That is fine as long as the "pulse train" is synchronous with the excitation. Hence this cannot be the case at frequencies much above about 2 kHz. The random nature of the pulses would obliterate these crossings at the higher frequencies.
Read the Griesinger papers on harmonic phases and add in the loss of synchronous capability above say 2 kHz. This would imply that the detection of Group delay should have a peak at about 2 kHz, which is the case as found in practice. Griesinger's paper explains a lot.
I ordered some EM 173s and WM 61As today. I hope they actually have them in stock..
FEL Connectors and Components
I was just reading one of the Griesinger papers and found a quote that I wanted to post in the hope that it will put to rest the discussion about the need for directivity control below 700 Hz - because our localization perception is already poor in that frequency region.
"Experiments by the author and with students from several universities discovered that the ability to localize sound in the presence of reverberation increased dramatically at frequencies above 700Hz."
Need I say more!
"Experiments by the author and with students from several universities discovered that the ability to localize sound in the presence of reverberation increased dramatically at frequencies above 700Hz."
Need I say more!
But we already knew that. The claim was that directivity control below 700 Hz, while less important than above 1 kHz, aids localization and improves clarity. And the above statement does not refute this claim. Can you provide a link so that we can understand the quote in its proper context?
Ah, but you see my claim was never that it made no difference, only that it is far less important than the critical range of 700 - 5 kHz. If I have to compromise on something it is certainly going to be < 700 Hz. There were some who claimed this region to be "more important", so if you are not one of those then we are not at odds.
Here is the link:
http://www.davidgriesinger.com/
Here is another quote that I really liked!
"Toole and Olive [5] studied the audibility of individual reflections in rooms, and in [6] Toole implies that if individual reflections are below the level of audibility with respect to the direct sound they can be ignored. This is clearly not the case. In small rooms individual reflections are almost always inaudible individually, but there are a great many of them, and their sum is highly audible"
This is exactly where I differ from Toole! I think Griesinger and I see many things alike!
Both quotes are from the series of three papers on "envelopment". Must reads IMO.
Here is the link:
http://www.davidgriesinger.com/
Here is another quote that I really liked!
"Toole and Olive [5] studied the audibility of individual reflections in rooms, and in [6] Toole implies that if individual reflections are below the level of audibility with respect to the direct sound they can be ignored. This is clearly not the case. In small rooms individual reflections are almost always inaudible individually, but there are a great many of them, and their sum is highly audible"
This is exactly where I differ from Toole! I think Griesinger and I see many things alike!
Both quotes are from the series of three papers on "envelopment". Must reads IMO.
Last edited:
Thanks for the links, will read up.
Right. If you want an engineered solution, control directivity above 1 kHz. If you want the cost/space no object solution, control directivity as low as you can.
The Toole quote is something I've experienced. If you keep on raising the volume of your speakers, there comes a point at which the music just becomes incomprehensible. You can actually hear the reverberated sound as a separate entity, and it's so loud that you cannot follow anything. This is where directivity helps. It increases the direct to reflected ratio, which increases clarity. And from my very limited experience, the increase in clarity improves with directivity control of the lower frequencies.
These are all unproven hypothesis
But this is a good place to throw out some ideas.
Right. If you want an engineered solution, control directivity above 1 kHz. If you want the cost/space no object solution, control directivity as low as you can.
The Toole quote is something I've experienced. If you keep on raising the volume of your speakers, there comes a point at which the music just becomes incomprehensible. You can actually hear the reverberated sound as a separate entity, and it's so loud that you cannot follow anything. This is where directivity helps. It increases the direct to reflected ratio, which increases clarity. And from my very limited experience, the increase in clarity improves with directivity control of the lower frequencies.
These are all unproven hypothesis
But this is a good place to throw out some ideas.
Last edited:
Earl,
Though I understand that comment about our ability to localize increasing above 700hz, does that mean that we do not have any discrimination below that threshold? I am not sure that is what is being said here, but that our hearing gets much more acute above this range. I will have to read the paper to make that judgement.
Though I understand that comment about our ability to localize increasing above 700hz, does that mean that we do not have any discrimination below that threshold? I am not sure that is what is being said here, but that our hearing gets much more acute above this range. I will have to read the paper to make that judgement.
You make it sound like everything below 700Hz is meaningless because of this.
The Griesinger quote would make me think that anything below 700Hz needs even more attention if localization becomes so poor in the presence of reflections.
We localize only on signals > 700 Hz.
I completely agree with Griesinger on this point - and he makes a very strong case for this in his papers. I wish that I had seen these papers before. They agree with what I have been saying in almost every detail. Thanks Elias.
Below 700 Hz, I think that he would. To be clear, his interest is not in loudspeaker design, but in concert halls. But what he has found for concert halls is very applicable to the small room loudspeaker problem. He uses some pretty long time constants (like 100 ms.) and I am not convinced that in a small room shorter times might be more useful, but his basic premise and concepts are right on the money.
I plan to implement his metric for localization and envelopment. It is just what I have been looking for.
I have written him and will try and open a dialog to see if he sees the application of what he is doing to loudspeakers the same way that I do. We will see what he says (if he even responds. I wouldn't get into the audiophile realm if I were him. I'd just stay away. Too many wackos)
I believe that it is a very small aspect of the problem. We localize only on sounds > 700 Hz. so very early reflections < 700 Hz do not cause image problems.
Sorry but this is NOT what he is saying. He is talking about localization of natural sound sources in acoustically small rooms but our topic is sound reproduction in acoustically small rooms.
If our reproduction system needs to replicate poor localization performance of a small room then it should be capable to do so but it should also be capable of reproducing a room that doesn't exibit small room acoustics.
Wrong. At least, for some of us,
. High levels of audible distortion reduce the ability of the ear/brain to decode the acoustic information it receives, it then overloads, and gives up.My personal experience is that if the "transparency", the level of obvious distortion is sufficiently low, then the SPLs can be raised almost without limit without perception problems. Since the vast majority of systems aren't capable of this performance the process is not well understood by most ...
Have there been any studies done to quantify comprehensibility of sound as a function of loudness AND distortion levels?
Frank
Frank,
I don't know about you but there is a limit to the upper spl level that I can handle and as you approach that limit I can hear my own hearing start to distort. It is not the source material as I can increase my distance to the source and the distortion drops back out. So I would say that we do have some kind of spl limit that does warn us when we are approaching the danger zone.
I don't know about you but there is a limit to the upper spl level that I can handle and as you approach that limit I can hear my own hearing start to distort. It is not the source material as I can increase my distance to the source and the distortion drops back out. So I would say that we do have some kind of spl limit that does warn us when we are approaching the danger zone.
