http://www.theatlantic.com/magazine/archive/1969/12/what-perfection-sounds-like/8377/

Anybody familiar with this technology?

Odd, I thought headphones are really good at crosstalk cancellation (especially closed headphones) and that's the whole point of crossfeed, to add crosstalk and give a more natural feeling of soundstage. After all, real life has crosstalk, the delay between one and the other ear of a sound is how you can judge location... http://en.wikipedia.org/wiki/Sound_localization

Right now, to me it sounds like he's tripping balls.

It sounds to me like binaural recordings...but with speakers. Kind of curious how he did that.

I can't imagine it working without extremely precise room setup and positioning of the listener. Probably wouldn't lend itself well to lounging with buddies watching a movie. For binaural recordings in a dedicated audio room, I'd think the sweet spot would be not much bigger than the size of a human head.

It sure would be cool if it were more flexible though.

Edited February 16, 201115 yr by acidbasement

video with the prof

Sounds like he's using "reverse crossfeed" -- in other words, using crossfeed to cancel the sound from one speaker to the other ear, rather than what it usually does.

And yes, that video does a good job of explaining the exact answer to my question. Although he skipped "the third thing", which I suspect is frequency response.

I listened to this on my home system and the whole interview had a scarily centered image, anyone else experience this?

Edited February 16, 201115 yr by diebenkorn

I listened on headphones, which is just wrong.

My god man didn't you see the warning.

I did. Wasn't listening for examples of the technology, just wanted to hear the explanation. Will listen on speakers later.

The left speaker only contains cues for your left ear and the right speaker only contains cues for your right ear? That's so wrong. Maybe if you record everything with the binaural dummy head, but what about all the current music? Currently a 2 speaker setup is basically multiplexing however many original mics down to the two speakers, whereas if you do crosstalk cancellation it ends up being two non-ideal point sources at fixed locations, one for each ear. Again I go back to closed headphones as a great example of how normal recordings would sound with crosstalk cancellation.

So, they design crossfeed and all sort of tricks, digital or analogue (as in the Phonitor) to get your headphones sounding like speakers, then this guy designs a trick to get your speakers sounding like headphones, which supposedly improves 3D imaging. Amazing

So, they design crossfeed and all sort of tricks, digital or analogue (as in the Phonitor) to get your headphones sounding like speakers, then this guy designs a trick to get your speakers sounding like headphones, which supposedly improves 3D imaging. Amazing

Excellent point! Maybe he should call his headphone speakers the Oxymorons.

Edited February 17, 201115 yr by Voltron

The left speaker only contains cues for your left ear and the right speaker only contains cues for your right ear? That's so wrong. Maybe if you record everything with the binaural dummy head, but what about all the current music?

Exactly, I can't imagine he found a process to convert normal speaker imaging to binaural, then produce that from speakers.

And I also agree about the earlier point about narrow sweet spot -- I bet it's painfully small.

Also: Torpedo -- well put.

This is one of the things I love about my Polk SDA CRS speakers (Stereo Dimensional Array) which has the extra set of drivers to eliminate crosstalk by canceling out the left speaker signal reaching the right ear, and visa versa. It works beautifully.

I also once owned a Carver unit with a "Sonic Hologram" circuit that tried to do the same thing electronically with conventional speakers, and it also worked quite well.