stereo reproduction

[MJ Dijkstra]

Hello All

Imagine, you record a sound with a pair of microphones. The sound is closest to the right microphone. So the right microphone detects the sound first and after a time delay (X) the left microphone does.

Reproducing the sound with your speakers: The right speaker emits the sound first to your ears. Your right ear will be the first to detect the sound but after a short delay (y) the left ear receives the sound of the same right speaker.
After the delay X (caused by the microphone set-up) the left speaker emits the sound after both ears will receive the signal with a short delay (y).

Questions:

Am I right to conclude that there is an additional delay with normal stereoreproduction, not present in the original sound?

Does this degrade our imaging of sound (we all know we can pinpoint sound with good equipment so maybe our brains can compensate for these additional delays) ?

Does this point into the direction of ' beaming' speakers or headphones as the best way to reproduce stereo?

[Kurt von Kubik]

Originally Posted by MJ Dijkstra

Hello All

Imagine, you record a sound with a pair of microphones. The sound is closest to the right microphone. So the right microphone detects the sound first and after a time delay (X) the left microphone does.

Reproducing the sound with your speakers: The right speaker emits the sound first to your ears. Your right ear will be the first to detect the sound but after a short delay (y) the left ear receives the sound of the same right speaker.
After the delay X (caused by the microphone set-up) the left speaker emits the sound after both ears will receive the signal with a short delay (y).

Questions:

Am I right to conclude that there is an additional delay with normal stereoreproduction, not present in the original sound?

Does this degrade our imaging of sound (we all know we can pinpoint sound with good equipment so maybe our brains can compensate for these additional delays) ?

Does this point into the direction of ' beaming' speakers or headphones as the best way to reproduce stereo?

Yes you are!
But this delay is practically linear, so I would not worry to much about that.
As you say, the sound reaches one mike before the other, but that is not always so.
Time difference stereo is normally done with 2 omnidirectional mikes, which are spaced by i.e. 60 cm or more.
Coincidence stereo is done by 2 microphones placed as close as possible to the same point, but with cardioid mics.
There is also ORTF, Blumlein, DIN, bafled, NOS and even more technics to create stereo imaging, all of them working in different ways.

Time difference is the most natural way of recording, but is seldom used. Its name is AB stereo, and is the audio purists choise, but it works best with acoustic instruments, and real time recordings.

You can look at reproduced music in this way:
The point at which the acoustics stops, and the electronics starts, is precisely where the mics are placed. Reversely when playing back, it canbe said, that where the electronics stop, and the acoustics begin, is precisely at the acoustic center of your speaker. They will together try to recreate the two omnidirectional points where the mics were placed. Thus you get an impression of the room and instruments recorded.
One can not say that the instruments are brought into your room, but instead the room behind the speakers should open up into the recording venue.

Sometimes though you van hear "off axis" information coming from behind you, i.e. people walking on the stairs in the concerthall believing they cannot be heard, or other noises from the environment.

You can read about different recording technics at the major microphone manufacturers, such as DPA microphones, Shoeps and others.

[graaf]

Originally Posted by MJ Dijkstra

Hello All


(...)

The right speaker emits the sound first to your ears. Your right ear will be the first to detect the sound but after a short delay (y) the left ear receives the sound of the same right speaker.
After the delay X (caused by the microphone set-up) the left speaker emits the sound after both ears will receive the signal with a short delay (y).

(...)

Does this point into the direction of ' beaming' speakers or headphones as the best way to reproduce stereo?

yes, You are right, stereo basically sucks, and this delayed crosstalk is the main cause of this

so, what can be done?

well, read this first:
Ambiophonics 2nd Edition - Chapter 2

than the glorious four:
Favorite Threads and Posts in "Loudspeakers"

enjoy!

best,
graaf

[kstrain]

Following on from Ambiophonics, the OSD idea formulated by Takeuchi and Nelson is worth some study. There is a brief outline at

The Virtual Acoustics Project in the ISVR: Optimal Source Distribution

The key difference to ambiophonics is (from the above url)

" The inverse filters have flat frequency response so there is little colouration due to different head response nor at any location in the listening room, even outside the sweet area."

(With a simple stereo dipole, if a room-divider is used the signal is mono away from the sweet spot; if a cancellation (inverse) filter is used the response is very non-uniform away from the sweet spot.)

See my thread on the topic for outline details of a practical implementation.

Ken

[Pano]

That OSD is unique! Never seen that. Might be fun to try.
Has anyone here experimented with it?

[kstrain]

I tried it after reading a post by dwk123

Try Ambiophonics with your speakers

and my messing around led to the thread

Ambiophonic optimal source distribution experiment (part 1: Introduction)

which gets going in post #6 after a false start with the wrong filters (i.e. after reading the Takeuchi and Nelson work properly).

It surprises me that more people have not yet tried it.

Ken