Page 4 of 7 FirstFirst 1234567 LastLast
Results 31 to 40 of 63

Thread: Ambiophonics - An experiment and a How to using foobar2000.

  1. #31
    Car Audio Moderator durwood's Avatar
    Join Date
    Sep 2006
    Location
    Chicagoland (St. Charles,IL)
    Posts
    2,028
    Not really. Winter kind of puts a damper on any in-car experimentation. I've stock piled new drivers but just waiting for it get warm enough to work with fiberglass and such. I still find Howard's plugin concept the best option, however for tweaking the off center seating effect, I have a concept of my own, but I need to work out the details and see if I can create a plugin/setup myself. Otherwise, it's better to work on it from the acoustic side to create a wider sweet spot while still avoiding side reflections.

  2. #32
    FLAC WuNgUn's Avatar
    Join Date
    Sep 2007
    Posts
    1,351
    Wow..forgot all about this thread...
    I think with my dual core upgrade, I might be able to implement the ambio and keep my active corssover...

  3. #33
    Newbie
    Join Date
    Jun 2007
    Posts
    27
    Same here Durwood. I still have not been able to put together my CarPc yet but plan to as soon as it gets a little warmer. However, I did get my audio drivers, a E-MU 1616m, and some other things.

    Do you still recommend placing the drivers in the center of the dash as opposed to A-pillars?

    Also, do you think Core 2 Duo setup is a over kill? I'm thinking of selling my DQ35JO mobo and get a cheaper mobo along with a dual core cpu. Another question I had was, if I opt for a Core 2 Duo or a Dual Core cpu, will I need a M2 or M4 psu?

  4. #34
    Maximum Bitrate techy101's Avatar
    Join Date
    Sep 2004
    Location
    Minneapolis, MN
    Posts
    511
    Looking at this thread and links briefly I'm not totally sure how this is a new and exciting idea?

    It sounds like it's taking very simple concepts used in basic professional reinforcement applications and putting new names on them.

    The core of what this is calling "crosstalk" seems to just be a comb filter. And since comb filters are created by the combination of two components, phase (defined as frequency dependent time delay) and amplitude, proper alignment and calibration should automatically correct this, just like in prof. reinforcement applications. (Not to sound insulting, and I’m sure most of us know this, but since this in an internet forum I should clarify that phase is not the same as polarity. Polarity is a 180 degree phase change. Much like a square is a rectangle but a rectangle is not a square)

    The other thing it seems to be saying is something that pretty much every speaker manufacturer already talks about and that is proper aiming of the speaker. Just minimizing or eliminating overlap in the coverage pattern of each speaker.

    Since this element is almost impossible to alter in an automotive setting it can be ignored for this discussion. We will always have major overlap in speaker coverage due to the nature of the environment and the “poorly” designed speakers used in cars. I say poorly not as a slam but just to point out the lack of definition in coverage patterns.

    Back to what it seems like this thing is really doing, correcting phase and/or amplitude to remove comb filters. In live reinforcement we use time alignment to correct this. Not in the sense of using a measuring tape and entering the distance into a processor, but by using a good measurement system (or if you don’t have one, a clean click track and good ears). Since time alignment really just a glorified way of saying phase correction we come back to one of the two core components of a comb filter, phase.

    Since altering the amplitude of each speaker isn’t really an option that only leaves phase to tinker with. By correcting the phase of each driver to be in line with each other to the best of our ability, we create a “sweet spot.” Since phase is frequency dependent it’s clearly impossible to bring all frequencies in alignment at the same time, but you can bring each driver in alignment individually for an end product that is very close to a “fully” aligned system.

    When that sweet spot is created in the middle it also starts to cause changes with the other element of comb filters, amplitude. Since there’s only one “sweet spot” everyone obviously can’t be right there. But since moving away from an audio source decreases amplitude (~ 6dB every time you double the distance via. Inverse square law) the perception of comb filters remains fairly steady. As you move away from the “sweet spot” the phase correction becomes worse and worse, but because one speaker is getting louder and the other softer everything comes close to balancing out.

    This method works very well and is used in most applications. It’s also quite old. So what makes this different? If I'm wrong I'd love to know why. I always enjoy discussions about audio.


    I hope I'm making some sense, doing my best to not get into to much jargon, although I know there is a lot.

  5. #35
    Car Audio Moderator durwood's Avatar
    Join Date
    Sep 2006
    Location
    Chicagoland (St. Charles,IL)
    Posts
    2,028
    Quote Originally Posted by jiggad369 View Post
    Do you still recommend placing the drivers in the center of the dash as opposed to A-pillars?
    If you place the speakers too far apart, you cannot control the crosstalk properly (too much room effect) and all you will do is muddy it up even more. This basically turns stereo inside-out and attempts to convert it to binaural over loudspeakers.

    Quote Originally Posted by techy101 View Post
    The core of what this is calling "crosstalk" seems to just be a comb filter. And since comb filters are created by the combination of two components, phase (defined as frequency dependent time delay) and amplitude, proper alignment and calibration should automatically correct this, just like in prof. reinforcement applications.
    Time alignment is not the answer to fix combs. Time alignment may work in the vertical (what it was designed for) but not for correcting horizontal (we have two ears and space between that changes with azimuth angle-standard time alignment will never fix this).

    If you try to fix a horizontal problem with TA, all you do is change/shift the comb filtering to different frequencies. However, it is still there because you have widely spaced speakers (two sources) trying to create a phantom image between them. Not to mention, TA only works for one spot in space as you have elluded to.

    Back to what it seems like this thing is really doing, correcting phase and/or amplitude to remove comb filters.
    Yep, that is exactly what it is trying to do. Combine the spacial aspects of multichannel, with the tonality/timbre of mono. <insert rant about stereo destroying any tonal qualities >1.5KHz>

    I take it you are associated or in the know on pro sound, so hopefully you understand the importance of mono.

  6. #36
    Maximum Bitrate techy101's Avatar
    Join Date
    Sep 2004
    Location
    Minneapolis, MN
    Posts
    511
    Quote Originally Posted by durwood View Post
    Time alignment is not the answer to fix combs. Time alignment may work in the vertical (what it was designed for) but not for correcting horizontal (we have two ears and space between that changes with azimuth angle).

    If you try to fix a horizontal problem with TA, all you do is change/shift the comb filtering to different frequencies but it is still there because you have widely spaced speakers (two sources) trying to create a phantom image between them. Not to mention, TA only works for one spot in space as you have elluded to.
    If amplitude remained the same I would agree with this, but it doesn not. Especially in an environment as small as a car, small movements have a dramatic effect on SPL. Even with <2dB offset between speakers the comb will be dramatically lessened.
    I think my point about TA only working for once location was taken out of context, because it was tied directly to the issue of amplitude variance.

    Since Combs are created by both Amplitude and Phase (50/50) a change to either creates a pronounced change in the comb. So if you TA (especially if you can view and bring the worst offending phase curves in line) then while that one spot may be the best, as you move to either side amplitude changes will be almost 2-fold, overpowering the comb. Since in a stereo application moving from the center half the distance to a speaker results in a 9dB change, even a few inches is noticeable. Heck, it's a big enough difference to be used in arena systems which require much larger movements for the same change to be noticed.

    Ignoring the amplitude aspect of comb filters is like only talking HP with a car. (loose analogy, not including weight, gearing, drag, etc...) HP and TQ go hand in hand and one by itself doesn't really mean much of anything.


    If you place the speakers too far apart, you cannot control the crosstalk properly (too much room effect) and all you will do is muddy it up even more. This basically turns stereo inside-out and attempts to convert it to binaural over loudspeakers.
    I have to disagree with this. Placement by itself has no real bearing on room effect. In a properly designed environment the room is an integral part of the sound. System design is supposed to incorporate the environment and not pretend like everything is an anechoic chamber. (which for the record actually sounds like . Got to tour the Bosch/Telex/EV/Midas/Everyone else R&D facility and spent some time in theirs. It was freaky, but also the speakers sounded terrible)

    And this use of binaural seems incorrect. Technically speaking every stereo system is a binaural system, since it "has or relates to two ears." So it doesn't matter where the speakers are placed. Binaural in this context seems to be in reference to binaural recording, which as the name implies takes place at the time of recording. (If you don't have it, rent the Shrek DVD, I believe it's the first, put on a pair of headphones and go into the special features. There's one specifically on this)

    Back to crosstalk, a properly aimed speaker will not interfere with the others in the system in it's primary or secondary fields. It will outside of that, but the amplitude drop off is so dramatic it becomes moot. That leaves reflections, which are usually a PITA, but can be controlled with proper system design and environment treatment, and can sometimes be very welcome. Letting a good environment warm a sound is almost always better than processing the warmth in.

    The one thing that has almost no bearing on reflections with properly aimed speakers is proximity and sound stage width. If the speaker is aimed partially at the wall, then the reflections become uncontrollable monsters, but set properly a well designed speaker can be darn near flat against a wall.

    I know that this doesn't directly deal so much with the car aspect, but since we are discussing general theory which doesn't deal with one environment alone it seems pertinent.

    Yep, that is exactly what it is trying to do. Combine the spacial aspects of multichannel, with the tonality/timbre of mono.
    If this is the case, then what makes this special? It seems like an unnecessarily complex and poorly performing band-aid for improper system design?


    As with before, I welcome responses. This is not one of those internetz I'M RIGHT YOU'RE WRONG, SO SHOVE IT' things. I really enjoy audio discussion

  7. #37
    Car Audio Moderator durwood's Avatar
    Join Date
    Sep 2006
    Location
    Chicagoland (St. Charles,IL)
    Posts
    2,028
    http://www.xlrtechs.com/dbkeele.com/papers.htm

    18. "The Effects of Interaural Crosstalk on Stereo Reproduction and Minimizing Interaural Crosstalk in Nearfield Monitoring by the Use of a Physical Barrier," Presented at the 81st Convention of the Audio Engineering Society, (Nov. 1986).

    Two papers there you can read. You were correct BTW, this is not a "new" concept. Technology just makes it easier to dig farther into it now. Those papers were written in the mid 80's, however it can be traced back even further back into the 50's/60's if you go digging in the AES archives.

    In addition, ambiophonics was a marketing term coined just as the word "stereo" is. It can also be called a "stereo dipole" if you like.
    http://www.isvr.soton.ac.uk/FDAG/VAP/

    ^Plently of good stuff there too including demos.

    I don't really have the energy to have real indepth discussions anymore about the pros and cons of different playback methods since the majority of people resist change anyway, but I do welcome them from time to time.

    You are talking about the time intensity trading mechanisms of our hearing, however it is too simple of a model for the human hearing, especially since it only deals with anechoic enironments in which we don't live. We cannot fix 3D space with 2D tools. Why do you think surround sound has become more popluar than 2 channel? How crucial is the center channel in a surround sound setup? Think about it and experience it yourself.

    You cannot directly compare pro audio to small room acoustics especially when talking about large venues- nearfield and farfield differs quite a bit, then there is the direct and reflected field as well-it get complex. Pro audio is mostly panned monophonic, not stereo, and it is far field, not nearfield (as in a car). Stereo is an effect that has to be achieved when the proper stars align. This is never going to happen in a car unless you drive an Mclaren F1, and even then everything >1.5kHz is still destroyed. This has been known from the early days of stereo.

    To achieve what "stereo" is designed to do, read up here
    http://www.linkwitzlab.com/stereo%20reproduction.htm
    and tell me how many of those points you can accomplish in a car.


    If you still want to hang on to stereo and the time intensity trading, at least consider PI-stereo/position independant stereo.

    I do like to borrow concepts from pro audio as far as acoustics go. It's too bad no more people apply it elsewhere outside of it...arrays, waveguides, power, headroom, active systems, etc etc. <-This is where the real big gains can be had, the rest is only minor improvements.

  8. #38
    Car Audio Moderator durwood's Avatar
    Join Date
    Sep 2006
    Location
    Chicagoland (St. Charles,IL)
    Posts
    2,028
    Quote Originally Posted by techy101 View Post
    If this is the case, then what makes this special? It seems like an unnecessarily complex and poorly performing band-aid for improper system design?
    Stereo IS an improper and poor performing system design. Mono is the only true reference, nature is nothing more than individual mono sources BTW.

    Anechoic recordings would sound good on a proper playback system equal to the environment it was created in. A pair of IEM's with good isolation in other words, or another anechoic room.

  9. #39
    Low Bitrate
    Join Date
    Feb 2009
    Posts
    62
    Not fully following the discussion, Like the idea so have subscribed and will read some more.

    Sound staging is always of interest and if we can find a way to create a soundstage that combines all the major seating positions in a vehicle that would be fantastic

  10. #40
    FLAC WuNgUn's Avatar
    Join Date
    Sep 2007
    Posts
    1,351
    I find that RACE setup thru AM gives of a LOT of reverb in the vocal ranges...
    It's okay on some music, but other music, it sticks out sorely!!

    What adjustments have you made to your RACE settings to make it sound okay in the car Durwood?

Page 4 of 7 FirstFirst 1234567 LastLast

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •