[Split]Basement technology - split from: Pear Cable CEO Calls James Randi's $1 Millio

[rsaavedra]

3D sound can be reproduced in earphones by using microphones imbedded in a sculpture mimicking the subjects head.

Yes, those are dummy heads, used to make binaural recordings, which are rather impressively realistic. They are intended to be heard from headphones. Check some legally downloadable binaural recordings on this sticky thread at Head-Fi.org.

Only problem with binaural recordings seem to be what regular stereo is best at: front imaging. Sounds coming from right in front aren't played back that well (as far as 3d location goes) in a binaural recording. Those sounds don't seem to come from several meters ahead right in front of you when listening to them on headphones, even from the best binaural recordings I've heard.

 

[Dan O.]

If you don't think that binaural recording present a perfect 3D rendition then your head is in the wrong shape.

 

[rsaavedra]

If you don't think that binaural recording present a perfect 3D rendition then your head is in the wrong shape.

Nope, mine has a normal homo sapiens' head shape. And the opinion isn't only mine, I've read quite a few observations about that, lack of perfect frontal imaging. Will try to find some and post some links.

 

[rsaavedra]

Many of the advantages of using MTB are similar to the binaural recording process: it is simple, it creates a realistic recording, it picks up a room’s natural reverb characteristics, it requires only 2 channels for playback, there are no problems with sweet spots, and it requires no special recording skills. On the downside, binaural playback requires headphones, there are elevation problems due to mismatched pinnae cues, the sound field moves if the microphone array shifted during recording, and there is front/back confusion as well as frontal collapse.

http://www.aessf.org/meetings/Reviews/2004_Reviews/Sept04_review.htm

The use of non-individualized HRTFs is limited by a lack of externalization (the sounds are localized in the head or very close to the head), incorrect perception of elevation angle, and front/back reversals. Externalization can be improved somewhat by adding dynamic head tracking and reverberation. Nevertheless, the lack of realistic externalization is often cited as a problem with these systems.

http://www.nae.edu/NAE/bridgecom.nsf/weblinks/MKEZ-695TMR?OpenDocument

 

[technoextreme]

Ok. Your point?

se

The equipment costs more than that house?

 

[Dan O.]

This is the problem when your head is the wrong shape.

The use of non-individualized HRTFs is limited...

You need to have your head reshaped to exactly match the dummy used for the recording. Then you will hear the perfect rendition (once your brain adapts to the change).

 

[rsaavedra]

Are you joking or just trolling? Noone needs the head reshaped. People could have the dummy head made after their head though.

In any case, where's your support to the claim that the 3D rendering is perfect in binaural recordings, in particular frontal and rear presentation? I presented my links, where are yours?

I also speak from personal experience; binaural recordings to me lack perfect frontal presentation. Do you speak from personal experience, so did you get your head reshaped then?

 

[Dan O.]

I speak from personal experience without the intervening recording and playback. The 3D rendering is excellent (up, down, front, back, center) using only 2 ears. If your experience doesn't agree then your hearing is defective and I can't help you further.

Try this experiment, with your eyes closed, have someone jingle some keys around you and see if you can describe where they are. Have them keep score. Then repeat the experiment while someone gently changes the shape of your ears.

Then tell me why you would expect to have perfect 3D rendering when listening through someone else's ears.

 

[rsaavedra]

And I see no links, so it was just your opinion as I knew all along.

 

[Dan O.]

The links were already posted in this thread before the split.

 

[rsaavedra]

The links were already posted in this thread before the split.

Right, because the original was about binaural after all I won't even waste my time looking for alleged links.

Binaural isn't perfect Dan, deal with it, stop advocating what you can't provide support for. Your posting your opinions and arbitrary statements online won't change the understanding and limitations of technologies that acoustic engineers have well studied for quite some time now. If you think current binaural technology perfectly recreates a 3D image, then you are simply deluding yourself.

 

[Dan O.]

I'm just trying to understand where our differences lay so they can be addressed. If you would just stand still for a moment we might be able to make some progress and resolve some of these differences.

What is your reasoning that binaural recording is not sufficient to exactly reproduce real world sounds? Most of us only have two ears. And the ear canal and working of the inner ear appear to only be capable of carrying a single time varying signal. If we reproduce the pressure wave function in the ear canal, why would it not be indistingushable from live sounds (not counting sub-sonics that don't bother following the air passage).

 

[Vic Vega]

The equipment costs more than that house?

You've obviously never looked at real estate in Newton, MA.

 

[rsaavedra]

What is your reasoning that binaural recording is not sufficient to exactly reproduce real world sounds? Most of us only have two ears. And the ear canal ...

Yes, good suggestion Dan. Let's stand still for a moment and consider... I'd like to consider why you'd like to argue and philosophise and try to elicit truth from your thoughts and opinions and reasoning alone, when there are known verifiable facts, like the ones indicated in my links. There are known shortcomings for binaural recordings and playbacks.

What is the basis for my reasoning? Those known shortcomings; factual information; assessments and knowledge backed by acoustic engineering papers (check the links again.) Plus the accounts by several other people who have tried binaural recordings; plus my own hearing.

Don't get me wrong, binaural sounds great, but it's simply not perfect front-image wise. Period. You won't improve one bit a technology by arguing or reasoning pointlessly about how/why it should be perfect when it simply isn't.

Now, what is the basis for your reasoning again? You'll simply start arguing a monologue trying to convince yourself out of reasoning alone? Nothing personal, but I might want to give a little more credit to what careful acoustic engineers have found and reported in their studies (which by the way is consistent with what my own ears tell me, and what several other people indicate.)

 

[Olowkow]

As I see it, the issue would be that when our two ears hear and locate a sound, it is by virtue of having an infinite number of point sources in 3 dimensions as data, with all the phase and amplitude information, to decode it. A spherical room with lots of small speakers distributed around it, and you sit in the middle. Then feed this with all the delays etc into separate channels. Now that's a stereo!

 

[Dan O.]

Don't get me wrong, binaural sounds great, but it's simply not perfect front-image wise. Period. You won't improve one bit a technology by arguing or reasoning pointlessly about how/why it should be perfect when it simply isn't.

Will you admit that there will be a difference between binaural sound recorded specifically for your ears and generic binaural sound recorded for the masses?

 

[rsaavedra]

Will you admit that there will be a difference between binaural sound recorded specifically for your ears and generic binaural sound recorded for the masses?

Yes there might be a difference. But I can't really say, because I've never heard a binaural recording made specifically for me. Have you heard a binaural recording made specifically for you?

We were talking about binaural recordings in general, as opposed to stereo recordings, not as opposed to individualized binaural recordings, which is just a special case of binaural recordings.

Binaural recording most people can listen to are the generic ones, those recorded using generic/averaged dummy heads, and at least those have the well known shortcomings already mentioned in previous posts.

PS. Also, I haven't read studies showing that individualized binaural recordings are "absolutely perfect" front-rear and elevation-wise, allowing perfectly natural 3D location of sound sources. They might be better than general binaural recordings, but I haven't seen studies indicating how much better or how "perfect" the individualized ones are.

 

[sol invictus]

PS. Also, I haven't read studies showing that individualized binaural recordings are "absolutely perfect" front-rear and elevation-wise, allowing perfectly natural 3D location of sound sources. They might be better than general binaural recordings, but I haven't seen studies indicating how much better or how "perfect" the individualized ones are.

They are not perfect, of course, but they're far better than anything else. Each person has what's called a HRTF - Head Related Transfer Function - which is unique. It depends on the shape of your ears, head, hairstyle, and probably other factors.

One technique to deal with this is to measure the HRTF for each individual, and then digitally process generic binaural recordings so they are customized for that individual. That way you don't have to make a new recording for each person. The US military has been experimenting with this, to try to provide soldiers wearing headsets while in action with an accurate 3D soundfield. For example when they hear a voice through the headset, they'll know immediately where it came from.

 

[Dan O.]

Yes there might be a difference. But I can't really say, because I've never heard a binaural recording made specifically for me. Have you heard a binaural recording made specifically for you?

Suddenly we are in a realm beyond our experience so we can all put away that pretense of knowing more than anyone else and explore this realm together.

In my earlier travels down this path, I've come across suggestions of how to test if simulated 3D sounds are indistinguishable from real sounds. But I haven't found the studies where this has been carried out. It isn't necessary either. For the simple enjoyment of recorded sound it isn't necessary to restore each instrument to it's precise geometric location but only to expand the sound stage so it isn't bunched up inside the listeners head or confined to two speakers.

We were talking about binaural recordings in general, as opposed to stereo recordings, not as opposed to individualized binaural recordings, which is just a special case of binaural recordings.

The individualized binaural recording is just a reference to show that the information necessary to represent the full 3D sound stage can be contained in a stereo recording. It isn't practical to distribute individualized recordings. But it may also not be necessary.

The brain is an amazing hierarchical learning correlation machine. It didn't come pre-wired for decoding the individual HRTF for locating sound sources in 3D. It had to learn through experience to do this and it has to continually relearn as the individual grows.

Binaural recording most people can listen to are the generic ones, those recorded using generic/averaged dummy heads, and at least those have the well known shortcomings already mentioned in previous posts.

Just as you can adapt to wearing glasses that distort your vision and eventually stop running into walls and tripping over shadows, wouldn't it be possible to learn to interpret the 3D relationship of sounds as heard through a different pair of ears. Your short experience to listening to binaural recordings is akin to trying on someone else's glasses. Your brain hasn't adapted to the new encoding.

I believe where the high-end stereophile has gone is to train their brain to decode the 3D information presented through their stereo system and listening room. However, where they think they are spending countless hours tweaking their system, they are actually tweaking the decoding in their brains. The end result though is what they are after; a better listening experience.

 

[Paulhoff]

I believe where the high-end stereophile has gone is to train their brain to decode the 3D information presented through their stereo system and listening room. However, where they think they are spending countless hours tweaking their system, they are actually tweaking the decoding in their brains. The end result though is what they are after; a better listening experience.

Very good, I think there is a lot to this, and I have said things like this to my wire believing friend.

Paul