One of those solid principles you are talking about is that single reflections have less perceptual impact in the presence of other reflections. So it's not really clearly defined that surpressed floor/ceiling reflections are a performance criteria per se.
I'd say what I always say at this point, "we don't know enough". We would have to study a multitude of reflection patterns (and recording techniques) to get a better picture. The task is quite complex. Naqvi did such studies but he didn't come far.
We would also need to look how adaption to the acoustics in the rooms we live in affects perception. The list goes on and on...
The only easy way out is having standardized (re)production rooms.
Markus, that's just what the Bech study was about: the effects of individual reflections in a complete sound field. Although it is not applicable in every situation.
I know the study but it doesn't provide the conclusive facts Earl was talking about. Not even all available studies on small room acoustics combined provide it.
I think the Soren Bech papers on this come close to what you describe. He simulated a speaker in in a room location with most expected reflections in the model. Each individual reflection was raised to find the threshold of perception (all are audible at some level). He found that only the first several had any likelyhood of being perceived, those including the floor bounce and rear wall bounce.
He did simulate a typical speaker dirctivity and room position (as you would expect him to do), but you can't claim universality to his conclusions. Still, floor (ceiling) and rear wall bounces seem to be the worst offenders in most situations. High level and similar arrival direction to the initial sound are plausibly the reason.
Note that the level drop of later reflections is pretty quick. The path distance is naturally longer and "wall transits" goes up. As designers we can design to deal with the first reflections. Some vertical directivity, bass radiation that uses the adjacent surfaces and judicious wall damping are all easy solutions.
David S
David, the problem is that the perceptual characteristics of a reflection pattern can change rather dramatically depending on room, speaker and listener location.
Markus
You have to start somewhere. Standardization, especially in the recording and mixing venues, would be great, but unlikely to happen. So what do we do? People are not going to design their homes arround the audio system. Heck, very few even modify a room for the system and even fewer design a room for audio playback. My interest is the later, although it is rare. But in the later case, there are principles that can be put in play.
You have to start somewhere. Standardization, especially in the recording and mixing venues, would be great, but unlikely to happen. So what do we do? People are not going to design their homes arround the audio system. Heck, very few even modify a room for the system and even fewer design a room for audio playback. My interest is the later, although it is rare. But in the later case, there are principles that can be put in play.
Member
Joined 2003
Here is one stab at using high vertical directivity to manage floor and ceiling reflections...
Here are the results comparing "new" and "old" speakers with 2, 20, and 100ms windows from the LP at about 4 meters. The "new" speaker is the upper trace in the pairs, the "old" is the lower trace.
The old speaker was a large compression driver WG with an XO of 800Hz to lower woofers, the new array with 380Hz XO
Here are the results comparing "new" and "old" speakers with 2, 20, and 100ms windows from the LP at about 4 meters. The "new" speaker is the upper trace in the pairs, the "old" is the lower trace.
The old speaker was a large compression driver WG with an XO of 800Hz to lower woofers, the new array with 380Hz XO
Member
Joined 2003
On-axis vs ~50 degree horizontal window for the listening area...
Here is a link to "what & why"...
A new center channel for the Octagon
Here is a link to "what & why"...
A new center channel for the Octagon
I'm impressed! If you know how to put it into a smaller package, I might want to implement something like that myself 😀 .
Hi Paul W,
Interesting system and good measurements clearly showing a reduction in the floor bounce effect.
You've combined a lateral CD flare with a vertical line array. The only question is whether you would get most of the sonic benefit (domestically) from simply a well behaved line array? The fundamental question(s) remain: do we need vertical directivity? Horizontal directivity? High total directivity?
David S.
Interesting system and good measurements clearly showing a reduction in the floor bounce effect.
You've combined a lateral CD flare with a vertical line array. The only question is whether you would get most of the sonic benefit (domestically) from simply a well behaved line array? The fundamental question(s) remain: do we need vertical directivity? Horizontal directivity? High total directivity?
David S.
Member
Joined 2003
Thanks Markus
Keiser
Unfortunately, size matters...a lot.
Hi Dave
My answer is "it depends".
This speaker is the center in a multichannel rig in a purpose built room. If I were building a speaker for simple stereo in a typical living room it wouldn't be this...it would probably be a wide dispersion monopole, or dipole if I they were to be placed 5' from any wall.
IMO, there are too many variables for a cookie cutter recipe.
Keiser
Unfortunately, size matters...a lot.
Hi Dave
My answer is "it depends".
This speaker is the center in a multichannel rig in a purpose built room. If I were building a speaker for simple stereo in a typical living room it wouldn't be this...it would probably be a wide dispersion monopole, or dipole if I they were to be placed 5' from any wall.
IMO, there are too many variables for a cookie cutter recipe.
Everything looks good on a 90 dB scale. There are 5 dB peaks and dips in that response.
And at any rate, whats wrong with a carpet and ceiling diffuser? Works for me.
Markus, you know what it would take to get what you want.
I recently tested an "elliptical" waveguide. It was "OK" (not great) in the horizontal plane, but a complete disaster in the vertical. Definately not worth the tradeoff. It was amazing how fast the horizontal collapsed into poor control as the waveguide was rotated. At about 20 degrees rotation it was as bad as the vertical. The overall DI was pretty bad.
Interesting...that waveguide is much like the one which is used on the fountek ribbons i have, but MUCH larger. I have been thinking of making a larger wave guide of the same type.
By the way, Earl, at what angles do you have your speakers set up? I'm currently experimenting with smaller angles (20°) and that seems to be beneficial - "natural" crosstalk cancellation?
I don't understand the toe in in front of the listener regarding acoustics. It seems like it would incease the chance of reflections from the opposite wall. And one probably doesn't get the same benefit from having diffusers on the rear wall in the room.
Is it to compensate for weaknesses in drivers or the waveguide?
If the speakers were angled directly toward the listener, one would still avoid reflections on the nearest sidewall and the chance for reflections from the opposite wall would be less.
Is it to compensate for weaknesses in drivers or the waveguide?
If the speakers were angled directly toward the listener, one would still avoid reflections on the nearest sidewall and the chance for reflections from the opposite wall would be less.
- Status
- Not open for further replies.