Ambiophonics - An experiment and How to...experience Realism [Archive] - DIY Mobile Audio

durwood

12-18-2007, 11:26 AM

Part I

A little experiment - First Try this. I'm not going to go into any details on what this is. I suggest you first just try the experiment, then go back and read about it. But if you must know what this is all about it's all here:

www.ambiophonics.org

What you need
-Pair of stereo speakers
-2 channel amp/receiver
-computer connected to receiver/amplifier or burn the pre-processed files to CD
-your ears

Place your speakers at a distance of 50cm apart from each other. Your listening position should be about 2m back from them so that they form ~+/-20deg angle between the line straight in front of you and the speaker.

Like this (but without the barrier)
http://www.ambiophonics.org/images/NewFig4.Gif

Then listen to the “left right width test” from this link (http://www.ambiophonics.org/Ambiofiles.htm) and find the spot at which you notice the widest stage width. For more setup info:

Setup & Calibration
Ambiophonics is optimized for one or two listeners on the medial plane of a pair of front speakers separated only about 50~75cm positioned at ear level about 2~3m from the listener(s). Choueiri CXC is optimized for separation of 49.5 cm positioned 1.9 m from the listener(s). Speakers should be matched pairs, known to have good phase response through any crossover networks. Correct speaker positioning, avoiding acoustical interference, and calibrating levels and timing are essential for successful crosstalk cancellation.
Your ears are the ultimate instruments of perceived quality. But first, calibrate speaker SPL using an SPL meter ($30 at Radio Shack) and this link to band-limited pink noise. Burn this file, the Left-Right Width Test, and any desired demonstration files above to an audio CD. With no crosstalk cancellation processing, the player set for loop repeat of the pink noise, and SPL meter fixed at the listening position, adjust each channel individually using level controls at the power amplifier inputs so that the signal (-18dB below digital Full Scale) produces precisely 85dB SPL from each speaker individually.
Still with no crosstalk cancellation processing, play pre-cancelled music excerpts above to verify speaker and listener positioning. Now with crosstalk cancellation processing, play pink noise through only the left input of the system and, if necessary, adjust speaker separation slightly relative to a fixed listening position so that at the listening position the sound is perceived at the extreme left only. Repeat iteratively using the right input only (for extreme right perception) until the widest stage width is achieved, up to 120° - double stereo's 60° - but with none of stereo's "hole in the middle" or distortion of tone color, especially for important central sounds.

I suggest you also listen to the other music files also on that page after you get good results with the width test.
http://www.ambiophonics.org/Ambiofiles.htm

Now I will list what you need to try it with all sorts of music.

What you Need

-Computer connected to a 2 channel stereo system.
-Foobar player
http://www.foobar2000.org/

-Foobar Convolver
http://www.foobar2000.org/components/foo_convolve.zip

-Ambiophonics Impulse File(s)-Method A or B.

Method A
Recursive Ambiophonic Crosstalk Eliminator (RACE)
http://www.ambiophonics.org/files/RACE/IR-RACE%5E15_2048sa2x2_46.wav
or
TacT RCS2.2XP in mode A-1 using settings 80us and "75" spread
http://www.ambiophonics.org/files/RACE/IR_A-1_080us-3dB_2x2.wav

Method B
44.1Khz
Choueiri Crosstalk Cancellation - CXC
http://www.ambiophonics.org/files/Choueiri/EYCv2Full.wav

durwood

12-18-2007, 11:26 AM

Part II - How to do this in real-time to any music you already listen to.(setup without pictures)

If you need screenshots I could provide later, but I hope this will make enough sense you can get it setup properly.

1) Install foobar
2) Extract/Unzip foobar convolver plugin to the foobar program folder under “components”
3) Run foobar
4) Under “File” menu goto “preferences”, then click on the “playback” drop down
5) Go to “DSP manager” under the “playback” drop down
6) Select “convolver” from the “available dsp” screen and double click on it to bring it into the “active dsp”
7) Highlight “convolver” and click on “configure selected”
8) Now go to the “impulse file” field and browse to find your impulse files you should have downloaded from the links above, either method A or method B.
9) Adjust the level by about -4db or so or use auto level adjust
10) Click ok and close all the dsp windows. Foobar is now ready to play any file you load into it and process it as an ambiophonic sound instead of stereo.

Added: You might need to swap your left and right channels. For some reason I had to do this on mine based on instruments being located on the wrong sides.

Play around with this and I'm curious to know what others think.

durwood

12-18-2007, 08:20 PM

Did I mention the stage width will literally kill stereo's stage width and the sense of "being there" in the room with the instruments and singers is greater than stereo too?

coffee_junkee

12-18-2007, 08:42 PM

Subscribed.

Weekend project! Too bad you couldn't have made this thread last week because I was snowed in all weekend. Also without internet, but anyway.. :D

durwood

12-18-2007, 09:02 PM

LOL. That's exactly why I posted it....because I was snowed in all weekend and had time to play.:D

Also did I mention a much larger sweet spot than stereo too?

durwood

12-18-2007, 09:43 PM

I like using the impulse file from method B for Eagles - Hotel California (live)

durwood

12-20-2007, 07:58 PM

More possible good news, this technique might just dismiss amplifier and cable effects because whatever differences there are, are eliminated during this methodology.

Not proven yet, but it might even level the playing field when you compare tube amps to transistor based amps.:o

Might work very well with underdash horns for you pro-audio driver people.:)

thehatedguy

12-20-2007, 09:28 PM

I doubt it on the tube/transistor thing.

The thing about this is you have to use specially burned CDs.

Not proven yet, but it might even level the playing field when you compare tube amps to transistor based amps.:o

Might work very well with underdash horns for you pro-audio driver people.:)

kappa546

12-20-2007, 09:39 PM

The thing about this is you have to use specially burned CDs.

doesn't seem like much to ask for...

thehatedguy

12-20-2007, 09:54 PM

It is if you want to use them in competition...:).

kappa546

12-20-2007, 10:25 PM

It is if you want to use them in competition...:).

hahaha i knew you would say that. i had it after the "..." but then decided i'd let you say it. anyways, the fact is most of us here don't compete. and even though you can't use them during competitions, you can still enjoy them outside the lanes.

Abmolech

12-20-2007, 11:07 PM

what area is this??:confused:

Oh thats right car audio competitions..:p :mad:

durwood

12-21-2007, 06:54 AM

I doubt it on the tube/transistor thing.

Like I said...it might. Ralph gives a small explanation on why, but specifically states nothing has been proven yet. He was only stating it based on theory. I know, it's not something you tube guys want to fathom.:p

The thing about this is you have to use specially burned CDs.

It is if you want to use them in competition...:).

Easy solution. Carputer as your source or processor.;) No special CD's needed.

Or you can build a physical barrier/absorber. If my understanding is correct all you need is the following equation. (I think Anthony Davis pointed this out ot me-focus on chapter 8 and abmolech reinforced it).

L=X(H+T)-D

L=distance from head to barrier (center console)
X=distance from barrier(console) to speaker
D=distance between center of speakers
H=distance between the ears (~6-7inches-depends on if you have a big head:D)
T=thickness of barrier (console)

use all the same units. It's algebra, so if most of your parameters are fixed, solve for the one you can vary or play with.

durwood

12-27-2007, 12:46 PM

After further investigation and experimentation, there is a problem with the way foobar's convolver works. I had some issues with how things sounded so I decided to try the software that was recommended on http://www.ambiophonics.org/Ambiofiles.htm

I wouldn't recommend trying the foobar method yet with your own music. Feel free to follow their guides and use their software although it's a bit more complex it yields very good results. I'll see if I can put together another easier method to try this out.

I doubt anyone tried this yet anyway, at least the first part.:(

pontiacbird

12-27-2007, 12:49 PM

thins isn't going to work at all.....you forgot one thing....

the flux capacitor!!!!! :p

interesting read, keep it coming!

durwood

12-27-2007, 12:59 PM

great scott! That's heavy doc. Where the hell are we gonna get 1.21 Jigawatts (Gigawatts)!:D

Ya I was all psyched about this but I thought something was weird like too much reverb and some phase cancelation....that was until I loaded up all their software and tired it their way. Then I was :)

Like I said first part is easy, speakers ~50cm apart (20inches) you sit back 2m ~ 6ft. Burn all these files to a disc and listen.

http://www.ambiophonics.org/files/Demos/OneTwo_RACE.wav
http://www.ambiophonics.org/files/Demos/Li%27lDavid_RACE.wav
http://www.ambiophonics.org/files/Demos/Handel_RACE.wav
http://www.ambiophonics.org/files/Demos/Parade_RACE.wav
http://www.ambiophonics.org/files/Demos/Lunchbreak_RACE.wav

durwood

12-29-2007, 12:05 PM

First 10 people to EMAIL me at hawbaked (at) msoe . edu will get a CD of their choice (Both I and you must own the original) processed ambiophonically that can be played back on any CD player as long as you setup your speakers properly as suggested in my first post.

Please be sure to give me your DIYMA username.

Format the subject DIYMA - [username] - ambiophonic CD.

Whiterabbit

12-29-2007, 09:22 PM

my speakers are 20 inches wide. Do I place them 2 inches apart or right next to each other? (cone centers 20 inches apart.)

Think itll still work without hte use of teeny bookshelves?

Whiterabbit

12-29-2007, 09:34 PM

pwnt by pat

12-30-2007, 12:02 AM

Can the carputer be set up to do the processing in real time, so that any ordinary cd can be used?

durwood

12-30-2007, 09:16 AM

im not sure what Im supposed to hear. rather than move the drivers closer I moved my position backwards so the speakers were at a 20 degree angle to my position as specified.

What exactly am I listening for again? I dont hear anything different or revolutionary.

If your L and R speaker cabinets have 20 inches of spacing between them, then you need to sit directly between them at about 78 inches. Or if you are sitting closer, then that spacing needs to be adjusted. (spacing = 2 * listening spot distance * tan(10deg)) or use ~7.5deg. You might need to move forward and backwards until you hear the widest stage possible when listening to the "width one two " track. Then after you find the sweet spot, listen to the other tracks. The stage width should be close to 120deg and you should have more depth, and everything SHOULD sound more real. Some people like it some don't. I think it just takes a while for you to adjust since we are so used to the stereo listening triangle. The effects are more apparent if you can compare it against real music you have become used to and can flip back and forth between stereo and ambiophonic listening.

Can the carputer be set up to do the processing in real time, so that any ordinary cd can be used?

Yes. You can use Console (or audiomulch) + farina Xvolver plugin and those impulse files. Did you get your soundcard issue worked out with console yet?

I've got 3 people on the list so far. 7 more spots open.

pwnt by pat

12-30-2007, 10:39 AM

Yes. You can use Console (or audiomulch) + farina Xvolver plugin and those impulse files. Did you get your soundcard issue worked out with console yet?

I've got 3 people on the list so far. 7 more spots open.

awesome. Gives me a reason to set up a second carputer, which all I'm missing is a sound card, and turn my second car into an audio car. Not yet on the sound card.

edit: does this technique take multiple listening positions into affect. Like, you say the listening triangle is 120 degrees. Does that mean that the effect works anywhere in that 120 degrees?

fr125s's are pretty cheap and would probably make a great startup, with a sub in the glovebox. :) Sorry, this isn't the car audio section.

durwood

12-30-2007, 10:50 AM

awesome. Gives me a reason to set up a second carputer, which all I'm missing is a sound card, and turn my second car into an audio car. Not yet on the sound card.

fr125s's are pretty cheap and would probably make a great startup, with a sub in the glovebox. :) Sorry, this isn't the car audio section.

ya, I'm thinking a small array under the dash with the peerless 2" or 3" drivers in maybe a combo ported and waveguide app.

pwnt by pat

12-30-2007, 11:21 AM

did you see my edit?

edit: does this technique take multiple listening positions into affect. Like, you say the listening triangle is 120 degrees. Does that mean that the effect works anywhere in that 120 degrees?

I would think minimilistic would work best for this. Although damn you for giving me ideas. I've already got an awesome dash idea set up in my head and my daily driver isn't even done yet.

Oh and I JUST figured out teh carputer. Here's a hint:
Don't run more than one instance of allocator at one time.

or not... just some way I had FA set up that was causing issues. Got two FA's loaded and working just fine :)

pwnt by pat

12-30-2007, 11:27 AM

I wanted to put this in it's own reply since it's not off topic.

All I can say is WOW. I can certainly tell a difference. ABing definately makes it more clear, and it's not psychoacoustical, at all (well I guess the technique is, but not what I'm hearing). The image is definately more center focused due to the close proximity of the speakers. In my room, Logitech z680 speakers, put in a corner, the best location for me was ~11 inches apart sitting about 8 feet back. up close, the crosstalk barrier is definately noticeable with a very.... wierd sound. But back in the listening area, it's bliss. Vocals perfect center (as they should be). Localizing instruments was suprisingly easy. It also felt as even in my improper listening enviroment, my listening room got a LOT bigger. It felt as I could feel the reflections caused by the recording studios walls.

I'm using my car computer (in my room) to test using a M-Audio 1010lt, console, frequency allocator, and eycv2full impuse file.

thehatedguy

12-30-2007, 04:48 PM

pwnt by pat

12-30-2007, 08:11 PM

expect ~500 + cost of the screen and sound card + am/fm tuner if that's your thing. I use the M-Audio 1010LT (135-150 on ebay). 8 analog in/out, coax digital spdif in/out

Winamp will run VST's and you can run this VST through there. If you're not doing any other audio processing on the carputer, that cuts software costs a lot. If you do want to, you can do anything on the carputer that any professional mixer can do.

I'm going to try to convert my buddy to doing this. Maybe two small full ranges in the center front of the dash up against the windshield? We'll see. I have the spare carputer parts here, minus the sound card. :) If you wanna pick up the parts, I'd be more than happy to sell, too.

GreekPower

12-30-2007, 10:29 PM

just came from pwnt by pats house w/e he did boy does it sound good. 300+this trick thing= awesome

excuse my noobness haha

thehatedguy

12-30-2007, 10:42 PM

I just wonder if the sound card can match the DACs and sound quality I get with the Rane. I dunno about a $150 sound card sounding better than a $2k proaudio processor.

But then again, I don't think you could do all of the room correction stuff if you did not use a sound card.

expect ~500 + cost of the screen and sound card + am/fm tuner if that's your thing. I use the M-Audio 1010LT (135-150 on ebay). 8 analog in/out, coax digital spdif in/out

Winamp will run VST's and you can run this VST through there. If you're not doing any other audio processing on the carputer, that cuts software costs a lot. If you do want to, you can do anything on the carputer that any professional mixer can do.

I'm going to try to convert my buddy to doing this. Maybe two small full ranges in the center front of the dash up against the windshield? We'll see. I have the spare carputer parts here, minus the sound card. :) If you wanna pick up the parts, I'd be more than happy to sell, too.

~thematt~

12-30-2007, 11:01 PM

You dont need a soundcard if your processor has digital inputs. Why you would use an external processor though, and not a straight DAC, is another question (because the Carputer is running all the processing)

Apogee DAC has Firewire800 inputs, and I think the Behringer or the Dolby Lake has USB2(?). Cant speak for the Rane. Most processors/DAC's these days come with Optic anyhow.

Whiterabbit

12-31-2007, 12:22 AM

Se7en

12-31-2007, 12:32 AM

durwood

12-31-2007, 01:37 AM

has anyone gotten this to work in the house yet for two listeners? sitting approximately 2.5-3 feet apart.

The instructions said it could work for one to two listeners. In my house the position was far too sensitive to the results to be workable for two listeners (driver and passenger).

Unfortunately, it probably will not work for a driver and passenger. It still suffers from a sweet spot like stereo, however they suggest it would work for two people with the second person positioned right behind the other as long as the first person continued the "barrier" for crosstalk. I think and this is a big I think, Vector based amplitude panning would allow for a more forgiving listening position but I know very little about how that works.

This is sort of why I posted this in the other/home audio, because it may not suit some of you two seater people.

Abmolech would tell you a single center channel (decently sized line array with multiple drivers) would allow multiple listening positions (which I can understand), but I'm not sure what he was doing on the processing end to spread out the stage if anything at all.

thehatedguy

12-31-2007, 01:37 AM

I have one digital input.

Better question is why would you use an external DAC? You would need one for each set of outputs...so for a 3 way setup, you would need 3 DACs. And then you would need some way to do volume control post DAC.

The Rane runs AES/3 digital inputs since it is a proaudio processor.

You could use the external processor for the ambiophonics, but not for the room correction stuff.

You dont need a soundcard if your processor has digital inputs. Why you would use an external processor though, and not a straight DAC, is another question (because the Carputer is running all the processing)

Apogee DAC has Firewire800 inputs, and I think the Behringer or the Dolby Lake has USB2(?). Cant speak for the Rane. Most processors/DAC's these days come with Optic anyhow.

durwood

12-31-2007, 01:43 AM

So, how much would a good carputer run to do something like this? And can you get computers or sound cards with digital coax outputs (gotta feed my Rane processor).

I don't think I can get my speakers that close together in the car...20" is almost as wide as the transmission tunned.

I have to hear this!

Apogee makes some excellent DACs from what I have heard. USB DACs lack the proper clocking to reduce jitter from recent posts I have read. Spdif is still better at lower jitter ratings because it has a better clock. Maybe you can get some type of "word clock" connect between a computer and the Rane, I don't know.

For a good carputer setup, AMD BE2350 processor, motherboard and DC-DC PSU (M2-ATX) won't set you back too much. That would be the route I would take if I decided my older P4 mobile wasn't enough.

durwood

12-31-2007, 01:47 AM

I have one digital input.

Better question is why would you use an external DAC? You would need one for each set of outputs...so for a 3 way setup, you would need 3 DACs. And then you would need some way to do volume control post DAC.

The Rane runs AES/3 digital inputs since it is a proaudio processor.

You could use the external processor for the ambiophonics, but not for the room correction stuff.

Maybe you could do the ambiophonics AND room correction in the carpc processor and then go two channel via digital out to your RANE, or just do everything in the carpc right to the amps...balanced even.

www.rme-audio.com makes some very nice pro-type soundcard interfaces.

thehatedguy

12-31-2007, 02:40 AM

backwoods

12-31-2007, 09:45 AM

durwood

12-31-2007, 11:40 AM

I'm still messing around with my pc as a processor. Been playing with it the last few days while I've been stuck in the house with this freaking cold.

Rane sounds mucho nicer. The PC has some background noises. For comparison, it reminds of the behringer pieces as far as sound quality.

To double check, I threw on my headphones, and checked analog outs and still was a little disappointed.

What soundcard are you using? Onboard is a pile of crap usually.

pwnt by pat

12-31-2007, 12:21 PM

after playing with the test stage width and location files, I've found ~24 inches and 8ft to be pretty good. I'm having some issues though. It seems, and I don't know which part of the chain it is, but I seem to have a slight right bias, which I think is tied to issue 2. I can get left stage extremely wide, but right is much compacteed. Left stage extends about five feet past the speaker, but right is only about two-three.

backwoods

12-31-2007, 12:32 PM

What soundcard are you using? Onboard is a pile of crap usually.

first tried using my soundblaster audigy x2, then tried through onboard.

I have an maudio card I'll try, and am waiting for the lynx card to arrive

durwood

12-31-2007, 01:58 PM

first tried using my soundblaster audigy x2, then tried through onboard.

I have an maudio card I'll try, and am waiting for the lynx card to arrive

Audigy cards resample everything to 48Khz and onboard usually has poor components or inexpensive parts and if both use windows Kmixer, then that impacts quality as well.

after playing with the test stage width and location files, I've found ~24 inches and 8ft to be pretty good. I'm having some issues though. It seems, and I don't know which part of the chain it is, but I seem to have a slight right bias, which I think is tied to issue 2. I can get left stage extremely wide, but right is much compacteed. Left stage extends about five feet past the speaker, but right is only about two-three.

Ok, I had similar issues and not sure why yet. I didn't follow their instructions completely by using the bandlimited pink noise and measuring each channel at 85dB and adjust gains/balance accordingly, so maybe that's why. Other than that I have no other idea. I just shift myself slightly to the right and that kind of takes care of it. They did also suggest to make sure you have the setup placed equal distant from the sidewalls, because reflections might skew the results. Ideally they want your room to be fairly dead and absorbant.

~thematt~

12-31-2007, 07:01 PM

At that point the Rane would be redundant and only could be used for EQing and volume control.

And going to the DAC to 2 channel mode post carputer would be pointless since you need a DAC for each channel of output, or you couldn't do time delay per speaker...only per side.
The Rane would be redundant, would it! :P

You can quite easily use something like the E-MU for cheap money, or the Apogee DA16x or Rosetta 800 for, well, not so cheap money, for full DAC requirements. Multiple 2-channel DAC units WILL NOT WORK because of the jitter clock independence of each one.

Carputer with Room Correction, linear Xovers, linear EQs (graphic AND para), phase modifiers, and potentially more processing power then any output processor = Outboard Processor redundant (and cheaper) :)

If you wanna keep the Rane, the Carputer will relegate it to a large and weildy DAC and volume control unit.

I've been investigating Ambiphonics since Ambolech raised it on our local forums a while back. Next step up? Quadraphonics. Better sound, for cheaper. Certainly puts the car 'stereo' community on its head.

durwood

01-03-2008, 08:10 AM

I'm changing my tune a bit. For those I have talked to already, I should have the discs out by the end of this week. For the remaining 7 slots, I have decided to just give you 3 discs of tracks of my choice, 1 disc will be the original stereo recording unaltered. Second disc will be all the same tracks but in Ambiophonics Method A with default parameters, and third disc will be same tracks but using method B.

I'd love to do customized demo discs for everyone, but it's more time consuming and I wish I had more time. So this is the best I could do for now.

email @ hawbaked (at) msoe dot edu

thehatedguy

01-06-2008, 01:50 AM

Abmolech

01-06-2008, 02:14 AM

It is very encouraging to hear you guys giving this a go.
So many people have dismissed this without trial, so you guys are some of the brave few.

Roger Waters uses ambiophonics and sometimes quadraphonics at his live shows, so this medium is very capable of producing a very large sweet spot.

I suggest we have been proving the limitations of stereo for a goodly number of years now. Please give this medium some degree of time and experimentation to achieve the best from it.

Enjoy your new adventure.:)

durwood

01-06-2008, 09:22 AM

Being the computer genius I am, I just realized I could do a quick listen to the wav files using my PC and computer speakers...just click to open and play. When you are in the right spot, things are pretty interesting- like width 2' past where the speakers are physically located. But finding that spot with these little POS computer speakers is tough and sound like poo....

Hurray me and technology!:D

Excellent! :You get a atta-boy and gold star for the day.:D

Discs are finally done. I wanted them done Thursday, then Friday, then Saturday but life > me time. They will be going out Monday for all the brave souls. :)

I took a few demo discs to tweeter since they have some better setups than my aiwa stereo setup in my computer office.:blush: The results were even better there and had a few nay-sayers pretty impressed.:) the width stretched almost to the edges of the room, and it was a pretty wide room.

Ge0

01-13-2008, 01:46 AM

Thank you Durwood for taking the time to provide these sample discs. This saved me much time, effort, and admittedly confusion of processing my own. It also opened my eyes to new possibilities.

For those of you who don't know me that well, I'm another guy out to create the ultimate I can achieve in SQ. Only I do this regardless of what I have been taught/learned in the past. Like most though, I have my limitations. Budget and how far I am willing to go with my daily driver (vehicle) are obviously factors.

I've been trying to beat the death out of stereo playback for years to achieve sonic bliss. I have been successful to some extent but have always felt there must be more. So, without further babbling, here it is.

I set up my Ambiophonics demo on my computer sound system. Built in PC sound card EMU chipset with Klipsch Promedia surround system. Surround speakers are set to replicate the fronts for 2 channel music and provide 4.1 (no center) operation for multichannel formats. Surrounds were turned off for this experiment.

I started by selecting a few tracks I had interest in amongst Durwoods selections for A/B comparions stereo vs Ambiophonic. I burned his disks to my hard drive for quick A/B study. Speakers set 20" apart and me about 6ft back. Sorry for the lack of accuracy...

What I found. It took a while to set up correctly but once I did. WOW ambiophonic processing make a huge difference! For a while there I though my surrounds were still active. I had to place my ears up to them to find out for sure. They were not.

Ambiophonoic method A (what does this mean?):
Super wide soundstage extending well beyond the speakers. The soundstage seemed set way back and exenteded well beyond the speakers. By how much, maybe 4 ft? The soundstage seemed like a wall in front of me. Center vocals were rock solid but seemed a little weak and deep. The soundstage extension seemed to be biased more left than right. Left extended more. But, as Durwood mentioned, if I position my butt a little differently I could probably correct this.

Ambiophonic method B (again, huh?): Similar results to method A only the soundstage seemed to wrap around me more vs. laid out more flat in front of me like method A. Same left side bias for stage width. Left was more extended. Perhaps I did something wrong?

Stereo: Well what can I say. Suffered from the same problem that has existed for me for years. The soundstage hardly extended beyond the speakers. However, to its defense, the speakers were only placed 20" apart. Later, once I put my speakers back into position the sound stage widenend, but not much past the boundaries of the speakers. What stereo does have going for it? The center image was superior. Much more prominent, albiet, closer to my listening position vs the Ambiophonic method. What would I prefer? Prolly a wider soundstage. That was hard to dismiss. It gave me wood.

My conclusion: I love the width of Ambio but prefer the center image I get with stereo. If I could do some sort of manipulation to get the wide stage of Abio mixed with the clarity of the stereo center image I would be VERY happy. Not totally satisfied (will I ever be?) but much better off than what I have.

The golden question? Can this successfully be implemented in a daily driver vehicle without ripping it up to a serious extent? I am more than willing to place a PC in the signal stream for real time procesing but the limitation on speaker placement may be a show stopper.

Hope this helps. I HIGHLY encourage others to try it. It only takes an evening (including the listening session) for God sakes!!!

EDIT: Tried moving slightly closer to the speakers thismorning and shifted towards the right. I did receive a nice balanced image which may even have been a little wider.

Ge0

Abmolech

01-13-2008, 02:50 AM

You could always do a VNC ambiophonics setup.:)
(Voice in centre)

I would consider this unnecessary, you need to run through the setup procedure to reduce the left bias etc.

You should be able to improve your stereo setup, by moving them away from the walls, and using a smaller equilateral triangle.

In car?

Depends a bit on the car.
IF you have a large centre console, it is possible to use the "kicks" on the centre console. You could try the dash, with a suitable array.

Possibly the best car options would consist of no or limited centre console, however I perceive Little reason to limit yourself to them. For a van interior, ambiophonics must provide a realistic solution compared to stereo.

If you have space use the console kicks as a waveguide (continue the dash to the firewall), and if you wish to reduce floor to roof reflections in the listening positions, use vertical arrays in the centre console kicks.

Have a think about arrays and waveguides and how they can be used to control dispersion AWAY from problem areas... hint hint.
Also how waveguides turn early reflections into an advantage (acoustic loading), rather than a problem.

It is quite feasible with the combination of ambiophonics, waveguides and arrays to have little or no problems with early reflections, late reflections in the listening position, and most of all the dreaded acoustic crosstalk in the intensity sensitive frequencies.

If you wish for ambiance (I suggest it is worthwhile) it is possible to add a rear ambiophonics setup for ...yep
quadraphonics.:D (food for thought)

Hope this helps. I HIGHLY encourage others to try it. It only takes an evening (including the listening session) for God sakes!!!

So true.

Again I appeal to your sense of fair play, and give yourself time to squeeze the most out of ambiophonics.

durwood

01-13-2008, 03:44 PM

Ge0

01-13-2008, 09:45 PM

GEO thanks for taking your time to try this.:)

I'm not sure about the bias thing but I have noticed this suffers from "the sweet spot" just as stereo does. If you move around back and forth along the dividing line between the speakers, you will find the spot at which your stage is the widest. YOu might also have to move yourself a tad to the left or right. Do either of these while listening to tracks 1 or 2. Once you find that sweet spot, then go on to listen to the other tracks.

You're welcome. I eventually found the sweet spot (after posting my review). I edited my thread to reflect this. I hope to do some critcial listening of your test tracks throughout the week before I decide if I want to go any further with this. It does show promise, but may not be for my vehicle due to speaker placement challenges.

Once I found the sweet spot, I thought the center image was even more focused than stereo, but that was my take on it.

When I say it was not as focused in the conventional sense. The image did not smear or wander. It just did not seem to be as rich tonally. If this makes any sense, it also seemed a little transparent vs opaque.

I had better results in a larger room, I think reflections were slightly skewing the results when I was in a smaller room. I had to go to Tweeter to try it though.

Well, I have my car. Not a large room, nor any room for that matter. My wife and kids took over my home theater setup. It's either Thomas the Train, Bob the Builder, Dancing with the starz, dating shows, etc... I have NO time available at home to sit back and kick up the jams.

To get the most out of ambiophonic playback, you almost need to be in a heavily damped room. If this is not possible, we have three solutions-

1) Room correction
2) waveguides
3) Tons of sound control/dampening products (but in a car you have highly reflective windows).:(

Again, I posted this in other audio, as to it's completely new territory for car audio (or even home audio for that matter).

Yes you did post this in "other audio". However I'm constantly in search of new things to try in my car. Might as well see if the application fits.

Please elaborate on waveguides a little. What exactly would you hope to accomplish? Focusing the wave front at the listener to minimize reflections through some type of horn structure?

I've discussed this with abmolech, and for us in the US, it's going to be hard to use the the center console kick area because we are left hand drive. However, I plan on using the dash area instead, which will also meet the criteria of a waveguide.

First question. What difference does it make if you sit on the right side or left? The vehicle is symetric. You would surely have the same problem on either side. I picture mounting drivers under the dash. One above the right leg, one above the left. Do this on drivers and passengers side so each listener has a decent stage.

What DOES constitute the criteria of a waveguide anyway?

I'm also lucky enough that I already have a carputer AND the processing in place to attempt this in a car. Let's just say I was impressed enough with this playback method that I'm going to gamble and try it in the car. ;)

I've been meaning to talk to you about a basic carputer. However, I'm not sure if that conversation would be better off on DIYMA or your other hangout MP3CAR.

Ge0

BassBaller5

01-14-2008, 03:15 PM

So i got my discs in today!! thanks man.

To start off, the recording quality of all the songs is phenominal.

I started by putting the "just for fun" disc into my truck on the way home and while my car setup is deffinatly not optimal for this experiment (20 degree axis listening position,) i was still able to notice the large sweet spot/ extended sound stage.

I was deffinatly blown away.

Getting back to my dorm and my very crappy altec lansing desktop speakers, i attmpted tweaking a little more. I setup the speaker positions as listed and first listened to the tracks in their original format since i already had most of them (eagles/dire straits) on my computer.

I then proceded to play the ambiophonics disk...

I actually had to open my eyes and look around for the extra speaker a foot outside my normal stage! I had a firm defination of where the instruments were located and man was that spot sweeeet.

I should be making a run up to our local hi end audio shop later to try this out on some of their products where i will hopefully notice the major results!

Thanks again Darin.

EDIT: im still not really sure what im supposed to be looking for in the differences between method A and method B?

durwood

01-14-2008, 04:52 PM

You're welcome. I eventually found the sweet spot (after posting my review). I edited my thread to reflect this. I hope to do some critcial listening of your test tracks throughout the week before I decide if I want to go any further with this. It does show promise, but may not be for my vehicle due to speaker placement challenges.

Understandable. You've already been playing with rear fill and using that to control crosstalk, so at least you have options.

When I say it was not as focused in the conventional sense. The image did not smear or wander. It just did not seem to be as rich tonally. If this makes any sense, it also seemed a little transparent vs opaque.

Ah, gotcha. That is kind of the "you-are-there" vs the "they-are-here" effect. It's one of those things you may or may not get used to.

Well, I have my car. Not a large room, nor any room for that matter. My wife and kids took over my home theater setup. It's either Thomas the Train, Bob the Builder, Dancing with the starz, dating shows, etc... I have NO time available at home to sit back and kick up the jams.

No worries. I was merely stating that as an observation I found. I don;t really have any larger rooms in my house so I'm a bit limited as well.

Please elaborate on waveguides a little. What exactly would you hope to accomplish? Focusing the wave front at the listener to minimize reflections through some type of horn structure?

I'm a n00b when it comes to waveguides, but we aren't necessarily trying to get a horn, we just want to "guide" the sound where we want it, and keep it away from bad things that wreck havok. Once the sound leaves the "guide", anything can happen to the wave. Ambiophonics is supposed to be best played back in a room with very short reverb times so it doesn't affect the controlling of the crosstalk and such. Waveguides and/or room correction and dampening helps this.

First question. What difference does it make if you sit on the right side or left? The vehicle is symetric. You would surely have the same problem on either side. I picture mounting drivers under the dash. One above the right leg, one above the left. Do this on drivers and passengers side so each listener has a decent stage.

This was my original idea and it could satisfy the placement criteria. You will still have to time align. Again, I have not attempted this in a car or sitting off center and time aligning, but in theory it should be less of a problem with comb filtering than "stereo" or panned mono. This location however is only about half a waveguide (dash) but nothing on the floorside.

What DOES constitute the criteria of a waveguide anyway?

The best way I would describe it is to continue the natural curve of the cone. That would be ideal, but in most cases it might not be what we have to work with. This is probably one of abmolechs favorite subjects and might deserve it's own thread. Here is a useful link he gave me. Also, trueaudio.com and steridians (http://www.trueaudio.com/st_spcs1.htm). We don't want to quite be in horn territory unless you like the nasal sound of them, so one step down in steridians is a good way to go.

http://sound.westhost.com/articles/waveguides1.htm

I've been meaning to talk to you about a basic carputer. However, I'm not sure if that conversation would be better off on DIYMA or your other hangout MP3CAR.

I'll send you a PM. Maybe I'll fire up AIM or something-haven't used it in years.

EDIT: im still not really sure what im supposed to be looking for in the differences between method A and method B?

No problem, I'm glad you could experience it, it's better than always trying to explain things that don't follow "the norm". As far as what to listen for between A and B, I can't say for sure. To me, method A seemed like the ratio of direct to reflected energy was quite high, whereas B seemed a little less reflected energy and more direct energy-you could call it reverb or ambiance possibly? I'll have to see if I can find a better comparison/description of the two.

For more details on RACE/Method A ->
http://www.ambiophonics.org/files/RACE/RG-RACE_AES123NYC0710.pdf
http://www.ambiophonics.org/files/RACE/RGRM-RACE_rev.pdf

Method B is supposed to be an improvement on RACE developed by Edgar Choueiri.

thehatedguy

01-14-2008, 07:46 PM

Got the CDs but haven't had a chance to listen yet.:(

Weightless

01-14-2008, 09:08 PM

I got mine in the other day. Unfortunately I have been swamped at work, so I haven't tried it out yet.

I will post when I do though.

durwood

01-14-2008, 10:14 PM

Excellent.

Enjoy!

Update, my second post in this thread was a howto play any file ambiophonically, however it had some flaws and didn't work right. But after a little more digging, I have a new way...that actually works and sounds similar to the discs I sent out. Now if I could only edit my second post, then everyone with a computer and a set of speakers can try this with ANY music they own.

EDIT: OOps, jumped the gun again, my new method that I thought worked is ambiosonics, not ambiophonics which is similar but not the same. Maybe that will be another experiment on it's own-which could be good because it builds on "stereo playback" and doesn't require you to move your speakers.

BassBaller5

01-14-2008, 10:24 PM

This entire theory is really interesting me however im a little slow at understanding it....

so i ordered this used for 5 bucks! :D

Ambiophonics: Beyond Surround Sound to Virtual Sonic Reality"
Ralph Glasgal; Paperback; $4.99

-ill let yall know how it turns out!

durwood

01-14-2008, 10:35 PM

real quick take away-cliff notes version.

-Your ears are direction finders above 1.5Khz.
-"Stereo playback" has to create a ghost or phantom image between the loudspeakers.
-For frequencies above 1.5Khz, a centered sound must be sent from both L and R equally to appear in the center.
-However, something inside your brains still knows it's coming from the L AND R even though it sounds centered. This is bad, if it's really supposed to be centered, like on a stage.
-Solution, launch it from center stage, and allow the ambiance captured during recording to determine which direction it comes from. Your ears and brain will be much happier. :)

I printed the entire site (same as your book) and read it page by page (sometimes more than once). It's good "office" material-that's where I do all my thinking.:D

BassBaller5

01-14-2008, 10:45 PM

Couple questions for you...

-Your ears are direction finders above 1.5Khz.

Then why does crossing a sub to a higher frequency (say 70Hz and above) allow us to precieve rear "localized" bass?

- launch it from center stage, and allow the ambiance captured during recording to determine which direction it comes from.

how does this special ambiophonics recording achieve a "launch from center stage?"

-I printed the entire site (same as your book) and read it page by page (sometimes more than once). It's good "office" material-that's where I do all my thinking.:D

Thats exactly where all my new audio literature are headed! :D

durwood

01-14-2008, 10:55 PM

Then why does crossing a sub to a higher frequency (say 70Hz and above) allow us to precieve rear "localized" bass?

Well that's a subject on it's own. New research is looking into "interaural phase difference" (IPD). Want more "office" material?

http://www.filmaker.com/papers/RM-2SW_AES119NYC.pdf

how does this special ambiophonics recording achieve a "launch from center stage?"

It's not so much in the recording, the recording is designed to fit the playback system-which is where your speakers are placed in relationship to you the listener.

Thats exactly where all my new audio literature are headed! :D
:)

BassBaller5

01-14-2008, 11:16 PM

Thanks man! im always down for some reading material! that just looks like ill have to read that one a few times to understand it all.

So im understand what we need to correct, i just dont understand how the ambiophonics cd's you sent me can correct for your brain "de-centering" the stage?

durwood

01-14-2008, 11:20 PM

When I processed the music, it was ran through a crosstalk cancelation filter. You maintain this crosstalk separation (only on the processed CD's) by keeping the speakers close together.

BassBaller5

01-14-2008, 11:32 PM

Thanks for clearing all that up man. Thanks again for the disks. deffinatly worth researching into

Ge0

01-15-2008, 09:31 AM

Understandable. You've already been playing with rear fill and using that to control crosstalk, so at least you have options.

I still don't understand how the rears "control" crosstalk. Could you dumb this down a notch and attempt to explain in laymans terms? Once I get the general understanding of things, reading the technical stuff seems to come much easier to me. But, If I don't have a clue about what I'm supposed to be reading then it is a fruitless effort.

Ah, gotcha. That is kind of the "you-are-there" vs the "they-are-here" effect. It's one of those things you may or may not get used to.

That pretty much sums it up. However, I've noticed something else. On some tracks vocals seem set back in the stage vs. the rest of the musical ensemble. Hotel California may be a good one to emphasize this point. Also a track on the same disc with a female vocalist I was not familiar with. I don't think I've ever been to a show where the lead vocalists was hiding in the back of the stage. They are usually up front and center so they can be idolized :). But, this might just be me and my perception. Either that, or I have done something wrong.

I'll send you a PM. Maybe I'll fire up AIM or something-haven't used it in years.

HAH!!! Me too. Someone asks me what my ICQ, AIM, etc... is. Ughhhh... I'll need to get back with you on that one :).

You know, Like BassBalls (:)) I have been toying around with this in my vehicle on my commute to/from work. Some of the stage cues are off. But, I'm wondering if some of the massive amount of tweaking I've done to get PANNED MONOPHONIC, NOT STEREO to sound good aren't screwing with it (like that one Abmolech?). I wonder if I create a profile in my Zappy software for ambiophonic records if I couldn't improve upon this and make it pleasureable to listen to as is with no drastic rip up of loudspeakers and their locations.

Ge0

durwood

01-15-2008, 10:29 AM

I still don't understand how the rears "control" crosstalk. Could you dumb this down a notch and attempt to explain in laymans terms? Once I get the general understanding of things, reading the technical stuff seems to come much easier to me. But, If I don't have a clue about what I'm supposed to be reading then it is a fruitless effort.

when you run bandpassed rears, and use a L-R or R-L setup, you are essentially canceling out the stuff that combine in the front stage between L and R (crosstalk). Add a little delay to that and now you've added ambiance or feel of space, too much and you get echo (must be somewhere between 20-30ms-read Haas effect). This affects the comb filtering that is a result of the HRTF (head related transfer function). Sorry, I hope that dumbed it down enough.

That pretty much sums it up. However, I've noticed something else. On some tracks vocals seem set back in the stage vs. the rest of the musical ensemble. Hotel California may be a good one to emphasize this point. Also a track on the same disc with a female vocalist I was not familiar with. I don't think I've ever been to a show where the lead vocalists was hiding in the back of the stage. They are usually up front and center so they can be idolized :). But, this might just be me and my perception. Either that, or I have done something wrong.

No, I'd have to agree. I'm not sure why it happens yet and Method B seems to suffer from that more. It could be recording specific or maybe I needed to tweak something, or maybe method B is best used with a pair of rears as well. Method A had more parameters I could play with whereas method B was pretty set in stone. I'm currently researching other ways to process it that allow for more adjustability. Either way, at least you are experiencing 3 demensional sound vs 2D, you can tell there is depth. :)

HAH!!! Me too. Someone asks me what my ICQ, AIM, etc... is. Ughhhh... I'll need to get back with you on that one :).

I'm getting busy and not sure how much time I'll have but I'll do my best to help out. :) Future wife is going to have my head on a platter if I don't help her with some planning soon. :o

You know, Like BassBalls (:)) I have been toying around with this in my vehicle on my commute to/from work. Some of the stage cues are off. But, I'm wondering if some of the massive amount of tweaking I've done to get PANNED MONOPHONIC, NOT STEREO to sound good aren't screwing with it (like that one Abmolech?). I wonder if I create a profile in my Zappy software for ambiophonic records if I couldn't improve upon this and make it pleasureable to listen to as is with no drastic rip up of loudspeakers and their locations.

I think if I had included some stuff processed ambiosonically (not ambiophonic but similar), you might be able to playback on your existing setup without any tweaking. I'm playing with this stuff too, but just recently. That would have been good to include with the packages but I missed the boat on that one.

Ge0

01-15-2008, 03:04 PM

when you run bandpassed rears, and use a L-R or R-L setup, you are essentially canceling out the stuff that combine in the front stage between L and R (crosstalk). Add a little delay to that and now you've added ambiance or feel of space, too much and you get echo (must be somewhere between 20-30ms-read Haas effect).

OK, I get this part and fully understand.

This affects the comb filtering that is a result of the HRTF (head related transfer function).

This is where the gaping hole in my understanding lies. Affects the comb filtering in the frontal wavefront, or, in the wavefront coming from behind? By eliminating the phantom center created using crosstalk in the rear speakers I no longer have a phantom center in the rear. But, how does this enhance the front?

I'm getting busy and not sure how much time I'll have but I'll do my best to help out. :) Future wife is going to have my head on a platter if I don't help her with some planning soon. :o

Oh, crap!!! I don't want to contribute to your marital strife even BEFORE marriage!!! You'll get enough of that after 10 years :).

I was just thinking that I would like to assemble a small PC to be inserted in the signal chain between my head unit and amps. I could do some general EQ type filtering, as well as, try some of these sound processing algorithms you and Abmolech have been talking about in real time. You know, apply processing to any music I throw at it... In addition, Abmolech sent me over to the DOOM9 forums just to find a whole slew of new stuff I could try. Including a number of surround formats derived from 2 channel stereo.

I think if I had included some stuff processed ambiosonically (not ambiophonic but similar), you might be able to playback on your existing setup without any tweaking. I'm playing with this stuff too, but just recently. That would have been good to include with the packages but I missed the boat on that one.

Uhhhh... I guess I'm getting lazy with my reading. I did not realize there was a difference. Ambiophonic, Ambiosonic, whatever it takes :).

Ge0

durwood

01-15-2008, 03:21 PM

thehatedguy

01-15-2008, 03:58 PM

I need to know about the carputer stuff...all of it.:o

durwood

01-15-2008, 09:09 PM

I need to know about the carputer stuff...all of it.:o

Is this because of the demo discs or because of 384db/octave filters...did I mention that plugin is free? Not sure what happens at that steep of a slope but I'm sure there has to be some ringing, afterall the plugin is called rubberfilter.

I also found the ambiosonics plugin that will enhance a normal car speaker setup, which is also free. Not quite ambiophonics but it's an improvement over typical "stereo".

Ge0

01-15-2008, 09:57 PM

When reading some papers and posts, trust me that whole ambiosonic/ambiophonic wording messed with my head. They are different. If I had an easy explanation of the differences, I would post it. But I'm going to admit this is new territory for myself and until I make heads or tails of the differences I'm goign to leave it alone. Doom9 focuses on ambiosonics as it relates to surround sound. I found a few posts on ambiophonics over there, but the links for the VST plugins are dead or gone. :( So now I'm stuck to play with ambiosonics in addition to this.:o

What I am starting to think is that perhaps a simulated surround sound derived from a 2 channel source might be on to something. This experiment I have been doing with rear fill, delay, difference signal has really amounted to a big improvement in perceived ambience. I was wondering if one of these simulated surround algorithms might beat what I have already.

for info on comb filtering, there is a good tech article by werewolf on phase vs time delay over at ECA. You can also search google and find a wealth of info on HTRF.

I'll check out what Wearwolf has to say.

ADDED: if you need help on a carpc used as a processor, feel free to email me at my email rather than my PM box, which is kind of full at the moment. The key to sucessful audio processing on the PC is the right soundcard and a VST host. I can recommend a few, since the list is quite short. I just found a rockin new crossover filter plugin that can do 384db/oct brickwall filters. :o

I'll put together a list of specs. Once I have this I'll contact you. This may take a while so I shouldn't need to bother you too much in your "busy" period of your life (I know how that one goes, wait until you have a few kids :)). In the mean time I'll browse through MP3car to get some ideas.

Ge0

durwood

01-15-2008, 10:11 PM

I need to know about the carputer stuff...all of it.:o

This will get you started in the audio tuning plugins area. I recently made some updates, but I need to add more info and maybe some more organization.

http://www.mp3car.com/wiki/index.php/Audio_Tuning_via_Software

What I am starting to think is that perhaps a simulated surround sound derived from a 2 channel source might be on to something. This experiment I have been doing with rear fill, delay, difference signal has really amounted to a big improvement in perceived ambience. I was wondering if one of these simulated surround algorithms might beat what I have already.

Very possible. This V.I. plugin I've been playing with I think could do it. Did you find the Stereo to 5.1 Guide list at doom9? It's massive. Tons of info.

I'll put together a list of specs. Once I have this I'll contact you. This may take a while so I shouldn't need to bother you too much in your "busy" period of your life (I know how that one goes, wait until you have a few kids :)). In the mean time I'll browse through MP3car to get some ideas.

If I were starting from scratch, Either duo core or AMD BE2350 processor. Then go from there. AMD is going to be your best bang for the buck and if you can fit a microATX board in your car, then go that route instead of paying extra for the smallness factor of some of the other form factors-only if you have the space though.

durwood

01-17-2008, 01:46 PM

Alright, upon some more research, I found one awesome site with animations showing wave animations of different setups. Maybe it will help visualize, and see what happens to the sound in each case.

For the entire list of animations check it..

http://www.isvr.soton.ac.uk/FDAG/VAP/

Here is mono

http://www.isvr.soton.ac.uk/FDAG/VAP/images/ani16.gif

Here is typical "stereo" with a 45 degree source of sound

http://www.isvr.soton.ac.uk/FDAG/VAP/images/anivir60.gif

Now here is ambiophonics with a 45 degree source of sound

http://www.isvr.soton.ac.uk/FDAG/VAP/images/anivir10.gif

Here's one for abmolech. :D

http://www.isvr.soton.ac.uk/FDAG/VAP/images/single_time_fast.gif

Ge0

01-17-2008, 02:56 PM

Durwoods animation post really help shed light on things for me. I always excelled in lab based classes vs pure theory since I could actually SEE something and fugger out how it worked. Lets see if all this acoustical theory mumbo jumbo is sinking in...

Stereo USES acoustical cross talk to create a phantom image. The region of optimum crosstalk (or the stereo sweet spot) is small. If you are sitting outside the sweet spot you are not hearing the stereo effect. THUS, the reason Abmolech talks until he is blue in the face about why we are not hearing stereo reproduction in a car. It is impossible to be in the sweet spot since we are sitting too close to the speakers. That, and we are sitting MUCH to close to one side. We are in the wrong location!!! We attempt to apply bandaids to balance left and right pathlength diffrences through time delay, BUT, one thing still goes uncorrected. In a conventional car we are still SITTING TOO CLOSE! I assume there are ways around this by chopping your vehcile all to shit to drastically change speaker locations. Look at the NASCAR Mark Elderidge built up. However, this is not practical for us hobbyists who drive daily drivers. I also suspect that proper crosstalk is compromised in a vehcile due to the massive amounts of reflections going on in the listening environment. But, I won't overstep my bounds for the time being.

Other, more modern, means of playback such as Ambiophonics, Stereo Dipole, etc... DO NOT USE the complex wavefronts created by accoustical crosstalk. As a matter of fact, they try to avoid crosstalk altogether. These schemes place the loudspeakers much closer together so the individual wavefronts eminating from each speaker combine and propegate as one big wavefront not nearly as complex as the wavefront created by stereo acoustical crosstalk. These methods of playback mathematically alter the signal routed to left and right speakers to position virtual images within the sound field?

Getting close?

Do I get a cookie:) or a beating:(.

Ge0

durwood

01-17-2008, 05:15 PM

Damn Geo. You can have tomorrow off for that answer. :D

I'd like to add one more tidbit about stereo and equal path lengths. Everyone always talks about equal path lengths to the drivers but never talks about the spacing between L & R. The spacing is equally as important because stereo relies on perfect crosstalk. This is why you hear the big dogs moan and groan about time delay/time alignment. Funny thing is, they never moan and groan about crosstalk. Then you hear them groan about one seat vehicles and claim you need equal pathlengths to have a two seat car.:( Wrong again, you still have a lopsided listening triangle with a sweet spot. Wouldn't it be nice if there were a way that you could still have you're lopsided triangle, sit too close (nearfield listening) AND have a better sounding system (although) still suffer from crosstalk but at least take the HRTF into effect? (*cough* ambiosonics and/or Vector Based Amplitude Panning *cough*). That deserves a whole other thread though. If you want an "all seat" car, nothign will beat mono, we just have to find a way to correctly process a non-mono recording, but let's get back to the topic I'm getting off track.

As you can see from the animations and what Geo hit upon, stereo relies on crosstalk yet look at what it does to sound right around the head compared to mono or even ambiophonics. It's very messy looking no? Anyone care to disagree?

I didn't post up the crosstalk animations but those are definitely good to check out too.

Ge0

01-17-2008, 09:22 PM

Wouldn't it be nice if there were a way that you could still have you're lopsided triangle, sit too close (nearfield listening) AND have a better sounding system (although) still suffer from crosstalk but at least take the HRTF into effect? (*cough* ambiosonics and/or Vector Based Amplitude Panning *cough*). That deserves a whole other thread though.

I remember while setting the demo up that ambiosonics was pretty touchy in relation to distance to the speakers, somewhat on my position in between them, and the distance between them. Are you sure about this statement?

I would think that you at least need to sit back far enough for the combined L/R wavefront to form. Isn't this the region where their spatial processing works? I found when listening to ambiosonics that if you sit to close, the ultra wide soundstage collapses rapidly.

Ge0

Abmolech

01-17-2008, 11:21 PM

You can adjust ambiophonics to allow for nearfield listening. (hint also use waveguides and arrays:) )

Welcome to the darkside.:D

Knowledge is power...IE the force.:p

Ge0

01-18-2008, 12:28 AM

You can adjust ambiophonics to allow for nearfield listening. (hint also use waveguides and arrays:) )

Welcome to the darkside.:D

Knowledge is power...IE the force.:p

I know this is the "other" DIY forum. However seeing how this community is generally geared towards car audio, and, this is my primary interest... Please elaborate.

Now that I have begun the process of enlightenment, I am EXTREMELY curious. How can Ambiophonics be adjusted for near field listening? Give an example of such a wave guide that you so highly recommend time and time again. An array might be out of the question in my automobile (http://www.diymobileaudio.com/forum/showpost.php?p=311307&postcount=272) :).

"Welcome to the Darkside" Well, I'm Glad to be here.

Ge0

durwood

01-18-2008, 09:02 AM

I remember while setting the demo up that ambiosonics was pretty touchy in relation to distance to the speakers, somewhat on my position in between them, and the distance between them. Are you sure about this statement?

I would think that you at least need to sit back far enough for the combined L/R wavefront to form. Isn't this the region where their spatial processing works? I found when listening to ambiosonics that if you sit to close, the ultra wide soundstage collapses rapidly.

ha-ha! I tricked you! No seriously though, ambiosonics does not equal ambiphonics. If it helps to differentiate so the wording isn't confusing, we will call ambiophonics a (2 channel dipole, some like to call it a "Stereo" dipole, but I'm going to avoid that word too hopefully for obvious reasons.)

VBAP (Vector Based amplitude panning) is a version of ambiosonics where you can literally tell the program where your speakers are via listening angle and elevation angle and it will correct timing and amplitude and other things. It's very cool and it also controls crosstalk but in a different way if my understanding is correct. I didn't mean to confuse since this topic was about ambiophonics. But that is a step up from "stereo", and ambiophonics is a step up from ambiosonics which is a step up from standard surround sound, 5.1/PLII.

If we were concerned about accuracy, the order would go.

1) Mono
2) Ambiophonics /Quadraphonics
3) Ambiosonics/VBAP
4) Typical Surround Sound
5) Stereo

Yep stereo at the bottom. Mono is not viable option at the moment, unless we can find a way to keep phase info intact from "stereo" recordings. I.e. you will lose all staging info.
ambiosonics and VBAP utlizes the HRTF.
ambiophonics relies on crosstalk canelation and control of crosstalk and extracts ambiance from the recording to transport you to another space.

That's my understanding of the two.

I know this is the "other" DIY forum. However seeing how this community is generally geared towards car audio, and, this is my primary interest... Please elaborate.

Now that I have begun the process of enlightenment, I am EXTREMELY curious. How can Ambiophonics be adjusted for near field listening? Give an example of such a wave guide that you so highly recommend time and time again. An array might be out of the question in my automobile (http://www.diymobileaudio.com/forum/showpost.php?p=311307&postcount=272) :).

"Welcome to the Darkside" Well, I'm Glad to be here.

Ge0

Continue the flow of the cone or speaker = waveguide. Is that a good definition? Look at the way the cone curves out. Here is a picture of what I plan to do. The windsheild and dash become the waveguide. This is a great plus...why...because I no longer have to fight with dash or windsheild reflections, they will help me now. In theory, they just became part of the speaker. :)

http://i191.photobucket.com/albums/z174/durwood79/install/IMG_2431.jpg

http://i191.photobucket.com/albums/z174/durwood79/install/IMG_2432.jpg

durwood

01-18-2008, 09:13 AM

One might consider the fact that a great recording company acknowledges this playback technique in their library. They might actually make recordings that work very well with this method. ;)

http://www.chesky.com/core/body_library.cfm
http://www.chesky.com/core/body_librarydetails.cfm?newsid=180

chad

01-18-2008, 09:26 AM

durwood

01-18-2008, 09:48 AM

Nice find chad! ;)

Geo-
If you are concerned about "the sweet spot", go to the website I linked that had those animations above and go to the "downloads" section, they have several wav demo files of their own. If you look at their readme link, you can space the speakers even closer together so that you can sit closer to the speakers (nearfield i.e. a car).

http://www.isvr.soton.ac.uk/FDAG/VAP/html/readme.html

Chad's link shows a pair of speakers that could almost be consider a single mono speaker, except you will notice the blocker between them and also read the description. The listening distance can range anywhere from 3-11ft :)

chad

01-18-2008, 10:09 AM

Chad's link shows a pair of speakers that could almost be consider a single mono speaker, except you will notice the blocker between them and also read the description. The listening distance can range anywhere from 3-11ft :)

Did you see the block diagram for the processing?

durwood

01-18-2008, 11:23 AM

Kind of, it's hard to see a few things but yes, I would say it's the same principle but it sounds like they added a few other nice compensation things as well after reading through. I'd liek to know what the symbol is on the left channel in the middle though.

I just went to their website www.xvisionaudio.com and they are gone or changed names so I can't read up on their technology, but everything on that page theory wise is almost exactly like what is discussed on the ambiophonics page.

Compare the two:

http://mixguides.com/studiomonitors/Reviews/Flow-E-Sound.WEB.jpg

Ambiophonics

RACE
http://www.ambiophonics.org/files/RACE/RM-RACE_F%2Bmidi.amh.jpg

Choueiri Crosstalk Cancellation - CXC using the Farina X-volver

http://www.ambiophonics.org/files/Choueiri/Mulch-XTC4x4_sm.jpg

I'd say it's probably closer to the CXC.

Weightless

01-18-2008, 01:03 PM

Continue the flow of the cone or speaker = waveguide. Is that a good definition? Look at the way the cone curves out. Here is a picture of what I plan to do. The windsheild and dash become the waveguide. This is a great plus...why...because I no longer have to fight with dash or windsheild reflections, they will help me now. In theory, they just became part of the speaker. :)

http://i191.photobucket.com/albums/z174/durwood79/install/IMG_2431.jpg

http://i191.photobucket.com/albums/z174/durwood79/install/IMG_2432.jpg

I take it those two drivers will be for the left channel and you are going to do a matching set for the right channel? I recognize the mid(TG9), but not the tweeter.What driver is it?

This is really interesting indeed. Unfortunately I am so busy with work, I haven't had the chance to demo this...

When do you plan on building the pods(?) for them?

Questions, questions, questions....

durwood

01-18-2008, 02:25 PM

Yep, the 3" was a TG9 and the little guy was actually an older Eclipse 2" center channel speaker I had lying around. I used them for reference when I was taking pictures and trying to decide what would fit up there. I'm actually not using either of those drivers.:p

My plan right now is to use at least two of the new peerless 2" drivers per side. I might increase this to as much as 4 per side or 3 per side or just add my LPG tweeters if they don't reach up far enough. I've got 8 channels of processing power to figure out what's the best use of channels.

Scans will remain either in the doors or be relocated to the kick panels to utilize the kickpanel as a waveguide.

I'm still in the planning stages, but I need temporarily test the TG9's on the dash to make sure this works in the car first, then I'll commit all the way. When I do commit, it's going to be tricky trying to figure out how to mount my "soundbar" on the dash with cutting or drilling any holes. Expect to see a thread asking for ideas in the future. For now I'm banking on attaching it to the a-pillars, who knows how well that will work. Unfortunately, this isn't going to probably happen until June if all goes well. :(

Weightless

01-18-2008, 02:56 PM

Will it be a horizontal array? Such as TMM, TMMM, TMMMM, MMTMM?

I was thinking about doing the same thing but with the Aura Whispers. I was going to mate them up with my LPG softies. YUM!

I will definitely subscribe to your "ideas" thread. I have thought about this a bit, and some ideas.

Can't wait to see(or hear) where this goes...

J~

Ge0

01-18-2008, 04:03 PM

I'm thinking the latest round of discussions in this thread warrant opening its own topic in the car audio discussion forum. What do you guys think? This is starting to go beyond experimenting with it in the home.

Ge0

durwood

01-18-2008, 07:22 PM

Ge0

01-18-2008, 11:47 PM

Durwood,

The Chesky link you provided offered an excellent explaination of the differences between Ambiophonics and Ambiosonics.

http://www.chesky.com/core/body_librarydetails.cfm?newsid=180

Too bad I fell asleep 3 times reading through it :). I can't think straight this late so hang around the forums and write mindless drool. I'll give it another go tomorrow when I am fresh. I understood the basic theory but got lost when the article got analytical.

Thanks a lot for your patience in being a mentor.

Ge0

Ge0

01-22-2008, 04:00 PM

Got the CDs but haven't had a chance to listen yet.:(

Please do and let us know your take on things. I'm still up in the air about this. But, it does show some promise.

BTW: I know you are simply an innocent bystander, but MAN did you start a shit storm over on ECA :).

Ge0

durwood

01-22-2008, 10:16 PM

Please do and let us know your take on things. I'm still up in the air about this. But, it does show some promise.

BTW: I know you are simply an innocent bystander, but MAN did you start a shit storm over on ECA :).

Ge0

Nah it was a great discussion, no excellent discussion. That ranks up there as one of my favorite threads ever. It was enough to convince me to sign up and post over there. :D

Ok, I am starting to thing there is some sort of left side bias. Originally I just thought it was my crappy test rig or my cluttered room, but I'm begining to think there is some issue on one channel with the setup, I have to agree with thehatedguy and you Geo, I think the right side is not extending out as it should. That's going to take some analysis to figure out why.

Ge0

01-22-2008, 11:49 PM

Nah it was a great discussion, no excellent discussion. That ranks up there as one of my favorite threads ever. It was enough to convince me to sign up and post over there. :D.

That WAS a great discussion? For fuck sakes, I've been following it for 3 nights. It is growing MUCH faster than I can read and absorb. It's not over. 25 pages and growing?

Side note: W can be a little harsh in his opinions. I thought A was bad :). I guess I haven't been around here long enough to notice. DAMN. It was sad to see him go, but...

Ok, I am starting to thing there is some sort of left side bias. Originally I just thought it was my crappy test rig or my cluttered room, but I'm begining to think there is some issue on one channel with the setup, I have to agree with thehatedguy and you Geo, I think the right side is not extending out as it should. That's going to take some analysis to figure out why.

Then I am not on crack? I did notice that when I repositioned myself at one point this bias went away (although the stage width shrinked a little). However, this was NOT at the recommended listening position and distance.

Ge0

Abmolech

01-23-2008, 02:41 AM

At least you may understand how large of a leap of faith you are taking.

Actually trying something.:)

werewolf's all good, not the first time we have locked horns..:D

Got give me some credit for keeping him going, how many times did he say "I'm done".. he-he.

Anyway, congratulations on your new adventure.

durwood

01-23-2008, 07:23 AM

That WAS a great discussion? For fuck sakes, I've been following it for 3 nights. It is growing MUCH faster than I can read and absorb. It's not over. 25 pages and growing?

Side note: W can be a little harsh in his opinions. I thought A was bad :). I guess I haven't been around here long enough to notice. DAMN. It was sad to see him go, but...

No it's pretty much done for now until others try ambiophonics and want to discuss that. Most of the thread was a debate over stereo vs mono and then stereo's short comings (in a car).

Then I am not on crack? I did notice that when I repositioned myself at one point this bias went away (although the stage width shrinked a little). However, this was NOT at the recommended listening position and distance.

Ge0

The distances recommended are only guidelines and not the end all be all. I think I found it the bias went away actually when you palce the speakers even closer and you sit closer. At one point I think abmolech recommended placing the speakers together, facing forward, then pivot them keeping the back edge of the cabinets touching and angle then to a 15-20 angle between them. IIRC, this gave me pretty good width even at close distance ( a foot or two). Obviously, the larger you can spread/space the speakers out and sit farther back, you can get a wider stage, but too far and it all colapses. I think the breaking point is 75cm and 3m back, but you need a larger room to keep wall reflections from destroying it.

You asked about creating a driver and passenger listening area, well the thing is that it probably won't work right that way, think of ambiophonics as more of a center monophonic channel. If you don't time correct to account for the offset, then essentially driver and passenger should have a very good stage. If you want to make sure the driver has the best experience, time correction will correct the offset. I tried it in my home.

also, might wnat to read what some others ideas were, just keep in mind their cars are different because they sit on the right side. (You will recognize one or two of the trouble makers) :)

http://www.mobileelectronics.com.au/forums/index.php?showtopic=88744&st=0

At least you may understand how large of a leap of faith you are taking.

Actually trying something.:)

werewolf's all good, not the first time we have locked horns..:D

Got give me some credit for keeping him going, how many times did he say "I'm done".. he-he.

Anyway, congratulations on your new adventure.

Ya you kept him in the discussion, otherwise I'm not sure if it would have gone on for 25 pages. Hey I brought him back in at one point after the first "I'm done".:) I really did read his vector stuff, I need to sit down and draw it out and do some math for it to sink in. It will probably prove useful for understanding VBAP.

durwood

01-23-2008, 10:25 AM

Will it be a horizontal array? Such as TMM, TMMM, TMMMM, MMTMM?

Horizontal Array- Either MM+MM, MMM+MMM, MMMM+MMMM, or MMT + TMM.

Midbass will remain in the doors or move to the kicks.

Ge0

01-23-2008, 10:35 AM

Horizontal Array- Either MM+MM, MMM+MMM, MMMM+MMMM, or MMT + TMM.

Midbass will remain in the doors or move to the kicks.

What frequency could one run a midbass up to before localization effects would start to ruin the illusion the Ambiophonic soundfield has created?

Those little bitty Peerless drivers you mentioned look promissing for experimentation.

Question, no activity in the DIYMA thread regarding this topic. Should we kill it and maintain the conversation here, or, move auto specific discussion over there?

Ge0

durwood

01-23-2008, 10:47 AM

What frequency could one run a midbass up to before localization effects would start to ruin the illusion the Ambiophonic soundfield has created?

Those little bitty Peerless drivers you mentioned look promissing for experimentation.

Question, no activity in the DIYMA thread regarding this topic. Should we kill it and maintain the conversation here, or, move auto specific discussion over there?

Ge0

I'm guessing the same as for stereo. :D Essentially, the longer the wave gets (lower frequency) the less the head can block it, the less a physical barrier can block it, and if my understanding is correct, the crosstalk cancelation in the software is bandpassed somewhere around 300Hz. I'd be happy if I can play the peerless's down to 200Hz but it would be nice if I can push them down lower to 150Hz. I won't know if it's going to be too much distortion or not or too much to ask from the little guys when I want higher output.

Move the discussion to where? ECA? I'd rather stay here. I don't think it will get any more attention over there vs over here and I need to keep track of another forum as much as I need another hole in my head.

durwood

01-23-2008, 11:12 AM

Ge0

01-23-2008, 11:31 AM

Move the discussion to where? ECA? I'd rather stay here. I don't think it will get any more attention over there vs over here and I need to keep track of another forum as much as I need another hole in my head.

ECA, heck no.

This area is DIY Other Audio. We started a thread in DIY Mobile Audio (more relevant to car talk) that has had no activity since it started. I was just pointing out that we seem to be talking a lot about car implementation in the Other Audio forum, not the Mobile Audio forum (both of which are here at DIYMA).

If it doesn't bother anyone then we can leave the discussion right here. Heck, I don't care. I just don't want to be impolite about bringing this thread off topic by discussing car implementation.

Ge0

durwood

01-23-2008, 11:32 AM

ECA, heck no.

This area is DIY Other Audio. We started a thread in DIY Mobile Audio (more relevant to car talk) that has had no activity since it started. I was just pointing out that we seem to be talking a lot about car implementation in the Other Audio forum, not the Mobile Audio forum (both of which are here at DIYMA).

If it doesn't bother anyone then we can leave the discussion right here. Heck, I don't care. I just don't want to be impolite about bringing this thread off topic by discussing car implementation.

Ge0

Ah gotcha. the playback/recording method thread.

Ge0

01-23-2008, 11:38 AM

Another possibilty and totally car dependant, midbass under the dash firing down. Sealing the underside would be near impossible however so this isn't really practical.

A final option is the floor. I have actually seen a car or two resort to this location before even in a stereo setup.

Neither solution is feasible for me.

1.) Under the dash. Well, electronic do-dads on one side get in the way. The HVAC gets in the way onthe other side. Besides, this is where I intend on trying the midrange drivers. I think they are small enough to where I can squeeze them in.

2.) Hole in the floorpan of my vehicle to place a speaker? Hmmm, perhaps not. This is my daily driver. Today I drove through tons of slush and road salts on my way to work. I'm sure a good scheme to seal this off from the elements could be devised. Even if we did lick that problem, there is still the issue of your feet resting above/around the speaker. Say you were not a clutz and never put your foot through a cone. Slush and road salts on your feet could wreak havoc on the driver from the other side.

Perhaps my unwillingness to go to extreme measures is limiting my chances at sonic bliss, but, there has to be some tradeoff I guess.

Ge0

Ge0

01-23-2008, 12:13 PM

You asked about creating a driver and passenger listening area, well the thing is that it probably won't work right that way, think of ambiophonics as more of a center monophonic channel. If you don't time correct to account for the offset, then essentially driver and passenger should have a very good stage. If you want to make sure the driver has the best experience, time correction will correct the offset. I tried it in my home.

I'm wondering if we are on the same page. Take a looksie at the quadrophonic setup here:
http://www.ambiophonics.org/AES%20Munich/images/Fig_20_AES%20Munice.jpg

Now, remove (or keep?) the rear speakers behind each listener. Next, drop the two sets of front speakers down to their knee positions instead of above their heads. Aim the drivers up towards the head. Finally, replace the older ladies with two hot looking models wearing skimpy patent leather bikinis.

Still think it will not work?

I'm still confused on how you intended on trying this. Two (or more if you are doing an array) speakers mounted smack dab in the middle of your dash that would generate the sound field for both listeners?

Ge0

durwood

01-23-2008, 12:28 PM

Yes, replace the woman with hot models.:D Was that on a site somewhere related to acoustics or is that some family vacation picture where you were on the tour bus with your mom and your aunt? :p

No I understand what you want to do. I thought of that myself too at one point. Maybe if you place another x-vovler instance to cancel crosstalk between R1 and L2, maybe it will work. Dunno?

Here is something else to consider if you want to include everyone. Quadraphonics (ambiosonics but for a wider audience) or even VBAP. I don;t know enough about either though...yet.

http://www.isvr.soton.ac.uk/FDAG/VAP/html/multi.htm

http://www.isvr.soton.ac.uk/FDAG/VAP/images/anec3b.jpg
http://www.isvr.soton.ac.uk/FDAG/VAP/images/f41000.gif

I guess this is why I started a playback method thread. To get peoples brains moving and consider other ideas. :)

You know I shouldn't have listed playback methods in order earlier in this thread. I can't say for sure what would be the best for everyone's situation. I'm selfish and only want a one seat car, but I'm open to other ideas.;)

Ge0

01-23-2008, 12:52 PM

I'm selfish and only want a one seat car, but I'm open to other ideas.;)

I agree, I'm selfish too.

However, I have high hopes that if I can give my wife a decent sound stage too, she might understand what I spend countless of hours fucking around with vs. watching mindless TV shows with her. So, to some extent I'd want a two seater.

Ge0

durwood

01-23-2008, 12:57 PM

I agree, I'm selfish too.

However, I have high hopes that if I can give my wife a decent sound stage too, she might understand what I spend countless of hours fucking around with vs. watching mindless TV shows with her. So, to some extent I'd want a two seater.

Ge0

LOL. Maybe I should re-think my selfishness. Maybe all of us should think about that. :D

audioman42

01-23-2008, 03:07 PM

I tried it, but I liked stereo better. Am I doing something wrong?

durwood

01-23-2008, 10:12 PM

I tried it, but I liked stereo better. Am I doing something wrong?

Ignore the foobar part, it doesn't work right. Stick with their demos and instructions. What didn't you like about it. Be more specific please.

pwnt by pat

01-28-2008, 01:11 PM

Durwood, I was thinking of your plans for a horizontal array (and mine). I don't think that would be ideal. Granted not much you can do about it. The issue I'm seeing is the unified wavefront. Having any kind of horizontal array might cause some comb-filtering crosstalk that might confuse the image with regards to stage width.

I think that the only way to reduce the amount of crosstalk would be to keep the speakers at extreme head-angles, which is opposite of what ambiophonics tries to set up. The wider the array is, the worse the crosstalk is going to be.

I think what's going to be ideal is a single driver, or combo woofer/tweeter. The woofer would have to be high-efficiency to achieve "good" sound pressure levels (windows down anyone?). Other than that, the only way to achieve the effect is going to be with vertical mounted arrays. Stage height can be compensated with increasing path length differences, crossfiring or "dummy" tweeters not being affected by the impulse plugin, or just straight playing with 6k+ output levels.

Perhaps instead of on-dash, go kick panel arrays? You could very easily keep a large format tweeter in the dash in the originally planned locations, and do a dual woofer vertical array in the kicks using 3 high efficiency drivers. Both locations would be ideal because of the wave-guide properties of both. The vertical array in the kicks would use both the under-dash AND floor bounce as the waveguide.

Only issue I could see would be phase relationships between tweeter and woofers and that's not going to be bad at all.

durwood

01-28-2008, 01:26 PM

Well, I'm no expert but I do have to test my theory before I can say it will work or won't work. It's uncharted teritory. The way I figure it, 2-3 2" drivers is no wider than a single 6.5"-8" driver and i do have time alignment and ampltiude adjustment. :)

Side kick panels will not work for ambiophonics. It would have to be on the sides of the center console and I don't have room for it.

Then there is another option: one of which I do not fully understand on how to make it work or how to make it fit my idea or if it will work with ambiophonics, but this was posted by abmolech.

http://www.extra.research.philips.com/hera/people/aarts/papers/aar01p.pdf

I'm going to test a couple of things in my car this weekend. I've got a few pairs of 3" drivers I'm going to use to test some things.

pwnt by pat

01-28-2008, 01:33 PM

I'm reading the pdf now.

The solution to "not being able to use kicks" is to sit farther back ;) To get a 20 degree listening angle would only move the seating position back another foot or so, at least in my car. Center-seat or rear bench...

edit:
rear bench ftw.... you can have "driver" and "passenger" and have each persons heads less than 12" from each other... heh

I'm not going to lie, the math is over my head. Sort of the explanation is as well... They took a set of speakers, added another set beside them, added some processing, and wa la? From the looks of it, each left/right woofer combo receives the same processing as it's adjacent. I don't understand how they increased pressure levels as you travel farther away, though? Through crosstalk manipulation?

I just thought of another idea how this effect might be able to be achieved without further processing. This is just an idea... so....

Main draw back to cone woofers is off-axis attenuation. beaming. What would be the effect if in addition to your standard processed main speakers, you keep another set of speakers extremely off-axis and slightly attenuated or in a "cylinder" waveguide (promote beaming). The farther you moved away from the center sweet spot, the louder the off-axis speaker would get. In effect, the increase in loudness should be equal to the increase of moving closer to the speaker on the side you're moving.... That's poorly explained. Move left, left speaker gets louder (as per that whitepaper). Move left, my "beam* speaker gets louder, to "equalize" the left speaker...

durwood

01-28-2008, 01:47 PM

Using my windsheild/dash areas gives me a distance of 43" and still places me inside the boundaries of the ambiophonic center channel. :D I've already played with time delay and level delay in the house and it will correct for a one seater no problem. ;)

Now if I can utlize the proper crossovers, I might be able to cover the entire front seat if not the entire car using the info in that link. That would be a HUGE plus.:)

As of right now, A one-seater is fine for me because that is all I have now. ;)

pwnt by pat

01-28-2008, 01:53 PM

check out my edit.

I was going to say I think this might be interesting because I'm always searching for the two-seat answer ;) I'm trying to approach it from a single center seat and work position outwards, though... :banghead:

Don't you think that the gauge cluster hump on the dash might give you some issues for the passenger, when you move in that direction?

Abmolech

01-28-2008, 02:01 PM

Acoustic crosstalk is caused by drivers out of phase, at a distance.
In this case with a vertical array, it is a matter of keeping the phase correct across the width. time aligment or actual curve (or a combination) rather than a straight should procurr this.

pwnt by pat

01-28-2008, 02:04 PM

perhaps I'm confusing crosstalk with comb filtering or something? Too many fancy terms, and not enough knowledge to conduct these fluid dynamic thought experiments.

durwood

01-28-2008, 02:05 PM

I have to admit, the math gives me nightmares of my college days, but it does not look completely foreign to me. Some where in there I hope to make sense of it an apply it's principles to what I 'm trying to do.

My thought's are use time alginment and the beaming factor to control dispersion. If live concerts/events can pull of arrays and make it work, why can't we?

Are you picturing two V shapes? One on the left formed by the left speakers and one on the right by the right speakers? No quite that drastic but hopefully I understand you and you understand me.

I think the goal in the paper is to use different frequency bands to create the ability to widen the dispersion pattern, but still maintain clarity.

durwood

01-29-2008, 11:49 AM

Almost hate to do this because of the negative attitude towards this company, but this is EXACTLY what I'm planning on doing. Th eonly difference is I won't have the hole in the midbass region. :)

Anyway here is a link that pretty much covers the visual asspect of my concept. Click on the "learn more" link towards the middle right side of the page.

http://www.bose.com/controller?event=VIEW_PRODUCT_PAGE_EVENT&product=321gsx_dvd_index

or try this one

http://www.bose.com/controller?event=VIEW_STATIC_PAGE_EVENT&url=/popup/tech_details/pop_techmod_321.jsp

Se7en

01-29-2008, 12:31 PM

I hate to seem completely ignorant here (I'm still trying to keep up with this thread) but I was wondeirng if an omni-directional approach like this decware driver would help or hurt in this case?

http://www.decware.com/newsite/mainmenu.htm

http://s173.photobucket.com/albums/w64/se7en_251/RL15sm.jpg

http://s173.photobucket.com/albums/w64/se7en_251/rl1back.jpg

durwood

01-29-2008, 12:37 PM

Interesting idea. No sure to be honest with you.

In it's most simplest form (a 2 channel version), ambiophonics is almost nothing more than a stereo dipole speaker. The difference is that it already adds in the delayed L info into the right speaker, and the delayed R info into the left speaker and "blocks" the acoustic crosstalk electronically. The reason it removes the crosstalk is because it's already giving your ears and brain what I wants via the delayed signals.

Se7en

01-29-2008, 12:52 PM

durwood

01-29-2008, 01:25 PM

So that I'm clear, based on the speaker placement suggestions, we're talking about the equivalent of a single stereo speaker right?

There are two speakers, however they are placed so close together that they can be thought of as almost a single speaker (almost like a single mono speaker if my understanding is correct-at least this is how our brains should perceive it). The spacing between them is determined by your distance from the "stage" or speakers.

http://www.ambiophonics.org/Ch_8_ambiophonics_2nd_edit.htm

Example Speaker spacing (Azimuth in degrees)
Distance/ 10 Deg/ 15 deg/ 20deg (everything in inches)
12.00/ 2.10/ 3.16/ 4.23
15.00/ 2.62/ 3.95/ 5.29
18.00/ 3.15/ 4.74/ 6.35
21.00/ 3.67/ 5.53/ 7.41
24.00/ 4.20/ 6.32/ 8.46
27.00/ 4.72/ 7.11/ 9.52
30.00/ 5.25/ 7.90/ 10.58
33.00/ 5.77/ 8.69/ 11.64
36.00/ 6.30/ 9.48/ 12.70
39.00/ 6.82/ 10.27/ 13.75
42.00/ 7.35/ 11.06/ 14.81
45.00/ 7.87/ 11.85/ 15.87
48.00/ 8.40/ 12.64/ 16.93
51.00/ 8.92/ 13.43/ 17.99
54.00/ 9.45/ 14.22/ 19.04
57.00/ 9.97/ 15.01/ 20.10
60.00/ 10.50/ 15.80/ 21.16
63.00/ 11.02/ 16.59/ 22.22
66.00/ 11.55/ 17.38/ 23.28
69.00/ 12.07/ 18.17/ 24.33
72.00/ 12.60/ 18.96/ 25.39
75.00/ 13.12/ 19.75/ 26.45
78.00/ 13.65/ 20.54/ 27.51
81.00/ 14.17/ 21.33/ 28.56
84.00/ 14.70/ 22.12/ 29.62
87.00/ 15.22/ 22.91/ 30.68
90.00/ 15.75/ 23.70/ 31.74
93.00/ 16.27/ 24.49/ 32.80
96.00/ 16.80/ 25.28/ 33.85

This picture is from chad's link earlier in this thread for an extremelly closely spaced pair that allows you to listen in the nearfield (as close as 3ft).

http://mixguides.com/studiomonitors/Reviews/EMES-OWL.MAIN.WEB.jpg

Se7en

01-29-2008, 02:57 PM

So, to go completely off the deep end, in theory would it be possible (I'll admit being absolutely ignorant of the actual physics involved) to create a virtual or phantom stereo speaker by using two sets of speakers (for the sake of the discussing the ideal, matched full range speakers, 4 in total). 1 pair would create a "mono" phantom right stereo channel and the other pair the left.

I guess that the goal would be to ensure that right/left information was equaldistant to both listeners. I.e the left/left mono pair would create a phantom left channel image dead center if if they were equaldistant from the listener (let's say one in the left corner and one towards the center of the car with a mirror image for the right (meaning left and right channels both in the center of the car but equaldistant to both listeners).

Sorry for the rant of insanity. I hope my meaning made at least a little bit of sense.