Not surprisingly I would describe the situation as the exact opposite. To me my speakers offer a very convincing "they are here". The "you are there" is never convincing IMO (for any speakers not just mine). It's the room that makes this effect more than the speaker - my rooms are very live. Dipoles, being somewhat directive and constant work better in live rooms than "typical" piston based speakers because of the lack of any directivity control. But control the directivity, and better yet, make it even narrower, and the speakers work very well in lively rooms.
Well, it's your idea so I'll defer to your judgment. If you want it the other way around, you could do it as 1-cos(theta). Somehow, I have an intuitive problem with the notion that the way off-axis stuff takes up more of the graph than the listening-window stuff but that's just me.
Last edited:
I am certain that none of you scientifically sophisticated and erudite posters would be devoting such energy and warmth to this discussion without trustworthy evidence of human perception underlying your strongly held assertions.
You know... blind or double-blind tests, behind-the-curtain loudspeakers, ABBA comparisons, a little stats, quantitative judgments of flute placements along the front wall, subjective ratings of the soundstage.....
Perhaps links to this evidence could be provided?
Or perhaps not.
is it fair of me to ask for evidence?
When Floyd Toole was a researcher at the Nat Res. Council of Canada, I read a fair amount of his work (and I was on a Board of Directors of the NRC in a related division). Truly impressive efforts for an engineer working as a psychologist. I'd say he made a far better psychologist than I would make a bridge designer.
You know... blind or double-blind tests, behind-the-curtain loudspeakers, ABBA comparisons, a little stats, quantitative judgments of flute placements along the front wall, subjective ratings of the soundstage.....
Perhaps links to this evidence could be provided?
Or perhaps not.
is it fair of me to ask for evidence?
When Floyd Toole was a researcher at the Nat Res. Council of Canada, I read a fair amount of his work (and I was on a Board of Directors of the NRC in a related division). Truly impressive efforts for an engineer working as a psychologist. I'd say he made a far better psychologist than I would make a bridge designer.
Last edited:
I am not inclined to answer such a post, but the current discussion is purely mathematical. There is no psychology involved. So what evidence do you need? That the area of an angular arc of dTheta is sin(theta) dTheta? Would you understand it if I gave it? Or maybe you mean that what we perceive in a room in the power response and not the direct sound? Personally, I am confused about what "evidence" you are looking for. Care to illuminate your post?
Maybe Arcsin(theta) was correct, I just have not followed it through. But I know that the angles should be closely spaced at 0 and widely spaced at 90.
But, yes, the off axis stuff really does take up more of the graph if its "power" that we are talking about. So what you are saying is no different than the other post. Is it the direct field that we hear or the reverberant field? Well its both!! It depends on what you are talking about. For imaging its almost all direct field, but for coloration and spatiousness its almost all reverberant. Its just not a simple situation. You need to get them BOTH right.
Last edited:
Honestly, you are the last person among these posters I would have expected to dismiss my question since Dr. Lee has some credentials in behavioral science, if I read the commercially-purposed prose of your website correctly. Can you please ask her what she thinks of my request? I gather from your website that she is a psycho-acoustics Ph.D. - what is her background and credentials that you are so justifiably proud to extol at your website? Sounds quite relevant and you say it is quite relevant.
It is absolutely clear that the entire purpose of this thread is to establish great music reproduction as perceived by human listeners, not niceties of abstract geometry. So I don't understand why you fault me for not allowing it is all about geometry.
Arcsin crashes if it's trying to work with a number greater than one.
Thinking more about the reverberant field, I don't really see why any angle is more important than any other. We aren't projecting forward onto a plane. Any given angle is going to reflect off the room's walls and they all contribute about equally after a few bounces so linear scaling of the angle axis is probably pretty good if that's what you want to look at.
FWIW, reading some of Sean Olive's recent papers, if you were going to do a single anechoic measurement to try to approximate the reverberant response, 30 degrees off axis would probably come as close as any. He doesn't say that exactly and it depends on lots of things but that's my take on simplifying plenty of complex math and statistics that he presents.
For properly setup accurate speaker, shouldn't the "you are there" or "they are here" impression depend on the recording environment and recording techniques?
Thanks.
People are readily self-deluded, esp. when they have some personal attachment to the device under examination. That is why YOU MUST have blind or double-blind testing to show what factors in directivity matter, if any. Or are we talking about the kind of wine assessments you read in the newspapers?
Many years ago while doing social survey research for a DOT, I made a startling discovery of great engineering, social, and economic significance that could change the shape of cities around the world as we know them. I discovered there is one and only one type of highway fence of reasonable size that was effective in blocking noise from highways. It is the fence the homeowner designs and builds themself. Works every time to some degree yet other engineered solutions invariably fail.
It is true that it's not so simple. If the original recording was done in an ordinary live performance environment, the room reflections in the recording environment will be recorded. However, there are a few things that limit revealing playback of these by creating a masking effect.
1. Room reflections and room modes: This is why we have to try to keep these under control.
2. Delayed release of acoustic energy from the drivers: This is why we have to try to keep CSD under control.
3. Interaural of two channel system: This effects everything, and there seems no simple answer how to control this using speakers.
I am the first to agree with your position, but I'm still a little unclear on the specifics. Is there any evidence that directivity matters? Absolutely - just read Floyd Tooles book. Or is there some other aspect of the discussion that you are alluding to. When you made your first comment we were talking about sound power and the kind of scale that would better show sound power versus angle. I did not see how this needed any supporting "evidence".
All researchers in sound now agree that Constant directivity is important. There is some solid evidence but a lot more circumstantial evidence supporting this fact. There is no, or scant evidence as to what the width of this controlled constant directivity should be. Floyd gives some opinions, but they are based on some relatively weak arguments, which even he admits to.
When it comes to directivity then, there is almost no evidence to go on, which, of course, is why there is such a wide disagreement about it. Do we then have to be careful about claims and beliefs - absolutely. But there is a lot of data about how we hear and if one uses this data it is possible to arrive at some substantial hypotheses about what the directivity should be. But the fact is that there are virtually no blind or double-blind tests on directivity to either prove or disprove any hypothesis. Should then we all just pack up and go home? I thnk not. Sure it would be great to do the basic psycho-acoustical studies to put this issue to rest, and if you are willing to fund it, or know someone else who will, then I'll get right on it.
The fact is that there is not likely to be "evidence" in this regard for a very long time. Floyd and I had a long talk about what such a test would even look like and concluded that it would not be trivial. Interaction with the room is a major problem and now its not just speakers that you have to test, but speakers AND rooms. Very quickly the problems become viurtually insurmountable.
Hence, for the forseable future all we have to go on is indirect evidence about how we hear and hypotheses about what kind of directivity best fits what we know.
That would be desireable. "You are there" or "they are here" - what parameters of a room's sound field correlate to which perception? Nobody really knows because no scientific data is available. All we know is that the reflection patter is the key.
Ditto.
Yep. And that leads to what? Best Practices? That's about all we can do in the real world. Nothing wrong with making those practices better informed, tho. Seems that is what we all strive for.
Precisely correct except for "who bases their product line on directivity theories", which is uncalled for, not accurate, and is certainly not possible from the exceedingly small volume of product that I make. But leave off the "slam" and its perfectly accurate.