I agree. Where are we supposed to sleep?
Have you ever tried sleeping on a vertical mattress?
The answer to this is subjective. In stereo if you move forward you eventually get a hole in the middle. If you move backward you eventually get mono. Most people in large room demos never notice this or don't care.
In Ambiophonics, if you move forward you eventually get normal (narrow) stage stereo and if you move back nothing much happens and the stage remains full width with depth etc. So some six people along the center line will hear the same vibrant stage that Ambiophonics can deliver and no center speaker is needed for dialog.
In stereo if you move to the side you normally localize to one speaker and so mainly hear just one channel. Again many listeners don't feel all that deprived since the stereo stage is not very realistic and narrow to start with.
In Ambiophonics, if you move to the side you still hear both channels as in good mono reproduction so all you lose is the stage width. Movie dialog or soloists are still fully audible even offside. However, most listeners feel quite deprived going from Ambiophonics, to mono, (or stereo for that matter) much more so than in the stereo-to-mono case so this is a problem. Ambiophonics is really not for a room full of people, but neither is stereo. I maintain that, on balance, offside Ambiophonics sounds better than offside stereo. A visual factor is that people do like to stay centered on a video screen.
Ralph Glasgal
Home Page
Thanks.
I did some quick testing and the effect worked.
Didn't tweak it much yet.
I did get a lot of room ambiance that "jumped out" and seemed a little too high in level compared to the rest. I guess 44100 didn't help as it was hard to get it stable with to little resolution in delay.
Have to try more later.
The speakers where slightly toed in.
I did some quick testing and the effect worked.
Didn't tweak it much yet.
I did get a lot of room ambiance that "jumped out" and seemed a little too high in level compared to the rest. I guess 44100 didn't help as it was hard to get it stable with to little resolution in delay.
Have to try more later.
The speakers where slightly toed in.
Usually the delay parameter is not that critical. You can also compensate mechanically by moving closer or further away from the speakers and seeing if this make any audible difference. If you use the AudioMulch version of RACE you can adjust the delay parameter a microsecond at a time. With the TACT Audio Processor 10 at a time.
The only time I have heard anyone mention stability was when the speakers were not moved closer together. The speakers should be about one third closer together than in the stereo case. About 15 to 30 degrees compared to 60 for stereo.
Ralph Glasgal
www.ambiophonics.com
I used the JAVA based converter linked above.
Anything but 45us and 68us caused a tinny sound. 96kHz should give me more than two options to choose from.
I used about 20* between the speakers.
Ill have to do some more testing in a bigger room where I can play a bit more with position and parameters.
I'm not in that much hurry as I have to build the final speaker first.
Anything but 45us and 68us caused a tinny sound. 96kHz should give me more than two options to choose from.
I used about 20* between the speakers.
Ill have to do some more testing in a bigger room where I can play a bit more with position and parameters.
I'm not in that much hurry as I have to build the final speaker first.
I find that if you must toe, toe out is better than toe in. Toe in with stereo is done to either minimize side wall effects from the room or a type of level adjustment for the upper end (or in the case of a certain brand of speaker to avoid a hole in the response). With ambio, assuming you have it properly setup, the speakers are even further away from sidewalls than you are with stereo, so toe out gets you more coverage just like a concert array. So if your stereo setup has the speakers close to the sidewalls and you move back to get a shallower angle, it is not entirely the same as if you physically moved the speakers. If you have a large room, well it might not matter.
The nice thing about off-center seating with ambio is that the time of arrival error is magnitudes less than with off-center stereo, just like mono.
See paper on Transaural
http://www.transaural.com/
Transaural/OSD is a very close cousin of ambiophonics/stereo dipole
The nice thing about off-center seating with ambio is that the time of arrival error is magnitudes less than with off-center stereo, just like mono.
See paper on Transauralhttp://www.transaural.com/
Transaural/OSD is a very close cousin of ambiophonics/stereo dipole
Last edited:
There is something seriously wrong. 45 and 68 microseconds are much too small to do much of anything. In every working system that I know about, the values are centered about 100 microseconds. Try the two higher values at 44.1 and use the 4 kHz bypass to start with. Often problems like this are due to major level imbalances in vacuum tube amplifiers or preamps. The transcoder and all RACE programs assume that the amplifier speaker chain they are feeding are identical in level and frequency response to say half a dB. It is not unusual for vaccum tube devices to be out of balance by as much as 3dB which is okay for stereo but not Ambiophonics.
Ralph Glasgal
Home Page
"so toe out gets you more coverage just like a concert array."
First of all that is a really bad habit. See link.
PA Systems
With proper toe-in you can get a larger area getting the same volume from L/R speaker depending on directivity.
Going much above 91us gives severe comb filtering and tinny sound.
Ill do some more testing.
In my quick test I used 70cm between the speakers and sat 1.5-2m from the speakers.
Without 4k bypass sound was horrible.
My test setup might be too small to work properly without better resolution.
Balance should be spot on, at least all the way down to speaker.
The player version might be easier to test with as well when you don't have as many IO to worry about.
Hmm, I seem to get the same tinny/comb sound from the player version into headphones.
It is suppose do be that way before it hit the speakers?
And then sum into a perfect signal when it hits the head?
First of all that is a really bad habit. See link.
PA Systems
With proper toe-in you can get a larger area getting the same volume from L/R speaker depending on directivity.
Going much above 91us gives severe comb filtering and tinny sound.
Ill do some more testing.
In my quick test I used 70cm between the speakers and sat 1.5-2m from the speakers.
Without 4k bypass sound was horrible.
My test setup might be too small to work properly without better resolution.
Balance should be spot on, at least all the way down to speaker.
The player version might be easier to test with as well when you don't have as many IO to worry about.
Hmm, I seem to get the same tinny/comb sound from the player version into headphones.
It is suppose do be that way before it hit the speakers?
And then sum into a perfect signal when it hits the head?
That reflects my own impressions with Stephan's Ambiophonic audio player on a Mac mini. The sound is very unnatural, it's neither "stereo" nor "realism".
Best, Markus
You can appreciate that trying to exactly cancel crosstalk at frequencies over 4kHz is very difficult and requires extreme accuracy. It is also not necessary since the ear relies on the pinna for localization at these higher frequencies and crosstalk is not a factor. Thus the bypass option.
You are hearing one possible result of trying to cancel crosstalk at higher frequencies. You get unpredictable peaks and dips. These are usually no more audible then the comb filtering produced by the stereo loudspeakers, but in stereo if you just listen to one speaker you can't hear it. In Ambio the crosstalk cancelling signals and the possible HF peaks and dips are already present in the signal going to the speaker which is why you hear it in earphones. In my experience however, if you are not right at a speaker, you hear a normal sounding combination of sound from the two close together speakers anywhere in the room. Also unlike stereo, you will hear both channels even when you are far off center, since both left and right signals go to each speaker. In stereo you will likely localize to one spaker and only really hear one side well.
You are also correct that on the center line between the speakers, these speaker signals acoustically add and cancel to produce flat response without the stereo combing and now with the correct ITD and ILD that was captured properly on the disc.
Ralph Glasgal
Home Page
You are still thinking of terms of flawed stereo where you are trying to get that pinpoint imaging from widely spaced speakers. Your link even mentions the "hole in the middle" problem. One major advantage of ambio is no hole in the middle because the speaker is there, just as in mono.Also, last I checked arrays are usually curved in the area of coverage, just like horns curve outward, not inward.
An externally hosted image should be here but it was not working when we last tested it.
You can't use ambio with headphones. The crosstalk is already removed the second you put the headphones on and instead of 60 deg angle, now you have 180 with a head barrier. The reverse has to be done when using headphones (add crosstalk back in) or the sound is stuck in your head. This is why you see crossfeed circuits in headphone amps.
Maybe this is why people toe in too, to attempt to adjust the amount of crosstalk, in addition to avoiding sidewall reflections.
Ambio is in the center of room, further away from sidewalls so it is less of a worry.
Last edited:
I really do not have much interest in this ambio stuff.
I've tried similar things before, and i am not convinced the gain is bigger than the time wasted.
Stereophonic is interesting enough though, but to be limited to headphones only...
I was wondering about this "hole in the middle" you speak of.
I can not for the life of me, hear that there are any holes between my speakers. I have a tendency to "know" where the hole is when i listen to most tunes, but as soon as i close my eyes to those same tunes, i hear absolutely nothing, it's like a 2,5m wall of sound, from stereo speakers.
And i have a hard time understanding that adding several active gizmos to the signal path just to add more speakers to the room will give me any benefit at all. You may gain the desired effect, but the sound itself will not be as clean as it should be, slight colorations all over the spectrum. I know that annoys the crap out of me.
I've tried similar things before, and i am not convinced the gain is bigger than the time wasted.
Stereophonic is interesting enough though, but to be limited to headphones only...
I was wondering about this "hole in the middle" you speak of.
I can not for the life of me, hear that there are any holes between my speakers. I have a tendency to "know" where the hole is when i listen to most tunes, but as soon as i close my eyes to those same tunes, i hear absolutely nothing, it's like a 2,5m wall of sound, from stereo speakers.
And i have a hard time understanding that adding several active gizmos to the signal path just to add more speakers to the room will give me any benefit at all. You may gain the desired effect, but the sound itself will not be as clean as it should be, slight colorations all over the spectrum. I know that annoys the crap out of me.
Tried the Ambiophonic player btw.
It was weird, listened to a self made tune that has slight distortion in the bass, with the ambio player the distortion felt like it was smoothed out, the whole tune was a little bit more smooth than it should be, and on certain bell sounds there was a strange reeverb effect on the very beginning of the sound. the detail felt like it was dumbed down a bit, dynamics where not as good as they should be.
On the music i heard it sounded like a smoothed-out-unnatural-sound-almost-regular-stereo version.
Btw, why couldn't i edit my previous post? Can i only do that when i haven't logged out and back in again?
It was weird, listened to a self made tune that has slight distortion in the bass, with the ambio player the distortion felt like it was smoothed out, the whole tune was a little bit more smooth than it should be, and on certain bell sounds there was a strange reeverb effect on the very beginning of the sound. the detail felt like it was dumbed down a bit, dynamics where not as good as they should be.
On the music i heard it sounded like a smoothed-out-unnatural-sound-almost-regular-stereo version.
Btw, why couldn't i edit my previous post? Can i only do that when i haven't logged out and back in again?
"You can't use ambio with headphones. The crosstalk is already removed the second you put the headphones on"
The point I was trying to make it that I got the same type of tinny/comb sound from both headphones and speakers, although some less for speakers at 68us but still present. The amount of ambiance that stood out was the same though.
I think size has to do with it a lot more when only running 44100 and I think this introduces a lot of artifacts. Especially with the short delays attached with a small setup.
Without trying a bigger setup I will never know.
"just as in mono" You are right, as that is the only thing you will get with a splayed line-array. No sound-stage at all.
But that is a topic for another thread.
One thing I noticed is that mono sources seems to get less output than ambiance. And echoes seems stronger with more emphasis.
I thought one should get the same balance of content except better imaging and wider sound-stage (where content allows).
"where you are trying to get that pinpoint imaging from widely spaced speakers."
No just equal loudness from both speakers over a wider area due to directivity. I always try to get the speakers 22.5* off center (compared to 30*).
I will try to get it to work with a normal stereo setup, that way I can turn it off when there is more people in the room.
"but the sound itself will not be as clean as it should be, slight colorations all over the spectrum. I know that annoys the crap out of me."
I agree, it has to sound natural and clean for me to want it.
Although I enjoy experimenting so nothing lost yet.
The point I was trying to make it that I got the same type of tinny/comb sound from both headphones and speakers, although some less for speakers at 68us but still present. The amount of ambiance that stood out was the same though.
I think size has to do with it a lot more when only running 44100 and I think this introduces a lot of artifacts. Especially with the short delays attached with a small setup.
Without trying a bigger setup I will never know.
"just as in mono" You are right, as that is the only thing you will get with a splayed line-array. No sound-stage at all.
But that is a topic for another thread.
One thing I noticed is that mono sources seems to get less output than ambiance. And echoes seems stronger with more emphasis.
I thought one should get the same balance of content except better imaging and wider sound-stage (where content allows).
"where you are trying to get that pinpoint imaging from widely spaced speakers."
No just equal loudness from both speakers over a wider area due to directivity. I always try to get the speakers 22.5* off center (compared to 30*).
I will try to get it to work with a normal stereo setup, that way I can turn it off when there is more people in the room.
"but the sound itself will not be as clean as it should be, slight colorations all over the spectrum. I know that annoys the crap out of me."
I agree, it has to sound natural and clean for me to want it.
Although I enjoy experimenting so nothing lost yet.
I agree. With Ambiophonics the mix seems to fall apart. Maybe it's because most recordings are produced to sound natural with a typical stereo setup.
Best, Markus
Hi,
I don't get this. With 91us the first comb filtering notch is at 5.5kHz, and all the other combing effects are at freqs higher than this.
If you use 4kHz bypass there is no comb filtering in Ambiophonics.
In stereo the comb filtering starts at much lower freqs and there is nothing to prevent it! Maybe you are used to sound of the stereo combs and consider it 'natural'?
- Elias
I don't get this. With 91us the first comb filtering notch is at 5.5kHz, and all the other combing effects are at freqs higher than this.
If you use 4kHz bypass there is no comb filtering in Ambiophonics.
In stereo the comb filtering starts at much lower freqs and there is nothing to prevent it! Maybe you are used to sound of the stereo combs and consider it 'natural'?
- Elias
Everybody is used to "sound of the stereo combs". That is even true for the peole that produce recordings. As a consequence Ambiophonics is not compatible with conventional stereo recordings?
must be doing something dumb, but downloaded and opened the Ambiophonic Transcoder , selected my input and output (I presume) but on -my admittedly crappy- computer speakers there was no change in sound at all.
should be pretty simple??
I tried the mosc ambio (?) quite a while ago, seemed to work but as I already have a 120* soundstage it seems that the benefit of ambio might be lost on my system?? Still, if I can, I am willing to give this new version a try.
Dunno what data you might need to be able to spot where I am going wrong, simply using media player in the computer and all the program seems to need is the correct input and output...and as I only have one soundcard I can't go too wrong can I??
should be pretty simple??
I tried the mosc ambio (?) quite a while ago, seemed to work but as I already have a 120* soundstage it seems that the benefit of ambio might be lost on my system?? Still, if I can, I am willing to give this new version a try.
Dunno what data you might need to be able to spot where I am going wrong, simply using media player in the computer and all the program seems to need is the correct input and output...and as I only have one soundcard I can't go too wrong can I??