That would probably be one of John's dipole designs.
Our brain is trying to make sense of all the information it receives, so it's not a great surprise that the sound is not perceived as coming from the floor if there are strong, early reflections from a higher location.
Who is this John who has designed floor dipoles with ceiling firing tweeter? Link?
I need to do wavelet analysis of my previous experiment with the floor box to really see the levels and timings of direct sound vs reflections. Then I'd like to compare that with the theory of the precedence effect.
- Elias
"Our" John k...
I was not necessarily talking about "floor dipoles" but about speakers that result in the same perception as the speaker you mentioned.
Yes, would be very interesting and helpful to get the numbers.
As to the perception, I think there are two aspects. First, the ceiling reflection has an impact on summing localization (to my knowledge there's only one paper investigating vertical summing localization). Second, an increased ASW that makes sounds seem to come from an area between the floor and ceiling and not from a single point.
Hello everyone,
Measurement time!
I did some measurement of my ceiling firing experiment from the post#1080
http://www.diyaudio.com/forums/multi-way/121385-loudspeakers-room-system-22.html#post2133755.
There I placed a two way speaker on the floor beside a side wall. The speaker is facing towards the ceiling. I was listening more than 2 metres distance. The listening experience gave a little bit of contradictory feelings that time.
So today I placed the same speaker at the same location and measured impulse response. Then I calculated wavelet transforms to see how reflections behave in time-frequency domain. It shows the sound energy between 500Hz-20kHz up to 20ms. I choose 20ms to fit most of the first reflections into the picture.
For a comparison I measured the same speaker placed at 1m height facing the microphone. This would be the normal listening configuration. In both cases the flying distance to the speaker was about 2.5m. The microphone is placed at the listening position so it is above a leather soffa, this will generate some extra reflections!
Here's the speaker placed on the floor facing the ceiling:
Of course there are many reflections. Notable is for example that above 3kHz the strongest sound is not the direct sound but a reflection just after 5ms. Also the reflections looks like quite random. Reflection density is high. Also notable is that there are strong reflections arriving at 20ms.
And here's the speaker at the 1m stand facing the mic:
Of course the direct wave dominates (at 0ms). Note: the reflection at 2ms is most likely due to the soffa. At 4ms should be the floor reflection, 7ms the ceiling reflection. Reflections have a trend to be attenuated fast as time passes during the 20ms observation period. High freq reflections die faster than in the ceiling firing arrangement.
The ceiling firing arrangement gives (relatively) more high frequency energy during 20ms than the conventional direct firing stand placement.
Ceiling firing sounds very spacious. That maybe the best feature of such an arrangement!
However, as I mentioned earlier, ceiling firing lost some of the small detail of the recording that could be heard when direct sound dominated.
Interesting!
- Elias
Measurement time!
I did some measurement of my ceiling firing experiment from the post#1080
http://www.diyaudio.com/forums/multi-way/121385-loudspeakers-room-system-22.html#post2133755.
There I placed a two way speaker on the floor beside a side wall. The speaker is facing towards the ceiling. I was listening more than 2 metres distance. The listening experience gave a little bit of contradictory feelings that time.
So today I placed the same speaker at the same location and measured impulse response. Then I calculated wavelet transforms to see how reflections behave in time-frequency domain. It shows the sound energy between 500Hz-20kHz up to 20ms. I choose 20ms to fit most of the first reflections into the picture.
For a comparison I measured the same speaker placed at 1m height facing the microphone. This would be the normal listening configuration. In both cases the flying distance to the speaker was about 2.5m. The microphone is placed at the listening position so it is above a leather soffa, this will generate some extra reflections!
Here's the speaker placed on the floor facing the ceiling:
Of course there are many reflections. Notable is for example that above 3kHz the strongest sound is not the direct sound but a reflection just after 5ms. Also the reflections looks like quite random. Reflection density is high. Also notable is that there are strong reflections arriving at 20ms.
And here's the speaker at the 1m stand facing the mic:
Of course the direct wave dominates (at 0ms). Note: the reflection at 2ms is most likely due to the soffa. At 4ms should be the floor reflection, 7ms the ceiling reflection. Reflections have a trend to be attenuated fast as time passes during the 20ms observation period. High freq reflections die faster than in the ceiling firing arrangement.
The ceiling firing arrangement gives (relatively) more high frequency energy during 20ms than the conventional direct firing stand placement.
Ceiling firing sounds very spacious. That maybe the best feature of such an arrangement!
However, as I mentioned earlier, ceiling firing lost some of the small detail of the recording that could be heard when direct sound dominated.
Interesting!
- Elias
Hi Elias!
Your measurements' results confirm those done earlier and posted in this thread by Oliver (el'Ol).
Let me draw Yout attention to some commercial flooders that I have discovered recently, especially to the two-way Carlsson inspired AudioPro "Qube", see:
http://www.audiopro.se/media.php?id_file=527
http://www.diyaudio.com/forums/full-range/162827-mpl-13.html#post2230218
http://www.diyaudio.com/forums/full-range/162827-mpl-13.html#post2230308
best,
graaf
ps.
As I have declared earlier above - I don't discuss in this thread anymore and I don't post here but I make an exception just for You to invite You to join in the other thread at Fullrange section
Your measurements' results confirm those done earlier and posted in this thread by Oliver (el'Ol).
Let me draw Yout attention to some commercial flooders that I have discovered recently, especially to the two-way Carlsson inspired AudioPro "Qube", see:
http://www.audiopro.se/media.php?id_file=527
http://www.diyaudio.com/forums/full-range/162827-mpl-13.html#post2230218
http://www.diyaudio.com/forums/full-range/162827-mpl-13.html#post2230308
best,
graaf
ps.
As I have declared earlier above - I don't discuss in this thread anymore and I don't post here but I make an exception just for You to invite You to join in the other thread at Fullrange section
Hi,
Very interesting measurements. llow me a few comments.
This shows some interesting faccetts.
First, let's ignore for a moment the reflections (we come to that).
In the direct sound we see a that below 1KHz (that is in the frequency range that covers the fundamentals and first formants of most acoustic instruments there is severe lack of energy, which is filled in at low frequencies and at higher frequencies. Clearly this speaker is anything but "controlled directivity".
Most of this missing fundamental energy is visible after 7mS, meaning it is almost wholly reflected from the ceiling and re-reflected after bouncing of the floor (14mS).
Much other energy echoes around the room all across the bands.
It would an interesting second experiment to set a digital delay plugin to have a 7mS delay and try for a similar effect "by ear". The frequency dependence of the reverb will be hard to incorporate, still, it should be possible to come somewhat close.
I'd be interested if your comments when using a high quality artificial reverb with your forward facing speakers would be similar to those when turning your listening room into a reverb chamber? Are you game?
BTW, before becoming a hopeless Audiophile I used to have my "mastering" system also as my listening system. And I had very expensive digital reverbs and equalisers at hand. And at times I'd use a bit of extra reverb to make overly dry recordings to sound more "wet", more spacious.
Again, this shows quite clearly the problems of using speakers with poorly controlled directivity, especially in the vertical plane. Looking for the ceiling reflection around 7mS shows it consists mostly of 2KHz - 5KHz. I suspect the speaker has a dome tweeter and a slightly large cone woofer seeing this.
The sonic effect of this reflection will be to promote mostly the "brightness" region in the human hearing and even though the level is quite low, it lies in the ears greatest sensitivity region.
So I think controlling the speakers vertical dispersion (foam or felt wedges will do nicely) will likely help the sound quality. Of course, it will not make "dry" recordings sound spacious.
Maybe adding a small upwards radiating full range driver with a limited passband and a sensitivity below the main system can add some of the spaciousness back, with a switch to switch this effect off when not needed?
Ciao T
Very interesting measurements. llow me a few comments.
Measurement time!
...
Here's the speaker placed on the floor facing the ceiling:
...
This shows some interesting faccetts.
First, let's ignore for a moment the reflections (we come to that).
In the direct sound we see a that below 1KHz (that is in the frequency range that covers the fundamentals and first formants of most acoustic instruments there is severe lack of energy, which is filled in at low frequencies and at higher frequencies. Clearly this speaker is anything but "controlled directivity".
Most of this missing fundamental energy is visible after 7mS, meaning it is almost wholly reflected from the ceiling and re-reflected after bouncing of the floor (14mS).
Much other energy echoes around the room all across the bands.
It would an interesting second experiment to set a digital delay plugin to have a 7mS delay and try for a similar effect "by ear". The frequency dependence of the reverb will be hard to incorporate, still, it should be possible to come somewhat close.
I'd be interested if your comments when using a high quality artificial reverb with your forward facing speakers would be similar to those when turning your listening room into a reverb chamber? Are you game?
BTW, before becoming a hopeless Audiophile I used to have my "mastering" system also as my listening system. And I had very expensive digital reverbs and equalisers at hand. And at times I'd use a bit of extra reverb to make overly dry recordings to sound more "wet", more spacious.
And here's the speaker at the 1m stand facing the mic:
...
Again, this shows quite clearly the problems of using speakers with poorly controlled directivity, especially in the vertical plane. Looking for the ceiling reflection around 7mS shows it consists mostly of 2KHz - 5KHz. I suspect the speaker has a dome tweeter and a slightly large cone woofer seeing this.
The sonic effect of this reflection will be to promote mostly the "brightness" region in the human hearing and even though the level is quite low, it lies in the ears greatest sensitivity region.
So I think controlling the speakers vertical dispersion (foam or felt wedges will do nicely) will likely help the sound quality. Of course, it will not make "dry" recordings sound spacious.
Maybe adding a small upwards radiating full range driver with a limited passband and a sensitivity below the main system can add some of the spaciousness back, with a switch to switch this effect off when not needed?
Ciao T
Hello Thorsten,
Thanks for the comments. Maybe I should emphasise the reason I did the measurement is I wanted to see how the early reflection pattern looks like in the ceiling firing placement, because I could not localise the speaker at the floor. In the normal arrangement facing the listener at the ear level there is no problems to localise this (mono) speaker in the room.
Some clarification of the reflections. As can be seen the first reflection arrives very soon only 0.6ms after the direct sound. This is because the speaker is placed on the floor beside a side wall. The tweeter distance to the side wall was only about 15cm.
The reason of no energy below 1kHz is because of comb filtering. This can be best seen from this wavelet where comb filtering ripples occurs most of the frequencies:
What comes to the "controlled directivity" the speaker is not that bad in horisontal direction rather actually is quite constant and smooth. However, the vertical pattern is worse (you never see those in marketing literature). The speaker is a two-way with 6.5" + dome in 12l box.
I don't think mimicing the reflection pattern articifially would sound the same because of different spatial distribution of the reflections. With reverb reflections would come from the same direction as direct sound. I do have some reverb units (not that high quality though), but I rather feed the reverb to 'surround' speakers located side back and not in front. The envelopement is maximised if the late reverb comes a bit sideways.
Overall I think better way to obtain more constant early reflections is to use omni at high frequencies. However, at midrange and low frequencies omni does not sound good to me and I rather use higher directive sources like dipole line arrays below about 1kHz.
This is, I belive, due to the fact that there are completely different mechanisms for a human to hear and process sounds below and above about 1kHz. Then why the deliverance should be the same for both regions?? There is no need, in my opinion. Better results can be obtained if both regions are optimised separately. This is what I'm trying to do.
- Elias
Thanks for the comments. Maybe I should emphasise the reason I did the measurement is I wanted to see how the early reflection pattern looks like in the ceiling firing placement, because I could not localise the speaker at the floor. In the normal arrangement facing the listener at the ear level there is no problems to localise this (mono) speaker in the room.
Some clarification of the reflections. As can be seen the first reflection arrives very soon only 0.6ms after the direct sound. This is because the speaker is placed on the floor beside a side wall. The tweeter distance to the side wall was only about 15cm.
The reason of no energy below 1kHz is because of comb filtering. This can be best seen from this wavelet where comb filtering ripples occurs most of the frequencies:
What comes to the "controlled directivity" the speaker is not that bad in horisontal direction rather actually is quite constant and smooth. However, the vertical pattern is worse (you never see those in marketing literature
). The speaker is a two-way with 6.5" + dome in 12l box.I don't think mimicing the reflection pattern articifially would sound the same because of different spatial distribution of the reflections. With reverb reflections would come from the same direction as direct sound. I do have some reverb units (not that high quality though), but I rather feed the reverb to 'surround' speakers located side back and not in front. The envelopement is maximised if the late reverb comes a bit sideways.
Overall I think better way to obtain more constant early reflections is to use omni at high frequencies. However, at midrange and low frequencies omni does not sound good to me and I rather use higher directive sources like dipole line arrays below about 1kHz.
This is, I belive, due to the fact that there are completely different mechanisms for a human to hear and process sounds below and above about 1kHz. Then why the deliverance should be the same for both regions?? There is no need, in my opinion. Better results can be obtained if both regions are optimised separately. This is what I'm trying to do.
- Elias
Hi,
In such cases it is better to place the tweeter as close as possible to the sidewall.
Yet as we can see from your measurements, the vertical dispersion is as important, if not more so than the horizontal one, in domestic conditions at least, usually with speakers at around 1m height, 2.4m high ceilings and carpeted floors at any...
I surmised as much. I think a 3-way with a smaller midrange may have done better.
Agreed. However I would still suggest this experiement.
First, if you have a PC based source, adding high quality reverb is trivial.
Secondly, funny you should mention independent speakers and piping delayed signals to them.
One of my "snakeoil & voudoun" tweaks is place a fairly large number of specific resonators that are tuned to resonate at a wide range of musical frequencies around the room. These resonate slightly in sympathy with sound being played.
Their contribution is at such low levels that it bis not possible to measure reliably, yet the subjective effect on perceived space is rather obvious. I primarily use them behind the speakers, to afford a subjective 'disassociation" of sound from the speakers, yet placing also behind the listener does add to sense of envelopement.
This is arguably the polar opposite of how things are done nowadays, but makes some sense. I personally would suggest to use instead a directive HF source that matches the directivity of the lower frequencies and to ADD a lower level omni source for high frequencies.
If we do this, we end up in effect with something like the Briggs/Warfedale baffle shown above, which was designed for mono playback and used a fullrange driver with helper woofer radiating forward and generally covering the audible range plus a ceiling firing tweeter.
I did try something like that in my own open baffle setup which used supravox fieldcoil full-range drivers (this did provide essentially a flat response throughout most of audible bandwidth, up to maybe 12...15KHz). I normally had a forward radiating "Super HF" unit which mainly extended the HF response. I did try this with a lower crossover radiating upwards and thought the results had some potential.
I believe the principle is somewhat similar to the Mark & Daniel add-on Omni Tweeter and the similar device from Elac in Germany.
Ciao T
In such cases it is better to place the tweeter as close as possible to the sidewall.
What comes to the "controlled directivity" the speaker is not that bad in horisontal direction rather actually is quite constant and smooth. However, the vertical pattern is worse (you never see those in marketing literature
Yet as we can see from your measurements, the vertical dispersion is as important, if not more so than the horizontal one, in domestic conditions at least, usually with speakers at around 1m height, 2.4m high ceilings and carpeted floors at any...
I surmised as much. I think a 3-way with a smaller midrange may have done better.
Agreed. However I would still suggest this experiement.
First, if you have a PC based source, adding high quality reverb is trivial.
Secondly, funny you should mention independent speakers and piping delayed signals to them.
One of my "snakeoil & voudoun" tweaks is place a fairly large number of specific resonators that are tuned to resonate at a wide range of musical frequencies around the room. These resonate slightly in sympathy with sound being played.
Their contribution is at such low levels that it bis not possible to measure reliably, yet the subjective effect on perceived space is rather obvious. I primarily use them behind the speakers, to afford a subjective 'disassociation" of sound from the speakers, yet placing also behind the listener does add to sense of envelopement.
This is arguably the polar opposite of how things are done nowadays, but makes some sense. I personally would suggest to use instead a directive HF source that matches the directivity of the lower frequencies and to ADD a lower level omni source for high frequencies.
If we do this, we end up in effect with something like the Briggs/Warfedale baffle shown above, which was designed for mono playback and used a fullrange driver with helper woofer radiating forward and generally covering the audible range plus a ceiling firing tweeter.
I did try something like that in my own open baffle setup which used supravox fieldcoil full-range drivers (this did provide essentially a flat response throughout most of audible bandwidth, up to maybe 12...15KHz). I normally had a forward radiating "Super HF" unit which mainly extended the HF response. I did try this with a lower crossover radiating upwards and thought the results had some potential.
I believe the principle is somewhat similar to the Mark & Daniel add-on Omni Tweeter and the similar device from Elac in Germany.
Ciao T
Wouldn't this be an issue, that the sound that would diffract around the cabinets is met with this "black hole" of damping? Wouldn't this "enhance" baffle diffraction (speaking of frequencies above schroeder, of course)? Would you need to use wedge style foam?
Hmm, I might have suspected, but I'm still trying to understand. Aren't wedges used in anechoic chambers because the sudden loss of energy causes a resultant backwave? and would it be even better to bring the foam forward in front of the baffle plane and even around the front edges a little?
At the object based approach becomes possible, subtracting the playback room acoustics for direct wave and first reflections. Reverberation becomes generate in different way, from the impulse response, and equalised separately.
The procedure is described in WFS-Holophony , chapter 4.2. In that case, loudspeaker and room really are one system.
The procedure is described in WFS-Holophony , chapter 4.2. In that case, loudspeaker and room really are one system.
Some unsorted thoughts on ceiling reflections and "sound from above"
---------------------------------------------------------------------
Our ear/brain has developed even before speech and music came into human live.
Our perception system has developed to survive.
It's first aim is the identification of objects: Their nature, their distance,
their direction, their speed and direction of movement ...
---
OK, let's imagine our stone age huntsman passing a narrow cutting ... he is
being observed by someone who is able to move very quiet ...
But that sort of cat is not perfect either. Some pebbles get loose and begin to roll
down over the crag, causing some transient clicking noise ... got it ?
Where ? Exactly where, please ?
That fractions of a second decide whether our huntsman's plans for his future live
will be of any use or not.
Over 10.000 years later, that clicking noise might indicate an unlocked overhead bin
in an airplane with you sitting right underneath.
And if you did not know already, than i might be allowed to tell you, that luggage
falling out of unsecured overhead bins is the oftenmost cause for serious
insuries of aircraft passengers. Don't be afraid of flying in an airplane, but have
a closer look on the luggage please ...
---
So far so good ...
Human live - as well as killing and defence strategies - have been perfectioned.
Speech and musical culture have an omportant place also in human live.
There is still that frequency range around 8 Khz, which is associated with 'brillance'.
It is important for our perception and our ear conches are designed to give
that frequency range a gain of over 10 dB if coming from above. That gain is
lesser if the elevation angle is lower that 90 degrees from the median plane.
That 'directional band' is called the 'above' band according to Jens Blauert.
File:Akustik - Richtungsbänder.svg ? Wikimedia Commons
---
Now mixing some of my own (almost trivial) thoughts into the rules of the game:
- An acoustical 'object' or 'gestalt' is not complete without that 'brilliance' information.
- Sounds known to us miss 'sparkle' without that frequency range.
- The frequency range together with the 'presence' region below that 'above' band
has importance in detecting positions of 'sound objects' rather fine and make them
'real' even on a proximity level of perception. An object with 'no proximity' is not
a real (or realistic) object. For the 'realism' in a painting there are similar rules:
An object is embedded into its context/backround by perspective, oherwise it will
not occur to be real or realistic at least.
---
In the culture of music buildings have been designed for the purpose of listening to
music. A large fraction of sound energy in those buildings is reaching the listener
by ceiling reflections. A music pavillon e.g. has the purpose of protecting the
musicians and their instruments from a drizzle, but there is also an acoustic
purpose which may be even more important:
http://img.fotocommunity.com/Ruhrgebiet/Essen/Musikpavillon-in-der-Essener-Gruga-a18887292.jpg
Structurae [de]: Bilder: ID 4226
http://www.stifter-bachmann.com/fileadmin/r_projekte/09_7_weissenbach/Pavillon-01.jpg
http://de.academic.ru/pictures/dewiki/77/Musikpavillon_Kurmuschel_Sassnitz.jpg
http://www.rwbilder.de/frame/Themen/2010/Juni2010/OffeneBuehne/Musikpavillon_01.06.2010_01.jpg
http://s2.tramino.de/oberstdorf/417/1200.jpg
http://www.freiotto.com/FreiOtto ordner/FreiOtto/FreiOttoZelteGross-Dateien/image002.jpg
How is that sound utilized, which is radiated upwards ?
Again i like to throw some unsorted thougts towards you:
- The sound reflected from the ceiling is utilized to give some notion of presence
and sparkle even to the more distant listeners, so the organizer can sell
some more tickets even for the distant seats at a reasonable price for both sides,
organizer and visitor as well.
- The ceiling reflected sound is utilized for 'spectral averaging' over a string
ensemble e.g. making the spectrum of sound more smooth as it would be the case when
sitting at the side of that ensemble without any reflections present ... you would
get the directional radiation patterns of the closest instruments in a dominant
manner then and the desired 'mixing', 'smoothing' and 'fusion' does not occur to
the desired extent.
Some agreement so far ?
---
Now the third set of unsorted thoughts, risking the leap over to our beloved topic
of discussion: The Loudspeaker (and its frequency dependent directivity characteristics.)
Today in the focus: The vertical plane.
How would a loudspeaker sound, which has balanced energy and also dispersion
say up to around 4 Khz and from there upwards has a significant narrowing
in vertical radiation angle thereby only keeping the sound pressure radiated
on axis constant ?
Instead of a verbose answer i send some pictures with listening impressions from
a rather 'wet' and small room:
---------------------------------------------------------------------
Our ear/brain has developed even before speech and music came into human live.
Our perception system has developed to survive.
It's first aim is the identification of objects: Their nature, their distance,
their direction, their speed and direction of movement ...
---
OK, let's imagine our stone age huntsman passing a narrow cutting ... he is
being observed by someone who is able to move very quiet ...
But that sort of cat is not perfect either. Some pebbles get loose and begin to roll
down over the crag, causing some transient clicking noise ... got it ?
Where ? Exactly where, please ?
That fractions of a second decide whether our huntsman's plans for his future live
will be of any use or not.
Over 10.000 years later, that clicking noise might indicate an unlocked overhead bin
in an airplane with you sitting right underneath.
And if you did not know already, than i might be allowed to tell you, that luggage
falling out of unsecured overhead bins is the oftenmost cause for serious
insuries of aircraft passengers. Don't be afraid of flying in an airplane, but have
a closer look on the luggage please ...
---
So far so good ...
Human live - as well as killing and defence strategies - have been perfectioned.
Speech and musical culture have an omportant place also in human live.
There is still that frequency range around 8 Khz, which is associated with 'brillance'.
It is important for our perception and our ear conches are designed to give
that frequency range a gain of over 10 dB if coming from above. That gain is
lesser if the elevation angle is lower that 90 degrees from the median plane.
That 'directional band' is called the 'above' band according to Jens Blauert.
File:Akustik - Richtungsbänder.svg ? Wikimedia Commons
---
Now mixing some of my own (almost trivial) thoughts into the rules of the game:
- An acoustical 'object' or 'gestalt' is not complete without that 'brilliance' information.
- Sounds known to us miss 'sparkle' without that frequency range.
- The frequency range together with the 'presence' region below that 'above' band
has importance in detecting positions of 'sound objects' rather fine and make them
'real' even on a proximity level of perception. An object with 'no proximity' is not
a real (or realistic) object. For the 'realism' in a painting there are similar rules:
An object is embedded into its context/backround by perspective, oherwise it will
not occur to be real or realistic at least.
---
In the culture of music buildings have been designed for the purpose of listening to
music. A large fraction of sound energy in those buildings is reaching the listener
by ceiling reflections. A music pavillon e.g. has the purpose of protecting the
musicians and their instruments from a drizzle, but there is also an acoustic
purpose which may be even more important:
http://img.fotocommunity.com/Ruhrgebiet/Essen/Musikpavillon-in-der-Essener-Gruga-a18887292.jpg
Structurae [de]: Bilder: ID 4226
http://www.stifter-bachmann.com/fileadmin/r_projekte/09_7_weissenbach/Pavillon-01.jpg
http://de.academic.ru/pictures/dewiki/77/Musikpavillon_Kurmuschel_Sassnitz.jpg
http://www.rwbilder.de/frame/Themen/2010/Juni2010/OffeneBuehne/Musikpavillon_01.06.2010_01.jpg
http://s2.tramino.de/oberstdorf/417/1200.jpg
http://www.freiotto.com/FreiOtto ordner/FreiOtto/FreiOttoZelteGross-Dateien/image002.jpg
How is that sound utilized, which is radiated upwards ?
Again i like to throw some unsorted thougts towards you:
- The sound reflected from the ceiling is utilized to give some notion of presence
and sparkle even to the more distant listeners, so the organizer can sell
some more tickets even for the distant seats at a reasonable price for both sides,
organizer and visitor as well.
- The ceiling reflected sound is utilized for 'spectral averaging' over a string
ensemble e.g. making the spectrum of sound more smooth as it would be the case when
sitting at the side of that ensemble without any reflections present ... you would
get the directional radiation patterns of the closest instruments in a dominant
manner then and the desired 'mixing', 'smoothing' and 'fusion' does not occur to
the desired extent.
Some agreement so far ?
---
Now the third set of unsorted thoughts, risking the leap over to our beloved topic
of discussion: The Loudspeaker (and its frequency dependent directivity characteristics.)
Today in the focus: The vertical plane.
How would a loudspeaker sound, which has balanced energy and also dispersion
say up to around 4 Khz and from there upwards has a significant narrowing
in vertical radiation angle thereby only keeping the sound pressure radiated
on axis constant ?
Instead of a verbose answer i send some pictures with listening impressions from
a rather 'wet' and small room:
Attachments
The "small line source" ribbon like tweeter is about 12cm in length.
The woofer has about 10cm diameter.
Crossover frequency is around 900Hz.
Since there is little directionality in the crossover region, that speaker
does not have the usual "2-way woofer/dome discountinuity" in the
crossover region and sounds rather homogeneous from midrange to
presence.
But it has a problem too: While the vertical directivity of that tweeter
increases with frequency, the horizontal directivity is not changing
significantly with frequency.
With tweeter on ear level, the speaker sounds as brillant at 180 degrees
as on axis, but mind the vertical plane ...
The poor guy who designed and built that little bugger around 10 years
ago had to find a compromise between flat energy response and flat
sound pressure on axis, so the speaker is designed to be listened
from a vertical angle around 10 degrees vertically relative to the
center of the tweeter.
If you listen to that tweeter on ear level if will drill a 'brilliance hole'
into your cochlea ...
If you adjust vertical angle right, it is quite balanced. That makes the
speaker a rather bitchy thing.
There is another problem: The brilliance information is there, but
it does not balance properly with the presence region, no matter
how you adjust angles or turn your head.
A measuring microphone 10 degrees vertical off axis will tell you
that everithing is pretty flat ... but that is only a small fraction of
the truth. My ears tell me, that the speaker is bitchy and it stays
bitchy.
The same speaker turned by 90 degrees, has a different flavor and
is far less bitchy. There is sparkle at nearly every listening position.
A deviation in horizontal listening angle from axis - which
was the super critical vertical axis before - can be put away
quite easily - subjectively.
Of course, everyone would mount such a tweeter vertically,
because of the more even dispersion in the horizontal
plane, imaging issues etc. ... right ?
There is only one from these speakers in the world, so i
cannot verify a stereo setup unfortunately.
Possible that the horizontal alignment would couse some
irregulararities in localization when listening out of symmetry ...
But nevertheless i fear the horizontal setup is superior under
that certain conditions in terms of homogeneity and simply
being less bitchy.
Kind Regards
The woofer has about 10cm diameter.
Crossover frequency is around 900Hz.
Since there is little directionality in the crossover region, that speaker
does not have the usual "2-way woofer/dome discountinuity" in the
crossover region and sounds rather homogeneous from midrange to
presence.
But it has a problem too: While the vertical directivity of that tweeter
increases with frequency, the horizontal directivity is not changing
significantly with frequency.
With tweeter on ear level, the speaker sounds as brillant at 180 degrees
as on axis, but mind the vertical plane ...
The poor guy who designed and built that little bugger around 10 years
ago had to find a compromise between flat energy response and flat
sound pressure on axis, so the speaker is designed to be listened
from a vertical angle around 10 degrees vertically relative to the
center of the tweeter.
If you listen to that tweeter on ear level if will drill a 'brilliance hole'
into your cochlea ...
If you adjust vertical angle right, it is quite balanced. That makes the
speaker a rather bitchy thing.
There is another problem: The brilliance information is there, but
it does not balance properly with the presence region, no matter
how you adjust angles or turn your head.
A measuring microphone 10 degrees vertical off axis will tell you
that everithing is pretty flat ... but that is only a small fraction of
the truth. My ears tell me, that the speaker is bitchy and it stays
bitchy.
The same speaker turned by 90 degrees, has a different flavor and
is far less bitchy. There is sparkle at nearly every listening position.
A deviation in horizontal listening angle from axis - which
was the super critical vertical axis before - can be put away
quite easily - subjectively.
Of course, everyone would mount such a tweeter vertically,
because of the more even dispersion in the horizontal
plane, imaging issues etc. ... right ?
There is only one from these speakers in the world, so i
cannot verify a stereo setup unfortunately.
Possible that the horizontal alignment would couse some
irregulararities in localization when listening out of symmetry ...
But nevertheless i fear the horizontal setup is superior under
that certain conditions in terms of homogeneity and simply
being less bitchy.
Kind Regards
Hi,
Just an equally unsorted ovbservation.
Certain sounds etc. cause a mild (one hopes) version of the fight/flight reflex, flooding the body with adrenalin, giving a high (one can also take certain drugs for the same effect magnified).
I recently talked with someone about audio who has tastes running rather counter to mine.
Now to me the adrenaline high combined with music that is not supposed to elicit it by itself is a "bad thing". at the same time this Gentleman considered it "always a good thing".
This may explain the different reactions people have to things like ceiling flooders... It would literally be an issue of emotional response and reaction to (brain internal) drugs.
Ciao T
Just an equally unsorted ovbservation.
Certain sounds etc. cause a mild (one hopes) version of the fight/flight reflex, flooding the body with adrenalin, giving a high (one can also take certain drugs for the same effect magnified).
I recently talked with someone about audio who has tastes running rather counter to mine.
Now to me the adrenaline high combined with music that is not supposed to elicit it by itself is a "bad thing". at the same time this Gentleman considered it "always a good thing".
This may explain the different reactions people have to things like ceiling flooders... It would literally be an issue of emotional response and reaction to (brain internal) drugs.
Ciao T
Taken from
The difficulties of audio reproduction
"
Conventional loudspeaker reproduction cannot create such an experience because the right box is only approx. 30 degrees off from the median axis. However, more tight spacing causes more correlation and is, consequently, less attractive. In the center of the concert hall the ceiling reflections are hardly contributing to an improvement of spatial impression. Such reflections in line with the center are counterproductive, normally. "
It seems to me, that proper localization of phantom
sources can hardly be achieved by designs which
radiate the presence to brillance region predominantly
towards the ceiling.
But: If a system lacks balanced dispersion in that
frequency range and the energy response drops, a
radiation of that range towards the ceiling might
increase the amount of presence and sparkle detected
by our ears and a more consistent and balanced
'color' of the sound allover the room might me achieved.
There might be also an improved "spaceousness" with
some systems as the speaker itself is less detectable
as a sound source. Nevertheless localization of phantom
sources will presumably suffer.
On the other hand, for the design of predominantly
direct radiating speakers, a consistent directivity
pattern in the vertical plane seems as important as
in the horizontal plane.
Reflections from the ceiling should have approximately
the same spectral "fingerprint" like the direct sound,
especially from upper midrange to brillance region.
If the localization of phantom sources can rely on
balanced direct sound (and uncolored lateral reflections),
a moderate presence of - also uncoloured -
ceiling reflections might improve "spaceausness"
but not seriously disturb localization of phantom
sources.
That amount of energy from above in "presence and
sparkle region" might also contribute in the listener
being less restricted to the sweetspot and give the
a more "homogeneous" experience allover the room.
IMO consistent ceiling reflection may also contribute in
'depth' of imaging.
Kind Regards
The difficulties of audio reproduction
"
Conventional loudspeaker reproduction cannot create such an experience because the right box is only approx. 30 degrees off from the median axis. However, more tight spacing causes more correlation and is, consequently, less attractive. In the center of the concert hall the ceiling reflections are hardly contributing to an improvement of spatial impression. Such reflections in line with the center are counterproductive, normally. "
It seems to me, that proper localization of phantom
sources can hardly be achieved by designs which
radiate the presence to brillance region predominantly
towards the ceiling.
But: If a system lacks balanced dispersion in that
frequency range and the energy response drops, a
radiation of that range towards the ceiling might
increase the amount of presence and sparkle detected
by our ears and a more consistent and balanced
'color' of the sound allover the room might me achieved.
There might be also an improved "spaceousness" with
some systems as the speaker itself is less detectable
as a sound source. Nevertheless localization of phantom
sources will presumably suffer.
On the other hand, for the design of predominantly
direct radiating speakers, a consistent directivity
pattern in the vertical plane seems as important as
in the horizontal plane.
Reflections from the ceiling should have approximately
the same spectral "fingerprint" like the direct sound,
especially from upper midrange to brillance region.
If the localization of phantom sources can rely on
balanced direct sound (and uncolored lateral reflections),
a moderate presence of - also uncoloured -
ceiling reflections might improve "spaceausness"
but not seriously disturb localization of phantom
sources.
That amount of energy from above in "presence and
sparkle region" might also contribute in the listener
being less restricted to the sweetspot and give the
a more "homogeneous" experience allover the room.
IMO consistent ceiling reflection may also contribute in
'depth' of imaging.
Kind Regards
Hello Thorsten,
i agree that dumping and reaction to brain internal
drugs may be a serious distinguishing mark between
individuals ...
Nevertheless concerning the ceiling flooder approach
i have to admit, that
@graaf: "there is something to it".
Not sufficient for me to make ceiling flooders in future,
but the discussion points to the importance of consistent
radiation even the vertical plane IMO.
Kind Regards
Hello,
How about this vintage idea?
Dipole on the floor, and ceiling facing tweeter?
Can this be regarded as a flooder?
http://www.troelsgravesen.dk/download/SFB-review.pdf
...
Hello,
depending on the crossover frequency i would say
it is a "presence to brillance flooder" ...
And i can imagine that design to have its own
qualities compared to products of the time ...
Kind Regards