HRTF is the most correct term.
Stereo mixes based on interchannel level differences deliver the most stable phantom localization. It doesn't get more "stereo".
Not to me. See post #266.
Not sure what you are trying to say. Adding a simple crossfeed to headphone playback will not externalize sounds. On the other hand, plugging one ear will not cause sound to be perceived within your head.
You missed my point. You cannot define a set of HRTF to use without first knowing your vectors. Would you use the same set of HRTF's for speakers at +/-10 vs +/30deg? How about 45deg? 180deg? 0deg?
Totally incorrect, it is actually the opposite. Stereo relies heavily on ITD and does poorly with ILD to acheive a stable soundstage. Move your head a couple inches and the sound changes drastically. Stereo starts falling apart >700hz and by 1.5-3Khz the errors are horrible. Or course it seems to "work" because human speech is within this range so some find it good enough, others reject it.
"2 loudspeakers and the listener form a equilateral triangle. "
If this is all you consider necessary, then good luck to you.
Last edited:
Quite honestly I don't understand what point you're trying to make. You said "Stereo requires a specific ammount of crosstalk to work." and that's simply not true. Ralph explained why in detail above.
Well, every mixing engineer will tell you that you're wrong. Phantom sources based on interchannel level differences will result in high localization accuracy which is typical for stereo reproduction. And yes, there is a "stereo seat" and this will always be a problem.
Localization accuracy of lateral phantom sources is typically pretty bad (see Blauert "Spatial Hearing"). You're again mixing up interaural and interchannel cues.
If this is all you consider necessary, then good luck to you.
That's the lowest common denominator and typical setup in control rooms. Sorry if you don't like what reality looks like.
Stereo as we know it depending on level differences is a completely artificial effect.
The brain only uses level differences if there are no phase differences for it to work with.
Anybody who ever had a head cold resulting in one ear being partially blocked should have noticed that: While listening to a stereo system is totally messed up we have no problems accurately localizing real sounds.
This has very solid evolutionary reasons.
(Sorry if this has been mentioned before but I did not follow this thread and reading through nearly 300 posts is too time consuming.)
The brain only uses level differences if there are no phase differences for it to work with.
Anybody who ever had a head cold resulting in one ear being partially blocked should have noticed that: While listening to a stereo system is totally messed up we have no problems accurately localizing real sounds.
This has very solid evolutionary reasons.
(Sorry if this has been mentioned before but I did not follow this thread and reading through nearly 300 posts is too time consuming.)
Here's a couple of AES papers on cross-talk in stereo
Keele
Part 1:
http://diy-audio.narod.ru/litr/Keele_1986-11_AES_Preprint_-_Interaural_Crosstalk_Part_1.pdf
Part 2:
http://diy-audio.narod.ru/litr/Keele_1986-11_AES_Preprint_-_Interaural_Crosstalk_Part_2.pdf
Keele
Part 1:
http://diy-audio.narod.ru/litr/Keele_1986-11_AES_Preprint_-_Interaural_Crosstalk_Part_1.pdf
Part 2:
http://diy-audio.narod.ru/litr/Keele_1986-11_AES_Preprint_-_Interaural_Crosstalk_Part_2.pdf
Any Polksters here?
Mathew Polk white paper on SDS/SRS technology and their new surround sound.
http://www.polkaudio.com/downloads/whitepapers/SDA_WhitePaper.pdf
Mathew Polk white paper on SDS/SRS technology and their new surround sound.
http://www.polkaudio.com/downloads/whitepapers/SDA_WhitePaper.pdf
it's amazing how some of this stuff gets recycled over and over. Give it a few years and someone else will come out something "new"
Position Independent Stereo very similar to polk's SDA.
http://www.extra.research.philips.com/hera/people/aarts/RMA_papers/aar01p.pdf
Carver Sonic Holography
Sonic holography - Wikipedia, the free encyclopedia
Unfortunately unless people keep up with trying to promote such ideas it will never take over and eventually people view the gadgets as gimics.
Does that really make you wonder? There has been so little progress in psychoacoustics within the last 80 years. We're still discussing things that Blumlein already knew (see last few posts about crosstalk in this thread) and still don't know how summing localization works.
If you want to hear what crosstalk free stereo sounds like, I have just posted some 39 short tracks at Ambiophonics Demos
These are 44.1 .wav files and you can burn them to a CD if you want to play them on an audiophile grade system. Half of them are recordings made with an Ambiophone and the others are just ordinary CDs. Try not to play them using vacuum tube amplifiers since these are seldom within a dB of balance.
Don't forget to move your speakers closer together or move back from them roughly two to three times the distance you are at now. You can also play the same 39 tracks in plain stereo and compare.
Ralph Glasgal
These are 44.1 .wav files and you can burn them to a CD if you want to play them on an audiophile grade system. Half of them are recordings made with an Ambiophone and the others are just ordinary CDs. Try not to play them using vacuum tube amplifiers since these are seldom within a dB of balance.
Don't forget to move your speakers closer together or move back from them roughly two to three times the distance you are at now. You can also play the same 39 tracks in plain stereo and compare.
Ralph Glasgal
thanks ralph..might be hard for most to move back three times as far!!!
will certainly give it a go (just bought a new computer which should allow me to download the program mentioned recently) which leads me to the next question.
assuming someone likes what they here, it might help if you explained how you did the files. For the moment I am assuming you used those programs linked to recently??
And for final clarification, no change needed to the speakers other than moving them closer together??
arrgghh, just looked at the link and you already explained what you used for the encoding (race..not that I understand it yet hehe).
So the new question is will the program you linked to earlier do exactly the same thing??
will certainly give it a go (just bought a new computer which should allow me to download the program mentioned recently) which leads me to the next question.
assuming someone likes what they here, it might help if you explained how you did the files. For the moment I am assuming you used those programs linked to recently??
And for final clarification, no change needed to the speakers other than moving them closer together??
arrgghh, just looked at the link and you already explained what you used for the encoding (race..not that I understand it yet hehe).
So the new question is will the program you linked to earlier do exactly the same thing??
Last edited:
It is better if you move your speakers closer together than just move back since having the speakers closer means less room influence, maybe better bass, and a good chance to stay in the near field. I have found however, that many audiophiles simply will not move their speakers and so never hear Ambio or worse.
You don't have to do anything except reduce the speaker angle to about one third the angle you use for stereo. Of course I do hope you will want to listen to your own collection of LPs or CDs so you can experience something that you like better and try the four speaker versions. Also you can better adjust RACE, while listening, to suit your preferences, system, and even a particular recording.
I used RACE 4.0 in the AudioMulch program to make the .wav files. This is shown and explained on the website. But I could have used any of the other Transcoders or Plugins. As a wise man once said, bits is bits. (I know audiophiles don't believe this.)
Ralph Glasgal
You don't have to do anything except reduce the speaker angle to about one third the angle you use for stereo. Of course I do hope you will want to listen to your own collection of LPs or CDs so you can experience something that you like better and try the four speaker versions. Also you can better adjust RACE, while listening, to suit your preferences, system, and even a particular recording.
I used RACE 4.0 in the AudioMulch program to make the .wav files. This is shown and explained on the website. But I could have used any of the other Transcoders or Plugins. As a wise man once said, bits is bits. (I know audiophiles don't believe this.)
Ralph Glasgal
All recordings I have tried benefit from ambiophonics, but some are simply jaw-dropping. Penderecky´s Seven Gates of Jerusalem on the Naxos label and the classic (it´s the one in Kubrick´s Space Oddisey) 1968 recording of Ligety´s Requiem conducted byMichael Gielen are magical. I have been trying to gather information as to the microphone configurations employed but I drew a blank. There is something unique about them that is not present on most other recordings and it would be interesting to find out what that is. Anyone who knows what I´m talking about could contribute with a list of their recordings that do this for others to try.
I want to say thanks to Ralph Glasgal and others who are contributing to the ambiophonics cause for this real and substantial leap in music reproduction.
I want to say thanks to Ralph Glasgal and others who are contributing to the ambiophonics cause for this real and substantial leap in music reproduction.
Alternative "hardware" implementations?
Hi everybody,
I've seen this page: Ambiophonics
Is the four speakers (110°) "WAF approved" version a valid implementation of the ambiophonics concept?
If it is, is it as good as the others? how is that supposed to work?
Shouldn't the 4 speakers be equal? (from the picture on that site it looks like they are completely different!)
Is there any relation with the Polk Audio Stereo Dimensional Array ?
...and what about this one:
Hi-Fi 3-way ambient speaker system
BTW: I always use careful speaker positioning in stereo setup to optimize imaging and always end up with >> 60° angle. Yet I feel no "hole" in the middle. Quite the contrary, I "see" clear & perfectly focused center image.
On the other end, I've recently experimented with "RACE" and have to say that it can be quite impressive.
Problem is, on my setup it makes the listening "hot spot" much smaller than what I have with stereo! (with my usual stereo setup -and some recordings-, I can even stand-up and move around the room without the "image" moving or being much altered, while with RACE the image can be amazingly good but only on a small area along the middle of the speakers).
Hi everybody,
I've seen this page: Ambiophonics
Is the four speakers (110°) "WAF approved" version a valid implementation of the ambiophonics concept?
If it is, is it as good as the others? how is that supposed to work?
Shouldn't the 4 speakers be equal? (from the picture on that site it looks like they are completely different!)
Is there any relation with the Polk Audio Stereo Dimensional Array ?
...and what about this one:
Hi-Fi 3-way ambient speaker system
BTW: I always use careful speaker positioning in stereo setup to optimize imaging and always end up with >> 60° angle. Yet I feel no "hole" in the middle. Quite the contrary, I "see" clear & perfectly focused center image.
On the other end, I've recently experimented with "RACE" and have to say that it can be quite impressive.
Problem is, on my setup it makes the listening "hot spot" much smaller than what I have with stereo! (with my usual stereo setup -and some recordings-, I can even stand-up and move around the room without the "image" moving or being much altered, while with RACE the image can be amazingly good but only on a small area along the middle of the speakers).
ideally the speaker are all the same...............
They basically are running the outer speakers as mono.
Another way is to run a mono speaker in the center.
Since the negative of the speakers is a common ground (usually), you take both of the negatives and connect them to a positive of the center loudspeaker, then run the negative back to the amp. You need to run a resistor pot around the center speaker, otherwise the center will be 6db louder than the fronts. The variable resistor doesn't work well due to a changing load of a loudspeaker, especially a full range driver.
I wouldn't like the smear of a tall full range driver line array. I'd run a focused for that application like this.
Norman
They basically are running the outer speakers as mono.
Another way is to run a mono speaker in the center.
Since the negative of the speakers is a common ground (usually), you take both of the negatives and connect them to a positive of the center loudspeaker, then run the negative back to the amp. You need to run a resistor pot around the center speaker, otherwise the center will be 6db louder than the fronts. The variable resistor doesn't work well due to a changing load of a loudspeaker, especially a full range driver.
I wouldn't like the smear of a tall full range driver line array. I'd run a focused for that application like this.
An externally hosted image should be here but it was not working when we last tested it.
Norman
Is the four speakers (110°) "WAF approved" version a valid implementation of the ambiophonics concept?
Shouldn't the 4 speakers be equal? (from the picture on that site it looks like they are completely different!)
Is there any relation with the Polk Audio Stereo Dimensional Array
Problem is, on my setup it makes the listening "hot spot" much smaller than what I have with stereo! (with my usual stereo setup -and some recordings-, I can even stand-up and move around the room without the "image" moving or being much altered, while with RACE the image can be amazingly good but only on a small area along the middle of the speakers).
The four speaker arrangement for full surround Ambiophonics has nothing to do with the 5.1 standard. The front pair does not have to be the same as the rear pair. The front pair and the rear pair should be at an angle of from 20 to 30 degrees or one third the stereo spacing. The rear and front angles or distance from the listeners do not have to be identical. You never need a center speaker front or rear. Playing a movie with a real rear stage, you will hear sound coming from a full 360 degrees unlike the standard 5.1 system.
Normally the listening area is larger than for stereo or 5.1. With Ambiophonics you can move along the center line between the speakers for yards, recline, stand up, lean, etc. If offside you get a nice mono signal everywhere in the room. In contrast, with stereo if you go offside you usually hear just one speaker. If you move closer, you get a hole in the middle. If you move back you get mono. In 5.1 you will localize to the rear or side speakers if you move out of the sweet area. But usually 5.1 video is just 5 channel mono so it works good enough. The other point is that when you move offside in stereo most listeners don't care much since it wasn't all that realistic to start with. In Ambiophonics you feel very much deprived if you are not reasonably centered. So yes you must remain between the speakers which is harder of course when they are close together. Ambiophonics is not for a room full of people.
There have been many attempts in the past to eliminate stereo crosstalk. The Polk speaker was one of them. Unfortunately the functions of the pinna, combing, and other considerations make it impossible to do XTC well this way. Same for Sonic Holography, Lexicon Panorama, SRS, 7.1, 10.2, and any number of psuedo surround gizmos. At the present time only TacT products incorporate Ambiophonic programs if you want a component rather than doing it in a PC.
Ralph Glasgal
Home Page
Hello,
Some questions about the RACE. I found this block diagram from
http://www.ambiophonics.org/Tutorials/RGRM-RACE_rev.pdf
Everything I see in this method is very simple! All there is are high/lowpass filters, summer, inverter, attenuator and delay. This indicates RACE can be done in analog domain with a few opamps. Have anyone tried to do this?
Do you know if the PC versions of RACE use excatly this same simple algorithm, without any extra HRTF filtering, or other extra processing?
What do you recommend for the high/lowpass filter order? 1st, 2nd or higher?
- Elias
Some questions about the RACE. I found this block diagram from
http://www.ambiophonics.org/Tutorials/RGRM-RACE_rev.pdf
An externally hosted image should be here but it was not working when we last tested it.
Everything I see in this method is very simple! All there is are high/lowpass filters, summer, inverter, attenuator and delay. This indicates RACE can be done in analog domain with a few opamps. Have anyone tried to do this?
Do you know if the PC versions of RACE use excatly this same simple algorithm, without any extra HRTF filtering, or other extra processing?
What do you recommend for the high/lowpass filter order? 1st, 2nd or higher?
- Elias
I was NOT talking about either 5.1 or the 4 speaker RACE Ambiophonics (with 2 front + 2 rear speakers).
Have a look at the diagram from the site I've linked, I was talking about this:
An externally hosted image should be here but it was not working when we last tested it.
Would that work? How?
sorry, but this is absolutely NOT anywhere like what I get in my setup (2 channel, unprocessed stereo signal sourced).
I can move all the way toward the speakers (even between them) and still get a nice center image (that gets all "around" me). And I can move all the way back to the rear wall, and still the image is there. Even better, if I'm far enough (> 2.5m or such) from the speakers, at least with some (not all) recordings I can move freely around the room (even off-axis from side wall to side wall!) without (almost) any alteration of the image. Which BTW is by no means limited between the speakers (as it is supposed to be with stereo), it definitely extends way beyond them in all directions.
I have no idea about how/why this is possible, but that's what I get here.
(again, this is with just two conventional speakers and unprocessed stereo signal. As said the only "unusual" thing is speaker positioning that's definitely not the "canonical" 60 degrees stereo triangle. The speakers are along the 4m wide back wall, > 1m from it but only ~ 50cm apart from the side walls. With really a lot of "toe-in").
Having just seen the paper on "Position Independent Stereophonic sound Reproduction" which have been linked here recently, I suspect that perhaps my arrangement uses a mix of speaker own directionality and listening room reflections to obtain a result which is somewhat similar to P.I. stereo (but that's just a guess).
Hi,
The side speakers at 55 degrees are for 'surround'. The angle 55 is supposed to minimize interaural cross correlation (IACC) and thus maximise envelopment.
- Elias
The side speakers at 55 degrees are for 'surround'. The angle 55 is supposed to minimize interaural cross correlation (IACC) and thus maximise envelopment.
- Elias