Why evaluating the sound of a single speaker is essential

[Sal1950]

I could go on.

Please don't, time to let it drop.
TIA Sal1950

 

[sceptical1]

To the door.

"There's no sense arguing with a college boy"
John Wayne-McLintock!

 

[mglobe]

View attachment 478221

LOL, I was going to ask why it’s not possible to like a post multiple times.

 

[mglobe]

To the door.

 

[Kal Rubinson]

Your comments were either strictly about aesthetics or you currently rent and modifications to the property are frowned upon.

I own the apartment but "the remaining options are not aesthetically acceptable."

 

[Sernyl]

I read a perfectly logical explanation that the stereo vinyl needles of those days had such crosstalk between the channels that the stereo panorama was quite realistic.

IMHO, this applies also to the amplifiers (where crosstalk is often badly documented)

 

[Jack B]

Funny...
I recently played around two very similar speakers (variations of the same model, comparable FR curve, etc., different cabinet material...devellopement) in mono L vs R, with the same physical posotion exchange etc...
one of them could subjectively be preferred...:
"more lack back", a little euphonious...
but in stereo, it's ultimately the slightly more rigorous in stereo that turns out to be the most interesting... and yet I imagine that many had to choose, would have valued the other pair with just this test/ comparaison...
;-)
(This is somewhat related to the topic of demo discs that work with just about everything... and test discs that "put things to shame"... The test with a speaker risks highlighting a subjective aspect (hard to describe)... but which won't be as relevant in stereo ;-) )

?Whaat are you saying (in straightforward English)??

 

[Jack B]

I also have a related question, which this thread has brought up.
In the minds of many and based on some replies, it appears that some ambiguity still exists about the 'psycho-' variables of psycho-acoustics.

The ASR test/evaluation of electronics are achieved through the use of a "golden" (reference) unit, namely the AudioPrecision system.
Similarly, for the ASR speaker test/measurements, the Klippel becomes that "golden" standard.
Wouldn't it, then, make sense to use a "golden" mono speaker for all A/B testing.
I think @amirm already has that third "golden" unit. (his Revels).
Such a 'mono-y-mono' (A/B) comparison and his conclusions may bring about a balanced soundstage for the those who fancy his subjective and/or objective views.

I could be wrong (I often am), but to me, your mentation is strange.

 

[xanalog]

[IMO] Windows "Insider Preview" should not be recommended (and/or relied upon) for use by even an advanced user.
The regular channels to update/upgrade is a lot easier/cleaner on the system.

Windows' "Insider Preview" was initially released circa 2014 [?].
After about 9months of deploying updates/upgrades as an Insider; I had opted out.
It has left quite the mess behind in my Registry, even to this day!

The best reason for my [IMO]: I am no beta guinea pig or an alpha lab rat!

 

[Hayabusa]

I read a perfectly logical explanation that the stereo vinyl needles of those days had such crosstalk between the channels that the stereo panorama was quite realistic.

Yea, like its out of phase and you create a kind of stereo expander ( which works by mixing R - =L*c and L -= R*c )

 

[RigorDude]

As explained in AES papers and my books, the most revealing music in multiple-loudspeaker comparison tests tends to have complex instrumentation - a dense spectrum - wide bandwidth - especially bass extension - and reverberation. The musical genre is unimportant.

For listening pleasure or demonstration the choice is yours.

What examples would you offer of high-fidelity recordings of "complex instrumentation - a dense spectrum - wide bandwidth..." that were engineered for playback in one channel? Robert Fine's early (pre-stereo) Mercury Living Presence orchestral recordings perhaps? Frank Sinatra's early Capitol sides? The Cole Porter songbooks with Ella Fitzgerald?

While those recordings are highly regarded, they are 70 years old. To my knowledge, since the mid 60s, no one has been making high-fidelity recordings in mono. To be sure, separate miking of the various musicians / sections of an ensemble is common in studios because it is a much easier way to get a decent result. But that isn't "complex instrumentation," it's essentially one mic feed at a time, followed by mixing these multiple feeds together. That is the mixing engineer's work, whether done well (e.g. the Sóley recording that @amirm used recently) or badly. But the mixes are always to two or multichannel.

If we take the left or right channel only of a stereo mix of mutli-miked instruments for single-speaker listening, we will be hearing somewhat of a mess: a bunch of stitched-together sounds, arbitrarily sliced in two down the left-right axis, with random microphone bleed for good measure. Try playing a stereo recording on your hi-fi with one speaker disconnected from the amps. The sound isn't just different because it's coming from one point; it is "off" -- badly balanced and boxy. This isn't the case if you play a good mono recording, such as one of those mid-50s Mercuries, through one speaker.

Perhaps there is a setting in Muse that somehow takes two- or multichannel-recordings and, presto, turns them into proper hi-fi mono. Or I suppose one could take a recording of complex instrumentation that was engineered and mixed for stereo or multichannel, and just stuff it all into one channel -- the mono button! Would we expect the result to be any good?

 

[NTK]

What examples would you offer of high-fidelity recordings of "complex instrumentation - a dense spectrum - wide bandwidth..." that were engineered for playback in one channel? Robert Fine's early (pre-stereo) Mercury Living Presence orchestral recordings perhaps? Frank Sinatra's early Capitol sides? The Cole Porter songbooks with Ella Fitzgerald?

While those recordings are highly regarded, they are 70 years old. To my knowledge, since the mid 60s, no one has been making high-fidelity recordings in mono. To be sure, separate miking of the various musicians / sections of an ensemble is common in studios because it is a much easier way to get a decent result. But that isn't "complex instrumentation," it's essentially one mic feed at a time, followed by mixing these multiple feeds together. That is the mixing engineer's work, whether done well (e.g. the Sóley recording that @amirm used recently) or badly. But the mixes are always to two or multichannel.

If we take the left or right channel only of a stereo mix of mutli-miked instruments for single-speaker listening, we will be hearing somewhat of a mess: a bunch of stitched-together sounds, arbitrarily sliced in two down the left-right axis, with random microphone bleed for good measure. Try playing a stereo recording on your hi-fi with one speaker disconnected from the amps. The sound isn't just different because it's coming from one point; it is "off" -- badly balanced and boxy. This isn't the case if you play a good mono recording, such as one of those mid-50s Mercuries, through one speaker.

Perhaps there is a setting in Muse that somehow takes two- or multichannel-recordings and, presto, turns them into proper hi-fi mono. Or I suppose one could take a recording of complex instrumentation that was engineered and mixed for stereo or multichannel, and just stuff it all into one channel -- the mono button! Would we expect the result to be any good?

Those "mono" mixes are test signals, and their artistic qualities are of little importance. See an earlier post by Dr Toole.

...
This is an obvious concern, and has not been ignored. We select stereo-summed recordings based on listening for obvious downmix artifacts. They exist, but are easily found. However, it turns out that in the multiple-loudspeaker mono comparison tests the program is really a test signal, not an artistic product, and even a purely technical signal like pink noise is an excellent at revealing resonances. The "best" loudspeaker is the one with the fewest audible resonances, the most neutral. See Post #6 in this thread. Whatever the soundstage and imaging, it is good to begin with loudspeakers capable of reproducing accurate timbre. Now they can be recognized in measurements.
...

 

[RigorDude]

Those "mono" mixes are test signals, and their artistic qualities are of little importance. See an earlier post by Dr Toole.

This is a head-scratcher. Previously I posted this quote from @Floyd Toole:

"the most revealing music in multiple-loudspeaker comparison tests tends to have complex instrumentation - a dense spectrum - wide bandwidth - especially bass extension - and reverberation."

When I asked for examples of such music, I was advised that it was really "test music" and that actually it would be perfectly fine to listen to pink noise. Why not cite a few examples of well-recorded music that still sounds decent in mono? Is that task so strenuous that we are compelled to seek refuge in pink noise?

Elsewhere, Toole says "We select stereo-summed recordings based on listening for obvious downmix artifacts. They exist, but are easily found."

"They" apparently refers to "obvious downmix artifacts." But such "downmix artifacts" -- i.e. crappy sound -- are not only "easily found" but ubiquitous with summed mono. It would be far more efficient to list the summed-mono recordings that don't suck -- don't have "obvious downmix artifacts" -- than to list those that do suck, because the vast majority of contemporary recordings do, in fact, suck when summed to mono.

@NTK maintains that the "artistic qualities are of little importance" in single-speaker listening. This isn't about the "artistic qualities" of the music -- Beethoven vs. Mantovani, Horowitz vs. Liberace -- but rather the quality of the recorded sound. On that score, does NTK think it would it be A-Okay (sorry, can't help myself) to use an iPhone recording of tunes from my car radio to evaluate a hi-fi speaker?

My guess is that the tracks used by pros such as Toole and @amirm for single-speaker evaluation consist mostly of club mixes, whether of pop, techno, dub, shoe-gaze or what have you. That's because club mixes are carefully balanced to never sound crappy, even when all the speakers put out essentially the same signal. So that when you go from multi- or two-track to mono with a club mix, only the spatial qualities are lost; the sound is otherwise consistent.

But all types of music are not available in club mixes. String quartet recordings aren't offered in club mixes. Nor are jazz piano trios, renaissance masses, or a myriad other genres of music. Which brings me back to my earlier question, but with an adjustment: What are some examples of recordings of "complex instrumentation" etc. that don't sound crappy when summed to mono -- and aren't club mixes?

 

[Newman]

This is a head-scratcher. Previously I posted this quote from @Floyd Toole:

"the most revealing music in multiple-loudspeaker comparison tests tends to have complex instrumentation - a dense spectrum - wide bandwidth - especially bass extension - and reverberation."

When I asked for examples of such music, I was advised that it was really "test music" and that actually it would be perfectly fine to listen to pink noise. Why not cite a few examples of well-recorded music that still sounds decent in mono? Is that task so strenuous that we are compelled to seek refuge in pink noise?

Just use any of the higher-scoring pieces listed in post #6 of this thread, played in summed stereo.

Elsewhere, Toole says "We select stereo-summed recordings based on listening for obvious downmix artifacts. They exist, but are easily found."

"They" apparently refers to "obvious downmix artifacts." But such "downmix artifacts" -- i.e. crappy sound -- are not only "easily found" but ubiquitous with summed mono. It would be far more efficient to list the summed-mono recordings that don't suck -- don't have "obvious downmix artifacts" -- than to list those that do suck, because the vast majority of contemporary recordings do, in fact, suck when summed to mono.

See post #6, linked above. They would have been played in summed-stereo mode and their efficacy in revealing resonant flaws is documented, in fact given a score.

@NTK maintains that the "artistic qualities are of little importance" in single-speaker listening. This isn't about the "artistic qualities" of the music -- Beethoven vs. Mantovani, Horowitz vs. Liberace -- but rather the quality of the recorded sound. On that score, does NTK think it would it be A-Okay (sorry, can't help myself) to use an iPhone recording of tunes from my car radio to evaluate a hi-fi speaker?

My guess is that the tracks used by pros such as Toole and @amirm for single-speaker evaluation consist mostly of club mixes, whether of pop, techno, dub, shoe-gaze or what have you. That's because club mixes are carefully balanced to never sound crappy, even when all the speakers put out essentially the same signal. So that when you go from multi- or two-track to mono with a club mix, only the spatial qualities are lost; the sound is otherwise consistent.

But all types of music are not available in club mixes. String quartet recordings aren't offered in club mixes. Nor are jazz piano trios, renaissance masses, or a myriad other genres of music. Which brings me back to my earlier question, but with an adjustment: What are some examples of recordings of "complex instrumentation" etc. that don't sound crappy when summed to mono -- and aren't club mixes?

You are overthinking this.

If you are going to actually compare speakers this way, remember that they need to be controlled listening tests, ie

• level matched
• blind
• short A-B selections rotated in fairly short intervals of 10-15 seconds
• speakers being swapped into the same spot
• preferably 3 or more speakers not just 2

Are you starting to see the issue? This is not a simple exercise. And that is the whole point: the work has been done for you (us) and the findings are in, specified as things to be spotted in spinorama measurements.

OTOH if you are just going to fool around, then it hardly matters, right? You haven't been terribly clear what your objective is, once you have chosen the music.

cheers

 

[janbth]

This is a head-scratcher. Previously I posted this quote from @Floyd Toole:

"the most revealing music in multiple-loudspeaker comparison tests tends to have complex instrumentation - a dense spectrum - wide bandwidth - especially bass extension - and reverberation."

When I asked for examples of such music, I was advised that it was really "test music" and that actually it would be perfectly fine to listen to pink noise. Why not cite a few examples of well-recorded music that still sounds decent in mono? Is that task so strenuous that we are compelled to seek refuge in pink noise?

Elsewhere, Toole says "We select stereo-summed recordings based on listening for obvious downmix artifacts. They exist, but are easily found."

"They" apparently refers to "obvious downmix artifacts." But such "downmix artifacts" -- i.e. crappy sound -- are not only "easily found" but ubiquitous with summed mono. It would be far more efficient to list the summed-mono recordings that don't suck -- don't have "obvious downmix artifacts" -- than to list those that do suck, because the vast majority of contemporary recordings do, in fact, suck when summed to mono.

@NTK maintains that the "artistic qualities are of little importance" in single-speaker listening. This isn't about the "artistic qualities" of the music -- Beethoven vs. Mantovani, Horowitz vs. Liberace -- but rather the quality of the recorded sound. On that score, does NTK think it would it be A-Okay (sorry, can't help myself) to use an iPhone recording of tunes from my car radio to evaluate a hi-fi speaker?

My guess is that the tracks used by pros such as Toole and @amirm for single-speaker evaluation consist mostly of club mixes, whether of pop, techno, dub, shoe-gaze or what have you. That's because club mixes are carefully balanced to never sound crappy, even when all the speakers put out essentially the same signal. So that when you go from multi- or two-track to mono with a club mix, only the spatial qualities are lost; the sound is otherwise consistent.

But all types of music are not available in club mixes. String quartet recordings aren't offered in club mixes. Nor are jazz piano trios, renaissance masses, or a myriad other genres of music. Which brings me back to my earlier question, but with an adjustment: What are some examples of recordings of "complex instrumentation" etc. that don't sound crappy when summed to mono -- and aren't club mixes?

Some of the music used by Harman to evaluate speakers are listed on their Art of Listening webpage:

Exclusive Selection of HARMAN’s Acoustical Experts Reference Songs for Sound Quality Testing

Free Selection of the perfect Tracks to discriminate and rate Sound Quality Attributes of Audio Systems and overall Sound Quality – at Home, in the Car, or on the Go

artoflistening.harman.com

 

[Floyd Toole]

Some of the music used by Harman to evaluate speakers are listed on their Art of Listening webpage:

Exclusive Selection of HARMAN’s Acoustical Experts Reference Songs for Sound Quality Testing

Free Selection of the perfect Tracks to discriminate and rate Sound Quality Attributes of Audio Systems and overall Sound Quality – at Home, in the Car, or on the Go

artoflistening.harman.com

I don't know who prepared that, but it seems to be aimed at demos not the very demanding double-blind, multiple-loudspeaker, monophonic, position substitution evaluations of loudspeaker sound quality that are very revealing of resonances and other detailed flaws. These are special exercises in which the music is really a "test signal" used to reveal resonances, not for pleasure. It is very reliable and very useful as a guide to understanding the meaning of anechoic measurements.
Listening in stereo or multichannel for pleasure is a totally different thing.

 

[janbth]

I don't know who prepared that, but it seems to be aimed at demos not the very demanding double-blind, multiple-loudspeaker, monophonic, position substitution evaluations of loudspeaker sound quality that are very revealing of resonances and other detailed flaws. These are special exercises in which the music is really a "test signal" used to reveal resonances, not for pleasure. It is very reliable and very useful as a guide to understanding the meaning of anechoic measurements.
Listening in stereo or multichannel for pleasure is a totally different thing.

OK, thanks, that’s interesting. I was of the impression that these were the tracks that were used in the double-blind evaluations, or at least some of them. It would be interesting to get hold of the actual lists of music for that, even if they were used as test signals!

 

[Newman]

OK, thanks, that’s interesting. I was of the impression that these were the tracks that were used in the double-blind evaluations, or at least some of them. It would be interesting to get hold of the actual lists of music for that, even if they were used as test signals!

See my post above your two.

 

[janbth]

See my post above your two.

I understand that a rigorous testing procedure is different from listening for pleasure, and that the music involved is "prepared" to provide test signals, but I'm still curious as to what music is used.

 

[Newman]

I understand that a rigorous testing procedure is different from listening for pleasure, and that the music involved is "prepared" to provide test signals, but I'm still curious as to what music is used.

And my post links to it.