I've wondered about this too. Conductors are much more in the direct field and less effected by hall reverberation, but how do they know what sounds good out in the hall when they are "sampling" from a much different place?
I do know that orchestra members are very critical of the characteristics of the orchestra shell. Ensemble playing depends on their ability to hear themselves and each other. Local diffusion is a good thing.
David S.
Perhaps conductors have the ability to go into the hall during some rehearsals, with some learned ability to mentally compensate to some degree for the empty seats having less damping. Some of the student conductors used to do this when I was in college. Also, I assume that conductors would be able to hear their ensemble led by a guest or assistant conductor or else another ensemble playing in said venue. During their training and ascent to their position, they may have had some external feedback/validation to calibrate their learning process and experience along the way.
Good survey - thanks
Clearly these results contradict any belief that early reflections are not an issue. There is a distinct difference in reflections above and below 10 ms and the direction of the reflections is extremely important. To not take these effects into account in both the speaker and room design is a serious failing.
Lidia and I reported this same effect, but for "reflections" that were so short in time that they would best be described as "group delay" - 1 to 2 ms. There the perception increased with level being inaudible at lower SPL levels.
If one thinks about this for a moment they will realize that this effect would sound very much like nonlinear distortion, but yet it is a completely linear effect. This means that our perception is nonlinear, which makes any discussion of the perception of nonlinear distortion extremely complex. How would one know that what they were hearing is a nonlinear effect as opposed to a nonlinear perception?
I assume that your calculator is correct, yes. Thats an excess or differential distance remember. That kind of time delay is extremely hard to do in any typical listening room unless one takes advantage of some very directional loudspeakers or makes the room completely dead. I much prefer the former to the later.
with all due respect but - is that so?
if "reflections in the same direction as the first incidence can be 5 to 10 dB above the first incidence before detection" and "reflections have to be 7 dB above the absolute detection threshold before they cause an image shift" and if "just above the noticeability threshold" they just "increase spaciousness" then where is the issue?
I have the same question. In my listening room front wall reflections have essentially no effect on "imaging", but do significantly increase "spaciousness" (and this is with full-range dipoles). Side wall (first) reflections are significantly reduced (in my room) so I can't comment on what effect they would have. Localization and "imaging" seem to be associated almost entirely with the direct wave, regardless the DI of the loudspeaker.
But there's more...
I find that my open baffle speakers sound great when they are at least 3 ft. out from the front wall of my room, but sound kinda bad when they are closer to the wall than that. It tells me that the rear radiation really needs to be delayed at least about 6-7mS before recombining at my ear, to give me the 3-D effect that I like. A psycho-acoustic phenomenon I guess. Above 10mS delay, I'd expect to sense it as an echo too much, rather than as an enhancer of any embedded reverbs.
I find that my open baffle speakers sound great when they are at least 3 ft. out from the front wall of my room, but sound kinda bad when they are closer to the wall than that. It tells me that the rear radiation really needs to be delayed at least about 6-7mS before recombining at my ear, to give me the 3-D effect that I like. A psycho-acoustic phenomenon I guess. Above 10mS delay, I'd expect to sense it as an echo too much, rather than as an enhancer of any embedded reverbs.
The question is, are these effects detrimental or are they perceived as a pleasant addition to spaciousness? Toole summarizes that no single reflection in a "typical" room is loud enough to have a negative effect (e.g. second image).
On the other hand there's interesting data from Cremer/Müller showing that several low level reflections from a similar direction are perceived with almost the same loudness as one single reflection.
Last edited:
Imaging vs 3-D spaciousness
I designed and built my open baffle speakers after hearing how OB spkrs created a sense of depth, individuation of instruments, and gave amazing life to embedded reverbs. I was then pleasantly surprised to find that my re-optimization of the Bob Carver Holographic Generator still added every bit as much spaciousness and image individuation as it did with closed box relatively directional speakers. It told me that you can artificially add in a bunch of delays above 5mS from the open baffle nature of the speakers, without damaging the trick that the hologram circuit does with its roughly 125uS delay.
I designed and built my open baffle speakers after hearing how OB spkrs created a sense of depth, individuation of instruments, and gave amazing life to embedded reverbs. I was then pleasantly surprised to find that my re-optimization of the Bob Carver Holographic Generator still added every bit as much spaciousness and image individuation as it did with closed box relatively directional speakers. It told me that you can artificially add in a bunch of delays above 5mS from the open baffle nature of the speakers, without damaging the trick that the hologram circuit does with its roughly 125uS delay.
I'm with you on the OB positioned at approximately 3 ft from the front wall where the delay is about 6ms. But if you put it 5.5ft away and you are at approximately 11ms delay and I'm sure it would still sound fine... I think one would have to move a fair bit further to hear an echo as illustrated (40ms would be a clear echo).
Thus I think the 10ms stuff quoted early is pretty darn close to reality. It wasn't stated that anything less than 10ms would sound bad... just that it is easier to hear such a sort delay using pink noise. Over 10ms and it would be 'easy' to hear the enhanced impression of a sense of 'spaciousness'.
I don't think anyone would argue with you that 'very early' reflections in the first 1-5ms should be avoided. The OB speakers also do a fairly good job of limiting the side and ceiling reflections which are 'very early'. (Yours do too of course 🙂)
It can be argued that a reflection off the front wall of the sound stage with at least 6ms delay (or slightly more) can actually sound quite pleasant. 😉
The first question, "to what degree does the direct sound mask the later sound" is a good one.
Regarding the later explanation, "how far down does the reflection need to be to have minimal frequency balance impact" I believe we are missing more significant questions;
a) how much delay is required before one perceives the reflected sound as a reflection rather than part of the original sound? and,
b) does the amount of delay required for our brain to integrate or perceive the reflected sound as such vary with frequency?
b) does the amount of delay required for our brain to integrate or perceive the reflected sound as such vary with frequency?
I think everyone would agree with most of these statements.
The vertical reflections especially are often only a few ms behind the first and would be very difficult to make an argument that we do not perceive them as part of even as a distortion of the direct sound.
This requires further clarification for me because I can definitely hear or "detect" attenuated reflections so long as the original sound is not ongoing (i.e the after effects of a loud hand clap in a room with lots of reverb compared to the same clap in a well damped room).
I'm unclear as to how the 'noticeability threshold' is defined. Is this defining the ability of our ears and brains to interpret the indirect sound as a reflection?
Why would lateral reflects be interpreted so much differently than vertical reflections? Is directional perception on the horizontal plane that much more acute than the vertical plane?
Again this definition of 'notice-ability' and/or 'detection threshold" eludes me.... And these explanations get even more convoluted if we consider that our hearing might perceive these reflections differently depending on the frequency range in question.
Sounds produced at over several hundred Hz would have generated multiple sound waves in the same 10ms and I can believe that our brains could distinguish between the direct and reflected sounds.
But in the case of the sound of a large kick drum which is producing much of it's primary sound at 55Hz, the sound wave is quite long and therefore even at 10ms delay it should still be difficult for our brains to distinguish the reflected sound from the direct sound because the reflection would be out of phase when compared to the direct sound.
Quote:
Originally Posted by graaf
with all due respect but - is that so?
if "reflections in the same direction as the first incidence can be 5 to 10 dB above the first incidence before detection" and "reflections have to be 7 dB above the absolute detection threshold before they cause an image shift" and if "just above the noticeability threshold" they just "increase spaciousness" then where is the issue?
OK. At what SPLs are these figures valid?
From an earlier post here:
Originally Posted by graaf
with all due respect but - is that so?
if "reflections in the same direction as the first incidence can be 5 to 10 dB above the first incidence before detection" and "reflections have to be 7 dB above the absolute detection threshold before they cause an image shift" and if "just above the noticeability threshold" they just "increase spaciousness" then where is the issue?
OK. At what SPLs are these figures valid?
From an earlier post here:
Is the subjective effect of reflections the same for a direct source at 70 dB as, at say, 85, 90, 95 dB? That is, does the just noticeable difference remain the same as the level of direct sound changes? If it does not then there is possibility of image shift, changes in spaciousness or perhaps even timbre, with changes in level.
Looking at the ear as a nonlinear device beyond 'equal loudness' gets ugly fast.
I'm sure everyone has heard the reflection off the front wall of a ML panel speaker and the broad image and poor speech intelligibility it can produce while the speaker seems nearly absent from the auditory scene. Something's up with those figures for sure. There would be no OB 'magic' if it wasn't for that front wall reflection. The strange part is that there is an OB 'magic' with comparatively little lateral reflections barring the presence bloom on some designs.
Dan
I'm sure everyone has heard the reflection off the front wall of a ML panel speaker and the broad image and poor speech intelligibility it can produce while the speaker seems nearly absent from the auditory scene. Something's up with those figures for sure. There would be no OB 'magic' if it wasn't for that front wall reflection. The strange part is that there is an OB 'magic' with comparatively little lateral reflections barring the presence bloom on some designs.
Dan
Last edited:
thank You for a lot of important questions that should be asked
best,
graaf
- Status
- Not open for further replies.