BACCH4Mac Pro Edition - For those considering BACCH

[kthulhutu]

@Theoretica Appl. Physics quick question for you guys.
In normal stereo listening, there is a dip at roughly 2khz in the phantom center vs using a real center speaker due to acoustic crosstalk.

Since BACCH is attenuating crosstalk and using a mono correction algorithm for the phantom center, does it also reduce/eliminate this colouration? Could this be one of the reasons why I prefer the phantom center with BACCH on?

 

[Theoretica Appl. Physics]

 

[Vapor9]

I recently purchased the BACCH4Mac software only version and I am very pleased.

For years I've run a Carver C-9 unit between my preamp and amp with generally good results. However, the Carver is 1980's technology and it was easy to hear it shortfalls on certain recordings with a kind of 'phase-ness' or slight frequency error. I never did think about taking it out of the loop however. I think I'm particularly sensitive to a good soundstage (or my system/room is very good at creating one). Its flaws were overlooked compared to its benefits. Also, my C-9 has been modified with upgraded power supply, new caps and a frequency compensation circuit (Gundry).

The BACCH software has taken the C-9 benefits through the roof without any side effects. The soundstage is just very natural, but individual players and voices are spread vividly across it. I'm surprised mostly about the added clarity that is given to soft vocals buried within a mix. I can much more easily understand the words in certain recordings. I'm sure this is due to the cancellation effect where my brain is less able to hear sound coming from two speakers and it is fooled into hearing it at one place in space.

The BACCH is certainly a keeper and I'm very happy that Professor Choueiri has used his knowledge to tackle this fundamental stereo problem.

 

[Dialectic]

I recently purchased the BACCH4Mac software only version and I am very pleased.

For years I've run a Carver C-9 unit between my preamp and amp with generally good results. However, the Carver is 1980's technology and it was easy to hear it shortfalls on certain recordings with a kind of 'phase-ness' or slight frequency error. I never did think about taking it out of the loop however. I think I'm particularly sensitive to a good soundstage (or my system/room is very good at creating one). Its flaws were overlooked compared to its benefits. Also, my C-9 has been modified with upgraded power supply, new caps and a frequency compensation circuit (Gundry).

The BACCH software has taken the C-9 benefits through the roof without any side effects. The soundstage is just very natural, but individual players and voices are spread vividly across it. I'm surprised mostly about the added clarity that is given to soft vocals buried within a mix. I can much more easily understand the words in certain recordings. I'm sure this is due to the cancellation effect where my brain is less able to hear sound coming from two speakers and it is fooled into hearing it at one place in space.

The BACCH is certainly a keeper and I'm very happy that Professor Choueiri has used his knowledge to tackle this fundamental stereo problem.

The greater specificity of imaging is the biggest benefit in my view.

 

[srrxr71]

In all seriousness can I just put a thin acoustic panel at my nose extending outward?

The software price is not bad but there is no upgrade path to some of the features in the next edition up.

Also just as a question are not the recordings mixed with crosstalk when they are produced?

 

[Dialectic]

In all seriousness can I just put a thin acoustic panel at my nose extending outward?

The software price is not bad but there is no upgrade path to some of the features in the next edition up.

Also just as a question are not the recordings mixed with crosstalk when they are produced?

Crosstalk in the recording chain is desirable. The crosstalk that BACCH eliminates is the crosstalk resulting from the juxtaposition of your two speakers in a room, attenuating the sound from the left speaker hitting your right ear and vice versa.

 

[srrxr71]

Crosstalk in the recording chain is desirable. The crosstalk that BACCH eliminates is the crosstalk resulting from the juxtaposition of your two speakers in a room, attenuating the sound from the left speaker hitting your right ear and vice versa.

I see the term crosstalk cancellation actually means crosstalk recovery?

The one thing in the middle is my nose. Otherwise my room is treated to the hilt.

I’ve also noticed that aiming the speakers at one’s nose sounds very good. Makes the sweet spot somewhat smaller but great within it.

 

[Dialectic]

I see the term crosstalk cancellation actually means crosstalk recovery?

The one thing in the middle is my nose. Otherwise my room is treated to the hilt.

I’ve also noticed that aiming the speakers at one’s nose sounds very good. Makes the sweet spot somewhat smaller but great within it.

I wouldn't call it crosstalk recovery, but by eliminating the spurious crosstalk in your listening room, BACCH processing does seem to "recover" all manner of spatial cues that are otherwise masked in playback.

 

[onion]

In all seriousness can I just put a thin acoustic panel at my nose extending outward?

The software price is not bad but there is no upgrade path to some of the features in the next edition up.

Also just as a question are not the recordings mixed with crosstalk when they are produced?

I think there is an upgrade path so you only pay the price difference from the edition you're upgrading from - it may be worth asking the Theoretica customer support team.
Thin acoustic panel probably inadequate - if you stick a mattress from your nose extending towards the mid-point between the speakers, that will replicate the effect.

 

[srrxr71]

I think there is an upgrade path so you only pay the price difference from the edition you're upgrading from - it may be worth asking the Theoretica customer support team.
Thin acoustic panel probably inadequate - if you stick a mattress from your nose extending towards the mid-point between the speakers, that will replicate the effect.

I mean that’s probably the effect the software may create but really is blocking down to 20Hz necessary? Possibly 150Hz. A 4” panel could handle that but the air gap makes it 6”. I have a bunch of them around me so I can always experiment with that to get a taste before I spend money on software.

I am glad the M1 Mac is supported so that helps.

There is a lot to read which I have not yet. Seems like I can use my existing apogee duet and a regular UMIK mic. That helps a lot.

Yes I think the team is the best to contact about these questions. Also the webcam integration. It’s just that from $980 the next jump is $5k and I don’t know how much of that is the RME interface.

 

[phoenixdogfan]

I mean that’s probably the effect the software may create but really is blocking down to 20Hz necessary? Possibly 150Hz. A 4” panel could handle that but the air gap makes it 6”. I have a bunch of them around me so I can always experiment with that to get a taste before I spend money on software.

I am glad the M1 Mac is supported so that helps.

There is a lot to read which I have not yet. Seems like I can use my existing apogee duet and a regular UMIK mic. That helps a lot.

Yes I think the team is the best to contact about these questions. Also the webcam integration. It’s just that from $980 the next jump is $5k and I don’t know how much of that is the RME interface.

I contacted them. The Babyface is around $600-650. They will take that much off. Me, I have a Motu M4, so if I bite, I'll just use that. I would start with the Basic, and work up to the Audiophile which is about another $3500. I already have a Smyth A16, so I would not go for the headphone unit. The top of the line upgrade is more for recordists and other content creators, I think. I will first have to see if I can get the Mac 1 Mini to assume the role of my regular HTPC from one of my Window's boxes. Primarily, I think the Mac will run JRiver and Dirac just fine, so all I need is PEQ (which I think is readily available) and a software based crossover which is, I think, a much harder nut to crack.

Also, I need to be sold a little further. Not completely sure of what it's doing, and whether it's really recovering information on the recording lost in the playback chain, or whether it's just creating a sound effect which dazzles initially by being different from what listeners are accustomed to hearing. Obviously, if it's the latter, it won't wear well for the long term. So yeah, I need further information.

Certainly would be helpful if it could be reviewed by ASR. Also why is a custom HRTF needed for this application? I can see why it's needed for the Smyth, but I'm not certain why it's needed for crosstalk cancellation, as well as the degree of benefit it confers, considering it's eye watering premium.

 

[srrxr71]

I contacted them. The Babyface is around $600-650. They will take that much off. Me, I have a Motu M4, so if I bite, I'll just use that. I would start with the Basic, and work up to the Audiophile which is about another $3500. I already have a Smyth A16, so I would not go for the headphone unit. The top of the line upgrade is more for recordists and other content creators, I think. I will first have to see if I can get the Mac 1 Mini to assume the role of my regular HTPC from one of my Window's boxes. Primarily, I think the Mac will run JRiver and Dirac just fine, so all I need is PEQ (which I think is readily available) and a software based crossover which is, I think, a much harder nut to crack.

Also, I need to be sold a little further. Not completely sure of what it's doing, and whether it's really recovering information on the recording lost in the playback chain, or whether it's just creating a sound effect which dazzles initially by being different from what listeners are accustomed to hearing. Obviously, if it's the latter, it won't wear well for the long term. So yeah, I need further information.

Certainly would be helpful if it could be reviewed by ASR. Also why is a custom HRTF needed for this application? I can see why it's needed for the Smyth, but I'm not certain why it's needed for crosstalk cancellation, as well as the degree of benefit it confers, considering it's eye watering premium.

Thank you for that information. Yes a lot to think about.

So I can perhaps understand why a custom hrtf is needed. Let’s say a speaker on your right is supposed to send some signal to your left ear it has to cross your nose and face. This is different from having a speaker in front of you.

I suppose it adds something to the left speaker to compensate for that. That compensation being based on your face shape.

If both left and right are adding this filter to both channels then is it just basically a pEQ with perhaps some minuscule delay?

I may come across as ignorant here but doesn’t the mixing artist in a sense compensate the mix for his/her face/ears? Which then translates to the listener as the listener hears everything through their own face/ears?

Anyway I’ll just say I would need further information myself. In the meantime I have that acoustic panel I can hold up to my face to try this. No doubt the image will be wider and probably envelop you all the way to the side of your face but what reference does that target? I can see this as a way to test binaural recordings on speakers. With emphasis on test. So many things we can test to see if they improve our enjoyment reference or not.

I am reading from some what seems to be it gets rid of all cross talk and others say only the “bad” crosstalk. So it’s very hard to say what it’s doing. Seems like a black box that does stuff to change the soundstage and some people will like it. At $5k i’m going to want to know exactly what it’s doing. What parameters are adjustable. How exactly it’s cutting out some of the crosstalk and which parts and to what extent and in what frequency ranges.

One other thing that comes to mind is simply showering the sides and back areas with L-R difference signal. It was pleasant sounding when I tested it once. Again whatever floats a person’s boat I guess.

 

[phoenixdogfan]

Thank you for that information. Yes a lot to think about.

So I can perhaps understand why a custom hrtf is needed. Let’s say a speaker on your right is supposed to send some signal to your left ear it has to cross your nose and face. This is different from having a speaker in front of you.

I suppose it adds something to the left speaker to compensate for that. That compensation being based on your face shape.

If both left and right are adding this filter to both channels then is it just basically a pEQ with perhaps some minuscule delay?

I may come across as ignorant here but doesn’t the mixing artist in a sense compensate the mix for his/her face/ears? Which then translates to the listener as the listener hears everything through their own face/ears?

Anyway I’ll just say I would need further information myself. In the meantime I have that acoustic panel I can hold up to my face to try this. No doubt the image will be wider and probably envelop you all the way to the side of your face but what reference does that target? I can see this as a way to test binaural recordings on speakers. With emphasis on test. So many things we can test to see if they improve our enjoyment reference or not.

I am reading from some what seems to be it gets rid of all cross talk and others say only the “bad” crosstalk. So it’s very hard to say what it’s doing. Seems like a black box that does stuff to change the soundstage and some people will like it. At $5k i’m going to want to know exactly what it’s doing. What parameters are adjustable. How exactly it’s cutting out some of the crosstalk and which parts and to what extent and in what frequency ranges.

One other thing that comes to mind is simply showering the sides and back areas with L-R difference signal. It was pleasant sounding when I tested it once. Again whatever floats a person’s boat I guess.

Well, if it's getting rid of all crosstalk, isn't it just turning your speakers into headphones? If that's all its doing, seems to me you could get that selfsame effect for cheaper by buying a quality headphone.

 

[srrxr71]

Well, if it's getting rid of all crosstalk, isn't it just turning your speakers into headphones? If that's all its doing, seems to me you could get that selfsame effect for cheaper by buying a quality headphone.

This one in particular: https://www.headfonia.com/old-champ-the-akg-k1000/

Sort of. I heard since it tilts backward it creates a soundstage behind your head instead in front. Quite a fly in that ointment.

I heard the BMR headphone used in the valve index headset is probably the modem version of a contactless headphone. I should test that for music use.

 

[phoenixdogfan]

"So I can perhaps understand why a custom hrtf is needed. Let’s say a speaker on your right is supposed to send some signal to your left ear it has to cross your nose and face. This is different from having a speaker in front of you.

I suppose it adds something to the left speaker to compensate for that. That compensation being based on your face shape."

I can understand your head may add something to this equation, but I would think much more of this relates to the room, the wave launch and dispersion characteristics of your speaker. the distance of the listener from the sound source, etc. Ultimately, I would think in this instance the head is no more important than it would be for, say, a Dirac Live measurement. And Dirac doesn't bother with it at all. So I question exactly how much it adds, given all the other room related reflections, speaker interactions, etc that need to be filtered before the sound even arrives at the head to undergo a transfer function. And doesn't live music undergo a HRTF that isn't corrected for? So why would this be necessary, unless it was to tailor the sound to a specific formatted effect?

It's a little different with the Smyth Realizer which uses the HRTF to specifically match the sound of the headphones to the speaker playing in a specific listening space. Smyth is looking to create an exact copy of something and removing the head by measuring how the speakers sound right at the ear canal. Baach is just trying to remove the effects of cross talk in a room, most of which, I would think, comes from the room.

 

[srrxr71]

Well you can consider a fully treated room. I don’t think this product really deals with the room but I don’t know.

Other products like Dirac and Trinnov deal with the room.

Honestly to the best of my knowledge what this is doing is exactly the opposite of your smyth realizer. It is a reverse HRTF and probably not much more.

Considering what you went through to get the holy grail of headphone listening a proper HRTF I would have a hard time understanding why you would now pay $5k to do the opposite with actual speakers.


It’s like when they tell you it’s like a mattress between your ears it’s probably pretty much that. Nobody said it’s like mattresses all over your walls.

It all really seems like a reverse HRTF. Adding out of phase opposite side signal behind the ear. Protecting the opposite side ear from crosstalk. Again whatever floats peoples’ boats I guess.



BTW as an aside i can’t really afford to go down the Smyth Realizer path right now but I respect the concept. It’s holy grail for headphones if it is realistic enough.

I am however getting a 1 year license for Genelec Aural ID which is a software version of HRTF. You have to send in a 360 degree photo(s) of your head and shoulders to get your custom HRTF.

 

[phoenixdogfan]

Well you can consider a fully treated room. I don’t think this product really deals with the room but I don’t know.

Other products like Dirac and Trinnov deal with the room.

Honestly to the best of my knowledge what this is doing is exactly the opposite of your smyth realizer. It is a reverse HRTF and probably not much more.

Considering what you went through to get the holy grail of headphone listening a proper HRTF I would have a hard time understanding why you would now pay $5k to do the opposite with actual speakers.


It’s like when they tell you it’s like a mattress between your ears it’s probably pretty much that. Nobody said it’s like mattresses all over your walls.

It all really seems like a reverse HRTF. Adding out of phase opposite side signal behind the ear. Protecting the opposite side ear from crosstalk. Again whatever floats peoples’ boats I guess.



BTW as an aside i can’t really afford to go down the Smyth Realizer path right now but I respect the concept. It’s holy grail for headphones if it is realistic enough.

I am however getting a 1 year license for Genelec Aural ID which is a software version of HRTF. You have to send in a 360 degree photo(s) of your head and shoulders to get your custom HRTF.

I've made that observation on more than one occasion that this seeks to eliminate cross talk while the Smyth seeks to add it. The only answer I can think of, is Theoretica is claiming there exists some ideal level of which headphones have too little and speakers have too much. If so, how is that "ideal level" established, how is it measured, and where does there exist psycho acoustic research to support this conclusion?

 

[srrxr71]

I've made that observation on more than one occasion that this seeks to eliminate cross talk while the Smyth seeks to add it. The only answer I can think of, is Theoretica is claiming there exists some ideal level of which headphones have too little and speakers have too much. If so, how is that "ideal level" established, how is it measured, and where does there exist psycho acoustic research to support this conclusion?

I appreciate what this company is doing. That ability to steer the sweet spot and/or create multiple sweet spots is cool. Its advances the science.

This cross talk elimination product seems cool. It adds a “slider” (probably a bunch of settings) that let you tune to taste.

It gets into that territory of “I added X and my music sounds so much better now”

That X could be anything and if I mentioned specifically some examples of what could be X it would ruffle some feathers.

I suppose they would have to fund the ABX studies to produce this psychoacoustic research. After all the Harmon tests were all done by testing on real people. Until then it is all just opinion.

Over on the “what makes speakers disappear” thread we are discussing techniques to create L-R to feed into the left speaker and R-L to feed into the right speaker and then L+R to feed to the center speaker. This would amount to yet another “slider” that makes things sound “better” while obviating the need for HRTF measurements but obviously needing an extra speaker. The slider part would be to adjust the levels of the 3 units.

Not the exact same thing but somewhat in the same vein I guess. There are other such experiments in that thread as well.

 

[Dialectic]

It would perhaps be a better use of energy to try to hear a BACCH system than to speculate endlessly on what BACCH does. Hearing BACCH is not like "hearing" a power cable.

 

[srrxr71]

It would perhaps be a better use of energy to try to hear a BACCH system than to speculate endlessly on what BACCH does. Hearing BACCH is not like "hearing" a power cable.

Of course not. Will they demo the software?