Others are seeing your comments as a complete dismissal of what I am saying. I don't think thats true, and I would suggest that for the most part you buy into the concept of minimizing the VER and maximizing the later reflections as being the best that two speakers can do. That you and others would like this to go further than what is possible with two channels is fine, but two channels is what I accept as the defacto standard and seek to maximize that scenario. Do you see alternative two channel designs that achieve both good image and good spaciousness (as I define them)? (I am well aware of your prference for multichannel, but lets not confuse the issues here.)
underline mine
yes, subjectively there is no ideal directivity pattern for stereo speakers, there cannot be, each has His own subjective ideal
it would have to be objective ideal, accuracy of reproduction can be no such ideal because of lack of standardization of production
realism of sound reproduction - in terms of sound source recognition - could be but this ideal is rejected
perhaps because of the simple fact that conventional stereo just doesn't sound real at all, real in the meaning of the word as Markus used it here:
yeah, and conventional stereo may sound more precise in comparison but essentially not realistic
Last edited:
Poor choice of words on my part. I really meant the problem doesn't apply at lower and higher frequencies. I'm thinking of the bookshelf speaker on a stand or the tower with the woofers near the top. Typically the floor bounce null is in the low midrange, below that frequency the woofer is sufficiently close to the floor for there to be no issue.
David S.
Last edited:
That's absolutely NOT my intention and it's also not what I've said.
Only a high/constant directivity design will create a reflection free zone without further room treatments/wall construction.
Any RFZ design will lack early spatial impressions. I would even go as far and say that it is the main attribute of most stereo recordings. There's nothing wrong with it, it is part of the art. It's a cultural phenomenon just like concert halls are.
But any "real" listening venue will also lack any reflections < 10 ms. Unless you can figure out how to crame an orchestra into a living room.
This is where I fail to see the argument - no "real" venues have VER < 10 ms., so how is it that in a playback situation they are deemed to make the sound more "real"? I don't get it.
Oh yea! Its that "It sounds good to me!" thing again. We don't need logic and science when subjectivism is the end goal.
I strongly believe that the strength, direction and arrival time of the earlier reflections are significant. What I am not convinced about is that there is one ideal polar curve that has an advantage over others. While you can knock down the level of earlier reflections with high overall directivity you can't ignore Tooles oppinion that a number of people will be unhappy with the dryer sound that comes along with it. I have experience that myself with an experiment to block sidewall reflections. The image became more between-the-speakers and artificial.
I'd really like to think that treating the room or designing room and speakers in concert to reduce all the early energy without reducing later energy at all, might be the best of both worlds (image specificity and sense of envelopment) but even that is unproven. I do think Ken Kantor's approach has been the only system that truly aligns itself with what we know about psychoacoustics.
What I don't think works are a wide range dipolar dispersion, a standard cardioid dipsersion pushed down to lower frequencies (resistance slots), a full omni pattern, a horizontal omni pattern, negative d.i. approaches (blocking the axial response) or a somewhat CD pattern with high directivity for the upper range and totally conventional directivity below 1k.
The arguements made for all these are more about "polar idealism" than psychoacoustics.
David S.
Last edited:
But any "real" listening venue will also lack any reflections < 10 ms. Unless you can figure out how to crame an orchestra into a living room.
This is where I fail to see the argument - no "real" venues have VER < 10 ms., so how is it that in a playback situation they are deemed to make the sound more "real"? I don't get it.
Oh yea! Its that "It sounds good to me!" thing again. We don't need logic and science when subjectivism is the end goal.
I'd say most environments we spend most of our time in are in the <10ms range. It's not about recreating the acoustics of a concert hall but about creating some early spatial impression which adds realism. The fact that we're already adapted to our own listening room may play a major role.
I don't understand the argument about cramping an orchestra into a living room. A RFZ design cramps an orchestra into a even smaller space, the space between the speakers.
I think you have to take into account the acoustics of the listening room and how a given speaker is going to interact with that, when you go to define what the best polar pattern. Some rooms are good enough that an omni speaker may be the most enjoyable (see Linkwitz). Bad rooms may do much better with controlled directivity. If the inter-aural crosstalk is going to destroy imaging in the 100HZ - 700HZ range anyway, then side wall reflections may recreate a sense that there is wide stereo "spaciousness" in that range, and be preferred, even though the spaciousness is "fake". Given how ridiculous most recording techniques are, I don't necessarily think that using such techniques to create "fake" subjective improvements is blasphemy. In the end, it's about enjoyability factor. When someone asks me for advice on what speaker they should get, I take a close look at the room first.Where would the speakers go? Where would they usually sit for listening? Etc. Each speaker design has a right room, and vice versa.
Maybe in theory, but the phantom images are also in the median plane (or close to it)
Why we hear phantom images in the first place? How did they ever help human to survive in evolution?The freq bands are, incidentally(?), the same where ITD and ILD take place and dominate.
I also believe stereo is a huge misunderstanding !
- Elias
Hi Guys
In several earlier posts I have suggested an experiment where you take your system outdoors (if you can) and set it up with the same or similar geometry as in your home. The reason is what Earl points out, that the research being cited did not include a test involving very low levels of reflected sound.
To me, I don’t see how one can say X doesn’t matter if you never have the chance to judge with and without X and lacking speakers with significant directivity, the next best thing is normal speakers with no walls or ceiling to cause reflections. THEN one can hear the “without” case and decide how valuable close reflections are for a stereo image. .
I suppose lacking an outdoors, one could try near field listing indoors by setting up in the middle of the room with the speakers say 3 feet away.
As i also described, in the recording process, the environment is also one where there are an absolute minimum of close reflections either with horn loaded on absorptive soffit mounted speakers or small wide dispersion point sources very close to the mix engineer and far from any side reflections.
The whole point is that when a pan pot is centered, one should have the image directly in front of you and have it be assignable from left to right when composing a mix. In other words, to make it as real as possible.
After fiddling with directional speakers and stereo capture on my own, my conclusion is that close reflections harm the image between the speakers, that between the speakers is the ONLY image that can be recorded / reproduced with two channels without exploiting hearing trickery.
Also, with directional speakers, the reproduction of the stereo image in the room is a lot closer to being outside and the listening position frequency response is much more like the near field response than using wide angle speakers in the same location. I am a little surprised that the reflections that cause the great ugliness in the wide angle’s response at the LP aren’t thought of as a negative thing or clue.
If you want to make something spacious or enveloping with DSP this at least can be turned off when desired but room effects are always on.
Not only that but when you can make a solid phantom image anywhere between any two speakers, the solution to greater envelopment is having more channels.
Hey Earl, how have you been? A busy summer here, finally the mosquitoes have let up.
Best,
Tom Danley
In several earlier posts I have suggested an experiment where you take your system outdoors (if you can) and set it up with the same or similar geometry as in your home. The reason is what Earl points out, that the research being cited did not include a test involving very low levels of reflected sound.
To me, I don’t see how one can say X doesn’t matter if you never have the chance to judge with and without X and lacking speakers with significant directivity, the next best thing is normal speakers with no walls or ceiling to cause reflections. THEN one can hear the “without” case and decide how valuable close reflections are for a stereo image. .
I suppose lacking an outdoors, one could try near field listing indoors by setting up in the middle of the room with the speakers say 3 feet away.
As i also described, in the recording process, the environment is also one where there are an absolute minimum of close reflections either with horn loaded on absorptive soffit mounted speakers or small wide dispersion point sources very close to the mix engineer and far from any side reflections.
The whole point is that when a pan pot is centered, one should have the image directly in front of you and have it be assignable from left to right when composing a mix. In other words, to make it as real as possible.
After fiddling with directional speakers and stereo capture on my own, my conclusion is that close reflections harm the image between the speakers, that between the speakers is the ONLY image that can be recorded / reproduced with two channels without exploiting hearing trickery.
Also, with directional speakers, the reproduction of the stereo image in the room is a lot closer to being outside and the listening position frequency response is much more like the near field response than using wide angle speakers in the same location. I am a little surprised that the reflections that cause the great ugliness in the wide angle’s response at the LP aren’t thought of as a negative thing or clue.
If you want to make something spacious or enveloping with DSP this at least can be turned off when desired but room effects are always on.
Not only that but when you can make a solid phantom image anywhere between any two speakers, the solution to greater envelopment is having more channels.
Hey Earl, how have you been? A busy summer here, finally the mosquitoes have let up.
Best,
Tom Danley
This has not been my experience at all ! These high freqs are the one that kill the phantom imaging for me. If the speaker is high directivity at the treble range, be it a horn, a waveguide or even a direct firing dome, the image will be in the tweeter and remain there without wanting move out from there.
Maybe it's true for you, but the only way to know is to try it by yourself. Good luck !
- Elias
It's a good point to stop and think: What is sound? What is music?
The information in music is a set of modulations, AM and FM but mostly AM. Human demodulates the signal to extract the information.
My current understanding is that an optimum system is one which optimises the modulation transfer from the recording to the listening position.
It's something I try to do for example here:
http://www.diyaudio.com/forums/multi-way/189533-monopole-vs-dipole-bass-small-room-battle-continues-wavelets-rescue.html
I started from the bass, but I'm planning to see the other freq bands as well (as time allows).
- Elias
But this is the freq range where the cross talk is required according to the stereo theory ! Once again, the amplitude information from the two stereo channels form an interference field at the listening position, from which the ITD is extracted. This is in case of amplitude panning.
- Elias
There is no reason to hear phantom images ! It is a moulding defect in the human auditory system !
- Elias
That range of dB is quite close to correct par and parallels my findings. But I also find out the level can be quite critical too and thus the system can be challenging to tune.
Nevertheless, it rules out high directivity speakers with toe in, if spaciousness is required.
- Elias
Tom,
There's an easier way to perceive direction, detail and dynamics of a recording: use headphones.
I too don't believe that a conventional wider dispersion design has any advantage over a RFZ approach but things change when a design is capable of creating very strong ipsilateral reflections (-5dB and higher). This adds a spatial sensation which is absent in "normal" stereo.
The "right" way to do it is probably more speakers but this thread is specifically about stereo speakers.
There's an easier way to perceive direction, detail and dynamics of a recording: use headphones.
I too don't believe that a conventional wider dispersion design has any advantage over a RFZ approach but things change when a design is capable of creating very strong ipsilateral reflections (-5dB and higher). This adds a spatial sensation which is absent in "normal" stereo.
The "right" way to do it is probably more speakers but this thread is specifically about stereo speakers.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.