Estoy convencido de que ni la TAD ni la M2 darían más precisión que unas lsr308 o beta20 por culpa de sus respectivas "fase" (yo era el primero que hace pocos años estaba convencido en que la fase era muy relevante a la mínima variación, pero actualmente no pienso lo mismo)
En cuanto a la distorsión armónica (en esto creo pensamos similar), habrá diferencias audibles cuando como es lógico, una caja esté tan al borde de la compresión que sus drivers generen armónicos indeseados muy intensos respecto a la señal (pero eso sólo se dará por el limite físico del tamaño de los drivers utilizados cuando enrosquemos demasiado a la caja que menos SPL soporta)... pero no ocurrirá mientras las beta20 y lsr308 trabajen a niveles de SPL donde sus armónicos indeseados no son lo suficiente intensos respecto la señal. Que EMHO creo se podría resumir con la conclusión final de éste blind test:
http://www.gedlee.com/downloads/The%...Distortion.pdf
- Volviendo al tema fase, podrías enlazar algún estudio blind test donde se hayan controlado los factores ajenos a la fase en sí (para evitar influyan) y se contradiga lo que he colgado de Earl Geddes, Floyd Toole, o Linkwitz? Te lo agradecería, pues me interesaría mucho leerlos con detalle
Piensa que por ejemplo, Linkwitz describe en mi enlace anterior añade un filtro de 24dB/octava en 100hz que añade 45 grados de desfase y concluye que dicho desfase es inaudible en prueba controlada (controlando la respuesta en frecuencia apenas variara):
Cuelgo la prueba completa:
http://www.linkwitzlab.com/phs-dist.htmAudibility of allpass crossover phase distortion
On the Dipole Models page and under FAQ19 I indicated my preference for a low order crossover between woofer and midrange, because of its reduced phase distortion. In the extreme this would argue for a 6 dB/oct 1st order acoustic crossover, which is exceedingly difficult and costly to realize, because it places stringent demands on linearity and frequency response of the drivers used.
I had convinced myself that the phase distortion of a LR4, 24 dB/oct midrange to tweeter crossover is not audible, Ref.17. Those same tests showed that changes in the phase shift at the very low end of the spectrum influenced the character of some test tones. Together with some comparisons between earlier speaker designs, which seemed to favor a model with 12 dB/oct crossover over the same with 24 dB/oct, I concluded that the woofer to midrange crossover should be of low order. In retrospect I think that it might have been an increase of non-linear amplitude distortion in the midrange driver, which gave the impression of better bass.
Following are test results and a setup that you can readily duplicate, if you like to investigate for yourself the audibility of phase distortion of typical crossovers that have allpass behavior. The acoustic lowpass output from the woofer and the acoustic highpass output from the midrange add to a flat amplitude response, but the phase of the summed output changes non-linearly with frequency. This is allpass behavior and it distorts the time domain waveform.
Ideally the waveform is transmitted undistorted as with a 1st order 100 Hz Butterworth crossover, when lowpass and highpass outputs are added.
The 2nd order Linkwitz-Riley, 3rd order Butterworth and 1st order Butterworth crossovers form a 1st order allpass when the polarity of the midrange is reversed. The waveform is distorted. The high frequency spectral content forms a sharp spike of opposite polarity to the input, followed by the low frequency content with the same polarity as the input.
The 4th order L-R crossover forms a 2nd order allpass when woofer and midrange outputs are added. Again, the waveform is distorted, though differently from the previous case.
The spectrum of the waveform, that is applied to the 100 Hz crossover, is shown below. It has a 50 Hz fundamental and harmonics at odd multiples thereof. Each spectral component is transmitted in the above crossovers with correct amplitude, but their relative phase is changed, particularly between the regions below and above the 100 Hz crossover frequency.
The waveforms of A, B and C above are quite different and it would not be unreasonable to expect that they sound different. Yet, I have not found a signal for which I can hear a difference. This seems to confirm Ohm's acoustic law that we do not hear waveform distortion. At least it seems to apply to the phase distortion generated by typical allpass crossovers.
Invitation to verify test results
The waveforms above were generated with an active circuit that duplicates the phase behavior of the different crossovers, but has perfectly flat (within +/-0.1 dB) frequency response. The circuit is readily constructed using the WM1 printed circuit board. While the board was not intended for this application it can be adapted by adding a few wire connections and using jumpers instead of some components. Capacitors may be placed where normally resistors would go and resistors connected in parallel to as in the schematic with the WM1 board component designations below. Only one trace needs to be cut. The circuit design is based on the general allpass configuration
Here is what I would like you to do:
Build the circuit. Test it for flat frequency response and equal gain in all three sections. Trim component values if necessary.
For listening via a loudspeaker insert the circuit ahead of your power amplifier. The speaker preferably has no crossover near the 100 Hz region, or is a 2-way with good bass extension, to reduce the possibility that its own phase distortion masks the contribution from the test circuit, or that it pushes the total phase distortion above the threshold of detection. A full-range Quad ESL might be a good speaker candidate.
Quality headphones are probably even more accurate transducers, though observations derived from them may nor translate directly to a loudspeaker in a reverberant environment. Insert the circuit ahead of your headphone amplifier.
Listen in mono.
Listen to your selection of test signals or program material and record any audible differences or lack thereof as you switch between the sections. Record all test conditions so that others can duplicate them, if possible. Note the playback level. Phase distortion may increase the crest factor of the waveform (B above) and lead to non-linear distortion in the speaker and/or the ear producing a change in sound character.
E-mail your observations to me. I will give a summary of the various findings on this page.
I would really appreciate your participation in this investigation. It could help to settle one more issue in knowing what is audible. I am not trying to prove that all allpass crossovers sound the same, they do not in practice, and there are good reasons for it. I merely want greater certainty whether phase distortion is a contributor and, I think, so would you.
Thank you.
PS:
You might be interested in the mathematics and consider pre-processed test signals instead of building the circuit. We are dealing with three allpass transfer functions.
The reference to compare to:
F0(s) = 1
1st order allpass:
F1(s) = (s - 1) / (s + 1)
2nd order allpass:
F2(s) = (s2 - s 21/2 + 1) / (s2 + s 21/2 + 1)
where s = s + j w
First you would need to determine the impulse response for F1 and F2, then translate it from 1 radian to 100 Hz, and convolve it with the test signal time record. This can be done with MATLAB or similar software.
Test signals might be square-waves of different frequencies, or other artificial signals, and speech, and music samples. The results of the convolution are probably best stored on CD-R for comparative listening tests. While this approach makes the analog circuit construction unnecessary, it must be carefully executed not to introduce digital processing artifacts and it limits the easy choice of test material.
Conclusion
The phase distortion of a 100 Hz acoustic crossover with 12 dB/oct or 24 dB/oct is not audible based upon the above tests.
This is my conclusion. No one has come forth with either confirming or contradictory observations. I must assume that the whole question did not generate much interest, or that readers had already an opinion that was not to be tested. Maybe you just did not have the time to investigate.
At 100 Hz a difference in acoustic path length of 1/8th wavelength, causing 45 degrees of phase shift, corresponds to 43 cm (17 inch). At 150 Hz crossover this decreases to 29 cm (11 inch) and makes the placement of a separate woofer relative to the midrange that much more critical. I like to keep offsets to less than 1/16th of a wavelength which makes it pretty much mandatory at 150 Hz crossover frequency to integrate the woofer with the midrange cabinet. This is not optimal for the woofer/room interaction which is minimized when the woofer is placed near the side walls.
With the woofer separated from the midrange by some angle, a 12 dB/oct crossover seems preferable to me, because the wider frequency overlap gives a more distributed wave launch than a 24 dB/oct crossover, where the transition in output from woofer to midrange cabinet is more abrupt.
-Junio del 2015 (critica opinión de Floyd Toole en cuanto al diseño de cajas y la audibilidad de la phase):
I watched this because so many people are talking about it – and I even took notes.
floyd
The overall thrust of the lecture is, I would say, “objectivity works”. We can make measurements of the hardware, judge them against some fixed criteria, and then demonstrate that these correlate with real, human preferences in blind tests. Hurrah! The age-old debate is over, and we can improve our new speaker designs by building them to maximise their objective scores in the full knowledge that this would correlate with human preferences.
However, I am not convinced by the criteria that were specified in the lecture; it seemed to me that there were holes in the argument and that the case was rather circular. As I wrote before, simply through the design of the experiments and what they are testing for, biases may be baked into such results just as badly as the sighted listening tests Toole dismisses. The scientific method does not guarantee validity if you simply extrapolate (or allow other people to extrapolate) the wrong conclusions from your experiments. Perhaps the situation might be healthier if there were many other people competing to achieve real progress in the audio industry, but I don’t think there are.
What he showed:
Sighted listening tests can be flawed (implied: all sighted listening tests are unreliable).
Some measurements can be carried out on speakers in an anechoic chamber (dubbed ‘Spin-o-rama’) and munged together to create a performance index related to flatness of frequency response and smoothness of off-axis response. Transient response is not a factor. At all.
In listening tests, speakers with the ‘best’ Spin-o-rama score are usually preferred by listeners over the opposite (implied: all else being equal).
Mono allows maximum discernment of difference, and does not contradict stereo listening results in the above tests (implied: therefore mono should be used for all listening tests)
Trained professional listeners give the ‘statistically healthiest’ range of scores, and do not contradict ordinary listeners in the above tests (implied: therefore trained professionals should be used for all listening tests)
What he didn’t show:
That it is valid to use the Spin-o-rama score in reverse i.e. as a tool for designing a speaker. He implies it is, but does not prove that a poor speaker could not be designed that achieves an exemplary Spin-o-rama score.
That transient response doesn’t matter – it is simply ignored. The speakers tested may have had good transient responses, or not, but as most of them were of conventional design they may all have been much of a muchness.
That various speaker technologies are inherently better or worse than others i.e. no view on whether sealed cabinets are better than bass reflex, or active crossovers better than passive – and his performance index is indifferent to this, assuming that flat steady-state frequency response is all that matters.
That mono speakers and trained professionals are the best choice for all listening tests.
You know where I am going. As I have mentioned before, it is possible to produce different colourations related to phase shifts while still producing a perfect frequency magnitude response (the drivers may have their phases matched perfectly throughout the crossover but the phase is shifted relative to other components in the signal). Similarly, bass reflex configurations severely distort the time domain response while maintaining a perfect steady state sinusoidal magnitude response. Toole’s tests don’t address these issues, but they are talked of as the Holy Grail of audio design and will therefore define speaker design into perpetuity.
I have no doubt that flat frequency response and smooth off-axis response are essential, as he says, but might there be more to it than just that? Any unexplained deviations between the listening tests and the measurements (it isn’t a perfect correlation) could be explained by a multitude of factors including the speaker’s transient response which, after all, is a straightforward difference between what was recorded and what the speaker emits – it is just that someone around 1936 declared that ‘phase doesn’t matter’. Until recently it has not been possible to verify this ‘fact’. Comparing different speakers all of which have phase/time distortion and other problems, and finding that listeners cannot tell them apart (in mono using someone’s idea of ‘typical’ music), does not tell us that a speaker without those distortions would not sound better.
Correlation is not causation, but people are talking about the Harman method as if it is. So, if I were a speaker designer doing things by the Toole book, I would always use bass reflex without thinking, as this would have no effect on the Spin-o-rama score but would result in a smaller box. And I would be supremely relaxed about crossover design, ensuring only that it matched the phases of drivers through the crossover. Phase correction and sealed enclosures wouldn’t get a look-in because they offer nothing extra in terms of the Spin-o-rama score but cost more to manufacture.
My opinion is of no consequence, of course, but there are some serious people who do suggest that transient response matters*, and it would have been nice if the guru of gurus could have mentioned it, if only to dismiss it with reasons.
* Just writing those words seems absurd. How can transient response not be important, and yet in a forum debate on the topic of the Toole lecture this is what Amir Majidimehr (onetime corporate vice president of Microsoft’s Consumer Media Technology group, and now running companies involved in audio and video) has just said:
As to effect of phase, and time domain correction, there is little objective data to back any of those techniques. Indeed, listening tests shows that we are exceptionally insensitive in this domain. In sharp contrast, we hear frequency response variations readily. It is for that reason that I so emphasize frequency response measurements and results. Time domain does not enter my vocabulary…
You will be aware of my opinion of the listening tests to which he refers. Clearly the industry could do nothing to correct phase in the past, and in a perfect circularity, using (in my inconsequential opinion) questionable listening tests with uncorrectable speakers, ‘proved’ that we are insensitive to phase, therefore it doesn’t matter. This is now baked into speaker design forever!
http://www.diyaudio.com/forums/multi...uides-582.html
http://www.diyaudio.com/forums/multi...uides-582.html
http://www.diyaudio.com/forums/multi...uides-583.html
... aunque lo suyo EMHO es leerse el enlace entero (pienso que el debate fue bastante interesante)
- Personalmente, que conozca hasta la fecha de hoy (evidentemente no conozco todos los artículos publicados) con cualquier nueva prueba controlada (no es sencillo controlar bien todas las variables), sigue sin haber alguna que muestre que con señal musical provocar pequeños desfases sean audibles en un sistema cajas/sala (ni con altavoces con diseño acústico normales ni con los mejor diseñados): más bien los resultados que conozco es que en sistemas caja/sala la fase no es audible a no ser que la diferencia del desfase sea muy marcada (algo que no se produce variando pocos dB con un filtro IIR... ni siquiera trabajando en zona de no fase mínima)... al igual que en auriculares el resultado de los estudios es que el umbral de audibilidad de la fase es algo menos permisivo (aún así, pocos dB de EQ IIR tampoco afectan audiblemente a la fase con señal musical a través de auriculares).
Llevo mucho tiempo ecualizado y/o jugando con intentar "igualar cajas" en lo posible o intentando que la ecualizada supere a la que no... y al final los resultados se acaban repitiendo en comparativa ciega:
La caja ecualizada (incluso solo con filtros IIR) siempre es la elegida como la que suena más precisa en el foco y la escena vs la que no lleva EQ en todas las pruebas que he realizado o estado presente (incluida una misma pareja de cajas sin EQ vs la misma pareja ecualizada). Al igual que nadie ha sido capaz de diferenciar qué caja sonaba mejor o más precisa entre dos cajas dispares en sala una vez igualada curva de respuesta y SPL.... excepto en condiciones de muy mala posición de la caja y/o cajas con directividad muy dispar de la "típica promedio del monopolo" (caso de un dipolo de radiación en 8 muy marcada en sala poco neutra, un caso muy extremo)... pero en la mayoría de casos acaban sonando "tan parecidas" que nadie ha sido capaz de asociar todas las pistas de las prueba a la misma caja (si no llegamos a la compresión de una de ellas por llevarla demasiado al límite); si fuera tan evidente la audibilidad de incluso pequeñas variaciones de fase, EMHO no se repetirían dichos resultados. Ello va encajando con esos estudios que cada vez muestran que la mayoría de parámetros medibles no son audibles si no son muy marcados en un sistema cajas/sala.
- La espacialidad te la da la cola de reverb. de la sala + ciertos mínimos en la neutralidad de la propia sala + que cada canal te llegue con misma curva de respuesta al oyente + que la grabación contenga ciertos cues registrados sobre la msima.
Estos parámetros son mucho más influyentes que la fase de la propia caja en sí:
si mides sin suavizado en anecoica verás lo desfasadísimo que ya sale el sonido de por sí de cualquier caja; es lo que Floyd Toole comenta ocurre algo similar en lo que recogen los micros, o en la voz de cualquier persona en una sala al desplazarse el oyente pocos cm (la fase varía)... pero en cambio lo bien que nos suena todo ello.
Si pequeños desfases fueran audibles, cualquier sonido en el mundo real sonaría horroroso (pero no es así).
Un saludete