John_E_Janowitz said:
Once again the subjective sense of "speed" under discussion is NOT with regard to how quickly the driver can oscillate (or by result, how high the freq. extension is near the average).
That Adire paper is not only useless in this instance, it actually is a disservice to the discussion. (..and one in which I've admonished the author for here in this forum.)
The paper was specifically an attempt to say "Hey look here, our woofers can oscillate quickly thereby *proving* that our high mass high excursion low eff. subwoofers are "fast"!". Of course he knew fully well that what people were describing as "fast sound" had nothing to do with it's ability to oscillate quickly.As for the rest - again, its a general guide to achieve specific subjective sound. In this respect:
A. you don't seem to recognize what people are describing when saying a driver sounds "fast" or "not".
B. you are arguing a topic that "A" you appear not to fully comprehend, and that you are arguing that a "guideline" is inaccurate because the underlying principal is incomplete at best - which is irrelevant. The guideline isn't there to say the guideline is correct, rather it's there to say you are more likely to achieve sound "x". (..fundamentally that's a rather substantial difference - think about it.
)gedlee said:
Hi Earl,
We basically agree here. Increasing mass has to change the drivers Fs and thus the impulse response, although it won't be seen in the initial rise, and is probably difficult to see in the impulse tail. However, there is also the possibility of a resonance at the high frequency cut off of the driver where the of the driver mass may resonate with the voice coil inductance. I don't know how significant this is, but in theory it is possible. This could be apparent in the initial rise of the impulse.
ScottG said:
At last, a statement that I can agree with. It's not that I don't agree or disagree with the others, it's just that with terms like "speed" and "impact" - that clearly mean entirely different things to different people - whats being said doesn't make enough sense to agree or disagree with.
With and without copper sleve. Which one is better?John_E_Janowitz said:
An externally hosted image should be here but it was not working when we last tested it.
My take on this is the the idea of speed and impact as related to a driver are pretty much nonsense. Adding a series inductor is not the same as changing VC inductance. The series connection acts more like a shelving filter, lowering only the amplitude of the signal applied to the driver starting at about f = Rc/(Ls + Lc)/2/Pi where c = coil and s = series. But the effect is the same, increased group delay, attenuated high frequencies.
So what would the impulse look like if the woofer were low pass filtered as it would be in actual use? (rhetorical question) That is what is relevant. If the woofer is to be used only to 200 Hz or whatever, what matters is what the impulse of the filtered woofer looks like. That will be a function of the filtered response. If the filter is designed to a specific acoustic target, then any woofer with the same Fs and Q will have the same impulse when filtered to the same low pass acoustic target. So all that so called "speed" is irrelevant since it is an artifact of what the raw driver is doing above the corner of the LP filter. The woofer LP filter sets the "speed" of the impulse rise as used in a speaker.
It seems the audio will forever be a discipline where science is treated as the enemy rather that the path to progress.
So what would the impulse look like if the woofer were low pass filtered as it would be in actual use? (rhetorical question) That is what is relevant. If the woofer is to be used only to 200 Hz or whatever, what matters is what the impulse of the filtered woofer looks like. That will be a function of the filtered response. If the filter is designed to a specific acoustic target, then any woofer with the same Fs and Q will have the same impulse when filtered to the same low pass acoustic target. So all that so called "speed" is irrelevant since it is an artifact of what the raw driver is doing above the corner of the LP filter. The woofer LP filter sets the "speed" of the impulse rise as used in a speaker.
It seems the audio will forever be a discipline where science is treated as the enemy rather that the path to progress.
John_E_Janowitz said:
And if you check the frequency response you will see that the one w/o the shorting ring has the high frequency response rolled off sooner than the one with the ring. All this is saying is that "speed" = extended frequency response. I'm not at all concerned if a woofer goes out to 10 K Hz or 1 K Hz if it is to be used only below 500 Hz. That 500 Hz LP filter is going to set th "speed" of the woofer in the system.
soongsc said:
Put a 200 Hz LR4 LP filter on them both and look at it again.
gedlee said:
Science forever an enemy? I personally see a complex field of opinions on the usefulness of today's current crop of measurements and measuring tools, but rarely, if ever I should say, have I heard someone say science is forever an enemy (absolutely useless, presumably) to audio advance. Just being scientific about what I see and perceive ...
Please.
Not everyone is good enough in pure science (in anything) and can speak that way.
However, the differences in "speed" and "impact" of sounds are audible and should be common experiences among most of us. These terms are used for describing things, just not so precise and thorough as measured data.
After all, music (what we are getting from loudspeakers) is an emotional thing, at least a big part of it. So it's not totally wrong to decribe it with some subjective terms.
I wholeheartedly appreciate all experts here who manage to link the subjective feelings and the scientific parts of music playback. Just there are always people not so knowledgeable but eager to get some 'guideline' so they can follow more easily.
Thank you very much.
Not everyone is good enough in pure science (in anything) and can speak that way.
However, the differences in "speed" and "impact" of sounds are audible and should be common experiences among most of us. These terms are used for describing things, just not so precise and thorough as measured data.
After all, music (what we are getting from loudspeakers) is an emotional thing, at least a big part of it. So it's not totally wrong to decribe it with some subjective terms.
I wholeheartedly appreciate all experts here who manage to link the subjective feelings and the scientific parts of music playback. Just there are always people not so knowledgeable but eager to get some 'guideline' so they can follow more easily.
Thank you very much.
john k... said:
If you deny the very sense that this hobby is dedicated to, what is the point of the hobby? With that in mind:
Do your electrostatic speakers sound "faster" to you than more traditional dynamic speakers?
Do they have impart less physical impact than traditional dynamic speakers?
Also,
Is science being treated as an enemy with respect to audio?
On the science of audio - how much do we know now that we didn't a century ago? How much more might we know a century from now?
Might it not be the case that a description of "fast" or "impact" will be scientifically resolved in the future?
gedlee said:
gedlee said:
..and yet the descriptors still manage to "crop up". It isn't really about our agreement of them, we haven't progressed that far.
Rather, the acknowledgment that such descriptions are used - as a point of inquiry. i.e. what promotes someone to describe a loudspeaker as sounding "faster" than another?
..are we still in the dark ages? ..is the world flat? On this particular topic in human hearing and audio reproduction, perhaps yes. Should we ignore it because clearly "the world is flat" and using the description "fast" or "speed" or "impact" is utter nonsense?
BTW, if audio as a science isn't progressing fast enough for you (..and really, how could it ever), then check out the science of smell and taste - it might make you feel better about your pursuits in audio.
CLS said:Thanks a lot for your reply ScottG
However I'm somewhat confused between the 'speed' and 'impact' things.
But why does more sense of speed turn out to be less impact?
I think we all are confused about it.. but unlike a few others I don't see that as something to be derided.
I believe that a greater sense of speed resulting in less impact is significantly related to mms in relation to sd for a given freq.. In my experience it is not relegated to any one limited pass-band, but rather the entire audible range (though the perception of impact does tend to elicit different comments for different pass-bands).
People often describe a ribbon tweeter as having little "impact", or is dynamically "restrained" while still sounding "fast". The same for electrostatic tweeters. Now they *might* (depending on the design) have a huge sd, and as a result have much greater mass than a conventional tweeter or even a compression driver. Despite this the "lack of impact or dynamics" description usually remains.
An actively driven compression driver and waveguide however - that's usually described as both "fast" and "dynamic" and/or having "impact". Greater mms, smaller sd (though significantly more than most tweeters), LOTS of force reacting on that mms.
Adding mass usually lowers eff., and that tends to reduce the perception of "fast". The compression driver overcomes this through additional force on that additional mass, and with compression - all dramatically raising eff..
2 drivers in parallel having less impact than one? All else equal - usually they wouldn't, nor would they have *more* impact. The difference is in comparison to drivers that are (as a single unit) more efficient. Those more efficient drivers tend to have both greater "impact" and "speed" when compared to two drivers "driven" in parallel that *net* the same spl for 1 watt at 1 meter as the single more eff. driver.
Hello,
If my memory doesn't fail, it resulted from very old and long discussions on different forums that, independantly of the lowest and highest frequencies reproduced by the loudspeakear, the sensation of "speed" for a bass drivers ( plus its load) is related to their ability to keep the relative phase between harmonics of a same note the most similar possible as in the recorded note.
This has also consequence on the shape of the impulse and even its rise for sure.
Best regards from Paris, France
Jean-Michel Le Cléac'h
If my memory doesn't fail, it resulted from very old and long discussions on different forums that, independantly of the lowest and highest frequencies reproduced by the loudspeakear, the sensation of "speed" for a bass drivers ( plus its load) is related to their ability to keep the relative phase between harmonics of a same note the most similar possible as in the recorded note.
This has also consequence on the shape of the impulse and even its rise for sure.
Best regards from Paris, France
Jean-Michel Le Cléac'h
Hello,
Phase linearity is a nice design goal, but in real listening space (= a living room) it cannot be achieved at low frequencies in practise because even with only one reflection the relative phase of harmonics is heavily distorted. A bass note has an envelope and duration of the envelope contains some cycles of fundamental frequency. Duration of the envelope is thus very long compared to the few first reflections so they are summed to the original envelope chancing the relative phases of harmonics because different relation of reflection path length to the wave length.
Due to the room we don't have phase linear bass even with ideal speakers. Sad but true.
- Elias
Graham Maynard said:
Jmmlc said:
Phase linearity is a nice design goal, but in real listening space (= a living room) it cannot be achieved at low frequencies in practise because even with only one reflection the relative phase of harmonics is heavily distorted. A bass note has an envelope and duration of the envelope contains some cycles of fundamental frequency. Duration of the envelope is thus very long compared to the few first reflections so they are summed to the original envelope chancing the relative phases of harmonics because different relation of reflection path length to the wave length.
Due to the room we don't have phase linear bass even with ideal speakers. Sad but true.
- Elias
Hi ScottG
I find your obeservations interesting - but the *explanations* given might need some further development.
There is a constant misuse of the B*L or B*L in relation to Rdc in the sense that there is more or less force available to move mass.
This simply isn't the case.
If you are argumenting that line the force available to move the mass is related to the current flow multiplied by whatever constant - even if this "constant" isn't constant over freqeuncy (impedance).
Feed your spekaer with more or less current - and B*L (or whatever) does not matter any more - very easy..
I tend to go with Jean-Michel - possibly we only would have to analyse the final HP transfer function to track it all down to differencies in phase.
I also doubt that thermal distortion is involved too much - change stuffing of a speaker and speed and impact will change as well - no need to change any parameters of the chassis itself to get that sonic change. As JohnK is demanding - we have to look at the final speaker - not necessarily the chassis alone.
If so - we might even relate to different XO concepts (butterworth / bessel) also giving a different taste of speed / impact / flow - no?
Does not matter IMO as there is very sophisticated "pattern recognition" done by our ear brain sytem - for sure capable to detect phase differencies in the presence of reflections.
Michael
I find your obeservations interesting - but the *explanations* given might need some further development.
There is a constant misuse of the B*L or B*L in relation to Rdc in the sense that there is more or less force available to move mass.
This simply isn't the case.
If you are argumenting that line the force available to move the mass is related to the current flow multiplied by whatever constant - even if this "constant" isn't constant over freqeuncy (impedance).
Feed your spekaer with more or less current - and B*L (or whatever) does not matter any more - very easy..
I tend to go with Jean-Michel - possibly we only would have to analyse the final HP transfer function to track it all down to differencies in phase.
I also doubt that thermal distortion is involved too much - change stuffing of a speaker and speed and impact will change as well - no need to change any parameters of the chassis itself to get that sonic change. As JohnK is demanding - we have to look at the final speaker - not necessarily the chassis alone.
If so - we might even relate to different XO concepts (butterworth / bessel) also giving a different taste of speed / impact / flow - no?
Elias said:
Does not matter IMO as there is very sophisticated "pattern recognition" done by our ear brain sytem - for sure capable to detect phase differencies in the presence of reflections.
Michael
The main difficulty in communicating these expressions is that there is no reference to a specific passage from a specific music source. If we can express listening optinions on a specific music source, talk about the exact instruments in very speciffic passages, then it is possible for others to playback the same music source and also relate to measured data. A very complicated process. Everyone is going to experience something different based on the speaker design and speaker driver they use.
Once I asked someone to test listen to a speaker. I got the response that human voices had some "noise" sound. How would one normally relate that with engineering terms? Well, it was necessary to first acknowledge what the person meant by listening to the same music; then listen to similar music to see if there is any common trend in sound coloration; make an educated guess where in the frequency range it might occur, and then look at data in that range from various types of measurements to see if a problem area can be identified.
One of the reasons I look at CSD first, is that like the effects of a room, slow decay masks lots of the difficiencies in speaker drivers. Once this is improved, combined with a room with reasonable absorption, then we can start discovering the audible effects of other difficiencies more accurately. This is also why I would recommend at some point, one might need to conduct some listening tests in anechoic environment.
Once I asked someone to test listen to a speaker. I got the response that human voices had some "noise" sound. How would one normally relate that with engineering terms? Well, it was necessary to first acknowledge what the person meant by listening to the same music; then listen to similar music to see if there is any common trend in sound coloration; make an educated guess where in the frequency range it might occur, and then look at data in that range from various types of measurements to see if a problem area can be identified.
One of the reasons I look at CSD first, is that like the effects of a room, slow decay masks lots of the difficiencies in speaker drivers. Once this is improved, combined with a room with reasonable absorption, then we can start discovering the audible effects of other difficiencies more accurately. This is also why I would recommend at some point, one might need to conduct some listening tests in anechoic environment.
Hello,
Detection ability of reflection depends on frequency and delay. At low freqs ear can detect the phase difference on ear signals (used for direction finding) and possibly the phase relations of harmonics. However, if low freq reflection is arriving withing the first fundamental freq cycle, there is no detection of the original phase of that cycle, rather what is detected is the shifted phase. One can detect the harmonic phase shift due the reflection but cannot map it backwards to represent the phase relations of a non reflected case. Thus the room masks the low freq phase in an unretrievable way.
Probably the most advantageous way to improve phase linearity at low freqs is to minimise reflections. This can be achieved by using high directivity sources, like dipole line arrays.
- Elias
mige0 said:
Detection ability of reflection depends on frequency and delay. At low freqs ear can detect the phase difference on ear signals (used for direction finding) and possibly the phase relations of harmonics. However, if low freq reflection is arriving withing the first fundamental freq cycle, there is no detection of the original phase of that cycle, rather what is detected is the shifted phase. One can detect the harmonic phase shift due the reflection but cannot map it backwards to represent the phase relations of a non reflected case. Thus the room masks the low freq phase in an unretrievable way.
Probably the most advantageous way to improve phase linearity at low freqs is to minimise reflections. This can be achieved by using high directivity sources, like dipole line arrays.
- Elias
Hello Elias,
When discussion comes on the necessary of low phase distortion in audio then there always someone bringing such argument as "phase means nothing as we listen in a room inside which refelctions occur".
About the distortion of the envelop due to reflections, I disagree, 1) for the "speed" perception in the bass register, only an, initial part of the envelop has to be taken in account
2) for those low frequency (wavelength more than 2meters) the difference of phase between the reflected wave and the direct wave will be small.
Take 2 louspeakers having same Mms and diameter, the firts one loaded in a bass enclosure and the other one in bass-reflex, having the same low frequency cut-off. Listen them in the very same conditions, despite the room influence, the "speed" perception of the one loaded in close enclosure will be better than the one loaded in BR and this has to be related to the very different group delay curve at ferquency lesser than 150-200Hz.
This is an easy demonstration than phase is important in the perception of "speed" and that phase "mixing" due to reflection have not the importance that you said (even if has some influence for sure) .
Best regards from Paris,
Jean-Michel Le Cléac'h
When discussion comes on the necessary of low phase distortion in audio then there always someone bringing such argument as "phase means nothing as we listen in a room inside which refelctions occur".
About the distortion of the envelop due to reflections, I disagree, 1) for the "speed" perception in the bass register, only an, initial part of the envelop has to be taken in account
2) for those low frequency (wavelength more than 2meters) the difference of phase between the reflected wave and the direct wave will be small.
Take 2 louspeakers having same Mms and diameter, the firts one loaded in a bass enclosure and the other one in bass-reflex, having the same low frequency cut-off. Listen them in the very same conditions, despite the room influence, the "speed" perception of the one loaded in close enclosure will be better than the one loaded in BR and this has to be related to the very different group delay curve at ferquency lesser than 150-200Hz.
This is an easy demonstration than phase is important in the perception of "speed" and that phase "mixing" due to reflection have not the importance that you said (even if has some influence for sure) .
Best regards from Paris,
Jean-Michel Le Cléac'h
Elias said: