New Linkwitz "LX521" speakers..

Perhaps the excellent constant directivity that you achieved with the tweeters ...
There was no excellence with those tweeters. :( They showed "blooming" at its best. :mad:
You probably need to look deeper into the dipole polar response pattern at Electro-acoustic models and the importance of the D/lambda ratio. Maybe you want to read my "How open baffles work" PDF which you can download from Offene Schallwand (OB) . I don't know of any midrange drivers or tweeters that are small enough for a decent D/lambda ratio when mounted back-to-back. Putting them in a box is out of question.

Rudolf
 
There is scientific evidence that stereo speakers become localisable at high freqs especially if direct to reflection ratio is too high. See e.g.:
Bennet, Parker, Edeko: "a new approach to the assessment of stereophonic sound system performance"


[...] The solution is very wide directivity at the top end.

How could wide dispersion be a solution? Having a center speaker appears to be the only solution.
 
But I do agree that point source tweeter do tend to attract attention to themselves.

Yes it can really be a problem.

Why this issue is almost never adressed in speaker desing ? Instead, many seem to go to totally opposite direction for examply by using high directivity tweeter horns :eek:

I suppose people think the sound is supposed to come from the tweeters... pin point imaging, you know :rolleyes:

- Elias
 
Ya know, Elias is probably the only one on this site that I consistently agree with.
I guess its because we both understand that traditional stereophonics is the biggest hindrance in image reproduction.
Folks.....stereo just don't work!
We all have something called HRTF!
Until we can get recording engineers to go along with this idea...
http://www.diyaudio.com/forums/mult...cs-your-speakers-post2862500.html#post2862500
We will have to be content with a matrix algorithm, whether it be electrical or mechanical.
 
Ya know, Elias is probably the only one on this site that I consistently agree with.
I guess its because we both understand that traditional stereophonics is the biggest hindrance in image reproduction.
Folks.....stereo just don't work!
We all have something called HRTF!
...

In terms of attention-attracting tweeters maybe Uli Brüggemann's FLOW could be a part of the solution? In german it has been widely explored at
Link: aktives-hoeren.de • Thema anzeigen - AcourateNAS und AcourateFLOW
As to my knowledge, unfortunately there is no explanation of the background of this theory and the implementation to correct for it on the web in english. You might have to download the free AcourateConvolver trial SW and read there about FLOW within the english Help-Files. It's definitely worth it.
Link: AudioVero
Simon
PS: Sorry, this one might be somewhat a little out of topic.
 
At least it sweeps the problem under the carpet ;) Out of the perception, out of the mind.

There are many 'only' solutions available for these issues :D But yes, a speaker in the center is not bad !

Be it a SSS, 3 speaker linear stereo matrix or :

- Elias

Elias, Does a waveguide dome tweeter on the front (closer to constant directivity) and a flat baffle dome tweeter on the back (with the "desirable" diffraction effects) approximate a solution to what you describe?
 
we both understand that traditional stereophonics is the biggest hindrance in image reproduction.
It's like "democracy" that way . . . the worst possible system except for everything else . . .

If what you're looking for is presentation of totally synthetic acoustic "images" (as with movie soundtracks) then multi-channel-in-an-anechoic-environment is entirely plausible, and practical to implement, too, at least in theaters. On the other hand if you're trying to reproduce a symphony orchestra in your "living room" multi-channel introduces a whole raft of problems, from the beginning of the signal chain (where are you going to place your microphones to capture the correct "image" and perspective?) to the end (what are you going to do about the unavoidable reflections in that "normal", or even remotely acceptable to most people, listening environment?). Not to mention the inevitable "where am I supposed to put all those loudspeakers?" questions, and "how will it sound mixed down to ear-buds/headphones?".
 
How could wide dispersion be a solution? Having a center speaker appears to be the only solution.

It's not just wide dispersion, it's also the dispersion angle in relation to the listener's head.

Basically the higher freq. section of the loudspeaker needs to be placed closer together for the left channel to provide the right amount of intensity to the right ear (and right channel to left ear).

In this sense then "center" (as a speaker location) is every bit as important. HOWEVER absolute center is counter productive in this respect - extending too much intensity to L(ear) as opposed to R and R as opposed to L (..absolute center provides equal pressure to L & R). Absolute center tends to "bunch" images toward the center UNLESS it's a well-done multi-channel recording with multi-channel playback.


Unfortunately angle also plays a part in sound-field "expansion", where it's advantageous to hear an angle that is closer to 75 degrees (from the listener's center), with a resulting increase in intensity to that near ear.


This is why reviewers are always describing reproduction "image" (or a "vivid") nature vs. "sound stage" with a resulting diffuse image and poorer center localization (..or program material with higher freq. content "sticking to" the loudspeaker or near it).


I'd also note that the paper Elias linked to isn't about room contribution at all, only direct sound. Basically it's assuming a relatively fixed stereo triangle angle for speakers with an omni nature below about 1200 Hz - essentially arguing that only at those lower freq.s will channel balance (channel intensity difference) correctly "vector" images (or position them properly.. at least horizontally). Again though, it's horizontal polar dependent (for the loudspeaker), it's freq. dependent (for the loudspeaker's reproduction/source material), and it's angle (and/or loudspeaker "spread") dependent in relation to the listener's head.


Getting back to the LX521, you can for example: Increase loudspeaker spread to achieve greater apparent channel seperation at lower freq.s.. and/or "toe-out" the loudspeakers for a similar effect (because of the slight loss in pressure off-axis due to the dipolar design). The problem is that you'll end up doing the same at higher freq.s..

Given this design, the recommendation would be to have the upper mid and tweeter with horizontal adjustment capability, so you could angle "in" the upper freq.s while having a fair bit of "toe-out" for the lower mid and bass section. Additionally, you would want a slightly closer seated position than SL recommends, and as a result a somewhat closer "spread" between the loudspeakers.
 
Last edited:
Elias, Does a waveguide dome tweeter on the front (closer to constant directivity) and a flat baffle dome tweeter on the back (with the "desirable" diffraction effects) approximate a solution to what you describe?

I think it depends on the delay of the front wall reflections. I'm assuming the front wall is reflective. And talking about high freqs only.

The key to success is to have decorrelated reflections at the listening position, where the decorrelation distance is just little bit over the interaural distance, few ms maximum. If reflection delay is too long, they cannot fuzz the direct sound enough.

Of course this is a rudimentary method, but seems to be obligatory due to the lack of better system level options than the 2 speaker stereo triangle offers.


- Elias
 
Well markus...people believe different things.
My beliefs come from my experience, not white papers and other people's theories.

You make it sound like I wouldn't have any experience and would base my opinion only on other people's beliefs and white papers. Rest assured that I do have listening experience with virtually all the mentioned reproduction techniques.

I believe that you know that Ambiophonics doesn't solve the HRTF issue.
 
Last edited:
You make it sound like I wouldn't have any experience and would base my opinion only on other people's beliefs and white papers. Rest assured that I do have listening experience with virtually all the mentioned reproduction techniques.

I believe that you know that Ambiophonics doesn't solve the HRTF issue.


I didn't mean to sound like that, but a lot of people here are always mentioning literature that others publish to make their points. We have to remember that the people that write those papers are people with opinions just like us....... Toole, Linkwitz, D'appalito, Haas, Glasgal.... and the rest of them. I think referring to these same old theories are whats keeping us talking about the same crap. So I say from my experience Ambiophonics is best. I have listen to stereo, multichannel, mono, pro logic matrixing, OSD.

I don't know of anything else that comes closer than ambiophonics in appealing to HRTF, regarding 2 channel sources. Of course it doesn't solve the pinna problem, but it puts it pinna cues in a better location.

We all know that the best way to realism is perfect wave propagation as found in high order ambisonics, but we have to keep it in context of 2 channel recordings.

Actually, we might all be without our loudspeaker obsessions soon if Binaural audio keeps gaining traction!
;)
 
It's like "democracy" that way . . . the worst possible system except for everything else . . .

If what you're looking for is presentation of totally synthetic acoustic "images" (as with movie soundtracks) then multi-channel-in-an-anechoic-environment is entirely plausible, and practical to implement, too, at least in theaters.

No dubbing stage(where soundtracks are created), and no movie theater is Anechoic in nature. Their acoustics are just better defined by standards from THX and SMPTE.

On the other hand if you're trying to reproduce a symphony orchestra in your "living room" multi-channel introduces a whole raft of problems, from the beginning of the signal chain (where are you going to place your microphones to capture the correct "image" and perspective?) to the end (what are you going to do about the unavoidable reflections in that "normal", or even remotely acceptable to most people, listening environment?).

We have been doing multichannel recordings for so long now, the problem you mention here are completely irrelevant. We know where to place the microphones in a acoustical environment - we plan for that in pre-production. We also have plenty of rehearsal time with an orchestra before the recording is ever done. I am also pretty familar with the recording venue based on the rehearsal we have done. The standards for speaker placement, subwoofer placement, and calibration of the monitoring system have also been firmly established.(see ITU -R BS-775-1)

You cannot account for the individual listening space, and all of the acoustic issues that come with them. You can only create the best sounding multichannel mix, and let the chips fall where they may in the home environment.


Not to mention the inevitable "where am I supposed to put all those loudspeakers?" questions, and "how will it sound mixed down to ear-buds/headphones?".

Where to put all those speakers is information you can easily gather on the internet. It is not hocus pocus, or hard to obtain information. Both Dolby and DTS have primers on 5.1 and 7.1 channel speaker placement. DTS even has a primer on where to put a ceiling speaker as well to blend in with a 7.1 setup.

There are very good 5.1/7.1 to 2.0 mixdown algorithms out there, and in the film industry we have been using them in conjunction with Dolby Digital since it was introduced in the home back in 1994. DTS also has a great mixdown algorithm that comes with the DTS encoder. The few movies that I have used both of the algorithms were monitored after the encoder mixdowned the multichannel to stereo mix.