I think it is wise to post articles on how we hear (i.e., humans) vs. issues with multiway loudspeakers. Tom Danley of Danley Sound Labs (the maker of Synergy multiple entry horns) recently talks about the issues with having vertical or horizontal separation of drivers in a loudspeaker and how that interferes with the perception of stereo imaging:
I recommend watching the entire 10 minute video to understand before commenting or asking questions.
Chris
I recommend watching the entire 10 minute video to understand before commenting or asking questions.
Chris
Great topic Chris! Fascinating points about how we hear especially the head "shadow" and notch in hearing for left to right and up/down.
An important takeaway for me is the speaker announcing its distance overlaying/confusing the recording attempting to convey distance.
In this slightly older video he makes more explicit that it is the lack of "summing" of the drivers in a conventional speaker which announces its location/distance.
One question that comes to mind is what is the hearing mechanism that perceives distance? I would think the ratio of direct to reflected sound would be part of it.
An important takeaway for me is the speaker announcing its distance overlaying/confusing the recording attempting to convey distance.
In this slightly older video he makes more explicit that it is the lack of "summing" of the drivers in a conventional speaker which announces its location/distance.
One question that comes to mind is what is the hearing mechanism that perceives distance? I would think the ratio of direct to reflected sound would be part of it.
Hi Mark.
I think the answer is fairly complicated, i.e., Toole references several reports in his first edition ("Perceptions of Distance", pgs. 174-175)--references that are absent in the third edition of his book - Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms. A quote from Toole's first book at the reference cited above:
I think that Danley has uncovered a relatively new component of distance perception with his observation of point source loudspeakers. removing an artificial cue to distance perception.
Apparently, it's not a simple answer based on notches in SPL response at the eardrums vs. elevation angle (that Danley described in the video). The fact that Toole removed all his references to distance perception (now called "localization" in the 3rd edition) says to me that this is an area of research that's still outstanding.
Chris
Last edited:
I think a portion of the effect that Danley reports is phase fidelity of the loudspeaker, i.e., getting the loudspeaker's phase response substantially flattened. If you look at the phase response of typical direct radiating loudspeakers not using DSP crossovers, the phase growth from 100 Hz to 20 kHz is typically 900-- 1400 degrees...or more (as reported by Toole).
I experienced this image deepening effect as part of the "subconscious auditory effect" of linear phase loudspeakers that also control their polar coverage angles down to the Schroeder frequency of the listening room. Here is a link to the thread where I first reported the rather startling subconscious effects of getting the phase flattened and corresponding excess group delay controlled above 100 ms using a pair of slightly modified Klipsch Jubilees (original edition):
https://community.klipsch.com/index...speakers/page/3/&tab=comments#comment-2379562
The yellow trace, measured using REW in my listening room at 1m, is shown below as compared to a Danley SH-50 (cyan trace)
![1062657154_TADTD-4002Jubileevs.DanleySH-50phaseresponse.jpg.4736f02d6c9d60a84ff90fd06beb3651[1].jpg](https://www.diyaudio.com/community/attachments/1062657154_tadtd-4002jubileevs-danleysh-50phaseresponse-jpg-4736f02d6c9d60a84ff90fd06beb3651-1-jpg.1110546/ "1062657154_TADTD-4002Jubileevs.DanleySH-50phaseresponse.jpg.4736f02d6c9d60a84ff90fd06beb3651[1].jpg")
I also lowered the K-402 horns 8" (20 cm) to get the 450 Hz crossover point well within 1/2 to 1 wavelength. What I experienced was large jump in the perceived depth of stereo images (and 5.2 array imaging), along with a dramatic increased perception of deep bass response/ naturalness and elimination of perceived harshness, as reported in the linked thread.
So there are several effects of cleaning up the phase and SPL response (which together are also called the "transfer function response"), along with substantially co-located drivers (within 1/2 wavelength of the Jubilee and 1/4 wavelength at the center K-402-MEH), and controlling early reflections via full-range horns down to ~100 Hz (about the Schroeder frequency of my listening room).
Chris
I experienced this image deepening effect as part of the "subconscious auditory effect" of linear phase loudspeakers that also control their polar coverage angles down to the Schroeder frequency of the listening room. Here is a link to the thread where I first reported the rather startling subconscious effects of getting the phase flattened and corresponding excess group delay controlled above 100 ms using a pair of slightly modified Klipsch Jubilees (original edition):
https://community.klipsch.com/index...speakers/page/3/&tab=comments#comment-2379562
The yellow trace, measured using REW in my listening room at 1m, is shown below as compared to a Danley SH-50 (cyan trace)
I also lowered the K-402 horns 8" (20 cm) to get the 450 Hz crossover point well within 1/2 to 1 wavelength. What I experienced was large jump in the perceived depth of stereo images (and 5.2 array imaging), along with a dramatic increased perception of deep bass response/ naturalness and elimination of perceived harshness, as reported in the linked thread.
So there are several effects of cleaning up the phase and SPL response (which together are also called the "transfer function response"), along with substantially co-located drivers (within 1/2 wavelength of the Jubilee and 1/4 wavelength at the center K-402-MEH), and controlling early reflections via full-range horns down to ~100 Hz (about the Schroeder frequency of my listening room).
Chris
Here is a researcher getting at the phase and reflection issues. Interesting assertion in his conclusions that "proximity"/ "coherence"/ "presence" are perceived by the inner ear through the "phase coherence of upper harmonics in the direct sound". And, that coherence is "randomized" muddied, made incoherent by "early reflections"
He is talking about concert halls but what might the lessons be for our listening rooms/ speaker designs/mods?
He is talking about concert halls but what might the lessons be for our listening rooms/ speaker designs/mods?
To answer my own question, part of the answer might be that for best fidelity/proximity/involvement one should only listen to binaural recordings on headphones that have been custom equalized for your ears. 
Eye opening assertion that when doing that custom equalization ordinary listeners vary in the critical midrange (not just in the highs) by as much as 8db!
Eye opening assertion that when doing that custom equalization ordinary listeners vary in the critical midrange (not just in the highs) by as much as 8db!
Would you hire a psychologist to design your bridge?
Would you adhere to the ramblings of an engineer pontificating on how hearing works? Amateurish gibberish. He doesn't even have the terms for describing what he seems to be trying to talk about*. (OK, Toole, trained in physics, is an honorary psychologist after 45 years study).
BTW, the last minute of the video endorses the value of dipoles (which spread the sound around) for great stereo localization. I have to keep my large panels just inches apart or if further apart the ping-pong stereo is too annoying for me. Counter-intuitive?
B.
*and seems he never heard of Tannoy dual-concentric drivers
Would you adhere to the ramblings of an engineer pontificating on how hearing works? Amateurish gibberish. He doesn't even have the terms for describing what he seems to be trying to talk about*. (OK, Toole, trained in physics, is an honorary psychologist after 45 years study).
BTW, the last minute of the video endorses the value of dipoles (which spread the sound around) for great stereo localization. I have to keep my large panels just inches apart or if further apart the ping-pong stereo is too annoying for me. Counter-intuitive?
B.
*and seems he never heard of Tannoy dual-concentric drivers
from Toole, 3rd ed. (2018), pg. 193, 7.4.7 Floor Reflections: A Special Case?
I also don't see why you find that ad hominem attacks are relevant to the subject of this thread. That's usually the tactic of those that have little to offer.
I have been amazed by the performance of Danley's MEH designs, regardless of his "apparent status" within the psychoacoustics realm. I trust his observations.
Chris
I don't believe that I would hire you to design a bridge...
I also don't see why you find that ad hominem attacks are relevant to the subject of this thread. That's usually the tactic of those that have little to offer.
I have been amazed by the performance of Danley's MEH designs, regardless of his "apparent status" within the psychoacoustics realm. I trust his observations.
Chris
I've referenced that Griesinger PowerPoint presentation several times when talking about clarity (not necessarily perception of distance). I didn't know that there was a YouTube video of him delivering the presentation. Thanks for that.
For me, it's probably the key insight to perception of the effects of phase growth in loudspeakers.
Chris
Thanks very much for an interesting thread like this. Not new material to me but that video of Mr. Griesinger is gold. I had read the papers and power points, but hearing him explain it all is much more fun and just might sway more people to pay closer attention to phase. 
The Mr. Danley talks are good stuff too, though not new to me, I'll admit to having read all of Mr. Danley's posts on this forum and most certainly on a few others as well.
I'd say low diffraction is one of the things that helps hide speaker position/distance... With a horn it can be the mouth termination, for instance the secondary flare. With direct radiators a smooth flowing shape of the enclosure instead of using sharp corners.
The Mr. Danley talks are good stuff too, though not new to me, I'll admit to having read all of Mr. Danley's posts on this forum and most certainly on a few others as well
.I'd say low diffraction is one of the things that helps hide speaker position/distance... With a horn it can be the mouth termination, for instance the secondary flare. With direct radiators a smooth flowing shape of the enclosure instead of using sharp corners.
Last edited:
There is an excellent book entitled "Why You Hear What You Hear" by Dr. Eric Heller. It's 500 pages of more than most people want to know about our hearing. I highly recommend it. Also, Leo Beranek touches on hearing in "Music, Acoustics & Architecture", Olson addresses the hearing mechanism in "Music, Physics and Engineering", and Knudsen & Harris also do in "Acoustical Designing in Architecture". There are numerous relevant papers in the archives of the Journal of the Acoustical Society of America. The AES might have some which address hearing as related to loudspeakers.
Personally, I'll offer that proximity (or lack thereof) to a natural acoustic source involves tonal balance (timbre). For example, take a person standing in front of you and talking, versus a person who is some distance away - the timbre of their voice is markedly different in each case. Same for musical instruments. In an enclosed space, the characteristics of the reverberant field also tell us if the source is near to us or farther away. Good mixing engineers can mimic these differences through judicious use of various controls to place instruments and vocals where they want them: up front, way in the back, or somewhere in the middle of the soundstage depth. Ditto for left / right placement.
Regarding loudspeakers, we've been living with multi-way systems with separated drivers almost since the beginning of time. People have been gushing over the sound of many models for decades. While point sources, such as the Tannoy models and the various Altec 604 incarnations, offer high quality sound without the inter-driver lobing, there are many non-point-source speakers which are excellent. If a speaker is well-designed, there is no reason why even a four-way system can't produce superlative sound quality.
Indeed since it takes both a flat phase and amplitude response to reproduce a square wave.I had read the papers and power points, but hearing him explain it all is much more fun and just might sway more people to pay closer attention to phase.
I'd say low diffraction is one of the things that helps hide speaker position/distance... With a horn it can be the mouth termination, for instance the secondary flare. With direct radiators a smooth flowing shape of the enclosure instead of using sharp corners.
Agreed, I learned this early on in my tweaking years of the Altec 511 reciprocating flare horn by damping the horn mouth to reduce/eliminate eigenmodes across it and back to the throat to modulate its mids/HF output (and much more, but strictly related to this highly resonant horn design). Cone/dome speakers did best overall with a baffle that was just wide enough to mount them, then slanted back at ~12 deg included same as for room walls to deal with 'slap' echo, though confirmed strictly empirically because I didn't want to take the time to do proper curves.
Over at headphones-amplification I reported experimental results on "open-wing crossfeed stereo sound" which relied crucially on reducing both LF and HF to get the singer to sing TO me with band behind on stage (tonality shift in agreement with Just Dave's comment above), as well as direction of sound entering both ears (slightly elevated front-L/R).To answer my own question, part of the answer might be that for best fidelity/proximity/involvement one should only listen to binaural recordings on headphones that have been custom equalized for your ears.
Eye opening assertion that when doing that custom equalization ordinary listeners vary in the critical midrange (not just in the highs) by as much as 8db!
https://www.diyaudio.com/community/threads/open-wing-headphone-crossfeed-stereo-sound.391630/
Unfortunately it has become very cumbersome to watch YT from where I am.
Last edited:
Griesinger is the researcher with regard to hearing (most particularly in concert halls, but overall including biological responses like ear and brain mechanisms where he is more like Toole as someone that synthesizes, and expands on, lots of work from many others).
While I like Tom Danley he doesn't even rate a comparison in this category.
Last edited:
Toole writes about tests with test signals while "Archimedes" research and Lyngdorf remarks were about tests with recordings of musical material
Why not take it one step further. Making sure you do get the right phase at the listening spot, if lining up the harmonics is important like Mr. Griesinger suggests, it would bring you closer to what's in the recording. That doesn't suggest that the recording itself is pristine enough to still have those harmonics lined up though. But you kind of are removing another veil, getting a little step closer to the source.
I'ts clear that Griesinger is talking about performers in a hall and Toole is talking about reproduction of a recording in a room, but if I have to agree with one of them, I'd pick Griesinger's version and avoid early reflections and get the phase right. Even in the home situation reproducing music.