I didn't browse through all the pages, so sorry if this already came up.
The German magazine Klang & Ton recently published a rather nice omnidirectional design. It uses a Tangband fullrange driver supperted by an oval Tangband woofer. The Pyramid cabinet must be a pain to put together, if I'd make it (perhaps cool for AV), I would use a classic rectangular cabinet with the same volume.
Have a look (and perhaps use google translate as well):
Metronom
The German magazine Klang & Ton recently published a rather nice omnidirectional design. It uses a Tangband fullrange driver supperted by an oval Tangband woofer. The Pyramid cabinet must be a pain to put together, if I'd make it (perhaps cool for AV), I would use a classic rectangular cabinet with the same volume.
Have a look (and perhaps use google translate as well):
Metronom
Not loud enough for me. My basic requirements for any speaker design is 105dB at the listening position. Not because I want to ruin my hearing but because it provides the needed headroom for home theater use. For music, the dynamic capabilities might be even higher (although nowadays most recordings are compressed to death) but we simply can't know because there's no reference level like in movies.
I listened to the Duevel Sirius. The coated titanium compression driver from eighteensound sounds very transparent and "crispy" if you like. I doubt that any JBL driver reaches that level. The spacial image is very wide and has not much height. Unfortunately I listened to German Physics in a very small and dead cabin, so I have no idea what the imaging of a Welsh driver is like in a normal room. The sound character of these speakers won't let you think about metal foil, it is rather soft, and not what one would call "crispy".
It's mostly an engineering challenge. (..and getting the pattern vs. freq. right from direct sound.) I should note though that my favorite "surround" is using *just* the Altec driver radial up against the wall (..sort of like a sconce for lighting). When placed like that it does extend the higher freq. response near the average.
A small tweeter with no real baffle to speak of (frontal placement) can be "omni" to almost 3.5 kHz (..or at least fairly uniform +/- 90 degrees horizontal). Still, that's not quite up to 7-8 kHz (by a little over a full octave).
There is also the challenge of combining it with a radial mid - smaller the tweeter, the higher the freq. for the high-pass filter (..and usually the steeper that filter is). Generally the tweeter's point of "omni" radiation is *below* it's high-pass filter. 😱
I think radials would be more popular if there was an inexpensive *good* Walsh tweeter that could handle a 1st order high-pass around 1.4 kHz.
Note Dan's SoundRounds:
http://www.htguide.com/forum/showthread.php4?t=28906
That at least provides an engineering solution to obtaining a higher freq. radial pattern for the mid.s (..which allows for a higher low-pass filter).
Last edited:
Unless the reflection is very near in time, I've personally found that my localization ability still comes from direct sound, not reflections. This is with a music signal in either stereo or mono (and usually tested in mono).
Try a mono source with a directive higher freq. output but with a more "omni" pattern 1 kHz and below (..like your B200's), aimed at the side-wall with a reflection more than 5 db above the direct sound from 1.5 kHz up. (..and with the loudspeaker 4 or more feet away from that wall.) Does it sound like it's coming from the speaker, or the wall? (..try this with a broad-band music signal, NOT a test tone.)
Last edited:
I'm currently running the delayed reflections with 10ms. The sounds tends to get muddy with shorter delays.
The image shifts to the wall, especially for high frequency, percussive sounds.
I do test with music only. This is what the current test setup looks like:
An externally hosted image should be here but it was not working when we last tested it.
Wow, we do hear differently then.
I just "tested" this again to make sure.. using a clock radio of mine (5.25 inch driver) in my bedroom and then again in my bathroom (music playing and treble boosted and not boosted) - listening more than 90 degrees from the driver's frontal axis but with the driver facing a side wall at varying distances from little less than 1 foot to more than 5 feet.
I localize the speaker, not the wall. In fact I generally localize the speaker when even in a different room (though the change in room acoustic overlay is obvious).
So, if you use an upward firing driver fullrange then it sounds as if sources are coming from the ceiling? *FREAKY*.
BTW, nice clean aesthetic. 🙂
Last edited:
That's the problem with sighted listening 🙂 The radio might also produce to much HF noise for it to work properly. Use a normal speaker and have the distances more like in a real speaker setup - speaker in the center pointing to the side wall.
Of course you won't localize the wall itself. Sound is heard as coming from the speaker location but the image spreads towards the reflections, with high frequency, percussive sounds coming from a direction somewhere between the speaker and wall.
Hi markus76,
Nice setup, are the panels against the front wall absorptive panels to create kind of a LEDE (life-end-dead-end) environment, and do you have any treatment of the back wall?
Regards,
Nice setup, are the panels against the front wall absorptive panels to create kind of a LEDE (life-end-dead-end) environment, and do you have any treatment of the back wall?
Regards,
I have a situation in my lounge where the TV is pointing towards a large mirror on the opposite wall, which reflects the sound towards the blinds, and then the sound is reflected back to the couch. On certain material, there is clearly a midrange "presence" coming from the blinds, which is 3m to the right of the TV. It sounds quite awful most of the time, but is interesting nonetheless. 🙂
Attachments
The panels are made from polyester with a thickness of 18cm. This acoustically removes the front wall. The room is about 10m long. There are no other treatments.
I have little if any bias based on sight. I also close my eyes when listening.
Your right though, it was a poor test. So.. I pulled a 6.5 inch fullrange speaker out of storage (with a "T-Amp") and tried it again. (..and it was NOT easy to get to the equipment. 😀 )
SAME RESULTS.. *except* this time when I moved all around the speaker I did get the impression of direction from the wall, and specifically from the wall, but only when:
1. approximately 150-210 degrees off of the frontal axis. (in other words pretty much directly behind the loudspeaker), AND
2. only when several feet away from the loudspeaker. This I attribute to sitting down basically at the vertical 0 degree plane (..vs. standing up where because of proximity I was able to hear a greater difference from my left and right ears).
Not once did I have the sensation of "panning" between loudspeaker and wall for imaging.
I tried this again with the radio - same results. (..I had not tried listening directly behind the radio before.)
Even remembering previous suboptimal placements of *stereo* loudspeakers where one loudspeaker was placed near a corner of the room and the other loudspeaker wasn't - it resulted in a "pull" to that one speaker, never a "panning" shift to the corner and side-wall. (..and in fact just adjusting the balance by increasing the spl from the non-corner loudspeaker re-centered the presentation.)
I also tried it again with the loudspeaker facing up to the ceiling (..just in case of a faulty memory).
Even sitting on floor I never had the impression of the sound coming from the ceiling, nor did I have any sensation of "panning" toward the ceiling.
The very same problems with an overly diffuse sound occurred again (though there wasn't any real imaging to speak of because it was mono), and yet there was still something that was clearly *better* despite the diffuseness.
I encourage others to try this if they have the ability - it would be interesting to see what others experience. 🙂
It all depends on the level (has to be pretty high), direction, delay and spectral content of the reflections. With my center plus wides setup it not easy to spot differences when levels are varied only gradually. But when switching the wides on and off, the difference becomes obvious.
Scott, I would caution against thinking you are w/o sight bias even with your eye closed. You do have a good memory for such things especially in a familiar environment. Markus's hearing agrees with experimental data FWIW. So does much of yours and I know you know your stuff. I've certainly heard dipoles where the image comes for the front wall at times. The other guy in the shop that that was "cool" and a sign of a superior speaker. Hmmm. These things can and do happen.
Dan
Dan
Yup, I recognize that there will always be some bias. The test isn't particularly well "controlled", BUT - I don't think it needs to be. It's not like we are making a definitive "textbook" statement for the populace. 😉
From what I can tell from what Markus has said, it sounds a lot like the conclusions from experimental data. i.e. "..This happens when reflections are -5dB from the direct sound or louder." And specifically experimental data that is only marginally related to what we are describing here. (..experiments based on the data I've read had very different conditions from what we would think of as "personal use".)
Basically I think such data creates it's own bias on expectation - and an expectation I've not found correct in this instance.
Again, it's not that difficult for each person to test this out. Hell, even for someone currently stuck with listening to headphones, I went to the trouble of digging through storage containers to "make it happen".
In the context of an "omni" design the seemingly preconceived notion (and it's potential resulting bias), is that louder higher freq. reflections shift images to that reflection - significantly. (..such as an increase in height for a ceiling reflection, or shifting in image width or placement to that reflection.)
I've just tested this repeatedly (and have done so previously, and under better controls), and I've *personally* not found any of it to be correct under any normal listening condition with what I believe is a reasonable probability of that conclusion.
Note that I'm not suggesting that others simply "take my word for it", instead rather - "try it yourself and see what you come up with - it's not difficult to do".
Despite all of the above, I do think that the pattern of DIRECT SOUND, diffraction, and very near reflections often does change image position, size, and shape to a significant degree under any normal listening condition. Isn't there a fair bit of data that confirms this? Aren't the conditions for those tests more in keeping with our own personal use?
Last edited:
I think it would have to be a very large spl difference between direct sound and reflected sound.
Now with "wides" - that's not a reflection, even if it's intended to synthesize one.
I've used several added-loudspeaker techniques before (both with processing and without any real processing), and it almost always seems to expand the sound both with respect to image "size" and placement (..and with more sophisticated process - the sense of venue).
I'm all for that, though I'm not particularly certain precisely how it should be accomplished for best effect. One of the better "minimalist" processing effects I've heard was via Dynavector in the late 90's. One interesting thing I found is that *three* (loudspeakers) frontal channel sound via synthetic center doesn't seem to improve upon a design when the L&R channels use a radial mid. (..about the only thing I found was that I could move L&R farther apart, but it didn't impact the presentation to do so.)
Last edited:
Your welcome! 🙂
You know, the more I think about - the more I believe this comes from the experiments dealing with synthesized reflections in anechoic chambers by using secondary loudspeakers. I think it's even referenced in Toole's book.
There are several problems that I can think of with such a test, but the two that really "jump out" at me are:
1. the size of the reflective "source" in a real room vs. that of loudspeaker.
2. the intensity of that reflective source relative to distance.
Both are interrelated:
#1 - as you move the loudspeaker away from the reflective wall (the "test" I mentioned to Markus where you "aim" the loudspeaker at the wall), the polar intensity becomes less confined. What might have been 1 foot wide at a quarter of a meter in front of the loudspeaker +/- 30 degrees essentially "expands" as you move the loudspeaker further away from the reflective wall - basically that +/- 30 degree window becomes a reflection that is many feet in diameter, and that's just +/- 30 degrees (it of course expands around to all reflective surfaces at lesser intensity). In comparison a loudspeaker used to synthesize a reflection at higher freq.s is often around an inch wide (..just in front of the tweeter).
2# - intensity is poorly localized with the reflection, it's all spread out in #1 across the wall and other surfaces. It doesn't behave like a small point source with the output in a defined area getting substantially louder and louder as you near it. The loudspeaker on the other hand does. Though I've not seen any studies on this, I've long thought that this was a major reason for our ability to better localize higher freq.s (at medium distances) - the wavelength itself is small, so "gradations" in intensity as you move toward the source are more apparent at closer listening distance.
..of course there are other problems as well such as the anechoic chamber removing the multiplicity of wall reflections, etc..
(..BTW, I've also seen demonstrations of highly focused sound applications (military) with not much more than a few degrees of dispersion, they actually do sound as if they are coming from the wall if pointed at it, and likely those reflections behave much more similarly to a loudspeaker used to synthesize reflections.)
Anyway - sorry to ramble on. 😱
Last edited:
I simply sum L+R to a center and the left wide gets L and the right R. And the effect is added spaciousness.
While I believe that angle of incidence is important, it is the relationship between overall level of all first reflections, arrival time and level of direct sound that defines perceived spaciousness.
Spaciousness in this context has nothing to do with the way we perceive the size of a room. I believe perception of room size is defined by late reflections.