That the omni speaker gets re-invented every decade is not proof that it is following any appropriate ideal. I think it is more that it is easy to market to the uninitiated: "90% of the sound in the concert halls has bounced off the walls..." "Response in every direction is the same."
Any discussion of ideal speaker directivity needs to answer the more basic issue of what sort of reflection pattern is ideal. What delay, what strength and what direction should relfections come from after in the direct sound from the speaker? I think there is plenty of consensus that later lateral reflections can be helpful and that early reflections, especially from angles near to the speakers will add coloration rather than space to the sound.
That being the case I don't see any of the popular omni speakers as being a solution.
David S.
Last edited:
I believe my reinforced paper cones are simply not stiff enough. Unfortunately I don't have the time and tools to do a thorough investigation with stiffer cones of different sizes and distances. All I can do is capture some basic data.
What do you mean by "shaped more as 360 degree radial horns"? Something like the Zenith speakers Tom has posted?
I wouldn't worry about materials. I used to make a lot of CD waveguides for tweeters and I always used manilla card stock for quick prototypes. I never saw any appreciable difference between the performance of that and the performance of the final molded pieces. At treble frequencies I think your material is stiff enough.
For a 360 degree horn I meant more like these two.
David S.
Attachments
I always love it when people draw how the sound paths will go, as if sound is somehow subsurvient to their desires and/or pencils. Horns suffer from this more than most I think. I still see people draw the sound waves in horns like they can somehow picture what it will do.
I always thought you were an advocate of wide-dispersion speakers. If early reflections add more tone coloration than spaciousness, how can you hold that position?
This looks like a major development effort. Furthermore I'm not sure that we would end up with a radiation pattern that is desirable. I believe hemispherical radiation that creates lateral, contralateral and ceiling reflections would be better. But how to get there?
Wide dispersion advocate? Not really. I consider myself a proponent of flat direct response "uber alles". I like CD approaches if they give better near on axis response. I think power response curves are overated and room response curves get you into trouble.
I used to think higher directivity was a good goal, but reading Toole's book has got me thinking that broader response or maybe a specialized room treatment could be the way to go. How about parabolic reflecters that catch sound at one wall and bounce it off of quadratic residue diffusors on the opposite wall? You can aim Earl's speakers at the parabolic reflector. Lots of late lateral sound that way.
David S.
The upward firing-into-a-reflector types will ideally give you cylindrical radiation. If you want ceiling reflection, then maybe the high upwards radiation you are still seeing is okay.
It seems like the only way this speaker type will give the donut shaped radiation promised, is if the reflector becomes huge to better block the upwards energy. That is, big enough to prevent diffraction around and wide enough that there isn't a direct upwards (at an angle) path.
Looking at some pictures of the Mirage Omnisat system, I don't see how they wouldn't have exactly the same problems. The reflector above the woofer isn't nearly big enough to block upwards radiation.
Thinking back several decades there was the Walsh "ice cream cone" tweeter that Infinity used. It had a lot of upwards energy also, in spite of what the marketing diagrams showed.
David S.
I get strong contralateral reflections with my Nathan's. This is a result of 45° toe-in and high directivity.
The problem with diffusors is that the typical listening room is too small for such a diffusor to create something really diffuse and not just spectrally questionable first reflections.
It's interesting to see that even somebody like yourself, an industry professional who has been in the speaker business for most of his life, can change his mind on this matter in just little over a year:
The Goals for an "Ideal Loudspeaker" - The Classic Speaker Pages Discussion Forums - Page 8
speaker dave said:
I think most industry professionals would agree about most of your criteria. However, the one thing about which opinions vary widely, is 'the ideal directivity', if there even is such a thing. My position has long been that many different radiation patterns can be made to work well, but they all have an inherent sound signature. Coming to think of it, I probably still think so, but I must say that I am very much impressed by what a higher than usual directivity loudspeaker can do.
Directivity deserves its own topic. I'll start it tonight (it's 18:06PM here, so that will be in a few hours 😉 )!
Last edited:
@Speaker Dave & Keyser,
The ideal directivity is very much a function of the intended use of a speaker.
One issue that merrits attention is the quality of the off axis response. Diffraction/interference ripples in the off axis response become audible through reflections. Therefore, I would argue that controlled directivity concerns the full 360 degrees.
vac
The ideal directivity is very much a function of the intended use of a speaker.
One issue that merrits attention is the quality of the off axis response. Diffraction/interference ripples in the off axis response become audible through reflections. Therefore, I would argue that controlled directivity concerns the full 360 degrees.
vac
Covered half the driver with a 45° cone:
0°-90° vertical, without reflector:
0°-90° vertical, with cone cap:
An externally hosted image should be here but it was not working when we last tested it.
0°-90° vertical, without reflector:
An externally hosted image should be here but it was not working when we last tested it.
0°-90° vertical, with cone cap:
An externally hosted image should be here but it was not working when we last tested it.
Interesting horizontal data for the cone cap:
An externally hosted image should be here but it was not working when we last tested it.
I didn't think that people would be searching my older posts for any inconsistancies. You might pull up this which give a fuller view of what I think about directivity.
http://www.diyaudio.com/forums/mult...speakers-not-more-popular-37.html#post2652828
"There are a lot of things in audio that I feel strongly about, that I consider as absolutes. But when it comes to speaker directivity I have to accept that it is a matter of taste. Some people like very broad directivity, a lot of interaction with the room, speakers that disappear and sound that is similar no matter where you sit. Vague imaging is okay as long as the sound is big and spacious. Others want the opposite: pin point imaging, no room interaction, headphone like stereo where you must sit dead center between the speakers.
Arguing absolutes here is like saying "you must go to a concert and sit front row center", or "no, no, no, the only good seats are back in the balcony". Different people have different preferences.
With only 2 channels you will have to pick your poison. With 5 or more you can start to create diffusion via a multiplicity of systems, even if they are individually quite directional. In that case you can choose between direct sound and diffuse sound."
(end of quote)
I've designed a lot of studio monitors in the past and most had higher than normal directivity, either through CD horns or dome tweeters with directional flares of some type. I generally like the higher clarity that comes from increased directivity, but I also ran across a number of people that really don't like the aspect where you can close your eyes and say: "all the sound comes from here, and here" (pointing at the tweeters). Those types want the speakers to dissapear and the sound to come from anywhere and everywhere.
Reading Toole I think he makes a good case that delayed lateral energy is good. Certainly Baron and others involved in concert hall design hae made a very strong arguement in the last several decades that lateral energy seperates good concert halls from bad ones. Equally Kates and Salmi and Bech make a compelling arguement that early floor and back wall bounces give colorations that the ear can't seperate from the direct response.
The conclusion from all this must be that a speaker that gives late lateral reflections but limited early reflections would be best.
If we look at popular alternative directional patterns, the dipole, the omni, the cylindrical pattern, the Bose 901 pattern, none of them make sense in that context. The highest ranked speakers in all of Toole's testing are conventional front firing systems with generally rising directivity and power response holes at each crossover point.
David S.
I wouldn't be able to do it at the same time, but it would be possible to do. I only have one set of mics that I can fix in my ears or the other dummy's. I'll look pretty darn funny standing on the roadside or at the beach with a dummy head in hand. You should see the wig I got for it.😱 Hmm, I wonder if it could be a conversation piece/starter for one of these California girls... Doubtful. I'll try and do something like that though Markus. I have some pretty good ones for my head with that type of stuff on there. I could dump it analog to my computer as I've been intending to do for some time now. Doing it with the dummy might be a bit awkward, but for the sake of science...
Yea, that's me alright. I think you just pimp slapped Neil Young. I'll take it as a compliment, that I certainly don't deserve. The thing I hate about recording is that you learn something every time you do one and there is so much I would do differently now. It's like building a speaker in that sense. Each subsequent one improves(hopefully). Even my DIY Dummy has improved let alone my techniques for recording/playing/singing. The noise on that recording is maddening. You should notice a distinct center of the head effect for the electric guitar and bass(the bass may be more vague) while the rest comes from in front of you and off to the sides through headphones.
Dan
ie. most music lovers, it's quite abnormal that the industry trends are dominated by biases of the pros and small part of the so called audiophiles
what about a flooder in Beveridge placement then? no significant floor reflection, significantly delayed front and back wall reflections, no magic, it is basic physics, yet nobody is interested, why??
limited in what sense? and why? could it be that only some early reflections should be limited?
Hello David,
i guess the dipole can be made meeting that preferred distribution of reflections
arriving in time and direction very well, if you place it in a way so that the rear radiation
(to significant extent) hits the side walls instead of the corner or front wall. After
"double bouncing" you can get contralateral reflections depending on room dimensions
and speaker/listener placement. (I think you proposed this alternative in dipole
placement yourself some months ago ?).
Alternatively one may use a prepared corner/frontwall section, which is able to
reflect energy to the opposite sides thus also achieving contralateral reflections
having sufficiently long delay ...
Kind Regards
Last edited:
Perhaps the moderator could transfer these last replys (from #196 except Dan's) over to the new thread rather than further hijack Markus' thread?
David
David
- Status
- Not open for further replies.