Dr. Edgar Choueiri explains BACCH

[doorofnight]

I can't see anyone saying that you should stop using BACCH in your system.

I was referring to this:

Crosstalk has nothing to do with this system and the same principle applies to two channel stereo legacy recordings but it gets confusing because normal stereo uses two channels as well, but if it wasn't recorded binaurally it is not the same thing and has nothing to do with crosstalk and should not be played on loudspeaker binaural.

I can't see anyone saying that you should stop using BACCH in your system. If you like the effect the BACCH program creates, you should keep on using it, but there should be no harm in knowing it will only act as an "effect plugin" for everything you play that is not a binaural recording.

I would not know binaural software on stereo recordings "only" act as an effect plugin when the effect makes music sound better. In sound reproduction, everything we do to enhance music is an 'effect.' I consider bass traps, acoustics, and controlled directivity of speakers all to be 'effect makers.' Some people do not care much about acoustics, while others do not care about controlled directivity. I care about all of them—it's the total sum that creates good sound reproduction. In general binaural software like bacch enhances everything most people like in sound reproduction.

 

[goat76]

I was referring to this:

It was pretty obvious that what he meant by ”should” was purely from a technical reproduction point of view, and not what people ”should” do as that is completely up to the individual listeners preferences of what effects is added to the reproduction system.

I would not know binaural software on stereo recordings "only" act as an effect plugin when the effect makes music sound better. In sound reproduction, everything we do to enhance music is an 'effect.' I consider bass traps, acoustics, and controlled directivity of speakers all to be 'effect makers.' Some people do not care much about acoustics, while others do not care about controlled directivity. I care about all of them—it's the total sum that creates good sound reproduction. In general binaural software like bacch enhances everything most people like in sound reproduction.

Yes, that may be your view and your preferences, but not necessarily everyone else’s view and preferences.

But if we allow us to talk about crosstalk cancellation in particular, if it was truly a ”fix” for a general problem, it would have worked equally as well no matter how the recording is done and what the recording contains. That is not the case.

 

[doorofnight]

Yes, that may be your view and your preferences, but not necessarily everyone else’s view and preferences.

Most people I know prefer uncolored sound with precise imaging, music that seems to float free from the speakers, and a soundstage that extends beyond them, creating an immersive "sound bubble." Audiophiles can spend hours discussing ways to enhance these effects. It's quite common.

 

[goat76]

Most people I know prefer uncolored sound with precise imaging, music that seems to float free from the speakers, and a soundstage that extends beyond them, creating an immersive "sound bubble." Audiophiles can spend hours discussing ways to enhance these effects. It's quite common.

Yes, and I’m completely fine with people trying to enhance their personal listening experience. But then we have those who have fully bought the idea that a crosstalk cancelation program like BACCH will solve some sort a general problem, as if it is meant that all sounds coming from the left loudspeaker is only meant to be heard by the left ear, and vice versa for the right loudspeaker for the right ear, but that is only the case if the recording is done that way which is only the case with binaural recordings.

 

[geickmei]

Are you comparing Record-Playback stereo (non-binaural/binaural) with Record-Playback surround sound?

Yes. Both are field type systems. Stereo just seems to be a binaural type system because of the mistake in history that left us with only two channels to work with. It SEEMS like the system is two ears/ two channels/ two speakers aiming their sound into your ears but it is not. The term "stereo" should be changed to "spatial audio" to try to eliminate the confusion. The idea with stereo is NOT signals for the ears, it is the placement of sounds in rooms. This sound needs to be as much like the original that was recorded as feasible, which means spatially as well as everything else like frequency response. Unfortunately the only part we have concentrated on is frequency response and distortion because we know much less about the spatial. This leads us to make very directional speakers because we think the idea is to shoot a tunnel of sound to the ears in the mistaken idea that all we want is the direct field from the speakers.

If we realize that legacy stereo is just a subset of surround sound as field type systems, we begin to analyze other factors of the reconstructed field such as radiation pattern, speaker positioning, the acoustic qualities of the room, and studying the spatial nature of the live sound that was recorded. It has a much greater reverberant field and spreads out way beyond the instrumental group with its early reflected sound and bass that seems to spread through the entire hall along the floor and everywhere. Some of us have noticed that some speaker designs do the spatial better, like dipoles and open baffle, because of a greater reflected sound output.

Surround sound just does it better because its spatial resolution is better with more channels.

Binaural is a different system. It aims to collect all of those spatial, spectral, and temporal qualities of the live sound with a direct sensory input to a dummy head shaped like a human head as if you could sit in on the concert as it happens. If you can then play that sound back directly into your ears you could eliminate the listening room from the equation and substitute the original, placing you at the concert where the dummy head was. There are problems doing this with headphones (IHL) and some problems with loudspeaker binaural. Both of these systems (spatial audio and binaural) should be studied separately to solve their problems and improve their advantages, but declaring crosstalk "the problem" with stereo does not help anything; just prolongs the confusion.

 

[Newman]

Surround sound just does it better because its spatial resolution is better with more channels.

...and from more appropriate directions.

 

[geickmei]

Well, yes but if we model the reproduction after the original and use the room reflections as additional (virtual) sources we can make 2 channel sound very good indeed. The trick in the future will be to make better use of the plethora of channels we now have available to us for the discrete placement of instruments and ambience and depth and three dimensionality. Imagine, if you will, two or three rows of direct channels up front to represent specific instrument placement in three dimensions in your room, or a set of wall speakers behind and around the stereo ones doing mainly the early reflected field from the recording, recorded separately. Imagine (finally) placing the center soloist in the center front speaker exclusively to solve the off side imaging problem. I already have audience applause coming from all around in my MS recordings rather than coming from behind the music in your room.

 

[doorofnight]

Yes, and I’m completely fine with people trying to enhance their personal listening experience. But then we have those who have fully bought the idea that a crosstalk cancelation program like BACCH will solve some sort a general problem, as if it is meant that all sounds coming from the left loudspeaker is only meant to be heard by the left ear, and vice versa for the right loudspeaker for the right ear, but that is only the case if the recording is done that way which is only the case with binaural recordings.

The main challenge with stereo systems playing stereo recordings is that most people want them to sound better

 

[iGude]

spatial nature of the live sound that was recorded. It has a much greater reverberant field and spreads out way beyond the instrumental group with its early reflected sound and bass that seems to spread through the entire hall along the floor and everywhere.

I couldn’t agree more and appreciate your contribution.

But I don’t yet understand why you think that the valid point that you make is in contradiction to crosstalk cancellation for recordings that contain contain enough crosstalk naturally such as recordings based on the ORTF or NOS method.

 

[geickmei]

I couldn’t agree more and appreciate your contribution.

But I don’t yet understand why you think that the valid point that you make is in contradiction to crosstalk cancellation for recordings that contain contain enough crosstalk naturally such as recordings based on the ORTF or NOS method.

Recordings do not "contain crosstalk." The term refers to binaural playback systems in which we cancel the opposite channel speaker at the ears. In live music there is no such thing as crosstalk, just normal spatial hearing. Playback in spatial audio systems such as stereo and surround sound uses specific speakers and channels for recorded sound placement plus phantom, or summing localization in between. One complication that is little understood is that for two channel field type systems played on very directional speakers the sound field stops at the left and right speakers unless we use a large ratio of reflected sound from the walls nearby, as with dipolar and open baffle and direct/reflecting speakers with designed-in radiation patterns. THAT principle in field type systems is what needs to be studied rather than switching over to binaural because binaural is severely limited in number of listeners.

 

[iGude]

Recordings do not "contain crosstalk."

The term crosstalk might be misleading. What I mean is recording techniques that have timing differences for off-axis signals similar to the human head such as the ORTF technique, i.e. both left and right channels contain the signal with time delay between them.
Wouldn’t stereo crosstalk from the speakers destroy this timing information?

 

[geickmei]

The system is not two channels piped to your two ears. You do not listen to stereo binaurally. We can make recordings based on either ITD or ILD, spaced or coincident, a stereo pair or multimikes. Summing localization seems to work with all of them with good speaker positioning and room acoustics. Crosstalk cancellation spreads the stereo image wider than it was when recorded, which is an interesting thrill but not correct if it was not recorded that way - with a dummy head.

 

[doorofnight]

Crosstalk cancellation spreads the stereo image wider than it was when recorded, which is an interesting thrill but not correct if it was not recorded that way - with a dummy head.

Yes, it does - and a bit more. You're trying to make something objective that is inherently subjective. Most audiophiles seek a wider stereo image and often feel limited and I do not think that they care how the song has been recorded. At least the effect of binaraul software is real, and people can decide for themselves whether it’s a thrill / just an effect - and give it the value they want.

And besides that: there's no way to determine whether the reproduction at home matches how it was recorded in a studio, concert hall, or other setting. A lot of music is no longer recorded using microphones. Additionally, most producers listen up close to monitors with less crosstalk then we have in our home systems, but not all. However, it would be an interesting experiment to invite some producers to my home and ask them whether they feel their music is being reproduced correctly or not.

 

[iGude]

The system is not two channels piped to your two ears.

Well, the direct wave from the speakers to the ears can be considered „piped to the ears“, right?
And this direct wave defines the position of phantom sources on the soundstage, right?
Please correct me if I’m wrong.

I do agree with all the rest of your comments though.

 

[geickmei]

Yes, it does - and a bit more. You're trying to make something objective that is inherently subjective. Most audiophiles seek a wider stereo image and often feel limited and I do not think that they care how the song has been recorded. At least the effect of binaraul software is real, and people can decide for themselves whether it’s a thrill / just an effect - and give it the value they want.

And besides that: there's no way to determine whether the reproduction at home matches how it was recorded in a studio, concert hall, or other setting. A lot of music is no longer recorded using microphones. Additionally, most producers listen up close to monitors with less crosstalk then we have in our home systems, but not all. However, it would be an interesting experiment to invite some producers to my home and ask them whether they feel their music is being reproduced correctly or not.

If you have ever been to a live concert you have a general idea of where instruments are positioned w respect to the whole soundstage and the room around them. I got into recording to witness Toole's "Circle of Confusion" and to be able to compare the live vs recorded. In general the instruments are in a group at about the typical stereo spread, not in a 180 degree arc around you. As for producers listening differently from a typical audiophile system, Toole's research found that even they prefer a good home system to the monitors they typically use. I use my actual listening room to mix my surround recordings to get immediate feedback on what it sounds like. But during recording just a pair of headphones.

 

[doorofnight]

It was pretty obvious that what he meant by ”should” was purely from a technical reproduction point of view, and not what people ”should” do as that is completely up to the individual listeners preferences of what effects is added to the reproduction system.

No, it's obvious that from a technological reproduction of point that there is no problem at all, playing stereo recorings on binaural systems. I can do it as often as I like.

If you have ever been to a live concert you have a general idea of where instruments are positioned w respect to the whole soundstage and the room around them. I got into recording to witness Toole's "Circle of Confusion" and to be able to compare the live vs recorded. In general the instruments are in a group at about the typical stereo spread, not in a 180 degree arc around you. As for producers listening differently from a typical audiophile system, Toole's research found that even they prefer a good home system to the monitors they typically use. I use my actual listening room to mix my surround recordings to get immediate feedback on what it sounds like. But during recording just a pair of headphones.

I have been to live concerts in churches and concert halls, and I find that stereo recordings played on standard stereo systems often feel limited in comparison. Binaural software helps bring the experience closer to what I hear in a live setting, but ultimately, it’s never the same. That said, I don’t dwell on this too much. After all, a lot of music is produced without microphones and isn’t necessarily created in concert halls.

 

[doorofnight]

Yes, and I’m completely fine with people trying to enhance their personal listening experience. But then we have those who have fully bought the idea that a crosstalk cancelation program like BACCH will solve some sort a general problem, as if it is meant that all sounds coming from the left loudspeaker is only meant to be heard by the left ear, and vice versa for the right loudspeaker for the right ear, but that is only the case if the recording is done that way which is only the case with binaural recordings.

It's also the case with stereo recordings, but yes, the effect is more intense with binaural recordings. I rarely listen to those recordings.

 

[MattHooper]

If you have ever been to a live concert you have a general idea of where instruments are positioned w respect to the whole soundstage and the room around them. I got into recording to witness Toole's "Circle of Confusion" and to be able to compare the live vs recorded. In general the instruments are in a group at about the typical stereo spread, not in a 180 degree arc around you. As for producers listening differently from a typical audiophile system, Toole's research found that even they prefer a good home system to the monitors they typically use. I use my actual listening room to mix my surround recordings to get immediate feedback on what it sounds like. But during recording just a pair of headphones.

There’s also messy stuff about whose experience you are trying to replicate.
Is it just the mixer or engineer or the artist themselves?

If it’s the artists, then those who employ mixing engineers often aren’t sitting in the mixing seat. Often enough the artists are lounging around on whatever other seating in the room such as a sofa further back from the speakers, and even when you have the artists involved in the mixing or sitting at the mixing desk, especially back In the earlier days, they would often be several people along the mixing board meaning people are getting different mixes of left and right signal. So exactly who’s Sonic experience are we trying to reproduce?

 

[doorofnight]

I really do not care if people do not like binaural software for stereo recordings on stereo sets.. If you don't like it, don't use it. But there's no way to tell other people that they should not play stereo recordings on binaural sets because of what they feel are objective reasons and claim that it's not correct.

 

[MattHooper]

Sorry, I haven’t read through all of this, or at least can’t remember what everybody has written but…

Are BAACH users here using the head tracking option?

Without the head tracking option, is this
“ head in a vice listening?” In other words does it mean really limiting the position of your head in order to gain the benefits of the BAACH processing?