People don't argue about 19th place . . . they fight about what should, or shouldn't be, #1. That has long been the nature of ORION threads, and surprisingly continues to be, even after it has been surpassed by its own designer.
And it's all good. Every "flaw" or "failing" found in the ORION design (and of course it has flaws and failings) advances the art, because "state of the art" is where ORION was in so many design areas. We all learn by seeing where it can be made better, and by building speakers that, in that area at least, are better. When introduced ORION was one of the few designs that even attempted to attain both "flat on axis" and "flat power" response . . . what we now call "constant directivity". It is still one of the leaders in that regard, despite its midrange failing, unless you choose, as some do, to ignore "baffle step".
It is one of the few designs that attempts, and actually succeeds, in providing both good "imaging" and enveloping "spaciousness" . . . attributes once thought to be mutually exclusive. We're still discussing, right in this thread, the "how" and the implications of that for other designs (incorporating room reflections).
And we're going that direction in discussion of "dynamics" too . . . realizing that maybe it's not the speaker itself, but how the speaker interacts with the room . . . and with that understanding perhaps learning that it is not necessary to make speakers behave like oversized headphones in other ways (sacrificing "spaciousness" and "realism") to maintain accurate "dynamics" as well.
Through it all speakers get better. We're well past "good enough" for most people in most applications, and dancing with the state of the art is yielding smaller and smaller returns. Still, that's what we (most of us, anyway) come here for. And it's all good . . .
I will also be able to get 10ms, once I treat my (low) ceiling. The first reflexions after ceiling are starting at 10ms actually. I did measurements last weekend, toe in clearly lowers the level of early reflexions by a very wide margin every 10 deg. I could even get more ITD, but then the listening position would be too close. Would -20db be low enough for you?
No it doesn't bother us at all, because we listen to real music recorded in real spaces, played in real rooms. And at this, the Orions are miles ahead..
But I can see why you are still desperately trying to convince people of the opposite.. And again, what separates us is maybe not the "right" and "wrong" side of things, where we could fight endlessly.. We just have different listening expectations and preferences, basta!
But you do have a good point, the Orions fail in directivity in a very important zone, treating the side walls improves things considerably. I see it more as a case for the importance of broadband spectral content of reflexions rather as one against all reflexions whatever they could be..
As a side note, I am going to buy a Berry 2030, for fun, and because of your recommendation.. So you see, we are open minded!
Last edited:
Really? I don't believe you.
Dave.
Let's say we all pretty much agree that it's impossible to recreate sound as heard in the recording environment...
Then where does that leaves us? That all "Auditory scenes" are illusions. If so then there is no right or wrong. Just preference. So throw any loudspeakers in the mix. Let's add Gedlee, or Elias Pekonen's, Unities etc. vs. Orion, Behringer, Nao Note etc. Completely different topologies, yet they are all projecting illusions. There is no one correct sound.
What SL himself said about his comparison of Orion and Behringer:
Quote:
(O=Orion, B = Behringer)
"The main difference between the two systems is in their spatial rendering and that is more noticeable from the closer seat A. The audibility of a different AS depends strongly on the sound track chosen. On some tracks the AS is confusingly similar for O and B. Where the AS is different for O and B it is not necessarily obvious which AS is more accurate or plausible, particularly when the difference is in the spatial distortion of the AS. "
Perhaps...
Last edited:
The metric was ill defined and ambiguous.
To elaborate on my point:
Lets say that we want to test the difference in weight between cats and mice. So we buy an electric scale, which is, unbeknownst to us, broken and reports random numbers. We then weigh 100 cats and 100 mice and average the results comparing the means and distributions and conclude that the two are statistically no different in weight.
How do we know that the scale is not broken? We need to test the scale before we do anything else. That was never done.
Last edited:
That's part of it. Another part is how you (and/or your room) responds to the "woofer bloom" inherent in box loudspeakers. Another is the problem of the vertical radiation pattern . . . and the effect(s) of lobing (high c to c spacing) at crossover on that ceiling reflection you've mentioned.
And there's always the problem, entirely out of our control, of how the sound is captured in the first place, and how it is "processed" in production.
There's perhaps something to be learned from musicians, who for the most part don't care about any of this. They listen to the music . . . and if the tone and timber is good don't much care, as long as the other "flaws" in the system are not too distracting. By focusing on the "details" of the system we make them distracting, and run the risk of losing the music in the process. Got to watch out for that . . .
Although we "fight in the same camp", I would have to disagree. You cannot generalize, SL himself in his test wrote:
System B can deliver a very satisfying AS and possibly a more accurate one than O
Considering that most recordings use microphone placements that are not representative of natural hearing locations, I can easily imagine that B renders an AS that is more true to the microphone location, than the more spacious and open rendering by O as for track #3, for example
This is quite true and why I don't think that musicians make good audio reproduction evaluators. The performance is everything. My piano friend, who I linked to earlier, was a very bad listener when I first met him. he could not even hear a section of his recordings that were clipping - until I pointed it out to him and how to detect it. Then he heard it. Even in his owns works, the performance was all he could hear. My brother is a musician and is the same way. Unless they try hard musicians are no better than anyone else at listening o reproductions, maybe even worse.
The issue that I am talking about is not about small numbers but about using a reliable gauge. Gauge capability, something that no one in manufacturing would ignore, is all but absent in audio testing. Once when Lidia and I were doing a subjective test we found that the "pro experts" gauge was less reliable than the guy off the street. Go figure!
So how would a proper test protocol look like to answer questions regarding spaciousness, dynamics, timbre, etc.? Would we need to rent Harman's test facility (if it would be possible)?
Markus
It is a whole lot easier to see the flaws in any given experiment than to design a good one yourself. This is certainly why so little is done. The questions that one asks have to be simplistic and well defined. Trying to "do it all" in one experiment is a sure road to failure. Harman's system would be great if it was available, but who would pay for that, and who would do the actual experiment.
The use of the phase "Acoustic Scene" was completely ambiguous and could mean anything to anyone. No example of good versus bad was available and the listeners were left on their own to define this metric as they saw fit. As soon as I heard that this was the metric, I said "No thanks". The "scene" was supposed to be a large hall. Kind of biases the results as well. So much of this kind of testing is done under the assumption that the goal is recreation of a concert venue. That is so limiting in scope that it ignores the reality of the marketplace today. There were no "points" for good or bad Imaging, or coloration, just "Plausible Acoustic Scene".
It is a whole lot easier to see the flaws in any given experiment than to design a good one yourself. This is certainly why so little is done. The questions that one asks have to be simplistic and well defined. Trying to "do it all" in one experiment is a sure road to failure. Harman's system would be great if it was available, but who would pay for that, and who would do the actual experiment.
The use of the phase "Acoustic Scene" was completely ambiguous and could mean anything to anyone. No example of good versus bad was available and the listeners were left on their own to define this metric as they saw fit. As soon as I heard that this was the metric, I said "No thanks". The "scene" was supposed to be a large hall. Kind of biases the results as well. So much of this kind of testing is done under the assumption that the goal is recreation of a concert venue. That is so limiting in scope that it ignores the reality of the marketplace today. There were no "points" for good or bad Imaging, or coloration, just "Plausible Acoustic Scene".
This is a bit out of left field, but having to do with reflection and directivity, A few years back I got to hear a very interesting experiment at a friend's house up in the high Sierras.
His listening room was his 2 car garage. The front wall (behind the speakers) was the garage door, about 16 feet wide. The first listen was with the door open. The dipole speakers had amazing depth. Harry James and his trumpet were out on the driveway, maybe 20 feet back. Depth and imaging were amazing with no front wall at all. Great depth, good location.
Push a button and down comes the garage door, quickly giving the room a front wall. Now the depth is limited to the door, more or less. Positioning not so precise. Everything is very much more ordinary sounding. On the plus side, the sound gained a good bit more energy in the lower mids, as so much energy was now coming back at us from the wall, and not lost out into the wilds of the Sierra Madre.
The same experiment with box speakers was less dramatic. Similar, but the effect was not as pronounced. This simple experiment cemented in my mind the need for diffusion behind the speakers. Break up the reflections, but retain the energy. THis wold be especially pronounced on open back speakers.
If you ever have the chance to set up an experiment like this, I highly recommend it. Removing the wall behind the speakers makes a big difference to the sound.
His listening room was his 2 car garage. The front wall (behind the speakers) was the garage door, about 16 feet wide. The first listen was with the door open. The dipole speakers had amazing depth. Harry James and his trumpet were out on the driveway, maybe 20 feet back. Depth and imaging were amazing with no front wall at all. Great depth, good location.
Push a button and down comes the garage door, quickly giving the room a front wall. Now the depth is limited to the door, more or less. Positioning not so precise. Everything is very much more ordinary sounding. On the plus side, the sound gained a good bit more energy in the lower mids, as so much energy was now coming back at us from the wall, and not lost out into the wilds of the Sierra Madre.
The same experiment with box speakers was less dramatic. Similar, but the effect was not as pronounced. This simple experiment cemented in my mind the need for diffusion behind the speakers. Break up the reflections, but retain the energy. THis wold be especially pronounced on open back speakers.
If you ever have the chance to set up an experiment like this, I highly recommend it. Removing the wall behind the speakers makes a big difference to the sound.
My conclusion would be a) make the front wall absorptive as much as possible or b) use in-/on-wall speakers.
Quite frankly I don't think that broadband diffusion is even possible in acoustically small rooms. Dimensions are simply too small.
It is not clear if this is due some distant reflections from outside that are being heard back ?
down to 300hz should be enough? That's feasible. but it would be large, hence ugly, so only for dedicated rooms.. We could maybe manage down to 6-800hz in living rooms?
I just finished reading one of Beranek's books on hall design. The feature that stroke me most (call me naive..) was that no sound leaking is permitted, all of the energy has to be kept and used, no absorption, only redirection. Which is understandable given the high audience. But why wouldn't it be the same for our rooms? Sound energy is precious!
On the other hand, I remember Rudolf's comment, that diffusion compared to strong reflexion made the image fuzzier but did not move it in space, or something like that..
I just finished reading one of Beranek's books on hall design. The feature that stroke me most (call me naive..) was that no sound leaking is permitted, all of the energy has to be kept and used, no absorption, only redirection. Which is understandable given the high audience. But why wouldn't it be the same for our rooms? Sound energy is precious!
On the other hand, I remember Rudolf's comment, that diffusion compared to strong reflexion made the image fuzzier but did not move it in space, or something like that..