Hello!
Analog cross-talk canceller:
Phase delay enhances 3D audio - Maxim
The circuit component values are for a handheld device dimensions, but should be implementable to Ambiophonics system as well.
- Elias
Analog cross-talk canceller:
Phase delay enhances 3D audio - Maxim
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
The circuit component values are for a handheld device dimensions, but should be implementable to Ambiophonics system as well.
- Elias
Some skeptics still find some of this hard to believe, but closely spaced speakers interact with the room more uniformly because they are positioned close to the same point in space within the room. You have to always be thinking of the speaker and room as a complete system. Therefore, once you restore the controlled crosstalk of closely spaced speakers, things are ok. Yes it does enhance the width, I measured before and after with a Goniometer and it definitely shows the signal going to the speakers has a wider angle than before if it was mastered to have width.
Mono fed into an ambiophonics processor either by one channel or both will still be mono, no big surprise here folks, that is how it should be. Stereo will still be stereo but wider with a more focused center image. That is the beauty of this, it's compatible with both.
Mono fed into an ambiophonics processor either by one channel or both will still be mono, no big surprise here folks, that is how it should be. Stereo will still be stereo but wider with a more focused center image. That is the beauty of this, it's compatible with both.
Last edited:
agreed! please take a look:
http://www.diyaudio.com/forums/multi-way/10962-stereolith-loudspeakers-question.html
agreed! please take a look:
http://www.diyaudio.com/forums/multi-way/121385-loudspeakers-room-system.html
ambiophonics is great but it is not the only way outside conventional stereo bermuda triangle
I have "Tried Ambiophonics with my speakers." Interesting, for sure. But I still prefer the old fashioned spread speakers.
My impression of it was that it gives you a "real" center image and phantom left-right, as opposed to the opposite of that in normal stereo. It's OK, but I was not crazy about it.
Try as I might, I could never get the 180° sound stage. Maybe 80~100° maximum - and I can get that with a normal setup. Usually it was much less. But it did work.
Could be that my speakers are just too big, or otherwise not suited to Ambiophonics. Fun experiment, tho.
My impression of it was that it gives you a "real" center image and phantom left-right, as opposed to the opposite of that in normal stereo. It's OK, but I was not crazy about it.
Try as I might, I could never get the 180° sound stage. Maybe 80~100° maximum - and I can get that with a normal setup. Usually it was much less. But it did work.
Could be that my speakers are just too big, or otherwise not suited to Ambiophonics. Fun experiment, tho.
Did you move your speakers to about one third the stereo spacing? What electronics or program were you using? The type of speakers should make no difference unless they are not symmetrical or out of balance.
Usually, the comment I get is that the stage is too wide and some listeners even hear things a bit to the rear (both easily correctable).
Ralph Glasgal
Usually, the comment I get is that the stage is too wide and some listeners even hear things a bit to the rear (both easily correctable).
Ralph Glasgal
hi,
I experimented with width controllers,
Have a look this one's ez to build and it works.
Stereo Width Controllers
I experimented with width controllers,
Have a look this one's ez to build and it works.
Stereo Width Controllers
another way is to use a mono center channel. Some use it for car audio by tying the grounds of the front speakers together then connect it to the positive of the center speaker then the center speaker's negative back to the negative on the amp (or head unit) and parallel the center speaker with a volume pot to make the center the same volume (if only speaker impedances were like resistors !).
The problem with normal stereo are the ambiguous pinna localization cues for the phantom center. They do not match the cues of a real sound source in the center. In Ambiphonics the center pinna localization cues are presented correctly whereas the cues for left and right are wrong.
When adding a center that derives its signal from the left and right channel signal, even more ambiguous pinna localization cues get generated. The best solution is to have 3 channels (left, center, right) and phantom sources between L-C and R-C. Feels like someone else proposed such a system before
When adding a center that derives its signal from the left and right channel signal, even more ambiguous pinna localization cues get generated. The best solution is to have 3 channels (left, center, right) and phantom sources between L-C and R-C. Feels like someone else proposed such a system before
Most acoustic or even virtual recordings are just two channel and normally have a reaonable amount of Interaural Level Differences and Interaural Time Differences unless a recording engineer has really destroyed this data. But the stereo triangle cannot reproduce the ITD and ILD that is actually on the disc properly. Note that the stereo loudspeaker effect is like an optical illusion it is not like everyday binaural hearing.
Consider, that in 60 degree stereo you barely have a good center image and that the localization is limited to the 60 degree angle and is mostly nonlinear. Now take three speakers and have a recording that is really three channel not just mono dialog in the middle. You are now facing the center speaker and trying to hear a stage extending over say the left 30 degrees. Try this with your stereo system. Put two speakers 30 degrees apart, face one of them and play your best recording. I don't think you will like this much. This is why 5.1 recordings normally do not use the center speaker for anything other than a mono signal like dialog or a soloist. Again, the stereo illusion only works over a limited angle and requires that the head be facing forward.
Ambiophonics in contrast is a loudspeaker binaural technology and mimics normal everyday hearing. Ambisonics and Wavefield Synthesis are other such technologies. Of course there is also earphone binaural.
I am not quite sure what you mean by overly long delays in recordings. But ITD delays up to 700 microseconds are needed to localize properly. Delays in the tens of milliseconds are heard as early reflections and give a sense of space not localization. Most recordings do pick up studio or concert hall reflections of this type. In advanced Ambiophonics this is not a problem. In stereo if too much such ambience comes from the front speakers one hears the sewer effect, so most recording engineers avoid picking up or concocting too much reverb.
Ralph Glasgal
Home Page
I'm not sure why left-center imaging would be any different from center-right imaging (which equals facing the left speaker in your example)?
Maybe Ralph is trying to say that 'phantom' source localisation occurs only with speaker arrangement where the head is facing between the speakers. So with regular L-R speakers there is a perceived phantom image but with L-C-R speaker arrangement you can't perceive phantom image so well between L-C and C-R. Is that correct Ralph? If so do you have any further reference about this?
Yes this is a very good way of saying it. As far as I know this basic attribute of stereo or 5.1 has never appeared in print in these words, but you can easily do the experiment yourself. Note that 5.1 was promulgated for movies, and movie theaters where five channel mono sounds great and people sit everywhere. It was never meant to used so extensively for music in the home.
Ralph Glasgal
Hi Ralph, thanks for asking.
Yes, I did move the speakers to the recommended +/-12°, which put them about a foot or two apart. (they're big, 30" wide by 60" tall)
I used the Java version and the VST plugin. The results were very similar. I wanted to try the electro-music version, but their ordering page seems not to be on a secure server.
The plugins did give me very good phantom images to the left and right of the speakers, right past the side walls. But I get that with my normal setup anyway, which is about +/-25°. My normal setup has a very strong, stable central image. Ambiophonics did give a somewhat better central image, right on the speakers. I felt there was not as even a spread across, tho. Like more lumped into L,C,R without the smooth spread of my normal setup.
My thought was that the radiating surface of my speakers is too large for the effect to fully develop. Perhaps point source speakers would do a better job. However, looking at the photos of the Ambiophonics "lab" I see rather wide panel speakers, so maybe my hypothesis is all wet.
I do hope to try this with smaller speakers, maybe even in a bigger space. I'll try some Meyer Sound UPA powered speakers or the QSC K8. I'll let you know how it goes. Dragging around 18ft3 speakers with a horn on top is no fun.
I'd like to add some thoughts to the quote below.
The overwhelming majority of recordings uses close miking. In the mixing process equalization and interchannel level and time delays are based on the needs of common 60° stereophony. Eberhard Sengpiel (recording engineer and university lecturer) and others have shown (here and here) that the interchannel levels and delays neceassary to place a phantom source at a specific location are higher than the interaural time and level differences found in listening to natural sound sources. Recordings made for 60° stereo are and have to be spatially and timbrally distorted when compared with listening to natural sounds.
Ambisonics and Wavefield Synthesis are trying to recreate the soundfield at the ears but Ambiophonics is more like reversed stereo. It presents only one correct pinna localization cue to the ear, the cue for sound sources coming from the front. Any other phantom sound to the left and to the right is presented with incorrect pinna localization cues. Stereophony presents two correct pinna localization cues for monophonic sounds in the left and right loudspeaker. All other phantom sounds in between are presented with incorrect pinna localization cues. I'm not sure which scenario is preferable.
Both techniques, Ambisonics and stereophony, are single seat solutions. A wider sweet spot delivering high quality sound to more than just a single person would be favorable.
The overwhelming majority of recordings uses close miking. In the mixing process equalization and interchannel level and time delays are based on the needs of common 60° stereophony. Eberhard Sengpiel (recording engineer and university lecturer) and others have shown (here and here) that the interchannel levels and delays neceassary to place a phantom source at a specific location are higher than the interaural time and level differences found in listening to natural sound sources. Recordings made for 60° stereo are and have to be spatially and timbrally distorted when compared with listening to natural sounds.
Ambisonics and Wavefield Synthesis are trying to recreate the soundfield at the ears but Ambiophonics is more like reversed stereo. It presents only one correct pinna localization cue to the ear, the cue for sound sources coming from the front. Any other phantom sound to the left and to the right is presented with incorrect pinna localization cues. Stereophony presents two correct pinna localization cues for monophonic sounds in the left and right loudspeaker. All other phantom sounds in between are presented with incorrect pinna localization cues. I'm not sure which scenario is preferable.
Both techniques, Ambisonics and stereophony, are single seat solutions. A wider sweet spot delivering high quality sound to more than just a single person would be favorable.
Last edited: