Dr. Edgar Choueiri explains BACCH

[STC]

When we are out in open spaces and we hear a fly buzzing around our head we can judge with tremendous accuracy the direction and proximity of that fly. It’s not learned, it’s not reliant on reflections and anyone who has experienced it knows what I am talking about. T

I made a mistake in my statement. What I meant to say was without reflection you do not have the cues for the space size. Regarding bees, at best you hear the faint buzz about 10 feet or so and it gets louder as it reaches closer to you. Will be 80 db? Probably when it flys right next to your ear. So you can know the distance based on the level difference. The direction of bee can be determined by HRTF and the spectral changes because the changing position will deliver different frequency response and level. Externalization of sound naturally happens. In anechoic chamber, even at 10 feet away the level could be 80 db so whatever level difference you hear will not match what you have heard in nature but I bet such tests in anechoic chamber would result a larger than the typical in head experience compared to the standard recordings used.

I have a binaural bee recordings that I use to demo. Depending on the listeners some could sense it is actually in the room circling around the head while for others stage is just on the front. For those perceive it to be circling around the head, I am believe it is just their imagination rather than what XTC could do.

 

[Justdafactsmaam]

I made a mistake in my statement. What I meant to say was without reflection you do not have the cues for the space size. Regarding bees, at best you hear the faint buzz about 10 feet or so and it gets louder as it reaches closer to you. Will be 80 db? Probably when it flys right next to your ear. So you can know the distance based on the level difference. The direction of bee can be determined by HRTF and the spectral changes because the changing position will deliver different frequency response and level. Externalization of sound naturally happens. In anechoic chamber, even at 10 feet away the level could be 80 db so whatever level difference you hear will not match what you have heard in nature but I bet such tests in anechoic chamber would result a larger than the typical in head experience compared to the standard recordings used.

I have a binaural bee recordings that I use to demo. Depending on the listeners some could sense it is actually in the room circling around the head while for others stage is just on the front. For those perceive it to be circling around the head, I am believe it is just their imagination rather than what XTC could do.

Binaural recordings through headphones are surprisingly limited in their effect. And does not work at all for some listeners. Edgar for one does not perceive spatiality with binaural recordings through headphones.

OTOH there is a test track from Chesky in binaural where David Chesky walks from 30 feet away off to the left right up to the ear of the dummy head while talking. The effect is so visceral and convincing I had one person literally jump out of the chair and exclaim “I felt him breathing on me!”

 

[STC]

Binaural recordings through headphones are surprisingly limited in their effect. And does not work at all for some listeners. Edgar for one does not perceive spatiality with binaural recordings through headphones.

OTOH there is a test track from Chesky in binaural where David Chesky walks from 30 feet away off to the left right up to the ear of the dummy head while talking. The effect is so visceral and convincing I had one person literally jump out of the chair and exclaim “I felt him breathing on me!”

Nothing magical with Chesky recordings they are also made similarly like mine. In fact, the whole idea of Chesky’s binaural plus was an early strategy for BACCH so nothing extraordinary feeling some breathing next to you with that particular recording. All XTC should do just that.

The reason I use my own binaural recordings along Chesky’s, I have a reference recording matching my ILD and ITD so that the XTC could be setup accurately at least for my HRTF.

 

[Lion♡]

Today XTC it is just an effect, not necessarily intended, adjustable with a knob at will.
Not very different in practice from the DRC where you are looking for a frequency response target that you like.

It's interesting.
May I know why it's effective?
When you have a listening distance of approximately 10m or more, the values in both ears fall similarly.
Then, is this natural phenomenon only an effect?
XTC's goal, including Bacch, is to minimize the characteristics of Crosstalk that appear strongly within a certain distance, erasing only the part that comes from the position and spacing of the speaker, except for the initial reflection and spatial characteristics of my space, your space, someone's space.

 

[STC]

XTC cannot replace multi channel. It is only confined to the stereo correction. If you want the recordings to sound like you are actually in the same space then you need several more speakers for the spatial recreation. XTC is just a small step to bring the best from the conventional stereo format.

5.1 format is better than 2.0. But 5.1 where the front and stereo channels is XTC then they are better than the standard 5.1. Having said that, you are going to get the immediate benefit just because you are doing XTC with your multi channel format. There you still need to address the level of the center channel or do you even want to use the center channel?

I have come to a point to accept that multi channel formats are best left to do what they are supposed to do and XTC is only for stereo or Quad formats with the virtual hall. But this involves at least 16 additional speakers and individual impulse responses. Too complicated for just an hour or two relaxation.

 

[Davide]

It's interesting.
May I know why it's effective?
When you have a listening distance of approximately 10m or more, the values in both ears fall similarly.
Then, is this natural phenomenon only an effect?
XTC's goal, including Bacch, is to minimize the characteristics of Crosstalk that appear strongly within a certain distance, erasing only the part that comes from the position and spacing of the speaker, except for the initial reflection and spatial characteristics of my space, your space, someone's space.

What is the spatiality you target?
There is no reference, nor metric to quantify it (*).
You adjust the intensity of the crosstalk reduction as desired.
So I call it an effect, like a common stereo widener.
I don't understand what the problem of calling it that is. Does it seem diminish? It is certainly not said in that sense.

(*) You could say that the target is the minimum crosstalk possible, but most of the audio tracks have not been conceived primarily for that, except binaurals/Type A/Type B.
But seems you still lack the way to quantify the spatial accuracy to a real event, independently from HRTF variable (unless the recording is done exactly with the listener's HRTF).
You can have more accuracy to the audio file, not necessarily to what heard from the mixing engineer (which will probably never be definable).
It is a matter of paradigm in the end.
But for this it seems to me this is an effect, as the crossfeed introduced in the headphones to compensate for the total lack of crosstalk and make the perception more realistic and enjoyable. And nothing wrong... indeed. Only some concerns about the cost. But this is nonetheless subjective.

But I repeat. Points of view ... it is not so relevant to call it in a way or the other. It's relevant what it does.

EDIT: My epilogue

 

[STC]

What is the spatiality you target?
There is no reference*, nor metric to quantify it.
You adjust the intensity of the crosstalk reduction as desired.
So I call it an effect, like a common stereo widener.
I don't understand what the problem of calling it that is. Does it seem diminish?

* You could say that the target is the minimum crosstalk possible, but most of the recordings have not been conceived exactly for that, only (but not necessarily) the binaurals that are a small part, and for which XTC is certainly indicated.
But you still lack the way to quantify how correctly it represents spatiality, so you only get a certain spatial effect, not necessarily what heard in the studio.
This is an effect exactly as the crossfeed introduced in the headphones to compensate for the total lack of crosstalk and make the perception more realistic and enjoyable.

But I repeat. Points of view ... it is not so relevant to call it in a way or the other. It's relevant what it does.

I think you misunderstood the purpose of XTC. The ideal XTC should give you the same feel like wearing headphones without the in head feeling.

 

[STC]

This is an effect exactly as the crossfeed introduced in the headphones to compensate for the total lack of crosstalk and make the perception more realistic and enjoyable.

Headphones listening is unnatural but today we accept them as normal because we have adapted to the sound. If you put on the headphones on a toddler for the first time you would notice the toddler would be looking up to find the direction of the sound but you don’t see such behavior with a person who got used to headphones sound.

Crossfeed introduces crosstalk which in typical stereo setup would introduce a 220 μs delayed sound. This would make the stage much narrower. It is nowhere the same as listening to loudspeakers because the role of pinna is bypassed with headphones.

 

[Justdafactsmaam]

XTC cannot replace multi channel. It is only confined to the stereo correction.

Edgar tells me he has an algorithm that allows the BACCH to process any multichannel recording and convert it to two channel output that the BACCH SP will process to recreate the multichannel imaging.

If you want the recordings to sound like you are actually in the same space then you need several more speakers for the spatial recreation.

Not true. I have personally experienced demos with the BACCH in which live recordings done with the in ear microphones used to create the filters match the original spatiality of the original sound 100%

You don’t need more speakers you just need the right recording

XTC is just a small step to bring the best from the conventional stereo format.

That is subjective. IMO it’s a major breakthrough. Opinions abound.

5.1 format is better than 2.0. But 5.1 where the front and stereo channels is XTC then they are better than the standard 5.1.

I have yet to experience any surround format that comes close to the BACCH in spatial reproduction.

I have not heard of a 5.1 system that works in conjunction with XTC. What format is that? Sounds interesting

And for all the folks looking for an argument I am simply expressing my personal opinions based on my personal auditions of various surround systems.

If others prefer 5.1 or any other surround sound format that’s cool. That’s your preference

 

[STC]

Edgar tells me he has an algorithm that allows the BACCH to process any multichannel recording and convert it to two channel output that the BACCH SP will process to recreate the multichannel imaging.

There are many such algorithms. Even the modest AVR can downsample multichannel for two speakers.

Not true. I have personally experienced demos with the BACCH in which live recordings done with the in ear microphones used to create the filters match the original spatiality of the original sound 100%

You don’t need more speakers you just need the right recording

two channel XTC can only make up for the frontal stage. Occasional you can make sound to appear behind your head but such effects often for a fleeting moment only.

I have yet to experience any surround format that comes close to the BACCH in spatial reproduction.

I don’t know what you mean by spatial production. Dolby got a definition. I was thinking it is about the actual audio scene of a particular venue. You cannot do that with 2 speakers. Once there was an institute that literally transports you to Sydney Opera House.

Binaural recording is only good if it is heard with headphones. Otherwise, you are producing the sound waves of what reaches your ears from your speakers. That alone dilutes the intensity of the sound and it further travels and to reach your ear.

I am sure you have many of not being satisfied with the sound of Chesky’s binaural plus. Although it can create a good 3 D impression but if listen to drums and other instruments you can notice the intensity missing.

 

[Justdafactsmaam]

There are many such algorithms. Even the modest AVR can downsample multichannel for two speakers.

That’s great. More ways two channel stereo can do what surround sound can do.

two channel XTC can only make up for the frontal stage. Occasional you can make sound to appear behind your head but such effects often for a fleeting moment only.

I’m not making this up. I have personally done the demo. The BACCH will accurately match the original sound in positioning from anywhere in the room. Anywhere

I don’t know what you mean by spatial production. Dolby got a definition. I was thinking it is about the actual audio scene of a particular venue. You cannot do that with 2 speakers.

Yes, you can. I personally hear it in my system.

Once there was an institute that literally transports you to Sydney Opera House.

Binaural recording is only good if it is heard with headphones.

No. It’s vastly superior with the BACCH. Again this is from personal experience

Otherwise, you are producing the sound waves of what reaches your ears from your speakers. That alone dilutes the intensity of the sound and it further travels and to reach your ear.

Not an issue if your speakers are up to the task

I am sure you have many of not being satisfied with the sound of Chesky’s binaural plus. Although it can create a good 3 D impression but if listen to drums and other instruments you can notice the intensity missing.

Not on my system.

 

[Davide]

It looks like a dog biting its own tail.
XTC conceptually helps the spatial perception minimizing the intrinsic interference of spatial information of system with limited physical source points, which by definition cannot render spatiality in a totally effective way for our hearing.
All this is welcome. No one is against, conceptually speaking.
But in practice what is gained seems not always quantifiable in objective terms (compared to the audio track it always is, but paradigm of fidelity is at stake).
To me, this should make it fall back into the sphere of subjective preferences.
And this is fine, at least, I am not against subjective improvements.
What shouldn't be left out is the fact that:
- It could be effective on a limited area, or still work as one person filter
- The system could be relatively complex
- The objective result depends on the room and the speakers
- The subjective result depends on the type of spatial cue of audio tracks
- The subjective result is not necessarily intended by the artist or producer
- The cost is relatively high and and in any case additional

If one believes that the compromises are acceptable, this surely could be a way to go.
Of course, to determine if the compromises are worth one can also hear the result on his system with the demo if possible.
But it is not strictly necessary if some compromises are already deemed unacceptable for the cost.

The usefulness of this discussion is also to highlight all these things.
It does not mean excluding anything, especially about the pleasantness of the result.

EDIT: My epilogue

 

[Lion♡]

What is the spatiality you target?

XTC and spatiality are not related.
Strictly speaking, the characteristics of each space vary depending on the intensity of the initial and late reflections and the time bands located, but it has nothing to do with the purpose of XTC.

So I call it an effect, like a common stereo widener.

While listening at a narrow listening distance of an equilateral triangle of 1m, you can feel that it has widened relatively when XTC is applied.
It's about erasing the gap between the speakers, but it's not widening, it's just erasing what our opposite ear defines as the gap.
But that's why I wrote down the above and asked.
What would it be if you listened to it at a distance of more than 10m?

* You could say that the target is the minimum crosstalk possible, but most of the recordings have not been conceived exactly for that,

It's the same as what I said above.
If I make and mixed sound source in a 3m space and listen to it by taking only the characteristic of extremely controlling initial reflection at a distance of 10m or 20m or more and making little difference in response to both ears, is that completely wrong?
Binaural sound sources or videos have their singularity and advantages, but that doesn't mean regular stereo sound sources don't fit.
It's just a constraint on the environment and the ears on both sides.

This is an effect exactly as the crossfeed introduced in the headphones to compensate for the total lack of crosstalk and make the perception more realistic and enjoyable.

After all, the headphones (or IEMs) are in full binaural condition.
But it's different from how we hear it, and there's no information about the crosstalk.
So as STC said, Bach (or other XTC) is to listen to the speaker like headphones, that's right.
But it's different because the way we hear speakers and natural sounds and headphones (IEM) without the information of the opposite ear are completely different.
That's why Bach HP also virtualizes headphones with speakers and then applies XTC again.

 

[Justdafactsmaam]

It looks like a dog biting its own tail.
XTC conceptually helps the spatial perception minimizing the intrinsic interference of spatial information of system with limited physical source points, which by definition cannot render spatiality in a totally effective way for our hearing.
All this is welcome. No one is against, conceptually speaking.
But in practice what is gained is not quantifiable in objective terms, nor, consequently, does it have a reference to aim for.

Yes it does. >25 db XTC

 

[Lion♡]

It's a picture that you can immediately understand why there are a lot of reviews saying that it was like a concert hall in the late XTC.
(Excluding the characteristics of large spaces, initial reflections from long distances, and late reflections)
The closer the distance is, the more the opposite ear constraints that define the spacing of the speakers are.
And it's the XTC that gets rid of that constraint, and it may feel like it's broadened, but it has nothing to do with the actual purpose and space.

 

[STC]

Yes it does. >25 db XTC

More than 25dB? at best you get that number at one or two specific frequencies.

 

[Davide]

Yes it does. >25 db XTC

I don't know where the number comes from but as an xtalk attenuation value it should be objectively valid to represent accuracy to the audio tracks. I don't know if it can be directly related to subjective preference regardless of the type of spatial information contained in it. It would be interesting to have more evidence. (obviously the fidelity paradigm is always at stake in this discourse).

XTC and spatiality are not related.

I suggest you to let go of the technicalities for a moment and look at the basic concepts.

 

[Lion♡]

I suggest you to let go of the technicalities for a moment and look at the basic concepts.

Maybe the words are a bit rude because I use a translator. Sorry if that was the case.
But it's hard to put XTC and defining space together. I think that creates more confusion.

It's a picture that you can immediately understand why there are a lot of reviews saying that it was like a concert hall in the late XTC.
(Excluding the characteristics of large spaces, initial reflections from long distances, and late reflections)
The closer the distance is, the more the opposite ear constraints that define the spacing of the speakers are.
And it's the XTC that gets rid of that constraint, and it may feel like it's broadened, but it has nothing to do with the actual purpose and space.

If you look at the picture that I attached, I think the intention of the question, "Would it be wrong if I listened to the mix in a few meters of space before at a certain distance and distance?"
We could say, of course, it's kind of an added effect depending on how you use the word, but it's not an addition, it's just a limitation of our hearing within a certain distance.

If you want the recordings to sound like you are actually in the same space then you need several more speakers for the spatial recreation. XTC is just a small step to bring the best from the conventional stereo format.

Agreed. Feels like supplementing more information about a particular area.
Fundamentally, you need information about each distance as well as the angle.
Nearby sounds need information on the near, far away sounds need information on the far away (but this is almost binaural sound source limited)
For example, when I listen to binaural ASMR (recorded by a YouTuber with a KU100 dummyhead), I hang XTC on both 90 degrees.
Normal headphones and IEMs sound only in the ears or on the auricles, but in XTC, they sound and feel as good as dummy heads, my body. Because I need that close information to stroke my head as the comb is when I brush my hair.

 

[STC]

No. It’s vastly superior with the BACCH. Again this is from personal experience

Superior to what? You are only comparing stereo or HT to BACCH. If you ask Carl ( I think he too is a member here) one University done comparisons with several XTC and his was declared the best.

Not an issue if your speakers are up to the task

If you have microphone, record close to the source and the same at the ears. listen to the recording through your speakers and you know what I am trying to say.

Not on my system.

I don’t disagree with you but that is still an opinion of yours.

 

[STC]

But it's hard to put XTC and defining space together. I think that creates more confusion.

Sometimes, you need to drag them to your place or go to their place to do the demo.

The problem with XTC is some making exaggerated claims which is beyond the scope of XTC.

End of the day, correct implementation is critical but sometimes I wonder if one deliberately setup poorly so that they can criticize. When it comes with stereo, we can have individuals perfecting placement for months but when it comes with XTC within few hours or days I get negative comments.