Some useful quotes/graphs on the subject:
These are all from NASA, ITU, or Olive/Toole stuff. Other studies are contained within them. They all generally/closely agree.
Dan
These are all from NASA, ITU, or Olive/Toole stuff. Other studies are contained within them. They all generally/closely agree.
Dan
Last edited:
But none of this has anything to do with imaging location, and apparent significant alteration of an image *between the dominate loudspeaker and the dominate reflection*.
In other words image localization between the left wall and the left speaker (or right wall and right speaker).
Now localization alteration between the center speaker and the left or right speaker - sure, that can happen. These are all direct sources, not reflections. IF the reflection left of left is loud enough (vs. right) it acts like a channel imbalance with the result that left speaker sounds louder than the right and imaging *between L & R* shifts to the left speaker (..and can simply be corrected by increasing the level in R.)
Also between the L speaker and the right wall, given enough intensity and time delay it very much can "stretch" out and image, or as acousticians say - increase apparent source width. This is fundamental with inter aural cross "contamination". 😀 ..But no change in image position laterally (..basically it just makes the image "bigger" and even more "diffuse").
I'd agree about the spaciousness thing. 🙂
I think Elias's room spectrogram highlighted this in the stereolith thread.
Basically a lower level of IACC is developed.
The left channel stays longer and louder on the left side of the room, and the right channel stays longer and louder on the right side of the room.
IMO this is a fundamental aspect that needs to be improved over current traditional stereo reproduction.
Last edited:
Sure, the "dominating" loudspeaker (where the image is centered) is the center speaker, in my setup. The left or right wide is altering the perceived size and spaciousness of the center image. Or do I misunderstand what you're trying to say?
I believe IACC is like RT60 - not a very useful indicator.
Last edited:
I'm agreeing with you, but I think you are misunderstanding what I'm saying. 😱
Yes, the Left speaker is modifying imaging from the center speaker (and the Right speaker).
Neither however is a reflection. (..actually there are reflections between the speakers themselves, but I'm specifically excepting that.)
In the "test" example I gave (and performed), it would be be the left speaker in relation to the left wall reflection. OR the left output of the center speaker in relation to the left wall (..assuming the center had a "left" out facing the left wall). And this is specifically regarding its impact on image location (and size, shape, etc.). Is this clearer?
Inter aural cross correlation as a concept, not as an acoustic measurement. 😉
Some useful quotes/graphs on the subject:
Olive/Toole..
Sound Reproduction 6.1.1
First link:
soud reproduction toole this is a begin at the beginning experiment - Google Search
There are a few critical statements made in this section, along with a few caveats (both express and implied).
1. Test: "..The listening environment is anechoic, the signal is speech, and only a single lateral reflection is examined."
2. Test Caveat: "..It is not data that can be applied to real-world circumstances listening to music.."
3. Test Condition figure 6.4: "..The anechoic setup uses a real loudspeaker to simulate the "mirror image" loudspeaker in the room situation. This is the experimental method that has been used in numerous experiments conducted over the decades."
4. Test Response, (lower sp-level) synthesized discreet reflection: "..listeners reported all of the sound as originating at the location of the loudspeaker that reproduced the first sound."
5. Test Response, (medium sp-level) synthesized discreet reflection: "..listeners reported hearing a change in size or position of the main sound image... ..This was called the "image shift" threshold."
6. Test Response, (medium) Caveat: "..In general, these changes were subtle and noticeable in these A vs. B comparisons, but it is doubtful that they would be detected in the context of multiple-image music.."
7. Test Response, (high sp-level) synthesized discreet reflection: "..at this level the listeners reported hearing a second source or image, simultaneously coexisting with the original one.."
What I've suggested previously is that the test response in #4 is the correct response for real-world conditions with respect to image localization and definition with near wall reflections.
The test responses in #5 and #7 just don't occur under *normal* "small room" reproduction conditions. (..at least not that I've heard.)
Last edited:
Ah, but I'm saying it isn't perceived as a reflection, because in large measure reflections aren't perceived. 😉
I would note though that some small element of "spaciousness" contributed in the L and R channels (of an LCR setup), probably is similar to a real reflection - because it actually increases the intensity of reflections (for the L & R's respective "sides" of the room).
BUT,
-that most of the effects, including spaciousness overall, are because you perceive those L & R speakers as additional *sources* (of direct sound).
A live experience with reflection and source localisation :
source = 10" driver, OB, perpendicular to the main room axis, 3 m from R wall, 6 m from L wall, 3 m from front wall, ceiling/floor nothing special
listener : 30° from the source, on room axis, around 3 m from the source
room : very alive, symmetrical boundaries
signal : a long 10 s sweep, quite high level.
results : under 700 Hz, I localize the driver as the source, over 700Hz, I localize the sound as coming from the left wall. Weird but it's like that. If both drivers (R & L channels) are running, the localisation stays in the middle. On music, I can seldom detect this on some transients, mostly percussions.
This reflection is perfectly seen in the impulse plot, the difference of path is around 22ms so it's not an early reflection. We can also suspect that the listener is in a dip caused by a bad lobing pattern in the direct radiation, but this experience works also with other drivers. We can wonder also why the back wave of this driver (OB) is not leading the localization as it comes earlier (14 ms) with almost the same level and spectrum.
It's more or less ok with Toole's statements but I think the subject is very complex.
source = 10" driver, OB, perpendicular to the main room axis, 3 m from R wall, 6 m from L wall, 3 m from front wall, ceiling/floor nothing special
listener : 30° from the source, on room axis, around 3 m from the source
room : very alive, symmetrical boundaries
signal : a long 10 s sweep, quite high level.
results : under 700 Hz, I localize the driver as the source, over 700Hz, I localize the sound as coming from the left wall. Weird but it's like that. If both drivers (R & L channels) are running, the localisation stays in the middle. On music, I can seldom detect this on some transients, mostly percussions.
This reflection is perfectly seen in the impulse plot, the difference of path is around 22ms so it's not an early reflection. We can also suspect that the listener is in a dip caused by a bad lobing pattern in the direct radiation, but this experience works also with other drivers. We can wonder also why the back wave of this driver (OB) is not leading the localization as it comes earlier (14 ms) with almost the same level and spectrum.
It's more or less ok with Toole's statements but I think the subject is very complex.
Last edited:
Cool! (..and thank you! 🙂 )
I'm having problems with the description.. particularly "perpendicular to the main room axis". Are you also "perpendicular to the main room axis? Or are you basically listening 30 degrees from the null at 90 degrees? Or something else altogether? 😕
Ah, just the drivers are side firing in this case, (an old tradition of the house), the listener is at the apex of the resulting equilateral triangle, ordinary situation.
I'm not sure that helped me. 😀 😱
So they fire to the side at the side wall or the 0 degree axis. They are dipole. You listen at about 120 degrees from that frontal axis?
i.e.
wall..............< L........R > ...............wall
.........................you.........................?
as dipoles, like this for the test
wall..............< L>........< R > ...............wall
............................me.............................
but in real life like this without the subs:
wall..............< +L>........< +R > ...............wall
............................me.............................
wall..............< -R>........< -L > ...............wall
............................me.............................
wall..............< L>........< R > ...............wall
............................me.............................
but in real life like this without the subs:
wall..............< +L>........< +R > ...............wall
............................me.............................
wall..............< -R>........< -L > ...............wall
............................me.............................
Ha! I was correct. I didn't think I had a chance of being correct. 😀
OK.. you do realize that this is not at all normal? 😛 In-genius perhaps, but not normal.
Remember this?
http://www.diyaudio.com/forums/multi-way/180301-lq-osd-acoustic-alternative-5.html#post2515343
😀
That's in March.
And I remember a newer picture in May, something different from the above. Now it's July, I believe there must be some other transformation!
http://www.diyaudio.com/forums/multi-way/180301-lq-osd-acoustic-alternative-5.html#post2515343
😀
That's in March.
And I remember a newer picture in May, something different from the above. Now it's July, I believe there must be some other transformation!
Reflections are additional sources. They are perceived as spaciousness, apparent source width, listener envelopment, repetition pitch, coloration, etc. When the reflection is too loud and/or too delayed it's perceived as an echo, a source separate from the direct sound. It really doesn't matter if the reflection is coming from a wall or from speakers. What really matters is the overall reflection pattern and the perception it causes.
Hi Radugazon,
how loud is that 22ms reflection? I found that precedence is easily overridden by percussive, high frequency sounds if they are loud enough. 22ms is pretty long and can cause echos with percussive sounds.
Markus, I agree :
I notice it on music only on a couple of records (the last Miles Davis for exemple), on congas, once in the disk. It's very fast and I have to be concentrated on it. Even if this could add a permanent fuzzyness, I don't feel it because I must be totally biased and this is classified as "spaciousness" (dn't laugh).
For the occasion, I did a sweep on the right midbass channel (mike in list spot, no filter, no Eq, front speakers only and some stuff in the middle but it's still representative). Impluse is enough to comment the phenomenon : it's a reflection museum here.
1: direct sound
2: 7.35 ms, probably the ceiling, small intensity, same polarity (I have diffusers) but almost full range
3: 15 ms, inversed polarity, it's the backwave of the dipole coming from the ipsilateral side, from the shape (and gated measurememnt) we can see that all that's after 1000 Hz is missing
4: 22.5 ms, normal polarity, it's the frontwave from the opposite wall, and I see today that it contains more HF than the direct sound (normal, this last is at 30° as the reflection is around 10° finally).
This is why I hear the HF part of the sweep coming from the wall. Simple and straight explanation. It's not precedence but directivity effect. About levels, except for the ceiling, they are all grossly the same.
Notice that with my too regular implantation (almost a 3 m multiple everywhere) the reflections have a pace of 7.5 ms. Not a good example, but I don't hear echoes, so...
Of course, in the listening configuration the roll off of the filter attenuates all this or make it disappearing. But this pattern is not reproduced by the
2" and 1" that I have as tweeters. The ceiling reflection vanishes (diffuser...) and the wall ones are reduced to symbolic glitches. Again, it's because their directivity is very different. IIRC, Rudolf once shown this on dipole tweeters, I listen probably in a lobe with very high level at 30°.
Now a good question would be "what is this adding or removing to the reproduction ?"
I notice it on music only on a couple of records (the last Miles Davis for exemple), on congas, once in the disk. It's very fast and I have to be concentrated on it. Even if this could add a permanent fuzzyness, I don't feel it because I must be totally biased and this is classified as "spaciousness" (dn't laugh).
For the occasion, I did a sweep on the right midbass channel (mike in list spot, no filter, no Eq, front speakers only and some stuff in the middle but it's still representative). Impluse is enough to comment the phenomenon : it's a reflection museum here.
1: direct sound
2: 7.35 ms, probably the ceiling, small intensity, same polarity (I have diffusers) but almost full range
3: 15 ms, inversed polarity, it's the backwave of the dipole coming from the ipsilateral side, from the shape (and gated measurememnt) we can see that all that's after 1000 Hz is missing
4: 22.5 ms, normal polarity, it's the frontwave from the opposite wall, and I see today that it contains more HF than the direct sound (normal, this last is at 30° as the reflection is around 10° finally).
This is why I hear the HF part of the sweep coming from the wall. Simple and straight explanation. It's not precedence but directivity effect. About levels, except for the ceiling, they are all grossly the same.
Notice that with my too regular implantation (almost a 3 m multiple everywhere) the reflections have a pace of 7.5 ms. Not a good example, but I don't hear echoes, so...
Of course, in the listening configuration the roll off of the filter attenuates all this or make it disappearing. But this pattern is not reproduced by the
2" and 1" that I have as tweeters. The ceiling reflection vanishes (diffuser...) and the wall ones are reduced to symbolic glitches. Again, it's because their directivity is very different. IIRC, Rudolf once shown this on dipole tweeters, I listen probably in a lobe with very high level at 30°.
Now a good question would be "what is this adding or removing to the reproduction ?"
Attachments
Last edited:
You disregarded the express context of "sources" as *direct sound*.
You are also discussing concepts regarding larger room acoustics, not small room acoustics with reproduced music, or even small room acoustics generally.
Radugazon's experiments actually parallel these larger spaces because his listening environment is far larger than normal. Additionally the *extreme* loss in pressure above 1 kHz of a 10 inch open baffle listening beyond the null will most definitely result in a vastly lower sp-level for the direct sound above 1 kHz, with a vastly higher sp-level for the reflection above 1 kHz.
Again, by and large I agree with your statements - it's just that they have little bearing for most people. They aren't *normal*, and unless I'm mistaken - they aren't even normal for you.
The thread's purpose is with regard to "omni.s", or at least mostly "omni" loudspeakers.
Within that scope I mentioned that for the most part, an "omni" loudspeaker does not produce a a significant effect with near reflections that would effect imaging in a normal small-room environment. Again, that's actual reflections. I even went as far as experimenting with conditions that were abnormal (moderately higher sp-level reflections at higher freq.s though still within a normal small room acoustic), and personally found that the results remained the same *except* when at a distance and virtually behind the loudspeakers.
If you want to continue talking about synthesized reflections, and large room acoustics.. that's fine with me, but please make clear note of this distinction BEFORE commencing on these topics, otherwise we are "talking at cross purposes". 😉
I don't talk about large spaces. Radugazon's space is not an acoustically small room. I talk about spaces that are highly directional. This is one of the major features of an acoustically small room.
Omnis generate lots of reflections. I've found that second order reflections from wall to ceiling and ceiling to wall can be pretty high in level (by the way, that's one of the reason why I think Earl's cross-firing setup does more harm than good). These reflections accumulate perceptually (see Toole). I try to learn how angle, time, level, spectrum and number of reflections affect perception. My center plus wides setup was just an experiment in order to get more insight and to get past dogmatics.
Last edited: