It should matters: 100 hz is very low... more or less in the middle of the chest impact feeling up to 200/250 hz according the authors (and as low as ? 60/80 hz... my reference to understand here is more 4 string bass of jazz artists than drums, last being harder for e due to their complex behavior).
good to know on the square-waves and CD ! 🙂 ....
good to know on the square-waves and CD ! 🙂 ....
Last edited:
It was not my intention to offend horn-loaded loudspeakers. I have not used the expression "bad sounding". The relationships apply to any medium, transistors, resistors, capacitors, whatever.
Mass is produced by complex interactions. In my definition, resolution refers to a system`s ability to reproduce details. Damping reduces that ability as well as the dynamic range. A larger surface area provides higher resolution.
The electrostatic principle is superior fundamentally due to the strong electric (electronic) property. Generally, "control" is a more appropriate term than "damping".
Mass is produced by complex interactions. In my definition, resolution refers to a system`s ability to reproduce details. Damping reduces that ability as well as the dynamic range. A larger surface area provides higher resolution.
The electrostatic principle is superior fundamentally due to the strong electric (electronic) property. Generally, "control" is a more appropriate term than "damping".
Transient response = frequency response.
20-20K, or as low and as high as you can go, at desired SPL.
Flat frequency response = perfect transient response (if also with flat 0 phase)
Where the 100Hz comes from, either sub or unity or some other horn or cone doesn't matter.
What matters ime, is that c-to-c distances are closed down to prevent lobing and diffraction. This is one concern i have with systems using multiple horns (along with separate mids and subs).
When c-to-c is closed down, and pattern control between driver sections is well integrated, that chest smack gets really awesome ime, along with pants flapping 😀
20-20K, or as low and as high as you can go, at desired SPL.
Flat frequency response = perfect transient response (if also with flat 0 phase)
Where the 100Hz comes from, either sub or unity or some other horn or cone doesn't matter.
What matters ime, is that c-to-c distances are closed down to prevent lobing and diffraction. This is one concern i have with systems using multiple horns (along with separate mids and subs).
When c-to-c is closed down, and pattern control between driver sections is well integrated, that chest smack gets really awesome ime, along with pants flapping 😀
The transcient should be related to the hability to reproduce the same signal form at te output than at the inputt ?
Musicaly speaking a transcient aka dynamic event is a time windows made by an instrument on a note between its lowest and biggest spl. Transcient suggests it can be a short event -short time windows-
As we often look at step response in a multi-ways loudspeaker, I'm not sure why a flat response is involving the ability of a driver to climb and damp as fast as the original signal is... in the recording.
Although I understand when there is a cutt-off sometimes harmonics are involved with some instruments... Is this not the waterfall and the step response of the filters overlaps that should tell us if a speaker has transients (electrically speaking) ?
Now I assume the coupling between ambient air an horn vs a flat ESL planar is indeed not the same...as of course behavior with the room and reflections. (are the horns air canon ? I mean pressurization sarbacans ? and differences is big enough for the tympans?)
To make it short I don't understand what the Fournier flatness as to do with transient but the ability to respect spl level and harmonics... which is saying perhaps few of the ability of the circuit to damp fast enough as it has to play not just an instrument but all the instruments in the recording that are in the same frequency windows (= break-ups = not transient perfect but for a sweep tone test?). Intuition says me than because the ESL is flat transducer and damp fast and still has to play a large width frequency windows then it doesn't "push" enough to produce the pressurization (whatever what is involved here or called) than a horn is producing vis à vis of a human tympan. I.e. ESL are good for micro details and horn canons good for transient boosts at low frequencies (or have an air impedance adaptation with air that permits that) ??
Well massively off topic, sorry for that.
Musicaly speaking a transcient aka dynamic event is a time windows made by an instrument on a note between its lowest and biggest spl. Transcient suggests it can be a short event -short time windows-
As we often look at step response in a multi-ways loudspeaker, I'm not sure why a flat response is involving the ability of a driver to climb and damp as fast as the original signal is... in the recording.
Although I understand when there is a cutt-off sometimes harmonics are involved with some instruments... Is this not the waterfall and the step response of the filters overlaps that should tell us if a speaker has transients (electrically speaking) ?
Now I assume the coupling between ambient air an horn vs a flat ESL planar is indeed not the same...as of course behavior with the room and reflections. (are the horns air canon ? I mean pressurization sarbacans ? and differences is big enough for the tympans?)
To make it short I don't understand what the Fournier flatness as to do with transient but the ability to respect spl level and harmonics... which is saying perhaps few of the ability of the circuit to damp fast enough as it has to play not just an instrument but all the instruments in the recording that are in the same frequency windows (= break-ups = not transient perfect but for a sweep tone test?). Intuition says me than because the ESL is flat transducer and damp fast and still has to play a large width frequency windows then it doesn't "push" enough to produce the pressurization (whatever what is involved here or called) than a horn is producing vis à vis of a human tympan. I.e. ESL are good for micro details and horn canons good for transient boosts at low frequencies (or have an air impedance adaptation with air that permits that) ??
Well massively off topic, sorry for that.
i humbly suggest you are overthinking transient behavior.
Frequency response alone won't tell you about transient response, but when you add in phase response and impulse response, the time domain elements, then all the information needed is complete.
Unless of course you believe there is more to transient response than measurements like square waves, impulse, and step response can tell us......
Frequency response alone won't tell you about transient response, but when you add in phase response and impulse response, the time domain elements, then all the information needed is complete.
Unless of course you believe there is more to transient response than measurements like square waves, impulse, and step response can tell us......
Well, I don't know really, that's what I ask !😀
I can agree indeed if you add phase & impulse (the same?) response as the ability to follow the dynamic gap in spl from the driver (anyhow music is so compressed today that my understanding is dynamic windows are in a little 7 to 10 db with nowadays casual recordings).
I'm asking myself if one can not mistake with the window of the -number of-cycles in a frequency and the ability for the loudspeaker to play it. If the Fournier curve is flat that means the loudspeaker can play the sweep tone at the accurate spl... that doesn't mean the drivers and/or the loudspeakers can handle the difficult musical message where dynamics behavior of several instruments are overlapping in the same frequency windows but at different dynamic behavior and all of that being reproduced by the same driver in a narrow time windows at different spl levels so with say dampings for the start of the next impulse response?!
If you ask me, to winn in precision we should have multichannels : one channel for voice, different channels that follow the behavior of real instruments.... just because music is involving very different dynamic behavior often in the same windows frequencies : typically 50 hz to 800/1000 hz for most acoustical instruments (o there is piano, triangla as well, etc...)
Am I wrong ? Or asking too much, maybe simplier in real life...and asking too much to loudspeakers !
At least, we indeed assume than the targett from a measurment point of view is that the drivers + the filters can pass any signals (I assume the higher the frequency between 2 notes, the most difficult is the damping of the transducer... which is often related to : with micro dynamics, many reviewer say ESL is a winner...
I can agree indeed if you add phase & impulse (the same?) response as the ability to follow the dynamic gap in spl from the driver (anyhow music is so compressed today that my understanding is dynamic windows are in a little 7 to 10 db with nowadays casual recordings).
I'm asking myself if one can not mistake with the window of the -number of-cycles in a frequency and the ability for the loudspeaker to play it. If the Fournier curve is flat that means the loudspeaker can play the sweep tone at the accurate spl... that doesn't mean the drivers and/or the loudspeakers can handle the difficult musical message where dynamics behavior of several instruments are overlapping in the same frequency windows but at different dynamic behavior and all of that being reproduced by the same driver in a narrow time windows at different spl levels so with say dampings for the start of the next impulse response?!
If you ask me, to winn in precision we should have multichannels : one channel for voice, different channels that follow the behavior of real instruments.... just because music is involving very different dynamic behavior often in the same windows frequencies : typically 50 hz to 800/1000 hz for most acoustical instruments (o there is piano, triangla as well, etc...)
Am I wrong ? Or asking too much, maybe simplier in real life...and asking too much to loudspeakers !
At least, we indeed assume than the targett from a measurment point of view is that the drivers + the filters can pass any signals (I assume the higher the frequency between 2 notes, the most difficult is the damping of the transducer... which is often related to : with micro dynamics, many reviewer say ESL is a winner...
Last edited:
Hi, N101N
Since I don't have a college degree in Physics or Magnetics, I can only give you my experiences with the speakers I have owned. I have read your post 243 and sill don't understand your conclusions. The moving mass of a compression driver diaphragm (or any other driver) is a fixed weight and does not change. Because I use field coil drivers, I have the ability to change the flux density in the gap by changing the voltage of the power supply. The higher the flux density, the more control ( damping ?) of the diaphragm. I have the ability to put 2.4 tesla + on a 4" beryllium diaphragm that weighs less then 3 grams. I don't think it is much more complicated then that. I have owned 3 different electrostatic speakers over the years and much prefer (overall) the sound of my compression drivers. Of course, this is MY opinion based on my many years of experience. What is your experience ?
Since I don't have a college degree in Physics or Magnetics, I can only give you my experiences with the speakers I have owned. I have read your post 243 and sill don't understand your conclusions. The moving mass of a compression driver diaphragm (or any other driver) is a fixed weight and does not change. Because I use field coil drivers, I have the ability to change the flux density in the gap by changing the voltage of the power supply. The higher the flux density, the more control ( damping ?) of the diaphragm. I have the ability to put 2.4 tesla + on a 4" beryllium diaphragm that weighs less then 3 grams. I don't think it is much more complicated then that. I have owned 3 different electrostatic speakers over the years and much prefer (overall) the sound of my compression drivers. Of course, this is MY opinion based on my many years of experience. What is your experience ?
If I understand sound power directivity curves correctly (is that the same thing as room power?) they are derived from symmetrical measurements. But Toole indicated "it is possible" an asymmetrical dispersion (favoring lateral reflections) might be advantageous in his paper titled, Loudspeakers and Rooms for Sound Reproduction—A Scientific Review* June 2006 (pdf attached.)
Toole also states early reflections are beneficial.
Early reflections disturb imaging, however, humans are terrible at localization in the 1,000Hz to 3,000Hz range and because that range is a fairly wide spectrum you'd wind up with beneficial "abundant reflections." I'd be concerned that a narrow band might be more prone to running into an audible comb filtering problem, but I don't know. Maybe there are enough reflections even in a narrow band that it doesn't matter.
As Toole stated, it's an area that needs more research. I don't know if anyone has done the research since 2006.
Last edited:
read this pdf about spinorama charts and you will get good info
https://www.sausalitoaudio.com/wp-content/uploads/2018/07/Interpreting-Spinorama-Charts.pdf
Yes, multichannel is a possibility. However, Toole said lateral reflections are beneficial. So you might want extra side channels instead of a center.
I feel a little silly even talking about this. I think what made me bonkers about it was that Earl Geddes said in terms of audio, "we live above 1,000Hz."
Now I'm all, "oh, it's spectrum above 1,000Hz, obviously I need to go crazy with it."
Well, almost 6 of the 10 audible octaves fully fit below 1,000Hz...
What to believe?...... what someone says, or what makes better sense?
Here's the complete quote to add context.
https://www.diyaudio.com/forums/multi-way/362214-major-frequency-ranges.html#post6391692
So you do it with a different set of priorities above 1kHz, and a different set of tools. Without the sub 1kHz you have nothing.. fixing that is pretty difficult too, so get to it.
Brad, many use the minimum early reflections approach. It's been called headphone like, except there is an abundance of reverberation to supplement it. I know of little discussion about obvious and specific holes in the response. If it were true this kind of system couldn't sound as good as it does.
Finally, the only difference between your suggestion and my suggestion, on compromising by allowing a change of directivity in a band that needs otherwise adjusting... is that mine doesn't allow the balance of early reflections to change (from none to none).
Thanks for that added perspective.
My take on the three frequency ranges described in that link:
up to 200Hz, 200-1000Hz, and 1000Hz and above...
is that 200-1000 Hz is no more forgiving as postulated, than 1000Hz and above when we approach correct response.
It might be more forgiving, in the gross sense of, which range causes the least sense of hearing 'wrongness' when there's an incorrect response.
But to that, my opinion is who cares about grading relative shades of worse....
(By correct response, i mean flat mag with whatever house curve, and a flattish smooth-ish phase curve.)
As far as 200Hz and below, in a room, it is so room dependent... i pass ....
Take it outdoors, and <200Hz is as equally important and unforgiving, as the other two described ranges...totally !!!!!!!
my 2c, fwiw 🙂
The change in tonality of the centre image when early reflections are reduced is very real and has been discussed in lots of places.
Fixing the Stereo Phantom Center
I have experienced this directly myself going from a reflective room with no treatment to a room where the early reflections were controlled through absorption. It does not make the system sound bad but it is quite noticeable, I use a Mid Side EQ filter to help reduce the effect but that introduces it's own set of compromises.
I like the clarity and imaging that reduced early reflections provides, I don't like the change in tonality at centre because that is exactly where I want to sit.
Bradley as mark said above do not forget the bass. In the Harman blind studies 30% of the rating of a speaker was due to the bass. Toole mentions this in his book too and points out the need to get the bass right to make a speaker well regarded. Nothing in your Geddes quote contradicts that.
