Horn vs. Waveguide

[amiklos]

amiklos,

you don't make use of the standardized stereo triangle. That's your choice, but it has no impact on the more general case of a equilateral triangle.

Markus, I apologize. I had assumed you were initially making reference to the placement that Earl has recommended for these speakers rather than the traditional stereo triangle, hence my discussion of a listener outside the triangle.

I've set mine up to be as close to Earl's recommendation as possible in my space and had not given thought to the "standardized stereo triangle" for use with these loudspeakers.

 

[ScottG]

I've set mine up to be as close to Earl's recommendation as possible in my space and had not given thought to the "standardized stereo triangle" for use with these loudspeakers.

I've suggested "inverting" the method Earl advocates, and listening relatively closely to the loudspeakers without a large amount of physical separation between the loudspeakers..

Such a position should radically increase soundstage proportions, especially width beyond the boundaries of the loudspeakers and overall depth, while some what diminishing "pin-point" image "focus".

Of course it's only for one person and domestically it's a "pig".

(..it does tend to highlight that one method does NOT fit all or is "absolute".)

 

[JoshK]

The stereo triangle is defined by a equilateral triangle (http://en.wikipedia.org/wiki/Equilateral_triangle).

Best, Markus

I've never seen anyone actually have such a setup, nor have I ever seen anyone advocate such a setup. I guess that I why I said what I said.

 

[ScottG]

I've never seen anyone actually have such a setup, nor have I ever seen anyone advocate such a setup. I guess that I why I said what I said.

Its rather close to what I advocate, though I also recommend what could be described as "toe-out". Further, I recommend a "triangle" shape that is relatively small - about 6-7 feet from each "point" (depending on the loudspeakers). Again, though - it usually isn't domestically acceptable IF its a "permanent installation".

I've not found a single listener that enjoyed precisely listening to the 0 degree axis, *except* occasionally at mixing consoles (and almost always to specifically enhance intelligibility despite an other-wise "off" sound).

Some prefer the "toe-in" method and others prefer the "toe-out".

 

[ZilchLab]

After spending a lot of time and (potentially) money, collecting data, analyzing it, coming to solid conclusions based on solid methods, it's frustrating to have others, even other authorities on the subject, argue with your findings when they themselves have not gone through the same rigor in your specific topic.

There would appear to be more posture and less rigor than we are supposed to imagine now in evidence here.

The only thing I'd add is that in my experience two or more parties disagreeing, yet each vigorously defending their positions because they believe they are right, often yields the most thorough examination of a given subject. The assumption here is that the parties are reasonably open minded despite their convictions, and perhaps that some more moderate force exists to push dialog in a constructive direction.

That's certainly how I see it, and the participants' posting innocuous "barbs" at each other only adds entertainment value, in my view. Sometimes leaving the room to give matters rational consideration has merit, but there's always the risk of a conspicuous absence being viewed by others as the equivalent of holding one's breath 'til turning blue, of course.

I've never seen anyone actually have such a setup, nor have I ever seen anyone advocate such a setup. I guess that I why I said what I said.

An isosceles triangle is the more common reality, and that works so long as the speakers are not toed-in, which, in the case of Earl's designs, moves their on-axis response anomaly into the direct field. Also, if you don't toe them in, the early ipsilateral (same side) first reflection is more prominent.

As others are suggesting, constant-directivity speakers not exhibiting that characteristic on-axis anomaly afford additional flexibility in establishing an optimum balance between imaging and soundstage width, according to room characteristics and listener preferences....

 

[markus76]

In the Geddes recommended alignment, the listener is outside the triangle; the speakers are cross-fired in front. He EQs at 22.5°. Draw it up, and it's easy to see what's going on.

An externally hosted image should be here but it was not working when we last tested it.

Best, Markus

 

[ZilchLab]

I believe we want to be further back than that, as the 15° green curve is on the edge of the anomaly zone, and any lateral movement, as in the person sitting next to us or at the end of the couch, is headed into the on-axis anomaly zone, effectively defeating the expanded image rendition:

http://www.diyaudio.com/forums/attachment.php?s=&postid=1862460&stamp=1245729650

Orange is 22.5°, and is itself not working all that effectively in terms of defined directivity above 4 kHz, at least in these measurements.

 

[markus76]

I believe at least for stereo listening we want a toe-in of 52.5° with Earl's speakers. On the other hand, stereo really isn't made for more than one listener. That's why Blumlein called it the "stereo seat".

Best, Markus

 

[ZilchLab]

I believe at least for stereo listening we want a toe-in of 52.5° with Earl's speakers. On the other hand, stereo really isn't made for more than one listener. That's why Blumlein called it the "stereo seat".

If so, that moots what Earl claims as a major benefit of this alignment.

By another view, it's merely an artifact, not the precisely contrived "feature" alleged, the primary purpose being to listen from outside the anomaly zone. There are other well-known and oft-practiced means to damp the early near-sidewall reflection. See Toole.

 

[markus76]

Why would toeing-in 7.5° more render Earl's claims moot?

 

[ZilchLab]

It doesn't. "Stereo being made for only one listener" does.

In truth, there's only one "true" sweet spot, equidistant from the two sources, but that doesn't preclude us from "enhancing" it via these and similar means....

 

[ScottG]

..but that doesn't preclude us from "enhancing" it via these and similar means....

But does "enhancing" in this respect have a "price"?

 

[ScottG]

As others are suggesting, constant-directivity speakers not exhibiting that characteristic on-axis anomaly afford additional flexibility in establishing an optimum balance between imaging and soundstage width, according to room characteristics and listener preferences....

That would be part of that "waffling" I commented on.

Is the on-axis response important or not? Is the increased off-axis image stabilization important or not? (.."which is it baby, spitz or swallows?".)

I think the main thing here is that it's a matter of degree. (..bad pun.) In this particular respect it is not deemed important until you reach a certain angle laterally that is *less than it might otherwise be*. Is it important? Depends on the user. Is the limitation prominently disclosed so that the user can make an informed decision? Not that I've seen.

 

[ZilchLab]

But does "enhancing" in this respect have a "price"?

Not really; most any CD horn/waveguide can do it, so long as you're willing and able to live with the "unorthodox" alignment.

If you implement it with those designed for the purpose, it works so well as to be disconcerting at first; we are accustomed to the image collapsing with lateral movement.

After a while, it becomes somewhat of a disappointment to listen without it, so we install a center channel, instead. I think that was a major reason it was abandoned -- more speakers to sell with multichannel.

I think the main thing here is that it's a matter of degree. (..bad pun.) In this particular respect it is not deemed important until you reach a certain angle laterally that is *less than it might otherwise be*. Is it important? Depends on the user. Is the limitation prominently disclosed so that the user can make an informed decision? Not that I've seen.

Which is why we have to be "students" of Geddes design to get the full picture. The caveats are there, but only in the fine print....

 

[ScottG]

Not really..

Having heard a several setups with different configurations I'd disagree.

In every case where the Jordan method is used that I've heard, you loose lateral spread and often some depth. Basically the hall-sound shrinks (significantly) with respect to upper freq. reflections from boundaries of real or processed recorded venues.

Some people don't seem to recognize this *or* haven't been exposed to enough variety to have had the opportunity to recognize it.

Some people do recognize it, but don't deem it as important. (..at least in the past, Jmmlc for one.)

Others recognize it and suggest surround sound (ambiance channels) to *try* to achieve it while still utilizing the Jordan method. (Earl if I remember correctly.)

Others both recognize it *and* deem it important, and do NOT think the surround sound "solution" is a solution for most 2-channel recordings.

I easily recognize it. Not only do I *like* a better defined "soundstage", but I also think that since it is part of the recording that it is more accurate to include it in playback as realistically as possible. I'm also willing to give-up image stabilization with significant single listener movement to get it. To me surround sound for most of my music doesn't work - venues typically become bloated and vague despite a nearly optimal set-up.

In the context of multi-listeners however specifically for Home Theater, it's all good.

Note that yet again, NONE of this mentioned, and being a "student" of Geddes design won't get you there - it requires something *beyond* the Geddes format.

 

[ZilchLab]

What's on recorded media?

Y'all'll NOT be takin' the Zilchster on that ride, nope.

Listener envelopment *can*, and often is achieved via good *2* channel reproduction. It's part of the reason people pay so much money for audio equipment or spend so much time in forums like these (..trying to find a way to do it for less).

Some of us DO recognize an imaginary reverberant field when we hear one....

Edit: Looks like my vertical lobe is somewhere between +7.5° and +10°:

 

[ZilchLab]

More Data ... Less Yak-Yak:

Could someone with better Photoshop skills than me overlay the following two images with maybe 10 dB of vertical offset, please?

I believe I have them scaled equivalently, but one is JPG and the other PNG, and I'm having difficulty making layers happen.

They are vertical polars +/- 10 dB from the lobe center, first for Markus's Nathan:

 

[ZilchLab]

And this one, my el-cheapo dumpster-diver Boston A60 Econowave. Note that my plot extends past 20 kHz in aligning them horizontally. Mine should go above Markus's, looks like, as the response doesn't collapse up there.

If you'd prefer to work with the "originals," they are third image down here:

http://www.diyaudio.com/forums/showthread.php?postid=1859757#post1859757

And this one here:

http://www.diyaudio.com/forums/attachment.php?s=&postid=1865902&stamp=1246087381

 

[Jmmlc]

Hello,


Here attached but not very pretty (allways difficult to work after a JPEG conversion...)

Best regards from Paris, France

Jean-Michel Le Cléac'h

And this one, my el-cheapo dumpster-diver Boston A60 Econowave. Note that my plot extends past 20 kHz in aligning them horizontally. Mine should go above Markus's, looks like, as the response doesn't collapse up there.

If you'd prefer to work with the "originals," they are third image down here:

http://www.diyaudio.com/forums/showthread.php?postid=1859757#post1859757

And this one here:

http://www.diyaudio.com/forums/attachment.php?s=&postid=1865902&stamp=1246087381

 

[ZilchLab]

Thank you Jean-Michel!

I was stunned when I first compared them; it should be apparent to all, now.

[There's a soundfield thesis in those plots.... ]