Of course it can... but the How is in the details. The cleverness of the FFT is in knowing the calculus is only good for linear (waveforms) - and nonlinear math is too crazy and hard or impossible-who knows- so by making the time between the points on the curve so short that it appears as a straight line (linear) between sampled points was the brilliance of solving the problem of how to solve non-linear problems with linear math and get an accurate solution; Breaking it up into many short linear pieces and calc.
But one needs to be very careful when sampling that the sampling frq is fast enough... especially in complex waveforms with sudden changes. Most math Plug-in-Play software have this under rudimentary control. For audio, at very low distortion levels, would larger bit and higher sampling freq give more accurate results? Might not matter much? -RNM
Last edited:
Give-me a loudspeaker with a flat response, group delay and impedance from 10 to 100 000 Hz, and a total distortion of 0.001%, and i will have a look to insulation material of loudspeaker's cables or apply transmission lines rules on them.
Did your enclosures present, as mine, a flat impedance all along the bandwidth ? Did they have a flat group delay ? No ?
Why don't you start to optimize the things witch have a > 1% effect before to care of the 0.00001 %.
This loudspeaker cable question look like a pure nonsense to me, and if something is unpleasant in the way my (imperfect) equipment reproduce the (imperfect) records i like, i will try to correct-it with effective methods.
Last edited:
I never found a piece of wire that was 'virtually perfect' except one: The silver wire in the CTC Blowtorch. AND it had to be first broken-in on the spool for at least 30 days before being cut and directionalized from the spool, for signal flow. Then another 30 days after being soldered into the CTC with selected SN62, and the whole unit burned in as a whole. That is the way MY personal CTC Blowtorch was made.
OK Bear, let's let 'em have it! '-)
This is just personal experience folks, this is the climate that I operate in. That is why we could take on the WORLD with the CTC Blowtorch, and usually, if not always, WON!
OK Bear, let's let 'em have it! '-)
This is just personal experience folks, this is the climate that I operate in. That is why we could take on the WORLD with the CTC Blowtorch, and usually, if not always, WON!
Connecting in series 10ths to hundreds of amplifiers and testing the end signal was done on lab (Bell Labs) and in practice (cable telephone amplifiers/repeaters) since early 1900. So, correct. There is nothing extreme or innovative about that.
Ref. Distortion measurement, perception and masking:
Alex. Voishvillo (now with JBL)
ALMA 2009 Europe
http://www.almainternational.org/assets/PapersLibrary/2009EuropeSymp./voishvillo%20-%20nonlinearity%20and%20acoustics.pdf
I would find more realistic if he had opted to “gate off” signals with an amplitude lower than 0.005 (-46dB) instead of the 0.05 (-26db) that he chose for his experiment if he had the desire to reach signals close to "zero crossing"
George
Love to see it.
Because this is an electronics forum.... not a speaker forum. But, I'd love to see something better than this antique way to move air... at least for the critical mid-high range. The Quad does pretty good... lower distortion than some popular amps... work on those to improve them or planar or ribbon -- anything. Today we can eq to flat so enough with that... we need lower distortion drivers.... and not at 1M/1W ! FCOL. But, thats for another forum.
Because this is an electronics forum.... not a speaker forum. But, I'd love to see something better than this antique way to move air... at least for the critical mid-high range. The Quad does pretty good... lower distortion than some popular amps... work on those to improve them or planar or ribbon -- anything. Today we can eq to flat so enough with that... we need lower distortion drivers.... and not at 1M/1W ! FCOL. But, thats for another forum.
How does the wire know its direction, what if it decides to switch, and what about AC it goes in both directions?
If you think about it the whole concept is rather silly, like well, the idea is the music flows frm the CD player to the preamp. The AC signal goes both ways, very silly indeed.
"high end" designers aren't reading Doug Self?
why is anyone building "high end" electronics worse than recording mic themal noise?
much less venue backgound noise?
or how about the "Gold Standard" of tape hiss or vinyl playback surface noise
in the past you couldn't deliver recorded music to home with inaudible noise floor with analog - only since digital audio can you potentially get 120 dB S/N
why is anyone building "high end" electronics worse than recording mic themal noise?
much less venue backgound noise?
or how about the "Gold Standard" of tape hiss or vinyl playback surface noise
in the past you couldn't deliver recorded music to home with inaudible noise floor with analog - only since digital audio can you potentially get 120 dB S/N
Last edited:
And what about effects of full moon effects on gravity Vs electron masses ?
Bridging the test gap -
This paper is excellent in bridging the gap between objective testing and listing/hearing loudspeakers and in particular- masking and deserves some comment here - maybe we need to use only tests for IM difference products which are "invisible to THD". - Need multi-tone stimulous.
- Masking is much less at low SPL listening levels. Higher unwanted harmonics are not masked.
- Weighting very appropriate for unmasked harmonics.
- masking isnt total... more like attenuation in many cases.... or we couldnt hear much going on in complex music, such as classical.
- The work is most useful with high distortion - like dynamic drivers. The conclusion I have a hard time with... to use the knowledge to sort of hide and cover up the poor driver performance. I guess if you have taken the dynamic driver technology as far as you can and its still pretty bad, you can use this approach.
Thx- RNMarsh
This paper is excellent in bridging the gap between objective testing and listing/hearing loudspeakers and in particular- masking and deserves some comment here - maybe we need to use only tests for IM difference products which are "invisible to THD". - Need multi-tone stimulous.
- Masking is much less at low SPL listening levels. Higher unwanted harmonics are not masked.
- Weighting very appropriate for unmasked harmonics.
- masking isnt total... more like attenuation in many cases.... or we couldnt hear much going on in complex music, such as classical.
- The work is most useful with high distortion - like dynamic drivers. The conclusion I have a hard time with... to use the knowledge to sort of hide and cover up the poor driver performance. I guess if you have taken the dynamic driver technology as far as you can and its still pretty bad, you can use this approach.
Thx- RNMarsh
Last edited:
the paper's results are usefully applicable to such grossly nonlinear systems as loudspeakers
test any of the mentioned metrics, thresholds, distortion levels against a audio power amp designed intelligently with at least the background knowledge of Doug Self, Bob Cordell's audio amplifier books and you will get Ron Quan's repeated table captions from his latest paper:
Table xx lists "X" amps that were at the residual level for
"Y test"
duh - read Cabot's AP distortion analyzer manual chapters, papers - defines SOTA in "conventional audio measurents" - includes multiple mutlitome formats from SMTE to FastTest
http://www.tmesystems.net/tmecommz/aug10/images/Fundamentals_Modern_Audio_Meas.pdf
we can build audio amps that measure perceptually transparent by all of the meausres outlined in the Voishvillo paper - and make the measurements with a modern AP analyzer
Last edited:
- Status
- Not open for further replies.