electro-music.com   Dedicated to experimental electro-acoustic
and electronic music
 
    Front Page  |  Articles  |  Radio
 |  Media  |  Forum  |  Wiki  |  Links  |  Store
Forum with support of Syndicator RSS
 FAQFAQ   CalendarCalendar   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   LinksLinks
 RegisterRegister   ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in  Chat RoomChat Room 
Live streaming at radio.electro-music.com

  host / artist show at your time
  Faux Pas Quartet and friends Music From Last Thursday
Please visit the chat
 Forum index » How-tos » Ambiophonic Sound Reproduction
Mosc's Ambio VST
Post new topic   Reply to topic
Page 1 of 2 [50 Posts]
View unread posts
View new posts in the last week
Mark the topic unread :: View previous topic :: View next topic
Goto page: 1, 2 Next
Author Message
mosc
Site Admin


Joined: Jan 31, 2003
Posts: 17557
Location: Allentown, PA
Audio files: 123
G2 patch files: 60

PostPosted: Wed Jul 23, 2008 12:01 am    Post subject: Mosc's Ambio VST Reply with quote  Mark this post and the followings unread

I have put together an experimental ambiophonic processor VST plugin using SynthEdit. I think this will only work on a windows system.

The algorithm is the same as the ChucK version, http://electro-music.com/forum/topic-27559.html

After much experimentation, I find that the recursion (feedback) in the RACE encoder adds distortion. This VST plugin uses my own wrinkle on the ambiophonic processing. I call it the MAP, Mosc's Ambiophonic Processor. Very Happy

Anyway, it would be great if some of you with Windows systems would try this out. I use it with the VST plugin adapter in Winamp. Then I can play virtually any type of audio file, and even use it with web radio.

Here's a screen shot.

Posted Image, might have been reduced in size. Click Image to view fullscreen.

Don't use this with headphones or with speakers wide set in the usual 60 degree separation. The speakers must be placed close together directly in front of the listener. Separation is about 20 degrees.

Balance is very important. If that's off then you won't get good results. That's why I added a balance control.

Normally, I just keep the width control all the way up.

This is the first VST I've done, so there is a possibility that it will have problems. Please give it a go and let me know how the VST works and how you like ambio playback. Remember, this will work with any stereo sources. Your CD and mp3 collection are now reborn.

Unzip the attachment and put it in your usual VST directory. Use it with your regular VST host. I'll post something later about how to use it with Winamp for those that would like to do that.

Many thanks to the master, Robin Miller. I have learned everything about this from him. Thanks too to Jack Tamul, Bachus, and Bill Fox who have given me a lot of valuable feedback.

I hope this VST will make it possible for many people to discover this exciting new way to listen to music. Many of us who have tried it will always never go back to conventional stereo.

Update: July 30, 2008 - I posted a little article on how to use a VST processor with Winamp and Quintessential Player. http://electro-music.com/forum/topic-27948.html

Update: July 31, 2008 - Here is the diagram of the MAP algorithm:


Maybe this could be clearer, the -1 in the circles is an inverter.


mosc-ambio.zip
 Description:
Mosc's Ambiophonics Processor - VST plugin

Implents the MAP algorithm. Tested with Winamp VST host plugin.

Download
 Filename:  mosc-ambio.zip
 Filesize:  882.78 KB
 Downloaded:  1377 Time(s)


_________________
--Howard
my music and other stuff

Last edited by mosc on Thu Jul 31, 2008 11:07 am; edited 4 times in total
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
kkissinger



Joined: Mar 28, 2006
Posts: 1212
Location: Kansas City, Mo USA
Audio files: 29

PostPosted: Wed Jul 23, 2008 7:32 am    Post subject: can't wait to try this Reply with quote  Mark this post and the followings unread

Howard,

I'll check this out when I get home.

Is it hard to write VST's?

Thanks for posting.

_________________
-- Kevin
http://kevinkissinger.com
Back to top
View user's profile Send private message Visit poster's website
mosc
Site Admin


Joined: Jan 31, 2003
Posts: 17557
Location: Allentown, PA
Audio files: 123
G2 patch files: 60

PostPosted: Wed Jul 23, 2008 12:11 pm    Post subject: Reply with quote  Mark this post and the followings unread

It's not too hard to write VSTs with SynthEdit, but it isn't all that ideal for DSP type operations. In this one I just made, I'm not sure how precise the delays are.

I'd be interested in what you think. I think pan ambio - four channel with an ambiopole in the rear - is really superb.

I'm also working on a four speaker two channel configuration suitable for sound reinforcement at electro-music 2008, but results are too preliminary to report at the moment.

_________________
--Howard
my music and other stuff
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
jack



Joined: Feb 06, 2003
Posts: 22
Location: East Coast USA

PostPosted: Thu Jul 24, 2008 12:57 pm    Post subject: VST Reply with quote  Mark this post and the followings unread

I'll try it out in sonar. It looks great.
Jack

_________________
JACK
Back to top
View user's profile Send private message
elektro80
Site Admin


Joined: Mar 25, 2003
Posts: 21969
Location: Norway
Audio files: 14

PostPosted: Thu Jul 24, 2008 1:02 pm    Post subject: Reply with quote  Mark this post and the followings unread

but..but... we gotta have this ported to ordinary plain computers... Shocked
Laughing

_________________
A Charity Pantomime in aid of Paranoid Schizophrenics descended into chaos yesterday when someone shouted, "He's behind you!"

MySpace
SoundCloud
Flickr
Back to top
View user's profile Send private message Visit poster's website
DrJustice



Joined: Sep 13, 2004
Posts: 2031
Location: Morokulien
Audio files: 3

PostPosted: Thu Jul 24, 2008 1:26 pm    Post subject: Reply with quote  Mark this post and the followings unread

elektro80 wrote:
but..but... we gotta have this ported to ordinary plain computers... Shocked
Laughing

We already have: here Razz

DJ
--
Back to top
View user's profile Send private message Visit poster's website
elektro80
Site Admin


Joined: Mar 25, 2003
Posts: 21969
Location: Norway
Audio files: 14

PostPosted: Thu Jul 24, 2008 1:50 pm    Post subject: Reply with quote  Mark this post and the followings unread

DrJustice wrote:
elektro80 wrote:
but..but... we gotta have this ported to ordinary plain computers... Shocked
Laughing

We already have: here Razz

DJ
--


Oh shit! I forgot that one! Laughing

_________________
A Charity Pantomime in aid of Paranoid Schizophrenics descended into chaos yesterday when someone shouted, "He's behind you!"

MySpace
SoundCloud
Flickr
Back to top
View user's profile Send private message Visit poster's website
durwood



Joined: Apr 25, 2008
Posts: 30
Location: Chicagoland

PostPosted: Tue Jul 29, 2008 6:16 am    Post subject: Reply with quote  Mark this post and the followings unread

I'll give this a go. I'm in search of a ambio VST that is adjustable because I am trying to adopt this to a non-ideal setup.

Really it would be nice to find one that could do one of two things (or both, but I don't think it's possible to acheive both at the same time):

1) Widen the sweet spot even if it meant a slight loss in accuracy

2) Be adjustable for a possible off-center seating position as in no directly central between the speakers.

#1 might create too much coloration. Mono allows for multiple sweet spots and since this resembles monophonic setup, then it could be possible.

I know #2 is an odd goal, but it seems possible to allow for individual channel delay adjustments, or some way to shift the sound image L or R if this makes any sense. Is this what the balance knob is for?

For example, let's say you were sitting off to the side of the speakers to the left. Let's call the four signals Ll=left speaker left ear, Lr=left speaker right ear, Rr=Right speaker right ear, Rl=right speaker left ear.

Obviously the azimuth angles are all screwed up creating different path lengths.

Ll is longer than it should be, Lr is shorter than it needs to be. Rr is longer than it needs to be, and Rl is longer as well. It would be cool if these signals were adjustable somehow. The RACE setup in audiomulch kind of allows this type of adjustment. Is there an easy way to do this via VST's?

I know this is asking alot considering the ideal setup is still trying to be perfected, but I figure nothing wrong with asking to see if my logic makes sense.
Back to top
View user's profile Send private message
mosc
Site Admin


Joined: Jan 31, 2003
Posts: 17557
Location: Allentown, PA
Audio files: 123
G2 patch files: 60

PostPosted: Tue Jul 29, 2008 11:57 am    Post subject: Reply with quote  Mark this post and the followings unread

The balance control is just a conventional volume adjustment of the L and R signals. Exactly like those on a preamp or something. It's just that most consumer products these days don't have this with digging through lots of menus. It's there just for convenience.

Your suggestion to allow changes of delay times is a good one. Have you had any luck doing this with the audio mulch RACE processor? The interesting thing is that delay is only adjustable in one sample increments, about 22.7 microseconds, no matter what the software says. What I'm suggest is that if you know of something that works, please let us know.

As for as widening the sweet spot. That is the holy grail of ambiophonics, after correcting for the timbre modification and the coherent signal attenuation.

I have been experimenting with a 4 speaker implementation that is quite good, IMHO. I'm not ready to post about it because I'm still experimenting. The sweet spot is still on axis, but the sound is "good" all over.

_________________
--Howard
my music and other stuff
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
mosc
Site Admin


Joined: Jan 31, 2003
Posts: 17557
Location: Allentown, PA
Audio files: 123
G2 patch files: 60

PostPosted: Tue Jul 29, 2008 12:00 pm    Post subject: Reply with quote  Mark this post and the followings unread

For reference, here is the impulse response of my MAP VST processor. This has no sonic properties - listening will just provide a couple of clicks. Laughing


map-vst-impulse-response.wav
 Description:
Impulse response of mosc's ambiophonic processor (MAP)

Not for listening - just an engineering reference.

Download
 Filename:  map-vst-impulse-response.wav
 Filesize:  1.18 MB
 Downloaded:  766 Time(s)


_________________
--Howard
my music and other stuff
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
kkissinger



Joined: Mar 28, 2006
Posts: 1212
Location: Kansas City, Mo USA
Audio files: 29

PostPosted: Tue Jul 29, 2008 5:23 pm    Post subject: Reply with quote  Mark this post and the followings unread

This works well in windows. Up to a setting of about '6' there is no coloration to the tone, above '6' there is a midrange bump outside the "sweet spot" -- in the sweet spot the bump isn't there.

I ran this on Windows XP with Cubase SX.

No problems downloading and getting it to run in Cubase.

I'm going to experiment with this plugin -- who knows, I may put it to use at electro-music Wink

_________________
-- Kevin
http://kevinkissinger.com
Back to top
View user's profile Send private message Visit poster's website
kkissinger



Joined: Mar 28, 2006
Posts: 1212
Location: Kansas City, Mo USA
Audio files: 29

PostPosted: Tue Jul 29, 2008 7:57 pm    Post subject: Reply with quote  Mark this post and the followings unread

kkissinger wrote:
Up to a setting of about '6' there is no coloration to the tone, above '6' there is a midrange bump outside the "sweet spot" -- in the sweet spot the bump isn't there.


OK, I just learned that the reason I heard the color change was that I didn't have my speakers positioned properly.

The left and right speakers need to be, literally, right next to each other!

When I repositioned the speakers as intended, I heard no coloration to the sound.

It definitely produces an enveloping, stereo field -- quite amazing, indeed!

_________________
-- Kevin
http://kevinkissinger.com
Back to top
View user's profile Send private message Visit poster's website
durwood



Joined: Apr 25, 2008
Posts: 30
Location: Chicagoland

PostPosted: Wed Jul 30, 2008 6:37 am    Post subject: Reply with quote  Mark this post and the followings unread

mosc wrote:
The balance control is just a conventional volume adjustment of the L and R signals. Exactly like those on a preamp or something. It's just that most consumer products these days don't have this with digging through lots of menus. It's there just for convenience.


Ok, thanks for clearing that up for me.

Quote:

Your suggestion to allow changes of delay times is a good one. Have you had any luck doing this with the audio mulch RACE processor?


It's kind of hard to make 4 adjustments separately and see if it works correctly by ear. The only easy way to make it easier is to do calculations and see if it agrees with my hearing. I thought by adjusting the head delays that would take care of it, but then I just realized I forgot to include the delays on the non-crosstalk signals. I need to retest it at some point to see if it works or not. Something that might make adjustments easier would be to figure out a way to have some type of a single azimuth angle adjustment that would create proper delays based on the seating distance from speakers and head/sphere size. I have a AES paper that investigated the vector math behind all of this and they briefly went into how an off-center seating position affects a standard stereo setup. I imagine these equations could be adopted to an ambipole since it's nothing more than vectors and trig anyway.

Quote:

What I'm suggest is that if you know of something that works, please let us know.


For a VST delay, Voxengo Delay will do delay adjustments by increments of 10uS. I also thought about using Voxengo Pristine space and separating the matrix impulse into 4 sections and then performing the delays there and recombine based on the RACE block diagram, but I'm not sure my logic makes sense this way. I thnk this is how Angelo Farina was doing it before he had the X-volver VST created.

I attempted to make a VST using synthedit but it didn't like the feedback loop and since it was my first attempt to use synthedit, I gave up on trying to create a RACE based VST. I thought there was another program that you use a GUI to make VST's but perhaps I'm mistaken. MAX/MSP? I don't recall. I'd love to try to code one myself, but I'm slow and coding makes me want to pull my hair out sometimes. I get impatient. Embarassed

For now, all my experimenting is done by stringing multiple VST's together in Audiomulch or Console.

Quote:

As for as widening the sweet spot. That is the holy grail of ambiophonics, after correcting for the timbre modification and the coherent signal attenuation.

I have been experimenting with a 4 speaker implementation that is quite good, IMHO. I'm not ready to post about it because I'm still experimenting. The sweet spot is still on axis, but the sound is "good" all over.


I'm wondering if the right match of acoustics can create a wider sweet spot. I'm experimenting with arrays to either narrow the dispersion to reduce room interaction, or widen to get better coverage then see which one works best with this type of ambiopole setup. I had decent results with a wider sweet spot by placing the speakers so they were touching, then only spreading out the front of the speakers so they form a V shape with the back edges still touching forming a 15deg angle between them.
Back to top
View user's profile Send private message
mosc
Site Admin


Joined: Jan 31, 2003
Posts: 17557
Location: Allentown, PA
Audio files: 123
G2 patch files: 60

PostPosted: Wed Jul 30, 2008 11:19 am    Post subject: Reply with quote  Mark this post and the followings unread

I, of course, used Synthedit for this VST but I think SynthMaker might be a good program. It seems to be able to work at the sample level and can include feedback. But I don't care for the feedback.

I don't know how one can provide 10 microsecond delays with a 44100 Hz clock. All these programs have settings that take variable values, but what you get is something different.

Experimenting with speaker placement is very productive.

_________________
--Howard
my music and other stuff
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
durwood



Joined: Apr 25, 2008
Posts: 30
Location: Chicagoland

PostPosted: Wed Jul 30, 2008 11:35 am    Post subject: Reply with quote  Mark this post and the followings unread

My best guess is that it must upsample within the plugin to a higher frequency to be able to perform at that resolution and then downsample back to the host frequency. I'm not 100% sure, but IIRC impulse measurements I measured seemed to correlate even with small adjustments-but I honestly wasn't looking that close so don't hold me to that just yet. I wonder if someone at Voxengo could sheld light on it.

Using your block diagram, I was able to string together your setup using Flipper (inverter VST) and Voxengo Delays with some gain adjustments too, and on these crappy computer speakers at low volumes in my office, it sounds similar to your all in one VST. I'll have to get a better feel for this when I return home.

I'm starting to wonder if the feedback is even needed too. There is a a fairly detailed paper on similar experiments back in the mid 80's.

http://www.xlrtechs.com/dbkeele.com/papers.htm

See paper#18

The Effects of Interaural Crosstalk on Stereo Reproduction and Minimizing Interaural Crosstalk in Nearfield Monitoring by the Use of a Physical Barrier," Presented at the 81st Convention of the Audio Engineering Society, (Nov. 1986).
Back to top
View user's profile Send private message
DrJustice



Joined: Sep 13, 2004
Posts: 2031
Location: Morokulien
Audio files: 3

PostPosted: Wed Jul 30, 2008 12:03 pm    Post subject: Reply with quote  Mark this post and the followings unread

(re. sub sample period delays)
durwood wrote:
My best guess is that it must upsample within the plugin to a higher frequency to be able to perform at that resolution and then downsample back to the host frequency.

That is one method. I use four times oversampling in the yet to be published Chameleon version of the MAP, which gives 5.2µS resolution. Another possibility is using real fractional delay lines. They are a crucial part of physical modelling. These can be implemented using a variety of methods, usually involving a bandlimited interpolation filter to obtain sample values that lie between the actual acquired samples. There are many papers and discussions on this around the net.

DJ
--
Back to top
View user's profile Send private message Visit poster's website
mosc
Site Admin


Joined: Jan 31, 2003
Posts: 17557
Location: Allentown, PA
Audio files: 123
G2 patch files: 60

PostPosted: Wed Jul 30, 2008 1:23 pm    Post subject: Reply with quote  Mark this post and the followings unread

Durwood: Great post, thanks.

The link to paper 18 is outstanding. As you know, that's what the purpose of the ambiophonic processors is all about; to replace the physical barrier with an signal processor. That paper really shows and explains in detail the problem with cross talk in stereophonic sound.

I haven't done very much theoretical analysis of this, but a great deal of listening and experimenting. I built an extremely flexible processor on the Clavia G2 - a superb DSP testbed with multiple DSPs. Having weekly discussions with Robin Miller and listening to all of his experiments helped too. Very Happy

Maybe these processors shouldn't be called ambiophonic processors, but crosstalk canceling processors.

As for the delay times, in the crosstalk cancelers there seems to be no need for minute accuracy. Three or four 23 microsecond samples seems to do the trick. Even longer delays could be used but no advantage is gained.

_________________
--Howard
my music and other stuff
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
x_x



Joined: May 05, 2008
Posts: 215
Location: mother earth
Audio files: 4

PostPosted: Wed Jul 30, 2008 4:35 pm    Post subject: Reply with quote  Mark this post and the followings unread

It works on cubase sx alright, I need some proper speakers to try it out though.
Back to top
View user's profile Send private message
mosc
Site Admin


Joined: Jan 31, 2003
Posts: 17557
Location: Allentown, PA
Audio files: 123
G2 patch files: 60

PostPosted: Thu Jul 31, 2008 11:30 am    Post subject: Reply with quote  Mark this post and the followings unread

Keele article sited by Durwood above (part 1 page 12) wrote:
In a construction article which appeared in Radio-
Electronics, Cohen [3] described a "Stereo Image Expander"
which, by minimizing interaural crosstalk, would expand the
stereo image. A delayed and frequency-contoured stereo
difference signal (R-L) is fed into a phase inverter,
whereupon the output, L-R and R-L, is added to the right and
left channels, respectively.

One problem that occurs with the electronic methods of
cancelling crosstalk is that each delayed signal added to
cancel existing crosstalk generates its own crosstalk, with
its own added delay. For the above mentioned Cohen article,
six signals are actually received at each ear: the original
three [R + L(delta T) - R(delta T)] plus their crosstalk
equivalents. This cancelling process can go on in this way,
ad infinitum. Only direct methods that block the crosstalk
at the source can get around this problem.

The Cohen reference turns out to be:
    J. Cohen,"Stereo Image Expander," Radio Electronics, June 1982.

I can not find this article on the internet.

The algorithm that Keele refers to is very similar to, but not exactly the same as, my MAP algorithm. The signs of the L(delta T) and R(delta T) are reversed. Maybe that was in fact the original Cohen algorithm, maybe there was a typo in Keele's paper, or perhaps I don't understand this passage.

In any case it's interesting to note that Keele was aware of the problem of the canceling process going on indefinitely. It is just that fact that was the motivation for the RACE algorithm. To me, the fact that the recursive (IIR) part of the RACE algorithm causes distortion isn't surprising. What is surprising to me is that only one delay in the cancellation signal works so well.

My guess is that this is because the ear/brain is not a purely mathematical audio detection system. It is a psycho-acoustic system that is very adaptable. I think the well know precedence effect, http://en.wikipedia.org/wiki/Haas_effect , is playing a big part here. My characteristically fuzzy thinking goes: that in the same way the ear/brain doesn't find the negative effects of the inter-aural crosstalk in spaced stereo quite as bad as the math predicts, the simple single-delay crosstalk canceler is made more effective because the single canceling signal is all that is needed to clue the ear/brain to more effectively "tune out" the undesired cross talk.

Another detail with my algorithm is that the cencellation signal is a bit louder than it should be to exactly cancel the crosstalk signal. I imagine this tiny loudness increase tells the ear/brain to give the cancellation signal precedence. (It's amazing how much smarter my ear/brain is than I am. Wink )

_________________
--Howard
my music and other stuff
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
durwood



Joined: Apr 25, 2008
Posts: 30
Location: Chicagoland

PostPosted: Thu Jul 31, 2008 11:48 am    Post subject: Reply with quote  Mark this post and the followings unread

maybe we should call your plugin a 1st order ambiophonic crosstalk cancellation plugin. ???

It would be intersting then to try to create a 2nd cancellation addition, then a third, and so on and so on...

Then analyze each order and do some statisical analysis subjective tests to see what the test subjects perfer.

Perhaps, the ambiophonics institute tried this I don't know. I do know this, there appears to be no loss of low frequency with your plugin/model as there is in the RACE or even the BACCH. I think most people who might have tired the older versions might find this one easier to digest at first. So this is Cool

I'm curious to see what changes with the RACE version G since it sound like he plans on addressing the "coloration" issue.

I compared my multiple plugin following your model and it's nearly identical as far as I can tell subjectively. You can use the adjustable time delay (~50microseconds-150microseconds) to adjust to head diameter (~16cm-22cm) and/or azimuth speaker placement angle 5-15deg.

Is your width control an attenuation or a timing adjustment of the crosstalk signals? I'm guessing attenuation?
Back to top
View user's profile Send private message
kkissinger



Joined: Mar 28, 2006
Posts: 1212
Location: Kansas City, Mo USA
Audio files: 29

PostPosted: Thu Jul 31, 2008 12:52 pm    Post subject: Reply with quote  Mark this post and the followings unread

mosc wrote:
As for as widening the sweet spot. That is the holy grail of ambiophonics, after correcting for the timbre modification and the coherent signal attenuation.

I have been experimenting with a 4 speaker implementation that is quite good, IMHO. I'm not ready to post about it because I'm still experimenting. The sweet spot is still on axis, but the sound is "good" all over.


What would happen if the left and right speakers are to the SIDE -- rather than left/right front? How would this affect your algorithm?

_________________
-- Kevin
http://kevinkissinger.com
Back to top
View user's profile Send private message Visit poster's website
mosc
Site Admin


Joined: Jan 31, 2003
Posts: 17557
Location: Allentown, PA
Audio files: 123
G2 patch files: 60

PostPosted: Thu Jul 31, 2008 1:25 pm    Post subject: Reply with quote  Mark this post and the followings unread

durwood wrote:

Is your width control an attenuation or a timing adjustment of the crosstalk signals? I'm guessing attenuation?


Yes, it just mixes in the delayed cancellation signals.

It's great that you have been able to reproduce the algorithm by other means.

Would you let us know how the delay time effects speaker separation and head width?

I've found the if the speakers are closer together than the width of a head it reduces the effectiveness, and if they are wider than 20 degrees the same thing happens.

_________________
--Howard
my music and other stuff
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
mosc
Site Admin


Joined: Jan 31, 2003
Posts: 17557
Location: Allentown, PA
Audio files: 123
G2 patch files: 60

PostPosted: Thu Jul 31, 2008 1:36 pm    Post subject: Reply with quote  Mark this post and the followings unread

kkissinger wrote:
What would happen if the left and right speakers are to the SIDE -- rather than left/right front? How would this affect your algorithm?


If you are talking about the 2 speaker playback - putting the speakers on the side wouldn't effect the algorithm but it would affect the effectiveness. It works best when the 2 speakers are up front and very close.

In most stereo recordings, the most important sounds come from the front center, or close to it. IMHO, the most important thing in a good stereo playback system is the quality of the center and sounds close to center. If the speakers were at the sides, then the center image would be a phantom like in spaced stereo. If the delays could be adjusted properly to cancel out the crosstalk it would at best sound like headphones - great imaging, but it sounds like it is inside one's head.

The ambiophonic processor with close set front speakers gives a very beautiful front stage and an ambient field that surrounds. Very Happy

If you are taking about a 4 speaker set up, I'd say it would be best to keep all of the speakers in line with the 2 in the center.

_________________
--Howard
my music and other stuff
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
kkissinger



Joined: Mar 28, 2006
Posts: 1212
Location: Kansas City, Mo USA
Audio files: 29

PostPosted: Thu Jul 31, 2008 1:46 pm    Post subject: Reply with quote  Mark this post and the followings unread

mosc wrote:
If you are taking about a 4 speaker set up, I'd say it would be best to keep all of the speakers in line with the 2 in the center.


Indeed, I was referring to a setup with an ambiopole in the center and left and right speakers to the sides.

Is it possible to add the left and right speakers to the drawing you did above? I'm curious to see how they fit (mathematically) into the setup.

_________________
-- Kevin
http://kevinkissinger.com
Back to top
View user's profile Send private message Visit poster's website
mosc
Site Admin


Joined: Jan 31, 2003
Posts: 17557
Location: Allentown, PA
Audio files: 123
G2 patch files: 60

PostPosted: Thu Jul 31, 2008 2:05 pm    Post subject: Reply with quote  Mark this post and the followings unread

durwood wrote:

It would be intersting then to try to create a 2nd cancellation addition, then a third, and so on and so on...

Then analyze each order and do some statisical analysis subjective tests to see what the test subjects perfer.

Perhaps, the ambiophonics institute tried this I don't know.


Ambiophonics Institute, I think I'm a member, if there is such a thing. Smile

Robin Miller built such a contraption in Audio Mulch with 15 delay sections. I help perform listening tests with this just about every week. The RACE algorithm came out of this. By adjusting the feedback control on a RACE processor, the attenuation of each successive reflection is increased.

My opinion; adding feedback does in fact increase the apparent width of the field, but it adds peak filtering (at the feedback frequency) and a reverberation sound. I personally find that the t recursion adds a "reverberant field" to the sound, and the extremes of the front stage are indeed wider than a 1st order processor (to use your term witch is dead on IMHO). But, while the width of the stage is wider, the sounds on the extreme left and right are, to me at least, in the same position as in the 1st order system! The recursion smears (distorts spatially) the right and left images; that extends the apparent width of the field. It's not a good trade off, IMHO.

Again, it appears to me that once you tell the ear/brain to ignore the crosstalk signals, it doesn't need and almost infinite reminder. I really should try to explain it because this is all speculation. I can only report on what I hear.

A very critical test for me is mono material. The processor should do absolutely nothing to the signal. MAP does not affect a mono (coherent) signal. The closely spaced front speakers are superb for mono compared to spaced stereo.

Quote:
I do know this, there appears to be no loss of low frequency with your plugin/model as there is in the RACE or even the BACCH. I think most people who might have tired the older versions might find this one easier to digest at first. So this is Cool

I'm curious to see what changes with the RACE version G since it sound like he plans on addressing the "coloration" issue.

I compared my multiple plugin following your model and it's nearly identical as far as I can tell subjectively. You can use the adjustable time delay (~50microseconds-150microseconds) to adjust to head diameter (~16cm-22cm) and/or azimuth speaker placement angle 5-15deg.

Is your width control an attenuation or a timing adjustment of the crosstalk signals? I'm guessing attenuation?

_________________
--Howard
my music and other stuff
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
Display posts from previous:   
Post new topic   Reply to topic
Page 1 of 2 [50 Posts]
View unread posts
View new posts in the last week
Goto page: 1, 2 Next
Mark the topic unread :: View previous topic :: View next topic
 Forum index » How-tos » Ambiophonic Sound Reproduction
Jump to:  

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum
emSynth

Please support our site. If you click through and buy from
our affiliate partners, we earn a small commission.


Forum with support of Syndicator RSS
Powered by phpBB © 2001, 2005 phpBB Group
Copyright © 2003 through 2009 by electro-music.com - Conditions Of Use