The bestiary of the most common (and debunked by science) audiophile myths

[Fidji]

I'm that case, I can see why you don't want to know about BACCH!

I don't find it worthwhile to engage with the wilfully ignorant, so I wish you well.

Good manners are in short supply lately, unfortunately

So I will be bit more straightforward - heard BAcCh in its iteration with camera on very expensive system. After initial “wow” you realize, that it is just crosstalk cancelation. Headphones are better in achieving the same. Guy that owns it stopped using it after couple of months.

But I fully trust, that it can impress those with limited experience re surround. And it is better than plain 2ch for casual listening.

 

[muza_1]

"Uhm, single dynamic driver IEMs cannot possibly do what multiple drivers do it's not fiscally posible" yeah right.

 

[Audiofire]

Audio power station in your home (PS Audio per se).

 

[Blockader]

It is true. I have experienced it first hand


I’m sorry but none of these tests involved the BACCH SP. Old studies sometimes become dated by new technology. It’s a thing.

Dr. Toole’s book is not a bible. It’s high time some folks on ASR come to grips with that.

Crosstalk cancellation is great, it eliminates the acoustic crosstalk problem of the stereo. But you what you hear is still a stereo recording which lacks the envelopment and spaciousness of a multichannel setup. I have tried BACCH too, it works with some recordings very well but it can't come close to a large immersive system even the designer of BACCH, Dr. Edgar Choueiri admits that.

For those who don't know what BACCH does:

BACCH transforms speakers into binaural sources through acoustic crosstalk cancellation, effectively turning them into something like headphones. Acoustic crosstalk is both a blessing and a curse in stereo systems, and BACCH is a solution to fix that without having *downsides. It can potentially enable two speakers to simulate any kind of recording by feeding the left and right ears with the correct interaural level differences (ILDs) and interaural time differences (ITDs), mirroring the way music would be heard in, for example, a concert hall. Before we delve a bit deeper, it's important to understand what ITD and ILD mean. We localize sounds based on the level and timing differences between our two ears. Stereo sound is perceived as stereo because the information meant for one ear partially reaches the other ear as well, creating a "phantom" center channel. However, this phantom center channel can also be achieved by ensuring that the ILD and ITD are equal. BACCH requires a measurement of the listener's HRTF to function because it relies on the differences between the two ears, it must gather detailed information about the listener's ear shape and size.(the transfer function of it in other words)

But here's the problem. Using BACCH on a stereo recording doesn't make a big difference usually.(even if it does, sometimes it doesn't feel *natural*). Potentially BACCH can make speakers simulate a concert hall but if the recording was stereo to start with, it can't do that. It can't add something that wasn't there. I think BACCH can work best if the recording was made specifically for BACCH/binaural recorded. Yes, it can make the soundstage larger than the room, yes it improves the clarity, yes it makes the imaging feel a bit more precise. But sometimes it just makes recording sound weird too. It all depends on the recording.

 

[Fidji]

I would add two audiophile mythical creatures:

- wife, able to hear smallest adjustment, while cooking in the kitchen
- old friend, violin player, confirming whatever myth.

 

[krabapple]

He can’t talk about it. People involved in developing proprietary technologies can’t talk about some things.

I know.

But I don’t need to talk to anyone. Your assertion that “two-channel, stereo, which will surely be obsolete within a few decades.” Is demonstrably false.

With two channel stereo we can achieve 100% spatial accuracy with the right recording.

We can achieve spatial proximity in any direction from inches away to extreme distances with some existing 2 channel commercial recordings.

We can achieve remarkable spatial impact with the vast majority of legacy stereo recordings.

And we can even decode existing and future multi channel recordings and play them back with remarkable perceptual accuracy over just two channels.

You might want to talk to JJ about all these claims too.

And btw, if you think the demo that will be clinging to two-speaker (and no, I don't mean headphones) playback in 2050 will be embracing BACCH ....I salute your faith.

 

[Emlin]

That stereo means two channels when it actually means 3D.

 

[DSJR]

It's also how Linn defined source , it was the table itself not pickup or arm , because that was what they manufactured back in the day what a surprise ...

No .. deck,then arm,then cartridge!

I do wish some would read exactly what I said...

I later heard a turntable NOT Linn which clearly out-performed it and cost less too, but I'm going into territory alien to most of you and maybe it's for the best as it's a territory I've long ago left...

 

[phantompowered]

Tubes are superior audiophile worthiness, because transistors were made to reduce manufacturing costs, with transistors, you lose analog warmth

Well, in a concession to tube lovers everywhere, this isn't patently wrong. That is, for a fact, why FETs were invented, but they're just not great at their job insofar as they trade cost and size for a bunch of problems tubes don't have to worry about.

 

[Blockader]

"Uhm, single dynamic driver IEMs cannot possibly do what multiple drivers do it's not fiscally posible" yeah right.

Sort of.

In conditions similar to a vacuum chamber, sound transducers show a roll-off of -12dB per octave above their resonance frequency. Earphones, which operate under conditions similar to those in a vacuum chamber, if they have a single driver, it has a resonance frequency around 3000Hz. As a result, iems with one driver can struggle to extend high frequencies above this threshold. (usually they put a solo driver which is very capable of delivering information above 3000hz and use acoustic filters around mids to tame its energy down afaik. Sennheiser does that a lot. )

To put it simply, if you place a driver with a flat response inside an earphone, its response would show the 12dB/octave roll-off beyond the 3000Hz cutoff frequency. This is a simplification, of course, intended to illustrate a point. To illustrate this point, I attached this image below. Assume the cutoff frequency is 3000hz and that's how a flat response driver would roll off if placed inside an iem case in ideal conditions:

You can definitely tweak a single driver to get a pretty even response across the board, using acoustic filters for extended bass and treble, and keeping the mids smooth. But if you really want smoothness and clarity up top, namely above 10kHz, a HF dedicated driver is necessary. Take the Samsung Buds2 Pro as an example – even with all the fancy DSP tech they use, they still have 2 drivers, despite with DSP everything being possible.

An IEM with 2 drivers can measure great on test fixtures, you can see their target curve fit Harman curve nicely, however when you do not have perfect seal, using the same 2 iem driver example, the bass response should roll off dramatically below the drivers resonance frequency(3000hz~) and your iems which measured like they have a lot of bass will sound bass light. Achieving a perfect seal with IEMs is close to impossible. This phenomenon I explained this paragraph is called as leakage tolerance and 2 driver iems have usually bad leakage tolerance. Amir doesn't take leakage tolerance measurements but thereare few people who takes such measurements(Oratory and headphonetestlab)
To counteract this issue, manufacturers often add an additional subwoofer driver dedicated solely to bass frequencies which has good leakage tolerance. (better than a mid driver)

In short 3 drivers are ideal, 2 drivers are good enough. One driver + DSP or One driver + careful tuning can come close to perfection too but it can't really be top tier. I can't see the point of using more than 3 drivers. It reduces the R&D costs of making a very capable HF / bass driver probably.

tl;dr, iems operate in vacuum chamber conditions, that brings limitations in HF extension and bass leakage tolerance. That's one of the reasons why we see iems with a lot of drivers and headphones with just one driver. Headphones operate under vacuum chamber conditions up to 150hz~ whereas iems up to 3000hz~.

 

[Justdafactsmaam]

Crosstalk cancellation is great, it eliminates the acoustic crosstalk problem of the stereo. But you what you hear is still a stereo recording which lacks the envelopment and spaciousness of a multichannel setup.

That is simply not true. It is recording dependent. With the right recording you get full envelopment.

have tried BACCH too, it works with some recordings very well but it can't come close to a large immersive system

If memory serves me you auditioned the entry level version. Not adequate, not nearly adequate to evaluate the system.

even the designer of BACCH, Dr. Edgar Choueiri admits that.

On the contrary, Edgar clearly and explicitly states the BACCH can take a multichannel recording and convert it to a two channel recording that will be indistinguishable from multichannel playback through the BACCH in two channels.

For those who doesn't know what BACCH does:

BACCH transforms speakers into binaural sources through acoustic crosstalk cancellation, effectively turning them into something like headphones.

Total misrepresentation. The only thing you got right there was the use of cross talk cancellation

Acoustic crosstalk is both a blessing and a curse in stereo systems,

Cross talk is an added distortion that creates conflicting spatial cues in the playback. Whether or not one considers it to be a blessing or curse is a matter of personal taste.

and BACCH is a solution to fix that without having *downsides. It enables two speakers to simulate any kind of recording by feeding the left and right ears with the correct interaural level differences (ILDs) and interaural time differences (ITDs), mirroring the way music would be heard in, for example, a concert hall. Before we delve a bit deeper, it's important to understand what ITD and ILD mean. We localize sounds based on the level and timing differences between our two ears. Stereo sound is perceived as stereo because the information meant for one ear partially reaches the other ear as well, creating a "phantom" center channel. However, this phantom center channel can also be achieved by ensuring that the ILD and ITD are equal. BACCH requires a measurement of the listener's HRTF to function because it relies on the differences between the two ears, it must gather detailed information about the listener's ear shape and size.(the transfer function of it in other words)

That part is accurate.

But here's the problem. Using BACCH on a stereo recording doesn't make a big difference usually.

That is ridiculously untrue.

(Even if it does, sometimes it doesn't feel *natural*).

That is subjective

Potentially BACCH can make speakers simulate a concert hall but if the recording was stereo to start with, it can't do that. It can't add something that wasn't there.

But those spatial cues are there on many stereo recordings. Doesn’t have to be binaural

think BACCH can work best if the recording was made specifically for BACCH/binaural recorded.

As is the case with multichannel

Yes, it can make the soundstage larger than the room, yes it improves the clarity, yes it makes the imaging feel a bit more precise.

The horrors!

But sometimes it just makes recording sound weird too. It all depends on the recording.

“Weird” is subjective. No doubt it will expand the soundstage of studio recordings, which are unnatural soundstages to begin with in ways that can be shocking. If one does not like that result with a specific recording one does have two options. Dial down the BACCH SP until the desired result is achieved or simply bypass the filter.

For these same studio recordings mixed in stereo what exactly is the superior multichannel option?

It’s a trick question given that the BACCH can simulate any multichannel option with the multichannel mix

 

[Blockader]

That is simply not true. It is recording dependent. With the right recording you get full envelopment.


If memory serves me you auditioned the entry level version. Not adequate, not nearly adequate to evaluate the system.



On the contrary, Edgar clearly and explicitly states the BACCH can take a multichannel recording and convert it to a two channel recording that will be indistinguishable from multichannel playback through the BACCH in two channels.



Total misrepresentation. The only thing you got right there was the use of cross talk cancellation



Cross talk is an added distortion that creates conflicting spatial cues in the playback. Whether or not one considers it to be a blessing or curse is a matter of personal taste.


That part is accurate.


That is ridiculously untrue.

That is subjective


But those spatial cues are there on many stereo recordings. Doesn’t have to be binaural


As is the case with multichannel



The horrors!


“Weird” is subjective. No doubt it will expand the soundstage of studio recordings, which are unnatural soundstages to begin with in ways that can be shocking. If one does not like that result with a specific recording one does have two options. Dial down the BACCH SP until the desired result is achieved or simply bypass the filter.

For these same studio recordings mixed in stereo what exactly is the superior multichannel option?

It’s a trick question given that the BACCH can simulate any multichannel option with the multichannel mix

You agreed with everything I said positive about it and you think I'm wrong based on your subjective conjecture when I say something *bad* about BACCH.
I do not think this conversation is going anywhere. Enjoy your BACCH, good for you. I decided to go with immersive setups though.

You can't call BACCH stereo by the way. You were saying that stereo can capture spatial details completely but BACCH is binaural. Stereo minus acoustic crosstalk = Binaural.
That's how it works.

 

[Justdafactsmaam]

I know

If you knew JJ couldn’t talk about it then asking me to ask him was just obnoxious and pointless

You might want to talk to JJ about all these claims too.

Again since he can’t discuss these things asking me to talk to him about them is just obnoxious and pointless.

And again I don’t need to talk to anyone about the performance of the BACCH SP. it is quite demonstrable. And testable. Funny how some folks here on a science based forum feel free to make assertions of facts about something for which they have zero objective data or actual experience.

If anyone wants to do some blind listening tests I’m all for it.

And btw, if you think the demo that will be clinging to two-speaker (and no, I don't mean headphones) playback in 2050 will be embracing BACCH ....I salute your faith.

I’ll cross that bridge when I get to it. The goal is to just get to it.

I wonder how you felt in the 90s about the odds of vinyl sales surpassing CD in 2023?
I bet you would have called it absurd.

We don’t know the future. And it’s not a question of faith.

But your misrepresentation of current facts was inexcusable.

 

[egellings]

I forgot to include this one in mine too.
While I was quite convinced that it actually made sense on the speakers, I always wondered, even without anyone pointing it out to me, what the heck should improve over time in a cable that carries electrical data

Do used cables need to be rebroken in by a new owner?

 

[Audiofire]

Well, in a concession to tube lovers everywhere, this isn't patently wrong.

"Half the Truth is often a great Lie." (Benjamin Franklin, 1758)

 

[Emlin]

Do used cables need to be rebroken in by a new owner?

Only if they are used in the opposite direction.

 

[Justdafactsmaam]

You agreed with everything I said positive about it and you think I'm wrong based on your subjective conjecture when I say something *bad* about BACCH.

You are getting some things right and some things wrong. I’m just pointing them out both ways.

I do not think this conversation is going anywhere. Enjoy your BACCH, good for you. I decided to go with immersive setups though.

That’s fine. Different strokes for different folks

You can't call BACCH stereo by the way.

The BACCH is a cross talk cancellation DSP that works with 2 channel stereo recordings. That’s what it is and that is what I have always represented it to be.

You were saying that stereo can capture spatial details completely but BACCH is binaural. Stereo minus acoustic crosstalk = Binaural.
That's how it works.

Binaural recordings are a subset of two channel stereo recordings.

Multichannel recordings are a subset of stereo recordings.

It’s all stereo unless it’s mono

 

[Talisman]

Only if they are used in the opposite direction.

You made me laugh

 

[Doodski]

So I would place “we have 2 ears, so 2 channels are enough” up there with cables and directionality of fuses.

2 ch will never go away I think because of the portability of it and some rooms are simply not conducive to surround sound and stereo is the way to go.

 

[Chrispy]

Back to the starting with an excellent source thing a bit....I always took that to be a high quality recording rather than a particular media player....