• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

NORMS AND STANDARDS FOR DISCOURSE ON ASR

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,321
Location
Albany Western Australia
You ask if there is anything of practical value in this thread. I have gone through themes of general, theoretical interest. But I will try and illustrate by a practical example too. Case studies are important!

What have we learned about vox populi in audio?

In Toole (2016), we have seen that by arranging a “market” consisting of 28 consumers who are to reveal their preferences for speakers, the designer of the “market” thinks he can infer a general theory of what constitutes the superior speaker, and what characteristics that make a good or bad speaker.

A closer look at the key word “preference” reveals that the author has a less strict understanding of this important term than experts on rationality and choice. It’s evident that “preferences” in audio change from one group to the next, which is problematic for many reasons, for example meaningful measurements.

So we have a situation where we measure something (“preference”), which is not fully defined and changes from one group to the next, possibly over time too, and we compare these unstable “preferences” to certain speaker characteristics which we cannot fully control for in a statistical sense. We have no possibility of doing ceteris paribus analyses. Do both definitions and measurement methods (i.e. a so called theorization and commensurability problem) deserve to be questioned? If so, it’s like opening a can of worms.

The author expects order, i.e. insight into relevant and irrelevant speaker characteristics to emerge out of what started as chaos based on his single interview with Mr. Market. That is optimistic, but obviously an enticing market story for some.

For comparison: The real masters of factor analyses, academics in financial economics, have access to vast databases of millions of datapoints over decades, sometimes over a century. Yet they’re criticized for creating a factor zoo, seeing patterns that are not there despite fancy p-values. In audio, it seems like one concludes strongly on a much thinner data material than in other sciences.

To me, some of the research in audio science seems to have a commercial end in sight. What is called “science” here is what one often associates with market research and surveys in other fields.

My point is, it doesn’t hurt to read audio science with a dose of skepticism. Some of our beliefs may rest on a foundation that is less robust than many thought.

I started replying to this and then realised you have been banned. However I'm sure you will still read this as a non contributing viewer however.

We have learned precisely nothing from your contribution in this thread. Apart from the fact that you have a spectacular lack of understanding and knowledge of the subjects involved. You clearly haven't done anything but skimmed the Toole et al research and allowed your own biased opinions to cloud pretty much everything you have written. It's so obviously agenda driven it beggars belief.

Couple this to your clear lack of self awareness and arrogance, your contribution, well more like diatribe, has been the antithesis of scientific. In fact its just been trolling in a pseudo intellectual form. Your repeated assertion that members don't intelligently consider the basis of information and data presented to them in research is insulting to everyone here.

I sincerely hope @Floyd Toole hasnt been put off from contributing to this forum due to this nonsense as he is someone we can genuinely learn from.
 
Last edited:

Cosmik

Major Contributor
Joined
Apr 24, 2016
Messages
3,075
Likes
2,180
Location
UK
All true, but since when "preference" is (or should) 100% overlap "quality"?
Because we're being told that preference in scientific listening tests *is* the same thing as quality - speakers with smooth dispersion are preferred by trained listeners... etc. Technically, maybe no one actually states the words that preference is 100% synonymous with quality, but the implication is clear. The fact that someone is doing the tests and that people are interested in the results is a strong clue.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,747
Likes
37,572
Because we're being told that preference in scientific listening tests *is* the same thing as quality - speakers with smooth dispersion are preferred by trained listeners... etc. Technically, maybe no one actually states the words that preference is 100% synonymous with quality, but the implication is clear. The fact that someone is doing the tests and that people are interested in the results is a strong clue.
Okay, so if you were starting from scratch how would you proceed to determine the correct accurate design for loudspeakers?
 

Cosmik

Major Contributor
Joined
Apr 24, 2016
Messages
3,075
Likes
2,180
Location
UK
Okay, so if you were starting from scratch how would you proceed to determine the correct accurate design for loudspeakers?
Now we're talking :). Just as with a microphone, ADC, DAC, amplifier. Decide what do we want the thing to do and then design and build it with as few remaining unknowns as possible.

In the case of speakers, we can't achieve the theoretically perfect point source that we might want, so we know we need to compromise, but we understand it and narrow it down to one or two variables.

We have no reason whatsoever to think that anti-phase from the back is a good thing unless it can be cancelled fully. We have no reason to build in resonant energy storage devices. (But if using a listening test-based system based on existing commercial speakers we just accept these because someone built them in for some unknown reason, and this adds variables to the mix).

We don't want the speaker to signal its presence with an unusual, arbitrary response, so 'quirky' dispersion is not desirable.

In the quest to take our living room to the concert and open its end up there is another variable that needs to be adjusted 'to taste', and that's the amount of side/rearwards radiation from the speaker we want / can tolerate and which may, also, be dependent on the room.

It leaves us a few variables to establish by preference - but we have narrowed them down and understand them. And we don't pretend it's science i.e. that there is "artistic preference" and that this is somehow different to "scientific preference" - they're the same thing.
 
Last edited:

Floyd Toole

Senior Member
Audio Luminary
Technical Expert
Industry Insider
Forum Donor
Joined
Mar 12, 2018
Messages
367
Likes
3,906
Location
Ottawa,Canada
Now we're talking :). Just as with a microphone, ADC, DAC, amplifier. Decide what do we want the thing to do and then design and build it with as few remaining unknowns as possible.

In the case of speakers, we can't achieve the theoretically perfect point source that we might want, so we know we need to compromise, but we understand it and narrow it down to one or two variables.

We have no reason whatsoever to think that anti-phase from the back is a good thing unless it can be cancelled fully. We have no reason to build in resonant energy storage devices. (But if using a listening test-based system based on existing commercial speakers we just accept these because someone built them in for some unknown reason, and this adds variables to the mix).

We don't want the speaker to signal its presence with an unusual, arbitrary response, so 'quirky' dispersion is not desirable.

In the quest to take our living room to the concert and open its end up there is another variable that needs to be adjusted 'to taste', and that's the amount of side/rearwards radiation from the speaker we want / can tolerate and which may, also, be dependent on the room.

It leaves us a few variables to establish by preference - but we have narrowed them down and understand them. And we don't pretend it's science i.e. that there is "artistic preference" and that this is somehow different to "scientific preference" - they're the same thing.

You said: "we can't achieve the theoretically perfect point source that we might want"
I'm not persuaded that the academic "point source" is a desirable goal. Relatively constant directivity above the transition frequency is a better objective, and it is practical. One can discuss the effects of different directivity indices, but if the "circle of confusion" is acknowledged as having merit then the forward-firing configuration would seem to be the default - as that is the norm for mixing and mastering.

What is needed to deliver a more credible sound field to listeners is a multichannel system. All else is compromise, especially two-channel stereo - so we play around attempting to extract from a directionally and spatially deprived system some sense of realism. Fortunately the adaptive human brain is a willing participant and we manage to find pleasure in the simplistic sounds that are delivered. So, listeners at best are able to offer opinions of "preference" of one compromise over others.

If we consider multichannel originals or upmixed stereo as a desirable basis, then the spatial cues exist in the (recorded or synthesized) direct sounds and one could argue that reduced indirect sound is desirable. Discussions would then revolve around the relative merits of upmixers, which, frankly, I think is a rewarding activity if we wish, as you say, "to take our living room to the concert".

Nice to be back to a rational discussion :)
 

watchnerd

Grand Contributor
Joined
Dec 8, 2016
Messages
12,449
Likes
10,414
Location
Seattle Area, USA
You said: "we can't achieve the theoretically perfect point source that we might want"
I'm not persuaded that the academic "point source" is a desirable goal. Relatively constant directivity above the transition frequency is a better objective, and it is practical. One can discuss the effects of different directivity indices, but if the "circle of confusion" is acknowledged as having merit then the forward-firing configuration would seem to be the default - as that is the norm for mixing and mastering.

What is needed to deliver a more credible sound field to listeners is a multichannel system. All else is compromise, especially two-channel stereo - so we play around attempting to extract from a directionally and spatially deprived system some sense of realism. Fortunately the adaptive human brain is a willing participant and we manage to find pleasure in the simplistic sounds that are delivered. So, listeners at best are able to offer opinions of "preference" of one compromise over others.

If we consider multichannel originals or upmixed stereo as a desirable basis, then the spatial cues exist in the (recorded or synthesized) direct sounds and one could argue that reduced indirect sound is desirable. Discussions would then revolve around the relative merits of upmixers, which, frankly, I think is a rewarding activity if we wish, as you say, "to take our living room to the concert".

Nice to be back to a rational discussion :)

Do you have any thoughts on the kind of "spray it around the room" directivity (or lack thereof?) being used by things like the Apple Homepod with its "beam forming tweeter array"?

I finally heard one the other day, and while it wasn't what I would call high end, it was better than I expected for $349, once my brain adapted.
 

Floyd Toole

Senior Member
Audio Luminary
Technical Expert
Industry Insider
Forum Donor
Joined
Mar 12, 2018
Messages
367
Likes
3,906
Location
Ottawa,Canada
Do you have any thoughts on the kind of "spray it around the room" directivity (or lack thereof?) being used by things like the Apple Homepod with its "beam forming tweeter array"?

I finally heard one the other day, and while it wasn't what I would call high end, it was better than I expected for $349, once my brain adapted.

It is a horizontally omnidirectional speaker for monophonic sound. Being active it should sound good - there are substantial advantages. The "beam" that is formed by the tweeter array is horizontally omnidirectional - call it a beam, but it is loose marketing jargon.

There are now and will be more surprisingly good sounding small speakers in the future. The science is mature.
 

Soniclife

Major Contributor
Forum Donor
Joined
Apr 13, 2017
Messages
4,510
Likes
5,437
Location
UK

Xulonn

Major Contributor
Forum Donor
Joined
Jun 27, 2018
Messages
1,828
Likes
6,312
Location
Boquete, Chiriqui, Panama
Physically, there can never be a perfect "audio system + listening room" for all types of music - the subject of searches for an audio Holy Grail by some audiophiles. From a scientific (logical?) perspective, the best that psychoacoustic researchers can do is to look for either "most preferred" for particular types of music, or try to identify compromised systems that might "preferred" by statistically significant percentages of listeners over a wider range of music genres.

My instincts and experiences- based on 65 years of observing and pondering human activities - lead me to believe that there are sets of ranges of preferences, and trying to go beyond that is like chasing rainbows.

Therefore, I think we need to be careful to debate "preferences" as regions in a matrix, e.g., those previously mentioned "sets" of conclusions. There are HUGE variances in real, live music across dozens of genres and types and configurations of venues - and there is a complete lack of a "live" reference for many studio recordings, where musicians can separated, wear headphones, and use direct feeds from amplified instruments to the mixing console.

Although we can strive for more "goodness" and "rightness" in reproducing recordings, and we can approach "almost like live" for acoustic solo and ensemble recordings, I view most recordings as a set of compromises plus a veneer of the sonic "preferences" of producers and engineers.

My first audio system and world-class live classical music experiences began within a year of each other in the late 1950s. When my father helped me to assemble my first monophonic HiFi system, I was a high school student working as an usher for Chicago's Orchestra Hall, and able to spend two years listening to live performances of Chicago Symphony Orchestra during the Fritz Reiner era. During that time, his collaboration with RCA switched from mono to stereo, and I was there for many of performances that are in the incredible 63 album Reiner/CSO "Living Stereo" series.

A Reiner strategy was to record pieces after they were thoroughly rehearsed for subscription concerts. This permitted long takes that would give the feel of a live performance, and notably, the finale of Rimsky-Korsakov’s Sheherazade was recorded in a single take without edits).

That entire CSO/RCA "Living Stereo" collection is still available from Sony for a list price of $1,000. I owned some of vinyl LPs and CDs years ago, and now have many of them as as FLAC files.

At the other end of the spectrum, I have experienced Santana in a Palo Alto fish and chips and ice cream parlor, the Rolling stones at Altamont, Pat Metheny at the U.C. Berkeley Greek Theater, Cat Stevens at the Berkeley Community Theater with a live string section, folk and acoustic guitar plus many jazz combos and bands in small clubs and halls, etc., etc., etc...

Because I am an expatriate retiree living in the mountains of Western Panama, I base my buying decisions on many years of experience, reading "between the lines" of reviews, the evolving findings (not "conclusions") of science and psychoacoustics, and conversations on internet audio forums. I am pleased with the results and find my current system to be very satisfying for casual (background) listening, dedicated listening sessions, and movies. I do not expect - and would not expect even if I had unlimited financial resources for audio - to be able to "accurately" reproduce all or even most of those sonic experiences. But I do accept the finding that the science of psychoacoustics can determine what "might be" the best compromise for my listening preferences. Indeed, I have been able to assemble a system that to me sounds quite realistic, and it only cost me about $1,700 including shipping to Panama. I probably could have gotten the same level of sonic performance for under $1,000 using the same loudspeakers, but chose to get some features, looks and include some "preferences" that are not related to sound quality.

I will be converting my second bedroom, where my computer desk and an unused bed are located, into a dedicated office/listening/TV room, but I will stay with my excellent 2-channel audio rather then a compromised multi-channel surround system. If I can afford it in a few months (not a sure thing), I would like to replace my Paradigm Atom v6 monitors with a pair of the gorgeous - and highly rated and recommended - Revel F206 "small" towers in walnut. Based on the combination of Harman research (including "preferences") and uniformly excellent reviews, I have absolutely no doubt that I will enjoy them immensely. I haven't owned a pair of floorstanders since my Klipsch Forte II's which I sold in 2005 - 14 years ago.
 

watchnerd

Grand Contributor
Joined
Dec 8, 2016
Messages
12,449
Likes
10,414
Location
Seattle Area, USA
There are now and will be more surprisingly good sounding small speakers in the future. The science is mature.

Which makes me think the way forward for multi-channel is via multiples of speakers like this (plus maybe some subs), arranged around the room wherever is good for decor, linked together via wifi, and let them do self-room EQ (they have microphones, after all), and let them via software turn mono, 2-channel stereo, n-channel multi channel content into software-defined multi-channel on the fly.

I'd much rather invest in a set of software-defined/grid-networked multi-channel via software speakers, than moribund formats like multi-channel SACD/DSD (which I'd have to buy new content all over again...), plus special purpose electronics/AVRs, passive speakers, etc, etc.
 

watchnerd

Grand Contributor
Joined
Dec 8, 2016
Messages
12,449
Likes
10,414
Location
Seattle Area, USA
Therefore, I think we need to be careful to debate "preferences" as regions in a matrix, e.g., those previously mentioned "sets" of conclusions. There are HUGE variances in real, live music across dozens of genres and types and configurations of venues - and there is a complete lack of a "live" reference for many studio recordings, where musicians can separated, wear headphones, and use direct feeds from amplified instruments to the mixing console.

Daft Punk's Random Access Memories, won a Grammy (in addition to Album of the Year, and many others) for Best-Engineered Album (Non-Classical).....yet it has no live reference at all.

Which makes one wonder how the judges know it's best-engineered?
 

digicidal

Major Contributor
Joined
Jul 6, 2019
Messages
1,984
Likes
4,844
Location
Sin City, NV
Daft Punk's Random Access Memories, won a Grammy (in addition to Album of the Year, and many others) for Best-Engineered Album (Non-Classical).....yet it has no live reference at all.

Which makes one wonder how the judges know it's best-engineered?

Maybe because it appears, in the information age at least, that most=best (in entertainment at least). Seems to be the case in movies as well. Always find it funny how many CG-fests win technical awards but rarely ones for acting... which I think it should be the other way around. If you can seem even somewhat "in character" while talking to tennis balls on sticks in front of a green-screen... then you're one hell of an actor. On the other hand if you just render impressive scenery and explosions - then you're not necessarily any more talented than an ordinary video game company.

Although, to be fair to Daft Punk... though definitely not my genre of choice - Random Access Memories is a very good album IMO.
 
Top Bottom