely as possible.
(...)

IMHO this test is a poor assessment of a system to meet objective 1 - the objective of sound reproduction is not just reproducing musicians in the acoustics of your listening room, but also replicating the acoustics of another very different listening space your room.

 

[morricab]

Despite our worthy and noble obsession with the "absolute sound" my Objective 1 and Objective 2 are theoretical and aspirational.

The discussion has proven that the only way to achieve Objective 1 is to experience an original musical event in your listening room and to record it on reel-to-reel tape. Then you can play back the recording on your high-end audio system in the same room and evaluate how closely the playback matches what you heard during the original musical event (Objective 1). You can assemble components and employ room treatment and other techniques to make that comparison match as closely as possible.

If you cut a lacquer and produce a vinyl record you can play the record on your system in the same room and compare it to the reel-to-reel tape (Objective 2).

I think I agree with Diapason and Peter A. and others who have commented that despite our noble efforts ostensibly and rhetorically in support of Objective 1 and Objective 2 what we really are doing is achieving Objective 3 by making educated guesses about what real voices and real instruments sound like from our live experiences with them, and then assembling systems which comport with our subjective, composite impressions of what we think recorded voices and recorded instruments should sound like on our systems.

I have actually done what you are suggesting here Ron and I will share my thoughts on the matter. I recorded my ex-playing 24 Paganini Caprices on a Stradivarius in my listening room and she was standing centered directly between my two speakers (Acoustat 1+1 at that time). I had a single mono condenser microphone positioned at the listening position at the height where my ears would be located. The microphone preamp was fed directly into a mid 1970s TEAC R2R analog tape recorder. The acoustic power of that single instrument was sufficient that it was difficult to capture the full dynamic range without overloading the tape. I had to deal with noise at the soft end and saturation at the high end. I managed but just.

Now to the sound. The recording is extremely present and alive, as one might expect from zero processing and using pretty good recording equipment. The acoustic is quite dry when played back though and you hear that it is a largely untreated, concrete block walled room. When played back in that same room, it is double trouble and it sounds too dry as you are getting a double helping of the same room acoustic. In other rooms, it can be amazing or too agressive. One thing is for sure though, it sounds more "live" than most other violin recordings I have.

I made a number of other recordings, some in stereo and some in mono at different venues but they were usually on DAT made with a portable Casio recorder. They were good but that big TEAC made for much better recordings overall. I still have the master tape but not the TEAC anymore...I burned a cd of the recording though for portability.

The realism of the recording is limited to some degree by the acoustic of the original event superimposed on a new acoustical space...that can matter only a little or a lot depending on the new acoustic space. My recording has a very strong acoustic space in the recording itself and it is heard no matter what room you play it back in.

 

[Folsom]

Look, I won't "go fer" the measurements for you...suffice to say they can be found easily in Sterephile and Soundstage...

I have never heard 3rd harmonic or lower designing...the only thing lower than 3rd is 2nd so that is an absurd way to phrase it. There are precious few amps that even come close to the design philosophy of having only 2nd and 3rd harmonic distortion products. It is not really possible with a multi-stage amp. In fact, based on the measurements I have seen out there, I am struggling to think of even one that matches this criteria.

"Why don't manufacturers advertise their harmonic distortion choices? Because it's about as interesting as a plumber designing a house's pipes to not leak. "

This is also an absurd statement. They do publish THD...but it is meaningless. Do they have the FFT data? Most likely , but as a few magazines that measure show, it is often not so pretty and has an impact on the SQ. They are just lucky that no one in the reviewing world is running their distortion data through a hearing prediction model or they might get some unfavorable press. If you would bother to go look you will see that every pattern under the sun is out there, so a lot of leaky plumbing can be found...even in 2017. Go look at the D'agostino Momentum monos distortion measurements in Stereophile...there is even evidence of zero crossing distortion!! So much for that being a thing of the past... It is always there at some level in all Class AB amps...it might be below obvious detection but it will manifest in the harmonic pattern.

I eagerly await your model that shows an amp with only 2nd and 3rd harmonic distortion generated. Boyk and Sussmann did computer modeling of devices and found that only a MOSFET, which has a quadratric transfer function, singled ended and running in Class A or push/pull running in Class A...with no feedback and no other circuitry (so a single stage) could Theoretically produce only low, even order harmonics (or for PP none!). However, that perfect theoretical device does not exist. Adding feedback or running the push/pull in other than Class A gave a myriad of harmonics. A bipolar transistor will make a myriad of harmonics no matter what because of its sharp non-linearity. Tubes will has make harmonics but are far closer to linear and therefore for the same drive level they make far lower levels.

Once you start adding stages like a differential input and a driver, all bets are off. Only 2nd and 3rd order?? I think not. Crowhurst demonstrated that adding negative feedback multiplied the number of high order components while reducing the low order ones in level. This continued indefinitely until you get what is indistinguishable from a noise floor but with the exception that it is not really noise because it is signal correlated and true noise is not signal correlated. An amp without negative feedback will not have this issue and will have less intermodulation possibilties with noise from the power supply becauase the signal only goes through the amp once.

People design for dominate 3rd or lower - and that mostly only means they need adequate slew rate for most designs. Should I say 2-3hd? Who cares. The basic rule of electronics is that voltage divides, so why would you expect to ONLY see 2-3hd? That's what's silly, if you ever want to see more than a few watts. And per your paper you linked, not only do the amount of devices play a role but so does the input voltage. And sometimes harmonics are redistributed for legit reasons (see below).

The D'agostino Momentum monos measurements are at 100w. At listening levels the harmonics may only be perceptible as 2nd order, including with measurement, as the power rises it trades off some harmonic order for lower distortion (divides the harmonics into higher order). So essentially during loud peaks (mostly bass?) it will change but otherwise it probably would be much closer to some Lamms. I don't really know why Stereophile measure and shows the results on such narrow information. Maybe I would agree with you more so if the information actually reflected problems in the first couple of watts, and showed different tests with a spectrum of harmonics.

Boyk and Sussmann observations are subjective with anecdotal support - and they're poking around in more territories than they seem to realize. They aren't proving anything, it's just some reasoning as to why they think their observations may be true.

Anyways I vote we don't continue this anymore, in this thread. It's not like I achieved some radical realization beyond common knowledge that the behaviour of transistors inherently are dominate in lower order harmonics...

 

[Ron Resnick]

I listened to this tonight....pretty incredible. Listen to the baritone sax,literally fills the room and the growl is really impressive,Harry Carney was a great one. I wish my room was bigger yet as the power of Ellington's band on some cuts is jaw dropping. The brass just soars.

View attachment 30668

How does this relate to the thread?

Please tell us to which audiophile objective you subscribe?

What is your musical preference?

 

[morricab]

People design for dominate 3rd or lower - and that mostly only means they need adequate slew rate for most designs. Should I say 2-3hd? Who cares. The basic rule of electronics is that voltage divides, so why would you expect to ONLY see 2-3hd? That's what's silly, if you ever want to see more than a few watts. And per your paper you linked, not only do the amount of devices play a role but so does the input voltage. And sometimes harmonics are redistributed for legit reasons (see below).

The D'agostino Momentum monos measurements are at 100w. At listening levels the harmonics may only be perceptible as 2nd order, including with measurement, as the power rises it trades off some harmonic order for lower distortion (divides the harmonics into higher order). So essentially during loud peaks (mostly bass?) it will change but otherwise it probably would be much closer to some Lamms. I don't really know why Stereophile measure and shows the results on such narrow information. Maybe I would agree with you more so if the information actually reflected problems in the first couple of watts, and showed different tests with a spectrum of harmonics.

Boyk and Sussmann observations are subjective with anecdotal support - and they're poking around in more territories than they seem to realize. They aren't proving anything, it's just some reasoning as to why they think their observations may be true.

Anyways I vote we don't continue this anymore, in this thread. It's not like I achieved some radical realization beyond common knowledge that the behaviour of transistors inherently are dominate in lower order harmonics...

One last comment and then I will stop. Zero crossing distortion is level independent and therefore a worse problem at 1 watt than at 100 watts. If you look at the THD vs power it is relatively flat...that means the relative distortion of the amplifier did not go down with power so the problems will track with power and they become more and more audible as the SPL level drops...that pesky masking again...

"Fig.9 indicates that the distortion is predominantly the subjectively innocuous second harmonic, though there is a suspicious-looking spike in the residual waveform almost coincident with each zero-crossing point. This spike is the result of a picket fence of higher-order harmonics (fig.10), though it's fair to note that these are all relatively low in level."

 

[PeterA]

IMHO this test is a poor assessment of a system to meet objective 1 - the objective of sound reproduction is not just reproducing musicians in the acoustics of your listening room, but also replicating the acoustics of another very different listening space your room.

Very true. It is an especially challenging exercise in my non dedicated, size limited living room. I am making efforts to lesson the effect of my room's acoustic on the sound I get at home, but there are limits to what can be done. This gets to the question of "is the performer in your room playing for you" or "are you transported to the original site of the performance". I prefer the latter, but it is a real challenge to achieve.