comparing live and recorded music

[johncarm]

I want to discuss the question: What is the right way to compare live and recorded music in order to select good audio components?
 
My other three threads can be merged by the mods if they want. They are, in essence, about the question "What do we need to know to design and test audio equipment?"
 
Here I want to focus only on the problem of comparing a live acoustic event with its reproduction.
 
Joe Bloggs stated a point of view which I think greatly impedes the ability to select good audio components in the context of acoustic classical music. Namely, he believes that audio reproduction is so different from live music that there is no meaningful comparison of accuracy; only preference testing can be used to make selections.
 
I attended some classical chamber music recording sessions in the 80's, and the question on everyone's mind, both musicians and engineers, was "How similar is the reproduction to the original?" We weren't asking what sounded "best," we were asking what was most accurate. Which of course makes complete sense if your goal is high-fidelity.
 
If you ask only "What do you prefer?" then you are retreating from knowledge about fidelity in reproduction.
 

 

[spruce music]

"What do we need to know to design and test audio equipment?"
 
If this is really your question, spend time on the DIY audio forums.  Its circuits and such you need to be learning about to answer the question.
 
"What is the right way to compare live to recorded music in order to select good audio components?"
 
The answer to this question is this is a poor method to select good audio components.   This assuming that fidelity is really your goal.  If enjoyment then personal preference and a pleasing coloration is perfectly fine.  Confusing those two goals is a big problem in some circles of the audio world. Joe Bloggs gave you good insight.
 
I have done a little recording as a hobby.  I think maybe Gregario is a pro in the business.  If you use some minimalist two mic recording for small groups and play it back in a similarly sized space you can get a decent fascimile of the original in the sweet spot.  I think people fail to realize how much suspension of belief or learning to suspend it goes on in stereo listening.  Similar to suspending belief for 2D movies.
 
Your first problem is getting live recorded music that hasn't been processed tremendously.  It is exceptionally rare in commercial recordings. Few are the recordings done with a couple mikes and no compression, no EQ, no reverb.  So once processing has been done you have a crooked yardstick.  You also likely will find such processing done expertly and with taste can make recordings that sound a bit more real than the high fidelity recordings.  So there is that problem.  So immediately you are off into a world of illusion. So many audiophiles assume they will prefer the higher fidelity recording and higher fidelity playback.  They believe they will easily recognize the difference.  That appears to be a myth.
 
Some components distort this or that aspect of playback and different people like different things.  The super dynamics of horn speakers, the effervescent clarity of panel speakers, and everything in between.  The smooth, luscious warmth of tubes, tape and turntables.  The sizzling rock concert power of big transistor amps pushing cone and dome speakers.  Yes these are somewhat gross colorations, but pertinent all the same.
 
It isn't hard anymore to get good electronics that are mostly colorless and honest. Speakers are the stickey wicket.  Pick a speaker with a character you most prefer, feed your preferred music, and then maybe do some room correction or simply digital EQ to taste.  The upstream components are simply not a big deal versus speaker variations unless they have been intentionally colored like some tube products. 

 

[johncarm]

  "What do we need to know to design and test audio equipment?"
 
If this is really your question, spend time on the DIY audio forums.  Its circuits and such you need to be learning about to answer the question.
 
"What is the right way to compare live to recorded music in order to select good audio components?"
 
The answer to this question is this is a poor method to select good audio components.   This assuming that fidelity is really your goal.  If enjoyment then personal preference and a pleasing coloration is perfectly fine.  Confusing those two goals is a big problem in some circles of the audio world. Joe Bloggs gave you good insight.
 

 
My goal is fidelity to the live performance. At the recording sessions I've attended, it's been pretty obvious how components, microphones, placement, room, speakers, etc. affected fidelity to the sound we were hearing in the recording hall. So why not use these observations? You seem to think it's a poor method. You stated a lot of stuff in your post but I couldn't really discern a reason for your opinion. So why do you think it's poor?

 

[spruce music]

   
 So why do you think it's poor?

Because stereo recording and playback has inherent limitations that prevent it from fully re-creating live sound.  Then throw in the lack of suitable recordings on top of that.  Then the inherent issues with speakers/headphones.  You might almost get around this with binaural and headphones.  You probably need custom fitted pinnae that match yours on the microphones for that to work just so. 
 
And the audiophile attitude you have about all this is not helping.  You want to pick an amp, or pre-amp or DAC by listening to recordings and seeing which sounds more like live.  All of those of merely decent quality are so transparent or close to it you won't find differences of concern unless you listen sighted and drink the koolaid.  Good DACs are easy to come by.  If you want real improvements look to speakers, the room and treating the room. 

 

[johncarm]

  Because stereo recording and playback has inherent limitations that prevent it from fully re-creating live sound.  Then throw in the lack of suitable recordings on top of that.  Then the inherent issues with speakers/headphones.  You might almost get around this with binaural and headphones.  You probably need custom fitted pinnae that match yours on the microphones for that to work just so. 
 
And the audiophile attitude you have about all this is not helping.  You want to pick an amp, or pre-amp or DAC by listening to recordings and seeing which sounds more like live.  All of those of merely decent quality are so transparent or close to it you won't find differences of concern unless you listen sighted and drink the koolaid.  Good DACs are easy to come by.  If you want real improvements look to speakers, the room and treating the room. 

 
But stereo recording and playback sounds similar to live. So why not pick what is most similar?
 
I don't know what you are talking about with respect to "audiophile attitude." I'm not talking about walking into a high-end store and playing my favorite recording. I'm talking about comparing the live sound, with my ears present in the studio, to the sound on the recording in the control room.
 
I don't see that you've given any reason why it's a problem to pick mic techniques, etc. that result in the most similarities to the live sound. What's the actual problem?
 
EDIT: by the way, who said I wasn't talking about speakers and treating the room? Who said I was just talking about DACs?

 

[spruce music]

Get a good sounding room and place some bi-directional mikes in a blumlein configuration.  Sit as close as possible to the mikes while the recording happens. 
 
Next, get speakers setup at a 90 degree spread from the listener's position.  Playback the recording.  Change speakers. Repeat the recording.  Rinse and repeat until you have a winner.

 

[johncarm]

  Get a good sounding room and place some bi-directional mikes in a blumlein configuration.  Sit as close as possible to the mikes while the recording happens. 
 
Next, get speakers setup at a 90 degree spread from the listener's position.  Playback the recording.  Change speakers. Repeat the recording.  Rinse and repeat until you have a winner.

How does this answer my question?
 
What's the problem with judging similarity of recorded sound to the live sound?

 

[icebear]

  I want to discuss the question:
What is the right way to compare live and recorded music in order to select good audio components?
...

 
Honestly what is your problem ...? You just listen

You go to a live concert, preferably with acoustic instruments, get a recording of the same musicians and program (ideally from the performance you attended) but same musicians and same program even from a different performance will be good enough. Then you listen with your audio set up at the (perceived) same volume and compare if the sound of the reproduction matches what you heard during the live performance.
 
If you want to go anal over this evaluation and have the means ... you rent a suitable room, audio equipment for recording and reproduction and the musicians with their instruments and a lot of patience. You have them play, record the session and then behind a curtain the reproduction will be played or the musicians play live again you sit on the other side of the curtain doing a blind listening evaluation.
 
I have arrived at a set up that is close enough to the live event.
I have seen Janine Jansen live, violin and piano sonata and have a recording of the same piece (different hall).
The sound of her violin is as close to the original live sound as it gets.
The only difference that I can perceive, is that I was closer to her when attending the live performance than when I listen to the recording via my speakers

 

 

[johncarm]

   
Honestly what is your problem ...? You just listen

s

I assume you're being sarcastic because you are just repeating the gist of what I said. It's Spruce and Joe Bloggs who have some kind of problem with judging similarity.
 
I do think we can one step beyond "just listening" and become curious about what patterns the musicians are noticing when they judge two things as similar.

 

[icebear]

  I assume you're being sarcastic because you are just repeating the gist of what I said. It's Spruce and Joe Bloggs who have some kind of problem with judging similarity.
 
I do think we can one step beyond "just listening" and become curious about what patterns the musicians are noticing when they judge two things as similar.

 
Usually a little bit of sarcasm swings in all my posts, got me there.

 
Are you asking about the judgment of the differences from musicians or from audiophiles?
You might get very different results as musicians listen more for content than for differences on the surface. They are often content with non audiophile equipment, maybe just a matter of means but in general it might be more like the artist doing a painting and later listening or reading about his work in the press, from curators, gallery owners and auction houses. The artist roles his eyes and checks get written.
 
I want true high fidelity in my set up. I want the reproduction sound as close as possible to the live event.
But then ... I love going to live concerts and the music of my preference is suitable to sound great when captured live.

 

[Joe Bloggs]

I attended some classical chamber music recording sessions in the 80's, and the question on everyone's mind, both musicians and engineers, was "How similar is the reproduction to the original?" We weren't asking what sounded "best," we were asking what was most accurate. Which of course makes complete sense if your goal is high-fidelity.


 


If you ask only "What do you prefer?" then you are retreating from knowledge about fidelity in reproduction.


 

You could modify the preference test by making the listeners score according to similarity to the original instead of preference.

As a matter of practicality, your goals can be better served by grabbing a hold of your sound engineer to learn miking and mixing techniques from him on the recording side and loudspeaker placement and room treatment techniques on the playback side, than coming up here trying to get us together to rewrite all of (psycho)acoustics theory. I suspect you haven't scratched the surface of what is possible with current technology.

 

[spruce music]

  How does this answer my question?
 
What's the problem with judging similarity of recorded sound to the live sound?

In your original post: What is the right way to compare live and recorded music in order to select good audio components?
 
I have read at least two papers where they recorded acoustic music with different techniques.  Had trained listeners or music school students on hand.  Then played back over high quality gear in the same space these recordings to judge which most accurately conveyed the live performance, the placement of virtual imaging of the musicians vs their actual place during the recording, and a judgement of overall fidelity.   In one case a Blumlein recording was judged most accurate.   In another Blumlein and I think it was NOS (could have been mid/side) were judged equally accurate with slightly different strengths and better than the other methods.
 
So research into the matter indicates a blumlein technique is most accurate at recording and playing back the recording.  Theoretically, going back to the 1930's the blumlein recording was thought to be the most accurate method of stereophonic recording. I have expressed the opinion speakers are the overwhelming component on the playback end that matters.    You wanted to select good audio components comparing live and recorded music.  I gave you what should be THE most accurate way to do that.  How did I not answer your question?  If you believe something other than speakers matter, I still have answered your question, you just swap other components than speakers. 
 
https://www.academia.edu/693259/An_Investigation_into_Stereo_Microphone_Techniques_and_their_Applications
 
Here is another paper I just turned up.  Also saying Blumlein is very good.  Paper is a thesis and about 100 pages.  You can skim the highlights though.

 

[spruce music]

If your desire is to simply listen to what you have and pick this or that well that isn't quite what you asked.
 
As already stated, musicians are notorious for listening to how other musicians are doing things and ignoring quality altogether.  If they can hear how chords were done or the rhythm things were paced and such that is all they are after.  Cassette tape, MP3 over a car system is plenty for that.
 
Listening to recordings typically you want a sense of the space.  Should not be 2D or cardboard cutouts.  Should be front to back depth.  Space with sense of a cushion of air around the performance.  A subliminal reality.  Of course you can only get that with a few recordings.  Not all that many will accurately make this possible.

 

[johncarm]

 
In your original post: What is the right way to compare live and recorded music in order to select good audio components?
 
I have read at least two papers where they recorded acoustic music with different techniques.  Had trained listeners or music school students on hand.  Then played back over high quality gear in the same space these recordings to judge which most accurately conveyed the live performance, the placement of virtual imaging of the musicians vs their actual place during the recording, and a judgement of overall fidelity.   In one case a Blumlein recording was judged most accurate.   In another Blumlein and I think it was NOS (could have been mid/side) were judged equally accurate with slightly different strengths and better than the other methods.
 
So research into the matter indicates a blumlein technique is most accurate at recording and playing back the recording.  Theoretically, going back to the 1930's the blumlein recording was thought to be the most accurate method of stereophonic recording. I have expressed the opinion speakers are the overwhelming component on the playback end that matters.    You wanted to select good audio components comparing live and recorded music.  I gave you what should be THE most accurate way to do that.  How did I not answer your question?  If you believe something other than speakers matter, I still have answered your question, you just swap other components than speakers. 
 
https://www.academia.edu/693259/An_Investigation_into_Stereo_Microphone_Techniques_and_their_Applications
 
Here is another paper I just turned up.  Also saying Blumlein is very good.  Paper is a thesis and about 100 pages.  You can skim the highlights though.

 
I see. I was asking you more specifically what's wrong with noting similarity of live to recorded music. As best as I can figure, you think that there must be a great deal of spatial resemblance before it makes sense to note similarities?
 
The problem with your suggestion is that it doesn't help recording engineers in 99% of situations. You still have to listen in the hall or studio, then step into the control room and compare. No way around it.
 
If your aim is high-fidelity, then you should be noting similarities and differences and choose the mic placement or whatever that gets you the most similarity.
 
I am not sure that's what most recording engineers are doing, because there is so much terrible recorded sound. Orchestra with the balance all wrong, strained dynamics, brass that isn't brilliant, woodwinds that aren't differentiated from each other, piercing massed strings, etc.
 
 

  If your desire is to simply listen to what you have and pick this or that well that isn't quite what you asked.
 
As already stated, musicians are notorious for listening to how other musicians are doing things and ignoring quality altogether.  If they can hear how chords were done or the rhythm things were paced and such that is all they are after.  Cassette tape, MP3 over a car system is plenty for that.
 
Listening to recordings typically you want a sense of the space.  Should not be 2D or cardboard cutouts.  Should be front to back depth.  Space with sense of a cushion of air around the performance.  A subliminal reality.  Of course you can only get that with a few recordings.  Not all that many will accurately make this possible.

 
It may be that some musicians ignore sound quality in recordings. But if a musician wants to make good sound, they have to be sensitive to details of sound. Makes sense, right? How can you make good sound if you never pay attention to sound quality? 
 
As a matter of fact, the guy from Sheffield Lab I worked with in the 80's was both a concert pianist and recording engineer. He said that musicians can become fabulous engineers in moments. All you have to do is ask them to listen to the recording as if it were a musician playing right in front of them. 
 
There is a myth that "sound quality" is a specialization and that "music" is a separate specialization, as if they were fundamentally different activities. Many musicians are told by the engineers not to concern themselves with the recorded sound because "that's the engineer's job." Total baloney.

 

[johncarm]

You could modify the preference test by making the listeners score according to similarity to the original instead of preference.

As a matter of practicality, your goals can be better served by grabbing a hold of your sound engineer to learn miking and mixing techniques from him on the recording side and loudspeaker placement and room treatment techniques on the playback side, than coming up here trying to get us together to rewrite all of (psycho)acoustics theory. I suspect you haven't scratched the surface of what is possible with current technology.

 
It sounds like you are referring to the myth that "sound is the engineer's job" and "music is the musician's job."
 
Both disciplines require what we call "an ear." That is, an ability to perceive patterns in sound. There is no reason that the patterns an engineers uses and the patterns a musician uses should be distinct. (The TECHNIQUES are distinct but not the EAR.)
 
I think they probably USUALLY ARE distinct and that's an unfortunate thing.
 
I also think that some musical styles lend themselves to distinct roles.
 
But acoustic classical music? The engineer and musician should be listening for very similar things.