BACCH4Mac Pro Edition - For those considering BACCH

[srrxr71]

Actually the way BACCH works you can have speakers in non symmetric locations and it will still be fine. It is not super specific like using sonic holography. The in ear mics allow the system to adjust for speaker position or the intro edition one just inputs measured the relative speaker locations or does it by ear. Don’t quote me please but I’m pretty sure Dr Choueri said he puts the speakers in a traditional configuration as it makes audiophiles happy. I also imagine that being in the optimal listen widow for each speaker helps(especially ESL’s).

I have actually used BACCH on an NAD VISO-1(if you know what that is you will laugh) while it did not create a true binaural experience the sound field was well defined several feet out from each side of the device. It sounded like a nice set of bookshelves with excellent imaging.

When the Jambox Bluetooth speaker was around it implemented a fixed version of BACCH which was called LIVE AUDIO and actually was pretty good especially considering the modest quality of the speaker.

Any of this may be wrong.

I feel like the ear mike version. The consumption version needs to be priced as such. Creators can pay $5 $7k $15k even. A nice mic as I learned on gearspace is $7k and according to some professional creators not a big deal in terms of cost. It can tax deducted etc.

Imho both head tracking and ear mic calibration would do well to be priced reasonably to increase the take up.

Education versions with limited file capability for experimentation would be very nice. Such to get used to any workflows associated in creation activities.

Even for consumption instead of a trial perhaps a version that only works on certain files but should have representation from a variety of genres and recording techniques.

It’s one thing to do research but it’s another to help people understand the value of said research.

 

[mcdonalk]

Let's be polite everyone. This is not how we discuss things here.

"This" is how I find things discussed throughout this forum more than any other audio forum that I visit.

 

[Gwreck]

I feel like the ear mike version. The consumption version needs to be priced as such. Creators can pay $5 $7k $15k even. A nice mic as I learned on gearspace is $7k and according to some professional creators not a big deal in terms of cost. It can tax deducted etc.

Imho both head tracking and ear mic calibration would do well to be priced reasonably to increase the take up.

Education versions with limited file capability for experimentation would be very nice. Such to get used to any workflows associated in creation activities.

Even for consumption instead of a trial perhaps a version that only works on certain files but should have representation from a variety of genres and recording techniques.

It’s one thing to do research but it’s another to help people understand the value of

"This" is how I find things discussed throughout this forum more than any other audio forum that I visit.

I feel a bit embarrassed. As after reading some several year old posts from this thread seems to be a trap for noobs who are BACCH fans to be mercilessly trolled.

 

[Dialectic]

I feel a bit embarrassed. As after reading some several year old posts from this thread seems to be a trap for noobs who are BACCH fans to be mercilessly trolled.

I got much more aggressive trolling about BACCH on WBF, before I disappeared there.

I love the Toole book, but on ASR, I find a misplaced certainty about the Toole/Harman conventional wisdom. The Harman approach to system building has a low performance ceiling.

 

[srrxr71]

Haha seems like a pro Harmon vs pro Princeton theoretica faction war in the works.

Just kidding.

It’s pretty simple. If 95% or even even 70% of my listening were of free space recordings I’d pretty much have to pay up. In this endeavor you have to utilize the highest known science for your playback pleasure.

If your listening pattern is more like mine then it’s understandable why that price is hard to justify.

I also think that you cannot call your system “end game” if you don’t have the ability to maximally “decode” all microphone arrangements available in our recordings.

I hope both factions can co exist. Hehe

 

[Sir Sanders Zingmore]

Haha seems like a pro Harmon vs pro Princeton theoretica faction war in the works.

Just kidding.

It’s pretty simple. If 95% or even even 70% of my listening were of free space recordings I’d pretty much have to pay up. In this endeavor you have to utilize the highest known science for your playback pleasure.

If your listening pattern is more like mine then it’s understandable why that price is hard to justify.

I also think that you cannot call your system “end game” if you don’t have the ability to maximally “decode” all microphone arrangements available in our recordings.

I hope both factions can co exist. Hehe

I pay very close to zero attention to the microphone arrangement of my recordings. My subjective experience is that in some cases BACCH dramatically improves the listening experience. In most cases it slightly improves it (well worth the entry level price in my opinion) and in a small number of cases I really can't tell much of a difference. There is not a single recording I can think of that is made worse.
Again, all subjective so feel free to do with that "information" what you will

 

[Cars-N-Cans]

The microphone techniques used to record binaural are completely different from what was used to make 99% of recorded music. It’s the sad truth.

I have my own binaural system (that uses passive techniques in the nearfield), and it works very well with all recordings. For best results both ILDs and ITDs need to be present. With, say, some jazz recordings that only have one mic per instrument/musician the imaging is still nice and each instrument is separate, but its still very much 2D since there is just panning alone, but with a more spacious sound. With good conventional stereo recordings there is excellent immersion and realism, tho. The only caveat for me is that there are no elevation cues since its directly from the recording, but the azimuthal position and distance are very accurate and realistic with good quality classical recordings. Basically the recording provides the ILDs and ITDs, and my presence in front of the speakers means these are converted to HRTFs and HRIRs. If the BACCH system also includes some form of individualized HRTFs or something similar it would be even better than what I can manage. As far as binaural recordings done with dummy heads go, these DO NOT work as the dummy's HRTFs get applied along with your own, so it causes severe distortion of the stereo image. But that is not a bad thing as it means pretty much any stereo recording outside of very old ones with hard-panning work. With hard-panning, the instruments end up at the extreme left and right positions. For mono the image ends up being a nearly 180 degree arc around the listening position. With proper downmixing, 5.1 also works with dialog being in the center and the front and surround forming an immersive image that extends out to the left and right extreme. But as far as stereo goes everything has the same imaging as what I get with headphones and Sennheiser's Ambeo recordings. I would say in that regard these are worth it if you can afford such as system and have the time to set it up properly.

 

[Cars-N-Cans]

Since a few people on this forum have asked the question “How does BACCH work?” we refer the technically-minded reader to the detailed theory behind BACCH presented in Chapter 5 of the textbook Immersive Sound: The Art and Science of Binaural and Multi-Channel Audio, Published by Focal Press. (Neither Theoretica, nor Prof. Choueiri, get any proceeds from the sales of this book.)

An excerpt of that chapter (with pages omitted for copyright reasons) can be downloaded from Theoretica’s web server. In particular, pages 160-164 describe the actual BACCH filter design method, and should answer some of the questions asked on this forum.

We refer the non-technical reader to Theoretica’s FAQ page on BACCH.

Regards,
Adam @ Thoeretica

I have a question, if you don't mind. Maybe this would get into some of the more proprietary aspects of the system, but does the processing do anything beyond applying the XTC filters (and head tracking)? Just curious if there is anything outside of the XTC that is done to enhance the imaging. I know that in-ear mics are used, but I was wondering if they did anything beyond measuring the impulse response.

 

[Gwreck]

I have my own binaural system (that uses passive techniques in the nearfield), and it works very well with all recordings. For best results both ILDs and ITDs need to be present. With, say, some jazz recordings that only have one mic per instrument/musician the imaging is still nice and each instrument is separate, but its still very much 2D since there is just panning alone, but with a more spacious sound. With good conventional stereo recordings there is excellent immersion and realism, tho. The only caveat for me is that there are no elevation cues since its directly from the recording, but the azimuthal position and distance are very accurate and realistic with good quality classical recordings. Basically the recording provides the ILDs and ITDs, and my presence in front of the speakers means these are converted to HRTFs and HRIRs. If the BACCH system also includes some form of individualized HRTFs or something similar it would be even better than what I can manage. As far as binaural recordings done with dummy heads go, these DO NOT work as the dummy's HRTFs get applied along with your own, so it causes severe distortion of the stereo image. But that is not a bad thing as it means pretty much any stereo recording outside of very old ones with hard-panning work. With hard-panning, the instruments end up at the extreme left and right positions. For mono the image ends up being a nearly 180 degree arc around the listening position. With proper downmixing, 5.1 also works with dialog being in the center and the front and surround forming an immersive image that extends out to the left and right extreme. But as far as stereo goes everything has the same imaging as what I get with headphones and Sennheiser's Ambeo recordings. I would say in that regard these are worth it if you can afford such as system and have the time to set it up properly.

The microphone measurements don’t really add any HRTF to the sound in BACCH so dummy head recording ms work well. In one version of BACCH there is a 3D mixer that uses HRTF to position a sound in 3D space. If you have your own HRTF it works better. Ideally instead of dummy head recordings one would use an ambisonic microphone array and then translate it to one’s own HRTF for listening. This is what the 3D mixer can do.

 

[Cars-N-Cans]

The microphone measurements don’t really add any HRTF to the sound in BACCH so dummy head recording ms work well. In one version of BACCH there is a 3D mixer that uses HRTF to position a sound in 3D space. If you have your own HRTF it works better. Ideally instead of dummy head recordings one would use an ambisonic microphone array and then translate it to one’s own HRTF for listening. This is what the 3D mixer can do.

Thanks. Thats what I was wondering. Is that the LiveAudio, by chance? Re: The speakers and HRTFs they need to be modified to take into account the users own HRTFs in my experience. The listers own HRTFs are being applied implicitly since the speakers radiate sound out into the room and not into the ears directly like headphones. Using speakers means you are applying them twice. But I think the recommended spacing for the speakers in the conventional BACCH XTC is something like 10-15 degrees, so the broadening from the extra HRTFs will probably enhance the imaging. For me it causes everything in the recording to be ~2x further apart than it should be, giving a “funhouse mirror” effect since the speakers are separated by about 75 degrees.

 

[Gwreck]

Thanks. Thats what I was wondering. Is that the LiveAudio, by chance? Re: The speakers and HRTFs they need to be modified to take into account the users own HRTFs in my experience. The listers own HRTFs are being applied implicitly since the speakers radiate sound out into the room and not into the ears directly like headphones. Using speakers means you are applying them twice. But I think the recommended spacing for the speakers in the conventional BACCH XTC is something like 10-15 degrees, so the broadening from the extra HRTFs will probably enhance the imaging. For me it causes everything in the recording to be ~2x further apart than it should be, giving a “funhouse mirror” effect since the speakers are separated by about 75 degrees.

Live audio was the licensed version of BACCH used for the Jawbone Jambox. I have read Dr Choueri state that BACCH does not require any special speaker placement. The headphones measure to get an impulse response of the speakers but not and HRTF. The 3D mixer part of BACCH is what requires an HRTF. It is a separate module of the program that is not really intended for audiophiles. I have my BACCH set up to use the 3D mixer to play back 5.1 audio with 2 speakers. I do not have a a personalized HRTF currently so the playback is not perfect but works pretty well.

 

[Gwreck]

Haha seems like a pro Harmon vs pro Princeton theoretica faction war in the works.

Just kidding.

It’s pretty simple. If 95% or even even 70% of my listening were of free space recordings I’d pretty much have to pay up. In this endeavor you have to utilize the highest known science for your playback pleasure.

If your listening pattern is more like mine then it’s understandable why that price is hard to justify.

I also think that you cannot call your system “end game” if you don’t have the ability to maximally “decode” all microphone arrangements available in our recordings.

I hope both factions can co exist. Hehe

As far as I have read and experienced BACCH works with all stereo recordings and the less artificial the recordings type the less artificial the playback on BACCH. Traditional stereo isn’t any different in that respect. The ideal recording for BACCH and sound field would be a binaural recording using your own head or ambisonic recording translated with your personalized HRFT.

 

[Cars-N-Cans]

Live audio was the licensed version of BACCH used for the Jawbone Jambox. I have read Dr Choueri state that BACCH does not require any special speaker placement. The headphones measure to get an impulse response of the speakers but not and HRTF. The 3D mixer part of BACCH is what requires an HRTF. It is a separate module of the program that is not really intended for audiophiles. I have my BACCH set up to use the 3D mixer to play back 5.1 audio with 2 speakers. I do not have a a personalized HRTF currently so the playback is not perfect but works pretty well.

Ok. Thanks for the clarification.

FWIW I was referring to his commentary in “Immersive Sound” that the preferred placement is around 10 degrees as it gives the largest sweet spot. But as you say it’s not a requirement. Regarding the HRTFs I have seen others in the audio field have the same comments about applying full HRTFs to speaker playback not working properly, but that is all dependent on the HRTFs that have been applied. If they are favorable when combined with one’s own then it will work.

In regards to the speakers, when you place them you are in essence selecting one particular set of your own HRTFs since their position is fixed with respect to you. I know in my system it was a balance between getting a strong enough head shadow, and correct imaging. The binaural effect starts at about 60 degrees of separation in the near-field, but too far and the image becomes too wide and eventually a hole forms in the middle of the soundstage due to the reduced spectral energy that is perceived there. That is my experience and understanding, anyway.

 

[Gwreck]

Ok. Thanks for the clarification.

FWIW I was referring to his commentary in “Immersive Sound” that the preferred placement is around 10 degrees as it gives the largest sweet spot. But as you say it’s not a requirement. Regarding the HRTFs I have seen others in the audio field have the same comments about applying full HRTFs to speaker playback not working properly, but that is all dependent on the HRTFs that have been applied. If they are favorable when combined with one’s own then it will work.

In regards to the speakers, when you place them you are in essence selecting one particular set of your own HRTFs since their position is fixed with respect to you. I know in my system it was a balance between getting a strong enough head shadow, and correct imaging. The binaural effect starts at about 60 degrees of separation in the near-field, but too far and the image becomes too wide and eventually a hole forms in the middle of the soundstage due to the reduced spectral energy that is perceived there. That is my experience and understanding, anyway.

That makes sense about the hole inthe sound field. I think BACCH is just not as picky as regular stereo. I had not heard the 10 degree recommendation before but it makes sense as other systems such as ambiophonics wanted a dipole for best performance( but I have not idea if it’s the same reasons).

 

[sweetsounds]

As an owner of the BACCH4mac I can definitely say that there is a significant difference between the intro and the audiophile version of BACCH. The audiophile version of BACCH include a babyface pro and the measurement microphones which is a fair bit of the cost.

I am a potential customer.
I am using Acourate meanwhile with very good success. Even Acourate was challenging to integrate into my system with Logitech Media Server.

General Concerns:
Price: BACCH-SP is overpriced for me, clearly looking to position it as a luxury item. So "BACCH4MAC-Audiophile+" and a good mic would be the right option, even though the price is also extreme exceeding 7k$.
Inflexibility: BACCH seems to be very much tied to Mac, the specific mics, the Babyface. Concerns, that it is hard to loop into an existing setup.
Performance: how well works the head tracking and other features.
Uncertain result: would be good to be able to try BACCH before purchase.


There are open questions before considering a plunge:
- how important is the quality of the binaural mic for the result? What is the spec difference between the mics BACCH-BM and BACCH-BM pro?
$2900 for the Pro alone is hefty and there is no measurement data on the quality: its SNR of 67dB doesn't sound superb.

- Is there data, how BACCH alters the frequency response over the spectrum? I hear, that it is almost tonality neutral, but see little data.

- Can BACCH be combined with room correction and how? Is DRC affecting BACCH negatively?

- Can I combine Acourate Filters into the Mac Mini and have Coax -> Mac/BACCH+DRC -> Coax?
If not combinable: Would one do "Source -> DRC -> BACCH -> DAC" or rather "Source -> BACCH -> DRC -> DAC"?

 

[Gwreck]

I am a potential customer.
I am using Acourate meanwhile with very good success. Even Acourate was challenging to integrate into my system with Logitech Media Server.

General Concerns:
Price: BACCH-SP is overpriced for me, clearly looking to position it as a luxury item. So "BACCH4MAC-Audiophile+" and a good mic would be the right option, even though the price is also extreme exceeding 7k$.
Inflexibility: BACCH seems to be very much tied to Mac, the specific mics, the Babyface. Concerns, that it is hard to loop into an existing setup.
Performance: how well works the head tracking and other features.
Uncertain result: would be good to be able to try BACCH before purchase.


There are open questions before considering a plunge:
- how important is the quality of the binaural mic for the result? What is the spec difference between the mics BACCH-BM and BACCH-BM pro?
$2900 for the Pro alone is hefty and there is no measurement data on the quality: its SNR of 67dB doesn't sound superb.

- Is there data, how BACCH alters the frequency response over the spectrum? I hear, that it is almost tonality neutral, but see little data.

- Can BACCH be combined with room correction and how? Is DRC affecting BACCH negatively?

- Can I combine Acourate Filters into the Mac Mini and have Coax -> Mac/BACCH+DRC -> Coax?
If not combinable: Would one do "Source -> DRC -> BACCH -> DAC" or rather "Source -> BACCH -> DRC -> DAC"?

The standard binaural microphones that come with it are more than adequate and it I have never seen anyone say otherwise.

Regarding frequency response the whole point of BACCH is to have crosstalk cancelation that does not color the audio. There are not any changes to frequency response that o can hear on my system.

I have combined BACCH with dirac and it seems to be fine but I prefer it without.

I do not know about accurate. But the BACCH can send the signal to any DAC and the babyface is mainly for microphone measurements.

 

[Sir Sanders Zingmore]

Uncertain result: would be good to be able to try BACCH before purchase.

You can try before purchase

 

[Cars-N-Cans]

- Is there data, how BACCH alters the frequency response over the spectrum? I hear, that it is almost tonality neutral, but see little data.

The XTC filters should be pretty neutral, but one thing I have not seen discussed (or maybe I missed it?) is how sensitive it is to room reflections. Yes the XTC filters will eliminate crosstalk, but the reflections, if they are closely spaced to the direct sound and not sufficiently attenuated, can have a similar effect of interfering with the spatial cues. I’d imagine at least some form of mitigation for the first 10-15 ms of reflections from the side walls and such would be advisable.

 

[Dialectic]

The XTC filters should be pretty neutral, but one thing I have not seen discussed (or maybe I missed it?) is how sensitive it is to room reflections. Yes the XTC filters will eliminate crosstalk, but the reflections, if they are closely spaced to the direct sound, can have a similar effect of interfering with the spatial cues. I’d imagine at least some form of mitigation for the first 10-15 ms of reflections from the side walls and such would be advisable.

It is sensitive to room reflections. With apologies to advocates of diffusion, absorbers are advised.

 

[phoenixdogfan]

The standard binaural microphones that come with it are more than adequate and it I have never seen anyone say otherwise.

Regarding frequency response the whole point of BACCH is to have crosstalk cancelation that does not color the audio. There are not any changes to frequency response that o can hear on my system.

I have combined BACCH with dirac and it seems to be fine but I prefer it without.

I do not know about accurate. But the BACCH can send the signal to any DAC and the babyface is mainly for microphone measurements.

I do not think either accourate or Audiolense work in a Mac environment. Dirac does.