The Need for Objective Metrics in Audio Reproduction

[amirm]

There are two parts to this forum:

1. The home page. This is where I will individually manage and the goal is to show and explain real science of audio as best as I can determine and relay. Currently that is work in progress but if you take one look at what is there, you see that is 100% about audio science and engineering. No one reading that and entering the forum would think this is just any other audio forum.

2. A forum. The forum is for you all. I play a small role in what is said in there. My core contribution to that is to say that when there is conflicting point of view, the compass to get out of that mess points to what is accepted by Audio Science. And by that, I mean what is considered norm in AES, ASA, IEEE Spectrum, and the like.

To that end, if you want to dispute Ethan, the path is to show him such references, scientific/engineering explanation, measurements, etc. That would then make the forum look like it is driven by reasoning of scientists/engineers as you state. Mere disagreement makes us look like any other forum to see who can shout louder. You saw me provide such documentation. That is my expectation from others.

And to be clear, I am not the guard for the forum. As the forum grows it will be way outside of my abilities to read everything and try to right every wrong. You all as the members have and know the core values and are the ones that set its overall direction.

 

[h.g.]

To that end, if you want to dispute Ethan, the path is to show him such references, scientific/engineering explanation, measurements, etc.

In dispute with Ethan about what? I was using him as an example of the difference between a technician's view and an engineer's view but it would seem to have fallen on stony ground like most, if not all, of my previous examples. There is nothing wrong with a technicians view but it is less efficient at dealing with tasks that require the understanding and knowledge of an engineer.

And to be clear, I am not the guard for the forum. As the forum grows it will be way outside of my abilities to read everything and try to right every wrong. You all as the members have and know the core values and are the ones that set its overall direction.

If you are not going to moderate to get what you want why do you expect the outcome to be different to that of every other unmoderated audiophile forum?

What are the core values? Are they written them down so that forum members have some guidance about what is in and what is out? Is lots of audiophile twaddle fine? Is the current proportion of audiophile twaddle too much, about right, not enough? What should be used to control the amount of audiophile twaddle? Teasing? Deletion? Polite requests to desist? Nothing?

I am not aware of a healthy home audio forum with only modest amounts of audiophile twaddle and suspect it will be rather difficult to achieve. Hydrogen Audio has had a go but the posting rate is modest and how audiophile twaddle is handled and/or should be handled is a problem for the regular posters that are left. Healthier home audio forums tend to be dominated by audiophile twaddle with some accepting a rational view and some not. I shall watch how it is handled and evolves here with interest.

 

[amirm]

In dispute with Ethan about what? I was using him as an example of the difference between a technician's view and an engineer's view but it would seem to have fallen on stony ground like most, if not all, of my previous examples. There is nothing wrong with a technicians view but it is less efficient at dealing with tasks that require the understanding and knowledge of an engineer.

You disputed his understanding acoustic science with regards to ringing. Then labeled him as a technician and left it at that. That doesn't demonstrate interest in scientific method. It just comes across as argumentative.

If you are not going to moderate to get what you want why do you expect the outcome to be different to that of every other unmoderated audiophile forum?

I am 100% moderating what happens on the home page. Those articles are then discussed with strong focus on the scientific area being addressed. That is what I want and cannot find elsewhere.

I am not aware of a healthy home audio forum with only modest amounts of audiophile twaddle and suspect it will be rather difficult to achieve. Hydrogen Audio has had a go but the posting rate is modest and how audiophile twaddle is handled and/or should be handled is a problem for the regular posters that are left. Healthier home audio forums tend to be dominated by audiophile twaddle with some accepting a rational view and some not. I shall watch how it is handled and evolves here with interest.

HA forum brings dictatorial point of view in audio. For example they do not tolerate posting of measurements to validate a point of view. To that end, what they are doing is not at all scientific or proper. And in that manner, not a competition to this forum.

But yes, let's see what unfolds here. For now if you are interested in more discussion of scientific method, I hope you show an example of it in your posts.

 

[h.g.]

You disputed his understanding acoustic science with regards to ringing.

I asked him twice I think what he meant by the term "modal ringing". He declined to answer. Without knowing what he means by it how can I or anybody dispute his understanding of it? I speculated he may be combining two separate things but he claimed confidence in his knowledge despite the questions he was asking so fair enough.

Then labeled him as a technician and left it at that. That doesn't demonstrate interest in scientific method. It just comes across as argumentative.

I used him as an example of thinking like a technician by looking for recipes rather than understanding as an engineer/scientists would in an attempt to illustrate similarities to your way of thinking. Didn't get through but then it is not always easy to get someone to recognize that others do not think in the same way as them. It had nothing to do with the scientific method. You may not like or agree with what I am saying but argumentative? Perhaps. I consider myself to be fairly idly chatting on the internet having got what I came for earlier plus an idea from chatting with Ethan to add to the to-do list.

HA forum brings dictatorial point of view in audio. For example they do not tolerate posting of measurements to validate a point of view. To that end, what they are doing is not at all scientific or proper. And in that manner, not a competition to this forum.

That seems an odd thing to say. Do you have a link? In my experience they over emphasize measurements demanding the results from blind tests in situations where predictions from established scientific knowledge would be accepted by those with the relevant background. But they do insist people follow their TOS when disagreements arise which can be a problem for a fair few.

But yes, let's see what unfolds here. For now if you are interested in more discussion of scientific method, I hope you show an example of it in your posts.

I am casually chatting on the internet about people, how they think, what words mean, forums and a few other things none of which seems to involve the scientific method although I haven't gone back and checked. To see an example you would probably have to look at my work rather than my posts. Off and on, but mainly off, I am working on a piece about cabinet radiation among a few other home audio topics. That might do but it is a few months off at the earliest and so you will need some patience assuming I ever finish it and write it up for a home audio audience.

 

[amirm]

That seems an odd thing to say. Do you have a link?

Sure. Here is the infraction I received with my post moved to their "Recycle Bin"

Above I was explaining how PLLs filter out jitter and how in the process there can be peaking at the corner frequency. It was an educational post met with a sanction for violating TOS #8:

"8. All members that put forth a statement concerning subjective sound quality, must -- to the best of their ability -- provide objective support for their claims. Acceptable means of support are double blind listening tests (ABX or ABC/HR) demonstrating that the member can discern a difference perceptually, together with a test sample to allow others to reproduce their findings. Graphs, non-blind listening tests, waveform difference comparisons, and so on, are not acceptable means of providing support."

Emphasis mine. By their notion we should all sit around looking at each other if we don't have a listening test to talk about -- not any method I have seen practiced in audio engineering/science.

 

[amirm]

Here is another violation of TOS #8 on HA forum. This time someone put forward an article by Meyer where he talks about differences in the sound of tube versus transistor amplifiers. I am quoting a part where he actually preferred the sound of the tube and that too was met with an infraction:

How could discussing an article the other side has put forward and quoting a section from it become a violation of a forum rule???

 

[Don Hills]

...
How could discussing an article the other side has put forward and quoting a section from it become a violation of a forum rule???

I believe your reputation preceded you. Summed up as "No-one likes a smartarse", which you came across as, judging by the responses to your posts.

 

[amirm]

I believe your reputation preceded you. Summed up as "No-one likes a smartarse", which you came across as, judging by the responses to your posts.

Me? Nah.

 

[DonH56]

Seems like we have traded audiophile snobs for engineer snobs, great...

 

[h.g.]

Here is the infraction I received with my post moved to their "Recycle Bin"

Above I was explaining how PLLs filter out jitter and how in the process there can be peaking at the corner frequency. It was an educational post met with a sanction for violating TOS #8:

"8. All members that put forth a statement concerning subjective sound quality, must -- to the best of their ability -- provide objective support for their claims. Acceptable means of support are double blind listening tests (ABX or ABC/HR) demonstrating that the member can discern a difference perceptually, together with a test sample to allow others to reproduce their findings. Graphs, non-blind listening tests, waveform difference comparisons, and so on, are not acceptable means of providing support."

Emphasis mine. By their notion we should all sit around looking at each other if we don't have a listening test to talk about -- not any method I have seen practiced in audio engineering/science.

Thanks for the post but without the context we perhaps need to be cautious about reading too much into the final post.

Jitter is one of the dafter audiophile beliefs in that low numbers are considered important even when orders of magnitude below audibility. It seems to be viewed very much in the same way as 0.001% versus 0.002% THD distortion into an 8 ohm load at 1 kHz used to be viewed for amplifiers in the 70s even though todays sophisticated audiophiles obviously have contempt for such an unsophisticated view!. It shifts hardware and so the manufacturers tend to provide what the punters want to buy and adapt their marketing and designs to suit. As a consequence rational people tend to have a pretty low tolerance for the subject and what is used to hold the interest of the uninformed.

Presumably the thread was about the audibility thresholds for jitter in some form? And you were bringing up some guff used to market jitter as important? Not necessarily incorrect guff but guff that was almost wholly irrelevant to the topic of the thread and irritating to those that wanted to stick to the subject of the thread (a stick to bash audiophiles?). Without the context I am guessing a bit.

If it was a thread about listening to a bunch of files then you would appear to be in violation of their rules on their forum. If you had listened to the files, heard a difference and that difference seemed to line up with what you were discussing then I would judge it unreasonable to bin your post otherwise it appears a reasonable thing to have done given their interests on their forum.

Having said that I agree with you that the rule forcing one to perform an experiment to talk about the result of an experiment goes against the whole point of science which is to predict the outcome of experiments without having to perform them. Unfortunately for this to work it requires those involved to possess the relevant scientific knowledge and know how to use it. This normally means something like a formal education in the relevant subject to gain the knowledge and a year or two using it for it to gel. Very few people on home audio forums possess this and so taking a step back and requiring the performance of an experiment which anyone can do does make some sense.

So what else might work? Well if I may use poor old Ethan again as an example then collecting the results of experiments without the predictive capabilities of scientific knowledge can work to a limited extent. It is inefficient but open to everyone with a strong enough interest. If he was more interested in acquiring the fundamental scientific knowledge of how sound is created, propagates and dissipates in a room then he would have used that to predict the answer to his questions rather than perform or seek the results of experiments. It is what science is all about and what enabled the technological explosion after people twigged how useful and efficient Newton's laws and relatives were in first sorting out stuff out on paper before rolling up the sleeves and getting on with it. Watching audiophiles claim one has to perform experiments to know anything or that experimental results are better than knowledge does rather distance them from engineers and scientists but they can still slowly get there if careful.

 

[h.g.]

How could discussing an article the other side has put forward and quoting a section from it become a violation of a forum rule???

I suspect that some of the moderators are perhaps not as even handed as others. However, to put things in perspective, a competently designed valve amplifier working within its rather limited effectively linear operating range is not going to be audibly distinguishable from a competent solid state amplifier doing likewise. For most rational people this is the end of the discussion.

Was operating a valve amplifier so that it audibly distorts and judging the quality of the distortion the point of the thread or was perhaps the audible threshold of the distortion the point of the thread? If the latter then binning your post would seem a reasonable thing to do although perhaps with a clearer reason.

 

[NorthSky]

Hi

This is something that is very dear/important to me. I came to stereo ast a very young age. 8~9 years old I was allowed/encouraged to us emy father's stereo system. I was hooked to music and its reproduction. In those times there was the notion of High Fidelity Hi-Fi... then came the High End Audio Revolution in the late 70's early 80's with the search for an Absolute Sound then the current it is all subjective mantra where whatever you like is good as long as it is expensive and "audiophile-approved" .. Bose doesn't qualify even if you were to like it, not expensive enough .
On the Other Hand there are the works of the AES and those of companies such as Harman and Bang & Olufsen with results to prove that there is an objective reality: That gears who measures well in some specific metrics are usually well received by a majority of listeners. Yet the High End Audio Industry and many (most?) audiophiles are not at ease with either the methodologies nor the results.
I believe that we need to encourage more in this direction . More works about what do we require from our systems and why do we like what we like. . What metrics should these be? What should an audiophile look in those measurements? What type of measurements would help? Who performs those?
Comments , discussions, etc are encouraged

Hi Frantz, Happy Easter Holidays.

Awesome post. I'm sure there are professional analyzers, measured people who open audio components and look @ all the parts and implementation of them all with the utmost scrutiny. Those people are the ones you're looking for.
The good stuff with less distortion and general matching with the most other audio components, duh, cost more money...commercially/generally.

So, where are those people we can rely on and trust for our audio transmission control? Everywhere, right here we have some, over there too, and elsewhere as well.
Our kids want good music, good quality audio, great sound recordings. So we start @ the source...the recording engineers; they are a good bet as to what audio products can reproduce best the quality music recordings (96/24).

High Fidelity is inside all the small parts of the audio electronics, so that there is the less contamination possible, zero distortion @ all volume levels, from the softest to the loudest. But we all knew that already.

Yeah, music recording engineers from the best world recording studios and record labels. ...The musicians too, they know best how their music sound when acoustic and when amplified @ various volume levels...different live space venues.

Pro audio gear, people who build it...they are the ones we're after.

1. High level of audio excellence; for no distorted sound.
2. No distortion @ all volume levels. ...Preamp, Amp, DAC, Speakers.
3. Inter-modulation and higher harmonic distortions...we look for those.
4. The machines. ...Electronic tools are the ones who perform the tests.

Sound is distorted air vibrations; the better control and less distortion the transmission waves to our ears...we want it. ...Like in real nature...there are no distortions between the bird's flapping wings of a hummingbird and our ear's flaps and vents. And where we are in time and in space with all the other stuff surrounding that natural space...trees, sky, wind, sunglasses, umbrella, leaves, flowers, spring smell, bees, other birds, distant natural sounds and not so natural...trains, planes, cars.

In the concrete basements of our homes, with thick walls that no outside noise can penetrate, no windows, bunch of pillows covering the walls and sleeping foams rolled down in the corners, disco ball light, we live in a totally unnatural audiophile world, trying to replicate the mid seventies. ...I think.

* Hydrogen Audio, Audio Science, diyAudio, Gearslutz, What's Best, http://benchmarkmedia.com/blogs/app...-dsd-provides-a-direct-stream-from-a-d-to-d-a ... etc., they all have some pros.

 

[fas42]

I believe that we need to encourage more in this direction . More works about what do we require from our systems and why do we like what we like. . What metrics should these be? What should an audiophile look in those measurements? What type of measurements would help? Who performs those?
Comments , discussions, etc are encouraged

The biggest problem as I see it is the lack of measuring methodologies that give numbers about what a complete system is doing, in terms of accuracy. Something like the software DiffMaker is aiming in the right direction, but it is too flawed, and limited in its current form.

 

[AJ Soundfield]

Here is another violation of TOS #8 on HA forum.

Let it go man. You've got your own forum here now to discuss audio science. Strange that it attracts fervent believer science denialists, but then again maybe not, as WBF proved.

 

[amirm]

Let it go man. You've got your own forum here now to discuss audio science. Strange that it attracts fervent believer science denialists, but then again maybe not, as WBF proved.

I am discussing it as to make clear what our mission is here in the formative stages of this forum. I left HA on my own and so "letting go" is in my DNA.

Our charter and mission here is clear: first to have fun and second to have a guiding light sourced by audio research/science. Don't want to see battles to death, personal accusations and information free posts.

 

[NorthSky]

AJ, I'm 100% with Amir here.

* I'll invite Arny ( https://hydrogenaud.io/index.php/topic,108987.0.html ) and Frank (F. Cook - electronic engineer from Toronto → https://ca.linkedin.com/in/frank-cook-56734325?trk=prof-samename-name ), Sean Olive & Floyd E. Toole here. ...Gene too (Audioholics).

Mike (Blizzard, Mivera Audio → http://www.miveraaudio.com/#!mivera-audio-forum/ycs0u/music ), Richard (Opus), ...are calibrated in the electronics application and implementation...solid audio science.
Amir too. ...He's the guy who started this ASR web.

♫ Test: doIEwzc6k_k

 

[Blumlein 88]

Hi Frantz, Happy Easter Holidays.

Awesome post. I'm sure there are professional analyzers, measured people who open audio components and look @ all the parts and implementation of them all with the utmost scrutiny. Those people are the ones you're looking for.
The good stuff with less distortion and general matching with the most other audio components, duh, cost more money...commercially/generally.

So, where are those people we can rely on and trust for our audio transmission control? Everywhere, right here we have some, over there too, and elsewhere as well.
Our kids want good music, good quality audio, great sound recordings. So we start @ the source...the recording engineers; they are a good bet as to what audio products can reproduce best the quality music recordings (96/24).

High Fidelity is inside all the small parts of the audio electronics, so that there is the less contamination possible, zero distortion @ all volume levels, from the softest to the loudest. But we all knew that already.

Yeah, music recording engineers from the best world recording studios and record labels. ...The musicians too, they know best how their music sound when acoustic and when amplified @ various volume levels...different live space venues.

Pro audio gear, people who build it...they are the ones we're after.

1. High level of audio excellence; for no distorted sound.
2. No distortion @ all volume levels. ...Preamp, Amp, DAC, Speakers.
3. Inter-modulation and higher harmonic distortions...we look for those.
4. The machines. ...Electronic tools are the ones who perform the tests.

Sound is distorted air vibrations; the better control and less distortion the transmission waves to our ears...we want it. ...Like in real nature...there are no distortions between the bird's flapping wings of a hummingbird and our ear's flaps and vents. And where we are in time and in space with all the other stuff surrounding that natural space...trees, sky, wind, sunglasses, umbrella, leaves, flowers, spring smell, bees, other birds, distant natural sounds and not so natural...trains, planes, cars.

In the concrete basements of our homes, with thick walls that no outside noise can penetrate, no windows, bunch of pillows covering the walls and sleeping foams rolled down in the corners, disco ball light, we live in a totally unnatural audiophile world, trying to replicate the mid seventies. ...I think.

It is my opinion, perhaps not widely shared though not unique, that everything between the microphone and amplifier input can be accurate enough to be fully transparent or audibly perfect. That isn't to say in between components don't sound different, but often those are preferences instead of accuracy. Amps need matching to speakers and speakers are of course where differences are at present impossible to overcome along with room interaction. Microphones differ as well of course. In between I think we can have complete accuracy needed for our ears.

The place where most damage is done is usually mixing/mastering and other processing. It need not be this way, but is. The next killer DAC or super-duper-certified-hi-rez digital format is not going to fix remaining issues. Truth is when done right we have what is needed except at both ends. I also think the mic end does far, far less damage than the speaker end. Of course such an opinion doesn't market worth a damn. No money to be made like that.

 

[AJ Soundfield]

I am discussing it as to make clear what our mission is here in the formative stages of this forum. I left HA on my own and so "letting go" is in my DNA.

Our charter and mission here is clear: first to have fun and second to have a guiding light sourced by audio research/science. Don't want to see battles to death, personal accusations and information free posts.

The quote in the HA thread have a link?
It appears to reference a blind listening test, which several members here completely reject. The very foundation of audio science. How are "discussions" in threads going to work there?

 

[NorthSky]

I agree; good synergy between electronics is imperative.
And looking @ the schematics is a start in the right direction.

And yes, the music recordings are the primary sources on what the entire audio reproduction chain is built around.
Good mics, good positioning mic techniques, good recording engineer's set of ears with great taste on "sounds" and imaginative sense of experimentation,
several listening testings, good recording machines, good hall's acoustics, good recording studios, good preamplifiers, good trim controls (volume pots, analog and digital...32-bit, 64-bit, 128-bit), good well designed amplifiers, good sources, faithful music replicators (audio files, mediums, bits, grooves, tape composition, disc composition, LP composition), solid and accurate transfers.

Longevity and reliability? Best with an upgrade path. ...Even with an easy way to replace specific audio cards.
...So; update from an Ethernet port or a USB port, plus easy access to remove and replace audio cards from the rear of the electronic source component.
...Like from the DAC section for example, the analog stage, the transport, the DSP circuit board, etc.
Because audio quality improvements are made every 3 to 6 months if not less.
And, in ultra high end audio, quality sound is related to state-of-the-art application and implementation of all audio parts.
Money is indeed a relevant indicator in the equation, a factor of consideration; and of course with a judicious balance between solid audio science design/engineering and audio product manufacturer's margin ($).

♥ I'll take a quality music recording from a top record music label over a DSD upsampling into the stratosphere from a six digits music playback source player.
_____________

♦ http://benchmarkmedia.com/blogs/application_notes

♫ _____________.___-------______ ♫---------------------------------
♫

♫

• http://benchmarkmedia.com/blogs/app...-dsd-provides-a-direct-stream-from-a-d-to-d-a

 

[fas42]

The place where most damage is done is usually mixing/mastering and other processing. It need not be this way, but is. The next killer DAC or super-duper-certified-hi-rez digital format is not going to fix remaining issues. Truth is when done right we have what is needed except at both ends. I also think the mic end does far, far less damage than the speaker end. Of course such an opinion doesn't market worth a damn. No money to be made like that.

Of course, one answer is not to listen to modern recordings ... all of the classical efforts leave me pretty cold: pretty terrible, sloppy engineering, with a fetish about the performer(s) rather than the music - and in all the rest of the conventional stuff most of the best works were done years ago. The interesting, more experimental current recordings are fine to go with, because they revolve around modern mastering methods.