The Path to Paradise . . . or the Road to Ruin?

[microstrip]

Genuine recorded acoustic is very rare, one would need a very simple recording ,perhaps just one pair of crossed microphones, most recordings are recorded 'dry' in a non reverberant booth and then reverb is added later to create the impression of natural acoustic.
I find that both my horns and active monitors do not produce any image behind the plane of the loudspeakers, but an image is projected into the room, and especially with the monitors indivuidual components are fixed laterally.
(...)

Although expertly enhanced by artificial means, many recordings have enough information to suggest a natural acoustic with correct layering behind the plane of the loudspeakers.

IMHO depth depends a lot on system, room and setup. I have listened to the same speakers (e.g. B&W 800 or Wilson's) imaging in front or behind the speakers in different systems.

 

[Purite Audio]

Hi Keith, I am trying to figure out how Lamm 2.2, thrax hybrid, and Ypsilon aelius will compare to each other and the spectral. Do you know on which page of Toole's I can find the answer?

ked you won't find the answer in Toole I m afraid, you will however find the answer within the specifications of the amps and of the prospective loudspeakers.
I will help guide you!
Keith.

 

[spiritofmusic]

ked you won't find the answer in Toole I m afraid, you will however find the answer within the specifications of the amps and of the prospective loudspeakers.
I will help guide you!
Keith.

In the week of the new Star wars film, this is VERY 1977.
So Keith, you are now Obi Wan to Bonzos's Luke?

 

[Purite Audio]

Although expertly enhanced by artificial means, many recordings have enough information to suggest a natural acoustic with correct layering behind the plane of the loudspeakers.

IMHO depth depends a lot on system, room and setup. I have listened to the same speakers (e.g. B&W 800 or Wilson's) imaging in front or behind the speakers in different systems.

If a recording of , for example an instrumentalist has been recorded in a booth with very little reverberation ,then there will be no natural acoustic whatsoever, reverb will be added to each musician afterwards to give the impression of a natural acoustic.
Keith.

 

[Purite Audio]

In the week of the new Star wars film, this is VERY 1977.
So Keith, you are now Obi Wan to Bonzos's Luke?

I am afraid Ked has already turned to the dark side!
Keith.

 

[bonzo75]

Amir by complex music he means acoustic music with 50 instruments rather than 4. Different voltages and frequencies applied simultaneously to different drivers through the electronics. Requires much more ability to separate, show transients, pppp to ffff stuff, dynamics, etc etc. Modern music with 4 instruments or so is not as complex

 

[amirm]

Amir by complex music he means acoustic music with 50 instruments rather than 4. Different voltages and frequencies applied simultaneously to different drivers through the electronics. Requires much more ability to separate, show transients, pppp to ffff stuff, dynamics, etc etc. Modern music with 4 instruments or so is not as complex

Yet that 50 instrument music is not revealing of loudspeaker performance differences. What a human thinks is "complex" is not what brings out the most in loudspeaker differences.

To wit, there is not one piece of music with 50 instruments or anything close to this complexity in MPEG reference clips for testing lossy codecs. Instead we find such simple pop clips as the start of Tom's Dinner by Suzanne Vega:

Hear the lack of clarity and "roughness" in her voice? Add background music to the same clip and it would not remotely be revealing as such. The damage to the transients will be covered by other instruments and hence, far less audible.

Again, read the article and references I provided to see the work that has gone into finding these revealing clips for loudspeaker testing, none of which are of the 50 instrument type. They are simple, pop music tracks. I have sat through tests with them and they are extremely revealing of the quality of loudspeakers and differences between them.

 

[Mike Lavigne]

Yet that 50 instrument music is not revealing of loudspeaker performance differences. What a human thinks is "complex" is not what brings out the most in loudspeaker differences.

To wit, there is not one piece of music with 50 instruments or anything close to this complexity in MPEG reference clips for testing lossy codecs. Instead we find such simple pop clips as the start of Tom's Dinner by Suzanne Vega:

Hear the lack of clarity and "roughness" in her voice? Add background music to the same clip and it would not remotely be revealing as such. The damage to the transients will be covered by other instruments and hence, far less audible.

Again, read the article and references I provided to see the work that has gone into finding these revealing clips for loudspeaker testing, none of which are of the 50 instrument type. They are simple, pop music tracks. I have sat through tests with them and they are extremely revealing of the quality of loudspeakers and differences between them.

so the testers got confused by more complex music. that explains a lot on any number of levels.

 

[bonzo75]

Suzanne Vega or Mahler 2? Hmm...

 

[amirm]

so the testers got confused by more complex music. that explains a lot on any number of levels.

It is not a matter of confusion. Your hearing system is not a mechanical recorder. It applies both physical and complex perceptual mapping to the waveforms that are arriving. A lot of this is driven by evolution and the need to communicate with one another. One can't apply lay logic to it and just go by the number of instruments and such. The makeup of human voice for example is far more complicated than a violin. The latter is highly harmonic whereas the former is not.

Here is another example of equal loudness sensitivity:

There is nothing remotely resembling a flat curve. Most sensitivity comes in 2 to 4 Khz as seen by the sharp drop. And then note how it is level sensitive. It is a dynamic system and doesn't lend itself to simple assumptions. Best to follow the people who have done the research into such things than to create our own assumptions of what is or is not good for testing.

 

[amirm]

Suzanne Vega or Mahler 2? Hmm...

We once hired an independent testing company to compare our 64 kbps codec to the CD. They set up double blind tests and recruited some 100 testers. For content they picked "audiophile recordings" which automatically translated into classical music. The results? Exactly what we wanted: some 90% of listeners thought 64 kbps was indistinguishable from the CD! Marketing press release went out with those test results and everyone was happy. Had they picked any of the MPEG test clips such as Suzanne Vega and the results would have been completely the opposite. We used long transform windows that are enemy of transients and non-harmonic content like voices.

It is not wise to confuse what is good music to listen to what is revealing of the system being tested.

 

[microstrip]

Yet that 50 instrument music is not revealing of loudspeaker performance differences. What a human thinks is "complex" is not what brings out the most in loudspeaker differences.

(...)

We are not just debating loudspeaker preferences for a significant percentage of the world population.

We are debating the preferences of a small group of audiophiles that know each other for some time, and have often expressed their views on the type of sound reproduction they prefer and the references they use for evaluation.

IMHO PeterA question was just in the sense that stereo music around vocals and a few instruments was less complex because it has less players and the number and exigency of the relationships between players, space and dynamics is not so critical as in classical music.

And yes, we will be disagreeing most of the time because some of us want to use acoustic instruments and live performances as a reference ...

 

[amirm]

IMHO PeterA question was just in the sense that stereo music around vocals and a few instruments was less complex because it has less players and the number and exigency of the relationships between players, space and dynamics is not so critical as in classical music.

That's an assertion, not data. You need to show just like the research that I reference that this is true. That actual listeners are able to better characterize loudspeakers this way as opposed to what was found to be true. Here is the list of tracks in one study:

· Tracy Chapman, "Fast Car", Tracy Chapman
· Jennifer Warnes, "Bird on a Wire", Famous Blue Rain Coat
· James Taylor "That's Why I'm Here", “That’s Why I’m Here”
· Steely Dan “Cousin Dupree”, “ Two Against Nature”
· Paula Cole, “Tiger”,” This Fire”
· “Toy Soldier March”, Reference Recording
· Pink Noise (uncorrelated)

Do you or Peter have any study that shows listeners are better able to evaluate loudspeakers using 50 instrument classical music?

Can you show that the above tracks are less complex than your 50 instrument orchestra as far as your system is concerned?

 

[DaveC]

Amir is right on imo.

Vocals are best for digging out the subtle differences, the difference between the two (large scale music vs voices) is voices are harmonically complex in and of themselves, it's not the same type of complexity as a symphony. So the term complex can mean different things and I think part of the issue here is "complex" is used interchangeably and means different things to different people.

Symphonies and more complex music with lots of instruments are in fact very valuable for testing some aspects of a system's performance. One I like to use is "A Moment So Close" on Bela Fleck's album "Live at the Quick". I've only heard a few systems that can do this piece justice. Most systems make for a muddy mess.

What I do is start out with material like simple vocals, instruments like piano, acoustic guitar, violin, cello, etc and evaluate for tone, resolution, the presence of grain, glare or harshness. Bigger music like symphonies or the Bela Fleck track tell you different things about the system, like it's ability to separate instruments in space, the ability to portray all the different parts with accuracy at the same time, soundstaging/imaging, etc...

 

[bonzo75]

That is exactly what I do. Start with vocals, arias, then piano and violin, and with pictures at an exhibition, scherazade, and Mahler 2. Also choir

 

[PeterA]

Why do you say modern music is less complex? It is actually the other way around...

Human voice for example is very complex. You want to tell an MP3 from its original? Listen to the vocals. Likewise guitar strings by itself are highly revealing. Why? Because it gives you the opportunity to hear what is before and after it. Those are not masked by other instruments.

I'll try to answer your question. Incidentally, I referred to contemporary, not modern music. Schoenberg, an early modernist, composed very complex music. Ron's list of music seems more contemporary, with few musicians performing at the same time. In that sense, it seems to me that there is less going on and fewer instruments to keep track of. I wrote that I consider it less complex, not that it is necessarily less complex. That is just my impression. Here is why: I attended four days of closed rehearsals at the Vienna State Opera. I listened from the edge of the orchestra pit. Yes, a solo soprano produces a complex sound, varying in volume and scale, and very nuanced. But when the conductor gradually brought in the full orchestra and other singers on the stage, the sound became more complex. There were many different instruments and voices to separate and try to understand. The scale, volume, numbers of performers and spacial cues increased dramatically the amount of information coming from the performance.

The other example I can use and which I experience almost daily, is listening to a single singer playing a guitar on my stereo system, be it Johnny Hartman or Shirley Horn My simple, two-way monitor based system does a pretty convincing job (read believable and natural) of reproducing a single voice or solo instrument. Even a small chamber ensemble or jazz group. However, this same system has a much more difficult time trying to reproduce in a convincing manner orchestral music.

The challenging areas are specifically spacial information like separation and position of the instruments on stage, the scale of instruments and their sounds relative to the fifty or other instruments on stage. The articulation of different wind instruments on stage and their individual timbers. Twenty violins, five violas and four cellos all playing at the same time different lines.

There is no doubt that a solo singer or guitar produces fundamentals and complex harmonics or overtones which are exceedingly difficult to reproduce in a convincing manner, but if one adds a second singer and a piano to the mix, is that not even more complex and difficult to reproduce? I have heard many instances of systems reproducing the girl with guitar. Many sound excellent and quite believable. I have heard far fewer systems that can reproduce Beethoven's Ninth Symphony or Mahler's Second as convincingly.

 

[microstrip]

Amir is right on imo.

Vocals are best for digging out the subtle differences, the difference between the two (large scale music vs voices) is voices are harmonically complex in and of themselves, it's not the same type of complexity as a symphony. So the term complex can mean different things and I think part of the issue here is "complex" is used interchangeably and means different things to different people.

Symphonies and more complex music with lots of instruments are in fact very valuable for testing some aspects of a system's performance. One I like to use is "A Moment So Close" on Bela Fleck's album "Live at the Quick". I've only heard a few systems that can do this piece justice. Most systems make for a muddy mess.

What I do is start out with material like simple vocals, instruments like piano, acoustic guitar, violin, cello, etc and evaluate for tone, resolution, the presence of grain, glare or harshness. Bigger music like symphonies or the Bela Fleck track tell you different things about the system, like it's ability to separate instruments in space, the ability to portray all the different parts with accuracy at the same time, soundstaging/imaging, etc...

Thanks! Fortunately aspects such tone, resolution, the presence of grain, glare or harshness are not an issue with most of the quality high-end speakers we are debating, we focus our debates on the second level.

 

[bonzo75]

Thanks! Fortunately aspects such tone, resolution, the presence of grain, glare or harshness are not an issue with the quality high-end speakers we are debating, we focus our debates on the second level.

Exactly, I take those as a given. If something has glare or brightness it is a rejection criteria, absence of it is not an acceptance criteria. Good equipment sounds good on a lot of music, but 95% of those breakdown when 50 acoustic instruments are out through them

 

[Purite Audio]

The very finest loudspeakers can sound harsh /strident in a poor sounding room.
Peter think about the differences between microphone placement for one singer or a small ensemble and for a full orchestra.
Keith.

 

[bonzo75]

The very finest loudspeakers can sound harsh /strident in a poor sounding room.
Peter think about the differences between microphone placement for one singer or a small ensemble and for a full orchestra.
Keith.

Why would someone have very large loudspeakers in a small room. Oh wait, what's the room size the liszt are in?

Again, we assume basic common sense. Not planning to her apogee full range, analysis amphi, or trios with bass horns in my current room