... AES ... paper ... . ... anecdotal evidence from owners of fullrange dipoles suggesting what time delay ballparks are acceptable for a full-spectrum reflection arriving from an angle not too far removed from the original source ...
Reflections from fullrange dipoles are certainly not full-spectrum. You can look at papers and anecdotal evidence from whichever perspective but one fact remains: we're still lacking some important basic knowledge.
Simple question: how does summing localization work? Guess everybody is aware of this: http://www.hauptmikrofon.de/theile/ON_THE_LOCALISATION_english.pdf
Reflections from fullrange dipoles are certainly not full-spectrum.
Well, it depends on the specifics of the design and how narrowly one defines "full-spectrum". Some dipoles come quite close to constant directivity over most of the spectrum (SoundLab's large faceted-curved 90 degree panels), and some are a fair approximation (three-way Maggies), and this is reflected in the spectral content of the energy bouncing off the wall behind them. Unless some frequencies in the backwave are deliberately attenuated (original Quads), most fullrange dipoles still generate a backwave reflection that is pretty close to full-spectrum, modified of course by the surface it bounces off of.
Duke
Last edited:
No, not with almost any loudspeaker. Such a loudspeaker needs to show constant directivity so the direct sound is still sufficiently flat. The majority of HiFi speaker designs is not CD.
As far as experimenting goes, the difference is proportional and largely trivial (for almost any speaker)... unless you absolutely insist on demonizing all non-Geddes products. Certainly just a coincidence that am I using religious metaphors again.
As far as experimenting goes, the difference is proportional and largely trivial (for almost any speaker)... unless you absolutely insist on demonizing all non-Geddes products. Certainly just a coincidence that am I using religious metaphors again.
What difference is proportional?
Constant directivity is an essential element in my proposal -- the off-axis spectral content and balance must remain the same over a defined beamwidth.
I am presently completing the ninth of an anticipated dozen non-Geddes constant-directivity loudspeaker project builds thus far this year, in a mere 24 pages:
Flex Your PCD Mettle: - Techtalk at Parts-Express.com
The saga begins just over two years ago in 166 pages here:
http://www.diyaudio.com/forums/multi-way/123426-horn-vs-waveguide.html
[We're on page 755 in the main thread over on AudioKarma.... 😉 ]
I am presently completing the ninth of an anticipated dozen non-Geddes constant-directivity loudspeaker project builds thus far this year, in a mere 24 pages:
Flex Your PCD Mettle: - Techtalk at Parts-Express.com
The saga begins just over two years ago in 166 pages here:
http://www.diyaudio.com/forums/multi-way/123426-horn-vs-waveguide.html
[We're on page 755 in the main thread over on AudioKarma.... 😉 ]
Last edited:
What difference is proportional?
You talk as if there is sharp bright knifeline drawn in sound by CD/Geddes (and we are talking just the tweeters, eh) and as if all other speakers are just plain ragged, uncontrolled, and of unknown directivity. Not so.
They ARE unknown until measured, mapped and published.
[There are surprises when that is accomplished.... 😉 ]
[There are surprises when that is accomplished.... 😉 ]
They ARE unknown until measured, mapped and published.
Great point!!

I don't know if everyone now agrees with your directivity map of Orion, but it certainly is not what I expected:
http://www.diyaudio.com/forums/multi-way/103872-geddes-waveguides-453.html#post2235918
http://www.diyaudio.com/forums/multi-way/103872-geddes-waveguides-453.html#post2235918
Last edited:
After Johns question, I thoroughly checked the data again and I am now confident in it (above 300 Hz. - I had real trouble below that since dipoles act so much different than a monopole that my usual techniques didn't work.)
I just wish that we had more examples so we could quite guessing about how well some designs work. But no one seems willing to submit any data. Maybe it is better not to know 😉
I just wish that we had more examples so we could quite guessing about how well some designs work. But no one seems willing to submit any data. Maybe it is better not to know 😉
Last edited:
I'm doing it every 5°, and it's a fairly intense accumulation of data done manually. That's 37 measurements if the baffle layout is asymmetrical. No way am it going to post-process all of that to meet your formatting requirements. If the text data files were sufficient, I could send you those, however.
As I rotate about the presumed acoustic center somewhere behind the baffle, I end up measuring in the shadow of the cab edge, ultimately, at 90°, looking at the side, and just seeing diffraction. How do you compensate for that? Forget the acoustic center and rotate about the front of the baffle? Calculate how far to move the mic off the acoustic center with progressively increasing angle as a compromise?
As I rotate about the presumed acoustic center somewhere behind the baffle, I end up measuring in the shadow of the cab edge, ultimately, at 90°, looking at the side, and just seeing diffraction. How do you compensate for that? Forget the acoustic center and rotate about the front of the baffle? Calculate how far to move the mic off the acoustic center with progressively increasing angle as a compromise?
Last edited:
all other speakers are just plain ragged, uncontrolled, and of unknown directivity. Not so.
Would you mind posting those measurements?
Would you mind posting those measurements?
Here's my first Google hit
Second hit
Googling isn't the way to find these, but here are two.
So what is the "way to find these"?
You claimed that "almost any speaker" can be used to "optimize the balance between imaging and spaciousness" "by varying the toe-in".
To prove that claim you now post data of a ceiling speaker (at only 8 different frequencies with unknown smoothing and gating) and a PA speaker (at only 6 different frequencies)?
This is how a typical 2 way HiFi speaker looks like (-90-90° in 5° increments, 1/3 octave smoothing):

Did you read Earl's paper on directivity? It should clear things up.
Last edited:

Sin bin time will result from further bickering. Perhaps you all need a time out.
bentoronto, you linked data of two different speakers to prove your claim that "almost any speaker" can be used to "optimize the balance between imaging and spaciousness" "by varying the toe-in". One speaker is shown at 6 different frequencies, the other at 8.
We don't know on what data those diagrams are based on. What we know is that 6 or 8 snapshots don't describe a loudspeaker's directivity characteristics sufficiently. To prove my point, I showed you higher resolution real world data of a typical 2-way HiFi speaker. This data clearly showes that the speaker is NOT CD and can't be used with toe-in.
We don't know on what data those diagrams are based on. What we know is that 6 or 8 snapshots don't describe a loudspeaker's directivity characteristics sufficiently. To prove my point, I showed you higher resolution real world data of a typical 2-way HiFi speaker. This data clearly showes that the speaker is NOT CD and can't be used with toe-in.
But you can EQ to get any curve you want. You should not think you have to buy or build warm speakers to get a warm sound. Atleast with CD designs you have some control over the off axis response and it creates a very smooth overall sound in room.
Hi All,
CD speakers are usually brighter than typical audiophile? This is my OSWG "Mummy" under 30 deg with driver resonances flatten. I did response falling a little. I feel more comfortable with a "warmer" speakers, maybe I need accommodation 🙂
This is pretty much universal in that we have all found this. There is a lot of discussion about it in these threads - somewhere. A CD speaker EQ'd flat on the listening axis will certainly sound bright.
- Home
- Loudspeakers
- Multi-Way
- Geddes on Waveguides