Linear Phase in Studio Monitors - the responses of Dr. Floyd Toole, Dr. Wolfgang Klippel, Andrew Jones, and James Croft

[thewas]

Imho quite a good read.
https://audioxpress.com/article/zero-phase-in-studio-monitors

 

[Absolute]

Fantastic! I also dig the comments below the article

I did actually do a blind test with and without phase correction when I had the Kii Threes. I couldn't reliably hear a difference when seated far away and close to back wall, but I could tell the difference on certain songs when both me and the speakers were clear of walls and in a near-field kind of distance. I don't remember what the results were, but it wasn't 20 out of 20. I needed to concentrate rather intensively and my ADD brain isn't particularly reliable for sustained concentration excercises.

If I recall correctly I think I noticed a smidgen less sharpness around the edges with phase correction on. I should have written it down.

Anyways, I agree with Putzeys. It's stupid to trade off other things for phase correction, but if it doesn't cost anything, why not?

 

[Frank Dernie]

I did some phase experiments about 45 years ago. I couldn't hear the difference. Mind you pretty well all speakers are not phase accurate then or now, nor is it likely to be important on a stereo mix down from multiple channels since this won't have any phase relevance anyway.
On a listening test a few years ago full DSP correction made a small difference on a pop music recording but was a notable gain of the stereo image on an old school simply miked Gilbert and Sullivan recording which may have had worthwhile phase information on it.

 

[thewas]

Also my experiences are similar at the Neumann head office where we could directly switch the O500C between minimal and linear phase or at my home setups where I use Acourate FIR, the differences are with normal loudspeakers only very small and not even audible with every recording, with big multiway horn speakers where the acoustic centres distances to the listeners ears differ substantially the differences might be larger though (haven't had yet the chance to test it at such).

 

[KSTR]

Anyways, I agree with Putzeys. It's stupid to trade off other things for phase correction, but if it doesn't cost anything, why not?

That's pretty much the bottom line. Correction to linear phase is the icing on the cake once everything else is done competently and the nice thing it is almost zero cost.
IME phase anomalies are more problematic at lower frequencies. A linear phase XO above 1kHz has only very subtle audible effect, if any. The worst offender is a subwoofer XO, LR4@80Hz, and basically any steep XO in the 50Hz to 300Hz range.

 

[KSTR]

nor is it likely to be important on a stereo mix down from multiple channels since this won't have any phase relevance anyway.

Sorry, but that's got nothing to do with it. It's like saying frequency reponse doesn't matter just because in the studio every single instrument and vocal track was subjcted to a hell of EQ'ing...

 

[Frank Dernie]

Sorry, but that's got nothing to do with it. It's like saying frequency reponse doesn't matter just because in the studio every single instrument and vocal track was subjcted to a hell of EQ'ing...

No it isn't. It is nothing like FR at all.

 

[Music1969]

So when I read the conclusion, it reads like using linear phase for DSP crossover is a good thing to do when used in in conjunction with driver time alignment + digital room EQ .

I use Audiolense for generating convolution files. Audiolense is doing driver time alignment. In Audiolense I can pick between lin phase and min phase for :

1. Target curve and
2. Crossover

I've selected lin phase for both.

Appreciate any feedback , if this is a good approach, when doing driver time alignment and digital room EQ

"As Croft pointed out, there are legitimate uses of zero phase that provide substantial advantages, such as in the case of the Trinnov room correction device found in recordings studios, cinemas, home theaters, and high-end two-channel home applications. However, the Trinnov is invoking zero phase in a room environment, which aids in the summation of multiple loudspeakers in a listening space. That however, is very different from offering zero phase embellishments to the anechoic response of a loudspeaker"

 

[anmpr1]

Thanks for the article link. I had not seen it.

I remember when Linear Phase speakers came to the surface in the mid-1970s when Technics introduced its SB series with the drivers physically time aligned and set back from the woofer baffle... Eventually, everyone seemed to figure out that this created as many problems as it solved, and the kangaroo baffle thing mostly disappeared from public view. (Dickason)

This was worked out (or an attempt was made to work it out) by Jon Dahlquist (Rectilinear and, later, his eponymous company); the former was supposed to be achieved with a 'special' crossover design, and the latter doing away with a front 'stepped' baffle in favor of small, multiple baffle-less mountings (the DQ-10). Many liked the result, but FWIW I could never enjoy the resulting sound (not that my opinion amounts to much).

Naturally I did some tests, reversing polarities of loudspeakers, and introducing phase shift to distort musical waveforms — listening for big differences. They weren’t there. At least not in the music I was listening to, through the loudspeakers I was using, from the musical sources I employed. Maybe I was simply unable to hear these things. Yes, there were times when I thought I heard things, but they were subtle, and hard to repeat. Changing loudspeakers made huge differences. Changing recording companies or engineers made huge differences. But, the anticipated “dramatic” event of inverting polarity appeared to be missing, in spite of how appealing the idea of waveform integrity is from an engineering perspective. (F. Toole).

No comment necessary, other than to reference another recent thread on ASR about absolute polarity, and the late Clark Johnson's claim that maintaining same is supposed to be a big deal. It's not (but would probably be nice if it could be done--at least on a conceptual level).

One intriguing approach was pioneered by Bang & Olufsen (B&O), the “filler driver” concept. This was basically a conventional two-way speaker with a filler driver that corrected the phase error. Ingenious but just as flawed as the other approaches. Why flawed? Because all implementations of phase linear speakers other than those using concentric drivers achieve their phase linearity on only one axis. Every other angle, both horizontal and vertical, must introduce added differential delay and so eliminate phase coherency. (A. Jones)

This was the 'phase-link' idea. I lived with a pair of S60 for several days and couldn't get them to 'sound right'. Again, my opinion FWIW. They were, however, nice furniture. We remember that it was 1976 or so, and B&O later made important contributions to loudspeaker design, including the 90 (which would certainly be more appealing, domestically, if it looked like furniture, and not a space alien).

 

[daftcombo]

Very interesting article, thank you.

 

[bennybbbx]

Imho quite a good read.
https://audioxpress.com/article/zero-phase-in-studio-monitors

in your link stand

It turns out that, for perfectly logical reasons, humans are not responsive to phase (i.e., waveforms), whether they are Beethoven, step functions, square waves, etc. A lot of highly motivated researchers have investigated this.

there is not know if they can hear ITD or not. maybe many acoustic research is only done with mono speakers. but for stereo the phase information is more important. I can hear on music that a phase align speaker give better sit inside the room sound. when hear mono i hear not much diffrence. I test with DSP and delay a ribbon tweeter 0.04 ms and there can hear clear the diffrence.

Here is a video that delay frequency

there can hear lots room changes depend on phase

another problem is also the speaker correction. dirac for example. sound very bad to me. this cause lots group delay changes also between left and right channel. thats lot hearable when can hear ITD. see this post with measure. https://www.audiosciencereview.com/...ad-for-your-health.29127/page-17#post-1188953

phase linear filters does not help much. maybe the pre ring have negativ effects too. group delay look good but did not sound so good(room width and clarity) as when use EQ settings that do less changes in group delay . so it can not easy compare

here is a very detail explain of step response with many examples. its only in german. maybe somebody translate to english. https://www.hifi-wiki.de/index.php/...rungantworten_-_Beispiele_aus_der_Messtechnik

 

[thewas]

there is not know if they can hear ITD or not.

Your ITD thesis was enough discussed here, no need to start it all over again:

step response is a important part to show speed of speaker that is good enough for ITD. See measures

I measure a headphone with Low Pass Filter 6 or 12 db at 1.5 khz to simulate a crossover. I do the test also with Phase linear filter, but it change not much in step response so i do not upload the results. there can see the headphone with the simulate crossover is much more faster as the...

www.audiosciencereview.com

 

[bennybbbx]

Your ITD thesis was enough discussed here, no need to start it all over again:

step response is a important part to show speed of speaker that is good enough for ITD. See measures

I measure a headphone with Low Pass Filter 6 or 12 db at 1.5 khz to simulate a crossover. I do the test also with Phase linear filter, but it change not much in step response so i do not upload the results. there can see the headphone with the simulate crossover is much more faster as the...

www.audiosciencereview.com

when you quote text that say phase is not important, then of course the answer to this is same. you and some other that write often in this forum seem the opinion what i write is wrong but then you should explain when phase is not important how can hear ITD ? . so it is possible that the theory you use is only check and verify on a single speaker and not in a stereo system. I have also a test video that show that phase changes sound lots diffrent also on 0.04 ms . just let other choose on the proofs i and you post which is more reality

 

[thewas]

when you quote text that say phase is not important, then of course the answer to this is same. you and some other that write often in this forum seem the opinion what i write is wrong but then you should explain when phase is not important how can hear ITD ?

You were given enough answers in that thread, but the discussion was not fruitful so the thread was locked.

 

[bennybbbx]

You were given enough answers in that thread, but the discussion was not fruitful so the thread was locked.

This link about the step response explain https://www.hifi-wiki.de/index.php/...rungantworten_-_Beispiele_aus_der_Messtechnik was not in the thread from me you link. maybe you read it too. as far i notice you are german and maybe you can translate that or do you think this is wrong ?

 

[thewas]

This link about the step response explain https://www.hifi-wiki.de/index.php/...rungantworten_-_Beispiele_aus_der_Messtechnik was not in the thread from me you link. maybe you read it too. as far i notice you are german and maybe you can translate that or do you think this is wrong ?

That link can be translated by anyone by with his browser https://www-hifi--wiki-de.translate..._sl=de&_x_tr_tl=en&_x_tr_hl=de&_x_tr_pto=wapp and mixes some known facts like the correlation of the shape of the step response to the amplitude/bandwith of the frequency response with some wrong and unproved assumptions, from what I see from the imprint it is a private page where no author is shown or open editing/commenting is allowed like in the real wiki.

 

[OWC]

Liking the practical remarks from James Croft in this article, but yeah this subject is quite well known at this point.

My side notes as for completely flatting the phase like is shown with RePhase.

There are some practicalities there as well.
Especially when one develops a bunch of loudspeakers products that can be mixed and matched by the customer.
Some of those products have different crossovers with different frequencies etc

In that case flattening the phase response can definitely help when different products are combined in a practical context.
When this is not being done some very deep and sometimes audible notches can occur.
(although depending where and how much, but notches in general are sometimes difficult to hear actually)

 

[OWC]

Anyways, I agree with Putzeys. It's stupid to trade off other things for phase correction, but if it doesn't cost anything, why not?

Literally speaking, yes I also agree.

Unfortunately for some people this "why not" becomes an object on its own, if not sometimes a major selling point for products.
Which leads to many people believing it, which leads to having this discussion every 5-10 odd years all over again and repeat the cycle.

 

[bennybbbx]

from what I see from the imprint it is a private page where no author is shown or open editing/commenting is allowed like in the real wiki.

have you look on this page with the desktop web browser and not with mobile ?. there can comment or edit the page same way as in wiki. BTW: I have not create or edit this page. on this page stand too that step response show time. and the frequency a speaker can reach depend on the rise and fall time and the linearity of this. step repsonse is also important measure for hydraulic systems. for hydraulic systems there is no frequency diagram show

 

[thewas]

on this page stand too that step response show time. and the frequency a speaker can reach depend on the rise and fall time and the linearity of this.

As it has been said the frequency response plot is a more expedient representation for such, as an example for a moving object someone could show the acceleration time function to indirectly show the velocity (by integrating it), but its easier to read the velocity plot directly.

step repsonse is also important measure for hydraulic systems. for hydraulic systems there is no frequency diagram show

As above because often the representation in time domain might be more expedient, that doesn't mean there cannot be a frequency response plot: