[analogsurviver]

It depends on the repertoire of classical music.

A Mozart symphony and Mahler symphony are two very different beasts - in many ways, loudness included.

There are louder classical pieces than Mahler's ... there is a piece for similar, but even larger orchestra/chorus than for Mahler's 2nd by Penderecki.. - it ends with WOODEN bells ...
Freaked me out on LP ( label : Polskie nagrania, Muza ) - hard to imagine just what effect this ultra loud ending has live in a concert hall.

And there are ultra quiet pieces - like for solo viola, aptly named Silence; - if it comes up to - 40dBFS, it is considered - rock'n'roll !

 

[analogsurviver]

They get it from quoting maximum volume levels of instruments and not the volume levels experienced by attendees. Either due to lack of understanding or having an agenda.

Nope.

Binaural recordings done from the audience reflect the exact true SPLs experienced by the attendees.

Of course, these levels do vary depending on the seat - but regardless of position, they are true to the actual levels of music heard live.

The loudest noise heard at the concert of classical concert ? Applause ... - it takes gargantuan orchestra hell bent on being loud in order to result in more dBs at listener - than the person clapping next to you.

 

[bfreedma]

Nope.

Binaural recordings done from the audience reflect the exact true SPLs experienced by the attendees.

Of course, these levels do vary depending on the seat - but regardless of position, they are true to the actual levels of music heard live.

The loudest noise heard at the concert of classical concert ? Applause ... - it takes gargantuan orchestra hell bent on being loud in order to result in more dBs at listener - than the person clapping next to you.

Is the audience sitting beside the musicians? Of course not, thus the maximum instrument SPL levels being used as deflection don’t represent the SPL levels of the recording and it’s repoduction. If the recording is done with full volume clapping at 6” from the mic, I wouldn’t be interested in hearing it.

 

[analogsurviver]

Is the audience sitting beside the musicians? Of course not, thus the maximum instrument SPL levels being used as deflection don’t represent the SPL levels of the recording and it’s repoduction. If the recording is done with full volume clapping at 6” from the mic, I wouldn’t be interested in hearing it.

It is the sound any concert goer would experience live. I do try to tell the people in proximity to applaud "on playback" ie WITHOUT actually making the sound - but if the rendition is particularly good, resulting in standing ovations, these recommendations are the first to go tossed out of the window ...

 

[bigshot]

Binaural recordings done from the audience reflect the exact true SPLs experienced by the attendees.

Of course, these levels do vary depending on the seat - but regardless of position, they are true to the actual levels of music heard live.

Binaural playback has nothing to do with sound levels. You don't know what you're talking about.

 

[Don Hills]

I recall a conversation quite a few years ago, I think on a recording and mastering engineers' mailing list, that we tend to listen to in-room reproduction at a significantly lower SPL than the original live performance. That is, when played back at a level that you judge to be equal to the performance you heard live, the actual SPL is 6 to 10 dB lower. I might have saved the discussion, if I find it I'll report back.

 

[castleofargh]

I recall a conversation quite a few years ago, I think on a recording and mastering engineers' mailing list, that we tend to listen to in-room reproduction at a significantly lower SPL than the original live performance. That is, when played back at a level that you judge to be equal to the performance you heard live, the actual SPL is 6 to 10 dB lower. I might have saved the discussion, if I find it I'll report back.

I'd expect that simply from the difference in ambient noise at the live event. our impression of how loud is the music once in quiet room would naturally be affected. do you think there are other reasons? with the premise that the listener tries to set his listening level to feel like the live event. otherwise there are many reasons why one would listen at a quieter level of course.

 

[analogsurviver]

Binaural playback has nothing to do with sound levels. You don't know what you're talking about.

I beg to differ. You are not the only one attending live concerts. Are you going to imply I can not tell if the playback level of the binaural recording is not within reasonable limits ( say 2 dB or less difference ) than the actual at the concert live !?!

Remember your complaint in the description of a concert you've been attending ( IIRC - one of Wagner operas ) , where you stated that on the recording "they did not put them on the back ( or something similar ) ".

Trouble is, for you to - finally - achieve that goal of reproduction of recording true to live performance, you would have - like it or not - transform into - my clone.

Binaural - done right - enables EXACTLY that. There are MANY hurdles towards this goal before it will achieve perfection, but many people are working on it. Just one of the examples of work in progress :

https://www.facebook.com/photo.php?fbid=10218126088414454&set=gm.340576729888417&type=3&theater

The reason why this nut is so desirable to crack is super simple and self evident. Compare the number of people with decent well set up surround speaker system with that with people with mid to high class smartphone and headphones priced at least say $100 ( China - fi ) or $200 if sticking to western better established brands.

Then compare the projected market figures for making ( and selling... ) any new prospective recordings ....

 

[gregorio]

Binaural recordings done from the audience reflect the exact true SPLs experienced by the attendees.

No they don't! They may represent the exact true SPLs that may enter an individual's ears but they do NOT represent the SPLs experienced by the attendees, which is one of the reasons it's never caught on!

Is the audience sitting beside the musicians? Of course not, thus the maximum instrument SPL levels being used as deflection don’t represent the SPL levels of the recording and it’s repoduction. If the recording is done with full volume clapping at 6” from the mic, I wouldn’t be interested in hearing it.

Exactly and in fact, a recording where the audience noise is at the exact relative SPL with the music as would actually exist for the audience at a classical concert is a primary indicator of a poor, amateur/hobbyist recording. A fundament fact to which analogsurviver has repeatedly demonstrated that he's completely oblivious!!

G

 

[KeithEmo]

Hey..... nobody really needs indoor plumbing, right?

Without "what if" people, we would probably still be living in caves.

 

[KeithEmo]

There's another important aspect of aliasing.

The sounds produced by most instruments contain at least some harmonics. Therefore, when playing music, a small amount of added harmonic distortion is usually heard as a slight change in the tonal balance, but isn't necessarily very noticeable. However, when aliases are folded back down into the audible range, they end up as sounds that are NOT harmonically related to the original content - which often makes them far more noticeable - even at relatively low levels.

(So, when deciding what levels may or may not be audible, they must be treated more like "extra noises that shouldn't be there" rather than as extra harmonics which may simply alter the balance of harmonics already present. If the aliasing is minimal, then all of the extra sounds will be at relatively high frequencies, which may not be especially noticeable, or distinguishable as separate tones.... but, if it is severe, then you're talking about "extra sounds", which correlate to the musical content, but are separate from it.... which can be highly audible depending on several specific factors.)

2a. Of course, but what is significant attenuation at 20 kHz? 1 dB? 2 dB? 3 dB? 6 dB? Something like 3 dB attenuation at 20 kHz due to a "soft knee" filter is probably quite insignificant.
2b. Yep.
2c. Higher than Nyquist frequency content being folded into audible frequency range isn't necessorily audible, but it's technically wrong and against the principles of digital audio.

 

[KeithEmo]

It's pretty widely accepted that we humans take distortion as a cue that something is loud. A relatively low-powered table radio sounds loud if it's heavily distorted... even if the measured SPL level isn't especially high. (We equate "loud" with "annoying".) Our ears and brains include an internal mechanism that acts like an automatic gain control. Because of this we actually aren't especially accurate at judging true SPL levels. Most of us who listen to high quality headphones have noticed a phenomenon where, after we turn up the music once or twice because it "doesn't seem very loud", we take off the headphones and notice that "the world seems very quiet" - until our ears adjust again.

I would suggest that most home systems simply cannot produce levels in the 120 dB plus range without a lot of distortion...
And, even when our goal is "to play music at realistic listening levels", most of us "stop turning up the volume when the system starts to distort"...
Therefore, that number is actually simply based on the maximum level a typical home music system can deliver without becoming noticeably distorted (in a way that we perceive as "being loud").

In one sense, this may be good, because it prevents us from inadvertently turning the volume up to dangerous levels...
But, in another sense, it prevents us from achieving totally accurate reproduction...

I recall a conversation quite a few years ago, I think on a recording and mastering engineers' mailing list, that we tend to listen to in-room reproduction at a significantly lower SPL than the original live performance. That is, when played back at a level that you judge to be equal to the performance you heard live, the actual SPL is 6 to 10 dB lower. I might have saved the discussion, if I find it I'll report back.

 

[analogsurviver]

It's pretty widely accepted that we humans take distortion as a cue that something is loud. A relatively low-powered table radio sounds loud if it's heavily distorted... even if the measured SPL level isn't especially high. (We equate "loud" with "annoying".) Our ears and brains include an internal mechanism that acts like an automatic gain control. Because of this we actually aren't especially accurate at judging true SPL levels. Most of us who listen to high quality headphones have noticed a phenomenon where, after we turn up the music once or twice because it "doesn't seem very loud", we take off the headphones and notice that "the world seems very quiet" - until our ears adjust again.

I would suggest that most home systems simply cannot produce levels in the 120 dB plus range without a lot of distortion...
And, even when our goal is "to play music at realistic listening levels", most of us "stop turning up the volume when the system starts to distort"...
Therefore, that number is actually simply based on the maximum level a typical home music system can deliver without becoming noticeably distorted (in a way that we perceive as "being loud").

In one sense, this may be good, because it prevents us from inadvertently turning the volume up to dangerous levels...
But, in another sense, it prevents us from achieving totally accurate reproduction...

That is PRECISELY WHY did Stax modify the venerable Lambda family of headphones/earspeaker beginning with L700 - it enabled, among other things, about 4 dB increase of maximum SPL prior onset of distortion - from 113 dB of which Lambda Pro ( and slew of versions following it ) has been capable - to the 117 or so dB L-700 can play at - CLEANLY.

It is the least expensive way to have full ( well, to 120 dB still miss 3 dBs ... ) dynamic range - and frequency range from well below 20 to well above 40 kHz.

I own Lambda Pro - and there are moments I would kill for the L700. These peaks last for < 0.1% of total music time -but failing to reproduce them takes away one hell of a larger % of the realism perceived by the listener.

 

[james444]

Where do people get the idea that volume levels at classical music concerts are extremely high? Have they never been to a classical music concert?

I have a habit of stuffing small pieces of tissue in my ears, whenever it gets too loud for me in a live concert.

The fact that I never ever felt the need to do that in well over a hundred classical concerts I attended, proves to me that these can't be serious offenders.

By comparison, at the last rock concert I attended (Muse), it was like 30 seconds into the first song until I reached for the tissue.

 

[71 dB]

There's another important aspect of aliasing.

The sounds produced by most instruments contain at least some harmonics. Therefore, when playing music, a small amount of added harmonic distortion is usually heard as a slight change in the tonal balance, but isn't necessarily very noticeable. However, when aliases are folded back down into the audible range, they end up as sounds that are NOT harmonically related to the original content - which often makes them far more noticeable - even at relatively low levels.

(So, when deciding what levels may or may not be audible, they must be treated more like "extra noises that shouldn't be there" rather than as extra harmonics which may simply alter the balance of harmonics already present. If the aliasing is minimal, then all of the extra sounds will be at relatively high frequencies, which may not be especially noticeable, or distinguishable as separate tones.... but, if it is severe, then you're talking about "extra sounds", which correlate to the musical content, but are separate from it.... which can be highly audible depending on several specific factors.)

This is true if you listen to music with one note playing at a time. I don't. I simple chord creates non-harmonic frequencies when there's non-linearity.

The fundamental frequencies of a A major chord:



The spectrum after hard clipping:

To have ultrasonic content folding on sensitive 3000-4000 Hz band, you need frequencies as high as 40-41 kHz. Content just above Nyquist folds on the frequency band where hearing in insensitive ( > 15 kHz).