BACCH4Mac "Absolute Sounds Product of the Year 2024"

[STC]

Yes. We're talking about the same thing.
Only a few angles worked in terms of enhancing the sense of envelopment (or in other words, adding information that was lacking).

The full information of venue acoustics will not be in the recordings. Such recordings would sound too muddy. Reverbs comes from all around the listeners by way of reflection of the original sound. They differ based on the point where the reflection takes place and I have not seen any one angles IR similar to another. perhaps, in a perfect hall that should be the case. so envelopment is something you do to recreate a venue matching the genre you are playing. The recordings could have been made in Sydney Opera house but what I try to reproduce with the IRs is for the recording to sound like playing in a concert hall with reverbs trails extending 2s or so. I only have St Cecilia impulse response so it may not sound like listening in Sydney Opera but at least some sense of “ We are there” in a concert hall.

Same story as above: the purpose was to enhance the sound in all directions, adding information about each angle, position, etc. And I already succeeded in my own way (from the length of the ear canal to more than 10 meters, all at the same time).

Wow! I hope you could make that available commercially.

As I mentioned before, the gap between an anechoic chamber-like condition and a normal room, except for the characteristics of the initial reflections and the spatial perception, was very small.

Not sure about that. A balloon burst in an anechoic chamber would sound just like a click. And with DCH, I use the click to sound like the balloon bursting in a concert hall or in room. The original click is left untouched. I don’t even use XTC for this demo as the burst supposed to come from one speaker only.

As I said before, even after XTC, there is still a very slight sense of what kind of space this is due to the initial reflections, but it's vague, so after that, you can overlay different angles of true stereo IR or personalized binaural IR or real BRIR (which is what I do) at different distances and different angles and play them back simultaneously.

This is too deep for me.

And because I'm using headphones or IEMs to play XTC, you might think it's a slightly different kind of thing, but it's really just BacchHP and C_Bacch's workflow.

Headphones/iem already without any crosstalk so by adding cancellation you are probably adding cross feed. I think I read in the forum some use that to create out of head or frontal perception. Anyway, this is a topic I am not familiar with. I hardly use iem except for testing occasionally.

Cheers!

 

[LIΟN]

Wow! I hope you could make that available commercially.

I'll polish it up a bit more and share it at some point (as a more generalized HRTF target, since this is the HRTF that works for me).

Headphones/iem already without any crosstalk so by adding cancellation you are probably adding cross feed. I think I read in the forum some use that to create out of head or frontal perception. Anyway, this is a topic I am not familiar with. I hardly use iem except for testing occasionally.

Replicate imagined speaker sound to headphones. (With Impulcifer)

Hi. I live in South Korea, and I enjoy exploring and studying various audio information from all over the world. I use a translator. I'm sorry the sentence wasn't smooth. I used the speaker set up in the in-room and was quite satisfied. But I had no choice but to dispose of the speaker because...

www.audiosciencereview.com

What I mean is that I (or some other users) did was to speakerize the IEM/headphones ---After this step, it's the same as the speaker.
And then apply XTC calibration to that speaker again. This is the speaker, not the crossfeed or IEM/headphones. just listening to it with IEMs/headphones.
BacchHP works the same way.

Btw, I'd love to see one comprehensive thread on XTC and stereo image and spatial cognition, not just Bacch.
I look forward to learning from and participating in discussions with talented people like you.

 

[STC]

What I mean is that I (or some other users) did was to speakerize the IEM/headphones ---After this step, it's the same as the speaker.

I see! this is something like

Btw, I'd love to see one comprehensive thread on XTC and stereo image and spatial cognition, not just Bacch.
I look forward to learning from and participating in discussions with talented people like you.

Ralph Glasgal dedicated his life doing research and implementing XTC. He was the first to write the algorithm for XTC and made it free for all. Search for his name in various forum where you can find everything you need to know about XTC and how we perceive stereo. I don’t think he wrote much here as many wasn’t receptive to XTC at that time.

 

[goat76]

Simply not universally true. The reason any hard panned sound can not extend beyond, behind or in front of the speaker is exactly because the crosstalk from that speaker masks the spatial cues that would allow it to extend beyond or in front of or behind the speaker. This is exactly why with the BACCH the object will come from any number of positions beyond, in front of, behind or even at the speaker position. It is allowing the listener to get the full effect of the spatial cues on the recording.

A hard-panned sound CAN of course extend behind the speaker with normal playback without crosstalk cancelation, how far behind the speaker a sound appears to be coming from depends on how far away the capturing microphone was placed from the recorded sound source. But as a hard-panned sound is an isolated mono sound emitting from just one of the speakers in a stereo setup, and there is nothing related to that particular sound played by the opposite speaker, there will not be anything that can move the sound object from the speaker's position on the horizontal plane.

Nope. The BACCH will in most cases move the sound away from the speaker with a hard panned sound. But it does NOT add a second or more sound. It simply cancels the crosstalk from the one speaker that is causing the opposite ear to identify the speaker as the location of the sound source

And how do you think that anything in the sound can be canceled out if there isn't a second related sound played by the opposite loudspeaker?

If a sound will be able to move from the speaker's position on the horizontal plane, at least a second speaker is needed to either pull the sound object towards the second speaker, or away from it, or even make the sound object appear to be coming from a point in front of the speaker. The more alike the second sound is to the first sound the more it will be pulled closer to the second speaker's position on the horizontal plane, but to make the sound appear to be coming from a point outside or in front of the speaker the second sound played by the opposite speaker must be manipulated in a way so that phase differences occur between the two signals.

A hard-panned sound is meant to be heard by both ears and the sound is meant to be heard from the position of the loudspeaker, just in the same way you would hear that particular sound object with both your ears if the real instrument was positioned at that same angle in front of you. it just happens to be the extreme position of the stereo field in mixing for normal stereo playback and the way the mixing engineer heard it while deciding to place the sound object right there in the mix.

But who knows how the sound engineer would have done it if she mixed it with a system with BACCH? I can tell you I would hardly ever want a guitar to appear to be coming from 30 degrees outside a normally positioned stereo setup with the speaker's setup with a 30-degree window. Instead, I would have likely panned the direct sound of the sound object to appear to be coming from about 30 degrees in front of the listener and used the extra 30 degrees width of the stereo field for reverb and/or recorded room sound of the guitar. But I say it again, the mixing engineer must hear what BACCH is doing to the particular mix, while mixing it, to be able to make the right decisions to reach the intended goal of how she wants the mix to sound. But with that said, it's of course completely okay if you as a listener prefer it to sound wider or more extreme, it's just not a more accurate reproduction of the mix.



Anyway, here is a hard-panned sound mixed for normal stereo playback, a sound that is supposed to be heard with both ears from the position of the left speaker. Notice that there is no information from the right channel.

In the picture below is the same hard-panned sound to the left channel but with the BACCH filter applied. BACCH is adding a strong second sound signal hard-panned to the right speaker to be able to cancel out the sound from the left speaker from reaching the right ear, but the problem is that the effect is way too strong and overshoots it, which will be very apparent when it comes to hard-panned sound objects which are supposed to blend acoustically and not sound isolated as they will do with BACCH. A hard-panned sound is simply dependent on crosstalk information to not sound too isolated, which is even more apparent when hearing such sound with headphones.

 

[STC]

A hard-panned sound CAN of course extend behind the speaker with normal playback without crosstalk cancelation,

Yes and it got to do how we interpreted distance based on loss of HF and reverbs.

But as a hard-panned sound is an isolated mono sound emitting from just one of the speakers in a stereo setup, and there is nothing related to that particular sound played by the opposite speaker, there will not be anything that can move the sound object from the speaker's position on the horizontal plane.

That’s so easy to do.

Crosstalk cancellation is dealing with sound arriving at you ears and how cancel out the unintended sound reaching the unintended ear in stereo.

 

[goat76]

Yes and it got to do how we interpreted distance based on loss of HF and reverbs.

Yes, and those things will also be true with the use of an actual real distance between the microphone and the sound source during the recording, the microphone will pick up a lower level of high frequencies, and the level of room reflections will end up with a higher ratio vs the direct sound waves from the sound generating source.

That’s so easy to do.

Yes, it's easy to make sounds appear to be coming from other places than the actual position of the speaker, you just need to manipulate it with phase shifts, which is the real reason why a hard-panned sound can appear to be coming from other positions than the actual position of the loudspeaker when using crosstalk cancelation.

Crosstalk cancellation is dealing with sound arriving at you ears and how cancel out the unintended sound reaching the unintended ear in stereo.

Yes, crosstalk cancellation deals with sounds that are unintended for the opposite ear, but that doesn't work when it comes to hard-panned sounds as they are dependent on the crosstalk between the speakers to sound as intended. That is the main problem with applying a crosstalk cancellation filter to different types of audio productions, as many modern music mixes contain a lot of hard-panned sounds that are dependent on crosstalk, otherwise, they will sound isolated in an unnatural way which in most cases was never the intention.

 

[STC]

Yes, crosstalk cancellation deals with sounds that are unintended for the opposite ear, but that doesn't work when it comes to hard-panned sounds as they are dependent on the crosstalk between the speakers to sound as intended. That is the main problem with applying a crosstalk cancellation filter to different types of audio productions, as many modern music mixes contain a lot of hard-panned sounds that are dependent on crosstalk, otherwise, they will sound isolated in an unnatural way which in most cases was never the intention.

The sound from a hard panned source will come from the speaker with XTC. So is that 30 degrees or 35 or 29 or 90 or 43 or 44 or 37 or 89 or 89.9 or 29.9 or etc etc. All you know is that it must come from the speaker BUT you do not know the exact placement of the speaker. When you do XTC, the shift of the image is slightly outward from the speaker. You can know exactly where it will emerge in space when you play the phantom sound and plug your other ear. So if your speaker is placed in an equilateral triangle it could be something like 35 or 40 degrees. But let’s say you are the recording,mixing and mastering engineer working with 60 degree separate monitors then you know the hard panned sound should emerge from 30 degree off centre. With XTC, you now position the speaker much closer so that now the hard panned sound emerge exactly from 30 degrees.

Having said that, this method not likely to produce true 3D enhancement. What you really want to achieve with XTC is to retrieve the encoded ITD and ILD without crosstalk and only those information delivered unhindered to the ears with crosstalk corruption for the placement where true 3D sound emerges. This is only possible when you place the speakers where our localization is poor and that will be directly Infront of you and humans are extremely sensitive to the direction of sound coming about 30 degree from the and XTC is less effective when speakers are placed at that angles.

 

[Justdafactsmaam]

A hard-panned sound CAN of course extend behind the speaker with normal playback without crosstalk cancelation,

Sorta kinda but not convincingly . The crosstalk anchors the image to the speaker at a certain threshold and partially or completely masks the spatial cues that would put the image away from the speaker

how far behind the speaker a sound appears to be coming from depends on how far away the capturing microphone was placed from the recorded sound source.

Not that simple. Spatial cues come in different flavors. But again the big problem is the crosstalk masks those other spatial cues

But as a hard-panned sound is an isolated mono sound emitting from just one of the speakers in a stereo setup,

“In theory there is no difference between theory and practice. In practice there is.”

Yogi Berra

As I said before in this thread out of the tens of thousands of commercial stereo recordings each took their own unique journey.

You don’t know what the artists or engineers heard so it fails as an objective refererence.

You don’t know their intentions unless they explicitly state them but since we can’t directly access their intentions they also can not serve as an objective reference

and there is nothing related to that particular sound played by the opposite speaker, there will not be anything that can move the sound object from the speaker's position on the horizontal plane.

Except when there is. You’re talking theory I’m talking practice
But again those qualities of real world stereo recordings are lost in the crosstalk

And how do you think that anything in the sound can be canceled out if there isn't a second related sound played by the opposite loudspeaker?

If a sound will be able to move from the speaker's position on the horizontal plane, at least a second speaker is needed to either pull the sound object towards the second speaker, or away from it, or even make the sound object appear to be coming from a point in front of the speaker. The more alike the second sound is to the first sound the more it will be pulled closer to the second speaker's position on the horizontal plane, but to make the sound appear to be coming from a point outside or in front of the speaker the second sound played by the opposite speaker must be manipulated in a way so that phase differences occur between the two signals.

That’s a technicality. The same effect can be achieved with just absorption. No added sound there. Same results only not good because it’s impractical.

A hard-panned sound is meant to be heard by both ears and the sound is meant to be heard from the position of the loudspeaker, just in the same way you would hear that particular sound object with both your ears if the real instrument was positioned at that same angle in front of you.

And that is something you continue to invent.
Unless explicitly stated you don’t know the intentions. Again Yogi knew what he was talking about.

it just happens to be the extreme position of the stereo field in mixing for normal stereo playback and the way the mixing engineer heard it while deciding to place the sound object right there in the mix.

No such thing as normal. Each recording took a unique journey. So all your arguments premised on “normal” or “what was heard” or “artists/engineers intent” fail on their face.

I can give you a number of classic examples of real world hard pans where the artists and/or engineers did not have the intent of landing the image on the speaker much less landing in on a speaker 30 degrees off center specifically

 

[STC]

Sorta kinda but not convincingly . The crosstalk anchors the image to the speaker at a certain threshold and partially or completely masks the spatial cues that would put the image away from the speaker

Sound from a single speaker is free from crosstalk and therefore it got nothing to do with crosstalk to localize where the sound is coming from as all cues of ILD and ITD correspond to what you hear in real sound. There is no stereo in real word sound. It is artificially created to somewhat render how the sound would be arranged in real space.

Not that simple. Spatial cues come in different flavors. But again the big problem is the crosstalk masks those other spatial cues

Which spatial cues you are referring to? If the recording have only two instruments and hard each panned then you are having a true 3D sound. As the cues reaching the ears will be matching real performance of two man standing at the same position of the speakers playing.

And that is something you continue to invent.
Unless explicitly stated you don’t know the intentions. Again Yogi knew what he was talking about.

I am sure Yogi would agree with goat76 on this.

 

[Justdafactsmaam]

Sound from a single speaker is free from crosstalk and therefore it got nothing to do with crosstalk to localize where the sound is coming from as all cues of ILD and ITD correspond to what you hear in real sound. There is no stereo in real word sound. It is artificially created to somewhat render how the sound would be arranged in real space.

Not if it’s a “left” speaker or a “right” speaker. When your left ear hears the right speaker and visa versa that’s crosstalk.

Let’s be careful not to conflate channels and speakers.

Which spatial cues you are referring to? If the recording have only two instruments and hard each panned then you are having a true 3D sound. As the cues reaching the ears will be matching real performance of two man standing at the same position of the speakers playing.

There will be other cues such as recorded reflections and reverb

I am sure Yogi would agree with goat76 on this.

I doubt it.

 

[STC]

Not if it’s a “left” speaker or a “right” speaker. When your left ear hears the right speaker and visa versa that’s crosstalk.

Let’s be careful not to conflate channels and speakers.

No. If you hear bell from 30 degrees to your left. the location is determined by pinna ( let’s ignore that for now), arrival time difference and level difference between left and right ears. that’s not interaural crosstalk errors that XTC meant to address. XTC meant to address the delivery of ITD and ILD in the recordings as each speaker supposed to deliver this information only for the respective ear but since sound can also travel to the other ear we now have contradicting information to the brain as the will be double ILD and ITD information presented to the ears. that what Iinteraural crosstalk is.

 

[Justdafactsmaam]

No. If you hear bell from 30 degrees to your left. the location is determined by pinna ( let’s ignore that for now), arrival time difference and level difference between left and right ears. that’s not interaural crosstalk errors that XTC meant to address. XTC meant to address the delivery of ITD and ILD in the recordings as each speaker supposed to deliver this information only for the respective ear but since sound can also travel to the other ear we now have contradicting information to the brain as the will be double ILD and ITD information presented to the ears. that what Iinteraural crosstalk is.

And that’s different from the left ear hearing the right speaker and the right ear hearing the left speaker how?

 

[STC]

And that’s different from the left ear hearing the right speaker and the right ear hearing the left speaker how?

A hard panned sound is from a single speaker. If this is a stereo recording then there will be crosstalk.

 

[Justdafactsmaam]

A hard panned sound is from a single speaker. If this is a stereo recording then there will be crosstalk.

Mostly. Not all media or stereo recordings have perfect channel separation.

Pay attention to Yogi

Also check out Abbott and Costello. This is starting to feel a lot like Who’s on first

 

[STC]

Mostly. Not all media or stereo recordings have perfect channel separation.

Pay attention to Yogi

Also check out Abbott and Costello. This is starting to feel a lot like Who’s on first

How many do Channel separation is needed? Here we are talking about hard panned sound which will be mono and come out from one speaker.

 

[Justdafactsmaam]

How many do Channel separation is needed?

30 db is all you need.

Who’s on first?

Here we are talking about hard panned sound which will be mono and come out from one speaker.

You can hard pan and still have a lot signal coming out of the other channel.

What’s on second?

 

[STC]

You can hard pan and still have a lot signal coming out of the other channel.

No.

 

[Justdafactsmaam]

No.

Then there was no such thing as a hard pan in stereo playback until CDs. None of those hard pans people talk about pre-digital recording were actual hard pans.

Which clearly isn’t true.

Go ask Yogi about it.

 

[goat76]

Then there was no such thing as a hard pan in stereo playback until CDs. None of those hard pans people talk about pre-digital recording were actual hard pans.

Which clearly isn’t true.

Go ask Yogi about it.

In the beginning of stereo recordings, everything was either hard-panned or positioned right in the middle of the mix as the mixing consoles only had three pan pot positions, L-C-R. Many mixing engineers still mix this way, Andrew Schepps is one of them who mostly mixes this way for stereo productions.

 

[tmtomh]

In the beginning of stereo recordings, everything was either hard-panned or positioned right in the middle of the mix as the mixing consoles only had three pan pot positions, L-C-R.

I had no idea that was the case! Interesting - it explains a lot about those early stereo mixes.