Could not agree more Dave (dlr),
What about feeding a driver under test with white noise and observing its spectral response variation at every ten degrees rotation from axis with 10dB increments of drive starting at say 70dB-1m. on axis ?
Any fricative/cone response must then show up as spectral variation at an angle it did not previously.
Here sines and time would not be relevent so surely this would show up any nature of EnABL treatments and yet not be overly complicated to set up either ?
Anything like this already been tried ?
Cheers .............. Graham.
What about feeding a driver under test with white noise and observing its spectral response variation at every ten degrees rotation from axis with 10dB increments of drive starting at say 70dB-1m. on axis ?
Any fricative/cone response must then show up as spectral variation at an angle it did not previously.
Here sines and time would not be relevent so surely this would show up any nature of EnABL treatments and yet not be overly complicated to set up either ?
Anything like this already been tried ?
Cheers .............. Graham.
Graham Maynard said:
White noise is not going to work for any objective results. That's partly the reason that the MLS measurement method was developed. Using any kind of signal such as white noise will be highly impacted by the room response. Floor, ceiling and wall reflections are going to impose their signature, that will therefore be room and sample point dependent. MLS measurements eliminate all room influences down to the limits imposed by the window requirements. A true anechoic chamber would otherwise be required and even in that case a sine wave sweep would be used.
This is even more problematic. Every change of position is going to impose its own, unique signature different from every other position in the room. This is separate from the natural off-axis response of the driver. This entire consideration is moot given the consequences described above.
How did you come to this conclusion? Sines (frequency) and time are fundamental to everything related to a driver's output.
On an aside, why the continual emphasis on fricative? It's not needed. This is a rather distinct and limited aspect of sound reproduction that has no distinct definition as to the actual spectrum involved. It's more confusing than helpful. Anything in the frequencies that impact the "fricative" sounds is part of the frequency response.
Dave
dlr said:
I believe it was mentioned earlier that these curves were generated by taken data from a large sample of listeners being interviewed on what they heard (ie no objective measurements)
dave
planet10 said:
Dave - Correct! Not to mention that the poor little #%#% weren't listening to treated (ie doped) speakers! Gee - why don't we do some testing that uses listeners that know how to judge good speakers!
Oh yeah - I forgot - that's a subjective method! 😉planet10 said:
Yes, they are the successors to the original Fletcher-Munson curves. It's all based on perception of loudness. It illustrates very clearly the reason why there can be such widely varying perceptions of speaker systems and teh impact of any changes. It also illustrates the necessity of valid, neutral and objective measurements.
Dave
dlr said:
It also illustrates the validity of Einstein's observation "Everything that can be counted does not necessarily count; everything that counts cannot necessarily be counted." or counted YET!
"Do you hear what I hear?"
One of the reasons for applying statistical experimental design to ANY test/listening experience/whatever is to develop a sense of what is common between the populations involved in the results.
No one ever wants to do statistical analysis because it takes a fair amount of work to set up trials, treatment combinations and criteria that sort of thing. Not to mention training of those involved on interpretation of the results.
None of the observations so far on this EnABL concept (whether subjective or objective) have been subjected to any sort of statistical management. None of the CSD plots, frequency sweeps, etc. have any conditions on reproducibility, confidence intervals, data variance, etc. Nor do the subjective evaluations of "loss of fricative content" have much meaning without some sort of criteria applied. Is it any wonder there is so little agreement as to what works, what it means, etc.?
As an example, perhaps Bud et. als could enlighten the rationale as to the pattern of squares, their relationship to each other and the purported "loss of fricative content", etc. through such evaluation.
One could imagine a fairly straight forward set of matrix trials wherein maybe 3 variables are manipulated (such as size, placement, and thickness of dots for instance) with criteria set for significance of response. I suspect Sy will probably do something like this, with different variables and criteria of course. Adjustments for confounded results could then be determined as well as significance of trial results and support for the hypothesis or null. Then the discussion could move forward in a coherent fashion.
"do you see what I see?" to paraphrase that old popular xmas song 😉
just a thought...
John L.
konut said:
One of the reasons for applying statistical experimental design to ANY test/listening experience/whatever is to develop a sense of what is common between the populations involved in the results.
No one ever wants to do statistical analysis because it takes a fair amount of work to set up trials, treatment combinations and criteria that sort of thing. Not to mention training of those involved on interpretation of the results.
None of the observations so far on this EnABL concept (whether subjective or objective) have been subjected to any sort of statistical management. None of the CSD plots, frequency sweeps, etc. have any conditions on reproducibility, confidence intervals, data variance, etc. Nor do the subjective evaluations of "loss of fricative content" have much meaning without some sort of criteria applied. Is it any wonder there is so little agreement as to what works, what it means, etc.?
As an example, perhaps Bud et. als could enlighten the rationale as to the pattern of squares, their relationship to each other and the purported "loss of fricative content", etc. through such evaluation.
One could imagine a fairly straight forward set of matrix trials wherein maybe 3 variables are manipulated (such as size, placement, and thickness of dots for instance) with criteria set for significance of response. I suspect Sy will probably do something like this, with different variables and criteria of course. Adjustments for confounded results could then be determined as well as significance of trial results and support for the hypothesis or null. Then the discussion could move forward in a coherent fashion.
"do you see what I see?" to paraphrase that old popular xmas song 😉
just a thought...
John L.
Re: "Do you hear what I hear?"
I absolutely see what you mean. For me the only significant "data point" in this entire thread is the number of people who report a 'positive outcome' when comparing a treated to untreated driver. I am neither qualified nor have the inclination to go through this entire thread and statistically quantify this figure. Suffice to say its overwhelmingly on the positive side.
auplater said:
I absolutely see what you mean. For me the only significant "data point" in this entire thread is the number of people who report a 'positive outcome' when comparing a treated to untreated driver. I am neither qualified nor have the inclination to go through this entire thread and statistically quantify this figure. Suffice to say its overwhelmingly on the positive side.
Re: Re: "Do you hear what I hear?"
If that works for you, that's great! This type of observation and interpretation is also known as observer bias, and is one of the variables a well designed analysis eliminates.
http://en.wikipedia.org/wiki/Observer-expectancy_effect
http://www.alleydog.com/glossary/definition.cfm?term=Observer Bias
John L.
konut said:
If that works for you, that's great! This type of observation and interpretation is also known as observer bias, and is one of the variables a well designed analysis eliminates.
http://en.wikipedia.org/wiki/Observer-expectancy_effect
http://www.alleydog.com/glossary/definition.cfm?term=Observer Bias
John L.
Re: Re: Re: "Do you hear what I hear?"
A much smaller number of experimenters have also dealt with this fact by presenting treated vs untreated auditions to people who were neither told what to expect nor given any details about the process itself, and similar positive comments were made. They could have been
or subjected to 
I have an 
Why don't you try it yourself and then you could be the 
auplater said:
A much smaller number of experimenters have also dealt with this fact by presenting treated vs untreated auditions to people who were neither told what to expect nor given any details about the process itself, and similar positive comments were made. They could have been
or subjected to I have an Why don't you try it yourself and then you could be the 
kaan said:
Kaan, that should be fine. I used the new modeller's acrylic "Revell Aqua Color" which is widely available in the UK. The model makers gave it good reviews - it has dense colour. And it is nicely packaged in an 18mL easily opened plastic cube. (88 ochre brown matt if you must know - bright red made my other driver look like a gypsy caravan!).
For the small centre rectangles I used a fine sable brush - very slightly moistened - which can be drawn almost to a needle point.
For the outside rectangles a larger brush will do a consistent rounded rectangle in a single short stroke.
Alan
Re: Re: "Do you hear what I hear?"
This is nothing more than someone hearing a change or thinking that they do. It's an unreliable way to determine what the changes are and how the changes are effected, the focus of this thread. Even a large amount of anecdotal input with absolutely no method of evaluating that input with no controls whatsoever is no basis for any reliable conclusions. Placebo effect considerations continue to be ignored. Thus the benefit of the objective aspect. It can work hand-in-hand with the subjective if the desire exists.
Any possible negative aspects have been largely ignored as well. Possible degradation of a perfectly good driver is low in the priorities of the proponents. It's also entirely possible that one person may like a change while another may not.
The next step after measuring and controlled testing would be to try to determine a correlation between the measured changes and the perception found in the controlled testing. That's the best avenue to improvement separate from the details of determining how and what changes are effected.
The overwhelming advice and urging here (regardless of the quality and/or cost of the drivers) can be summarized as follows:
"Try it, you'll like it!!!!"
I'm sorry, but that doesn't cut it for me. There are betters ways to pursue it.
Dave
konut said:
This is nothing more than someone hearing a change or thinking that they do. It's an unreliable way to determine what the changes are and how the changes are effected, the focus of this thread. Even a large amount of anecdotal input with absolutely no method of evaluating that input with no controls whatsoever is no basis for any reliable conclusions. Placebo effect considerations continue to be ignored. Thus the benefit of the objective aspect. It can work hand-in-hand with the subjective if the desire exists.
Any possible negative aspects have been largely ignored as well. Possible degradation of a perfectly good driver is low in the priorities of the proponents. It's also entirely possible that one person may like a change while another may not.
The next step after measuring and controlled testing would be to try to determine a correlation between the measured changes and the perception found in the controlled testing. That's the best avenue to improvement separate from the details of determining how and what changes are effected.
The overwhelming advice and urging here (regardless of the quality and/or cost of the drivers) can be summarized as follows:
"Try it, you'll like it!!!!"
I'm sorry, but that doesn't cut it for me. There are betters ways to pursue it.
Dave
one way street
Once again, seems to be a "one way street" wrt any understanding (or even an attempt at it) to engage in any rigourous method that could withstand realistic scrutiny... even in the discussion itself, much less doing any tests.
It sometimes can be helpful to actually read the references provided and ask questions if one is mystified as to their meaning....
Sigh....
John L.
Once again, seems to be a "one way street" wrt any understanding (or even an attempt at it) to engage in any rigourous method that could withstand realistic scrutiny... even in the discussion itself, much less doing any tests.
It sometimes can be helpful to actually read the references provided and ask questions if one is mystified as to their meaning....
Sigh....
John L.
konut said:I guess Fletcher-Munson is meaningless as well.
.
try googling "equal loudness contours" or "Fletcher-Munson", and report back
"...general indicators rather than a prescription as to what a single individual might hear."
http://www.webervst.com/fm.htm
John L.
Re: Re: Re: "Do you hear what I hear?"
...but you won't be the one doing the pursuing--of any kind. You'll just keep returning and rehashing your views... ad nauseum.
dlr said:
...but you won't be the one doing the pursuing--of any kind. You'll just keep returning and rehashing your views... ad nauseum.
Hi dlr,
Not everyone has the equipment you keep saying we should use !
But then maybe not everyone wants to use it for this investigation either !
You state >>Every change of position is going to impose its own, unique signature different from every other position in the room. <<
Yes indeed.
But we will still have only one pair of results to compare from each measurement position - nude, and treated - and if any white noise/amplitude generated changes arise, then these can have but only one driver related cause.
Dave also asks >>How did you come to this conclusion? Sines (frequency) and time are fundamental to everything related to a driver's output. <<
The only Avatar I've ever had shows what is important to me - not sines - other than when used for testing amplitude linearity, bandwidth, phase and propagation delay.
You can have a pure sine at any frequency, say at 5kHz, but if it and its near neighbouring sines are reproduced with a different group delay (due to series impedances and reactive loadings which modify voice coil current flow differently with frequency such that the drive and damping change in time with applied waveforms) compared to other pure sines within our audio spectrum say at 500Hz, then there is no way that examining the response at either 500Hz and 5kHz can tell us how accurately music waveforms will be reproduced during music time.
The same would apply to any fricative wave modalities generated via varying impedance modes and which subsequently act wrt music waveforms in music time, which no sine investigation could adequately reveal.
Will your MLS measurements check for this ? I don't know if it can.
The MLS process must operate with its own built in time/frequency limitations for it to NOT show up reflections etc., and whilst it studies the driver output within those limitations it is not actually recreating the same complex 'music' energised air flow motions we hear !
So I am asking in you return - - while we can all see pointers to driver damping and resonances etc. in what can be those wonderfully coloured waterfalls, do they reveal what we are looking for with regard to delayed impedance/reactivity modification of on-going music transduction ?
For at the present I am left wondering;-
Does Dave fire out these verbose challenges and rebuttals because he knows from experience that his favoured method of driver study does not show anything - because it cannot ?
Cheers .......... Graham.
Not everyone has the equipment you keep saying we should use !
But then maybe not everyone wants to use it for this investigation either !
You state >>Every change of position is going to impose its own, unique signature different from every other position in the room. <<
Yes indeed.
But we will still have only one pair of results to compare from each measurement position - nude, and treated - and if any white noise/amplitude generated changes arise, then these can have but only one driver related cause.
Dave also asks >>How did you come to this conclusion? Sines (frequency) and time are fundamental to everything related to a driver's output. <<
The only Avatar I've ever had shows what is important to me - not sines - other than when used for testing amplitude linearity, bandwidth, phase and propagation delay.
You can have a pure sine at any frequency, say at 5kHz, but if it and its near neighbouring sines are reproduced with a different group delay (due to series impedances and reactive loadings which modify voice coil current flow differently with frequency such that the drive and damping change in time with applied waveforms) compared to other pure sines within our audio spectrum say at 500Hz, then there is no way that examining the response at either 500Hz and 5kHz can tell us how accurately music waveforms will be reproduced during music time.
The same would apply to any fricative wave modalities generated via varying impedance modes and which subsequently act wrt music waveforms in music time, which no sine investigation could adequately reveal.
Will your MLS measurements check for this ? I don't know if it can.
The MLS process must operate with its own built in time/frequency limitations for it to NOT show up reflections etc., and whilst it studies the driver output within those limitations it is not actually recreating the same complex 'music' energised air flow motions we hear !
So I am asking in you return - - while we can all see pointers to driver damping and resonances etc. in what can be those wonderfully coloured waterfalls, do they reveal what we are looking for with regard to delayed impedance/reactivity modification of on-going music transduction ?
For at the present I am left wondering;-
Does Dave fire out these verbose challenges and rebuttals because he knows from experience that his favoured method of driver study does not show anything - because it cannot ?
Cheers .......... Graham.
Re: "Do you hear what I hear?"
20 years of empirical research (sort of like Edison coming up with the light bulb). This does not guarantee that EnABL is the best pattern, or that different drivers require different patterns aren't best for different drivers. What it is is the result of discarding a large number of patterns that don't work.
dave
auplater said:
20 years of empirical research (sort of like Edison coming up with the light bulb). This does not guarantee that EnABL is the best pattern, or that different drivers require different patterns aren't best for different drivers. What it is is the result of discarding a large number of patterns that don't work.
dave
Re: Re: Re: "Do you hear what I hear?"
The focus of this thread was always to let Bud pass on what he has learned, and encourage other people to try it and report back.
dave
dlr said:
The focus of this thread was always to let Bud pass on what he has learned, and encourage other people to try it and report back.
dave
Re: Re: Re: Re: "Do you hear what I hear?"
🙁
Quite agree! As a matter of fact, there has been lots of advancement in lighting technology. None of which people are asking for reasoning. Just trying and experiencing the results was convincing enough. Early on in this thread I have seen a few good ideas, but people contributing them have since moved on.planet10 said:
This is what I have known as well. However, there are just lots of people dreaming and talking instead of taking action. The same attitude is taking toll on the economy. As a matter of fact, the quality of this thread is going in the same direction IMO. Sad to see.planet10 said:
🙁
- Status
- Not open for further replies.