Of course phase can be measured. REW plots it along with every FR relative to the original source or normalized. You can see it with your eyes on the plot as clear as could be although it changes with location from the speaker. But humans can't hear phase. Not humanly detectable even though it is changing when you do a sweep tone even if the phase rotates dramatically. (Some trick research has shown polarity can be distinguished with music.)
Likewise polar response can be precisely charted as you say, esp with Toole's spinorama machine. But you don't hear polar response. You hear the frequency compass and polar response shows up as timbre or tone colour.
Toole emphasized the importance of polar profile in determining the SQ of a speaker. Because of it, you get different sound at different locations in a room sometimes even along the centre-line axis of the speaker.
An FR plot shows what you would hear at the location where your mic is placed as pure tones or as musical instruments played.
A directivity or polar plot shows what you would hear if you walked around your speaker for a single frequency or if you could have your arms filled carrying the entire set of different frequency polar plots. So you can hear polar plots in that sense, but you can't sit in your chair and hear it as distinct from tone colour.
Last edited:
Actually, you can hear your polar plot. Not that easy sitting in the listening spot and being all happy with yourself about the EQ tweaks you did there. But record it with a microphone at that listening spot and listen to it on headphones. Then it will be quite obvious, your brain will not mask the room that easily anymore. Still not convinced? Play back what you've recorded over your speakers, making it double the room effect.
You'll have a hard time to unlearn hearing the room from that point forward!
You'll have a hard time to unlearn hearing the room from that point forward!
Last edited:
Or locate a loudspeaker on floor, face pointing up and move around in circle with it playing pink noise. Quite informative.
The reason for Floyd Toole to hammer on the polar response is because we so easily mask the room with our brain. He was in the business to sell speakers so his main target was to find what works in most situations. Neutral speakers as far as the measured off axis plots go will sound neutral in the room too. Thanks to our brain hiding the coloration the room applies.
But if one gets speakers that have less of a well behaved off axis plots they never seem to really sit well in listening tests. Making a few songs sound wonderful and all others seem like garbage recordings. Just look around how many fault recordings for their non enjoyable listening sessions. Bad recordings do exist. But they are far more sparse than most in the audio scene would like you to believe!
But if one gets speakers that have less of a well behaved off axis plots they never seem to really sit well in listening tests. Making a few songs sound wonderful and all others seem like garbage recordings. Just look around how many fault recordings for their non enjoyable listening sessions. Bad recordings do exist. But they are far more sparse than most in the audio scene would like you to believe!
My system is DSP based, 3-way with active crossovers and 3 channel amps. So I was primed to agree with your assertion... But I was alarmed by the confrontational language of the first post in this thread. It seemed designed more to provoke an argument than to foster a good discussion.
Yep. Some people find arguing to be a form of entertainment, so they look for opportunities to be confrontational and start an argument.
I must admit I didn't read that, just enthusiasm from a hobbyist. I am sure if we were all round a table with beers we would have a great discussion. There are a huge pile of tools in the speaker designers toolbox and all have validity for discussion. In this case the first post noted that for rapid prototyping of an idea DSP beats almost anything else whatever you end up using for the final solution. I'm not sure that is contentious or provoking.
Everyone hears phase....all the time. It is inescapable.
Phase is not polarity; phase is not the particular rotational timing of any frequency or waveform component.
Phase is the relative timing between all frequencies. Relative.
Relative timing manifests itself in the form of tone, timbre, harmonics, transients...the whole audible shooting match...
How can we not hear it? 😉
I admit to being an admirer of Toole. Here's his 2019 opinion:
"Therefore, it seems that: (a) because of reflections in the recording environment there is little possibility of phase integrity in the recorded signal; (b) there are challenges in designing loudspeakers that can deliver a signal with phase integrity over a large angular range, and (c) there is no hope of it reaching a listener in a normally reflective room. All is not lost, though, because two ears and a brain seem not to care.
Many investigators over many years attempted to determine whether phase shift mattered to sound quality (e.g., Hansen and Madsen, 1974; Lipshitz et al., 1982; Van Keulen, 1991; Greenfield and Hawksford, 1990). In every case it has been shown that, if it is audible, it is a subtle effect, most easily heard through headphones or in an anechoic chamber, using carefully chosen or contrived signals. There is quite general agreement that with music, reproduced through loudspeakers in normally reflective rooms phase shift is substantially or completely inaudible."
Hope nobody finds that intolerably argumentative.
And speaking of lab practice (as I am an expatriate of Bell Labs), it may be all but unachievable to mess with phase without introducing other tell-tale changes folks can hear. So even if one or another experimenter says "Eureka", may not be because of phase per se. Just my feeling. Others may feel differently.
BTW, if you can set up your system with RePhase perfectly, you better not move your head from that one exact spot because it will change the phases you just set.
B.
"Therefore, it seems that: (a) because of reflections in the recording environment there is little possibility of phase integrity in the recorded signal; (b) there are challenges in designing loudspeakers that can deliver a signal with phase integrity over a large angular range, and (c) there is no hope of it reaching a listener in a normally reflective room. All is not lost, though, because two ears and a brain seem not to care.
Many investigators over many years attempted to determine whether phase shift mattered to sound quality (e.g., Hansen and Madsen, 1974; Lipshitz et al., 1982; Van Keulen, 1991; Greenfield and Hawksford, 1990). In every case it has been shown that, if it is audible, it is a subtle effect, most easily heard through headphones or in an anechoic chamber, using carefully chosen or contrived signals. There is quite general agreement that with music, reproduced through loudspeakers in normally reflective rooms phase shift is substantially or completely inaudible."
Hope nobody finds that intolerably argumentative.
And speaking of lab practice (as I am an expatriate of Bell Labs), it may be all but unachievable to mess with phase without introducing other tell-tale changes folks can hear. So even if one or another experimenter says "Eureka", may not be because of phase per se. Just my feeling. Others may feel differently.
BTW, if you can set up your system with RePhase perfectly, you better not move your head from that one exact spot because it will change the phases you just set.
B.
Last edited:
Imho, all Toole is saying is that there are engineering challenges in both recording and reproducing phase integrity.
I think that says little about phase audibility itself, and more about the technological limitations of recording and reproducing in reflective environments.
And also with limitations in designing loudspeakers that can deliver phase integrity, even setting aside reflections.
I dunno, seems pretty simple really...
There is the frequency domain, and there is the time domain......
The freq domain is a direct function of time itself; the time domain (phase) is the relative time between frequency components.
And all together it is called sound......again, how can we not hear it?
Oh, i find it is incorrect to say "if you can set up your system with RePhase perfectly, you better not move your head from that one exact spot because it will change the phases you just set."
I've found flat phase traces can hold up quite well for both on and off-axis measurements & listening.
Outdoors certainly, and even indoors some, with a high ratio of direct sound.
I find SQ to be much higher with flat phase than without...at numerous listening positions.
Outdoors really proves it....and for me largely negates Tooles' assertions that the brain doesn't seem to care about phase.
I think that says little about phase audibility itself, and more about the technological limitations of recording and reproducing in reflective environments.
And also with limitations in designing loudspeakers that can deliver phase integrity, even setting aside reflections.
I dunno, seems pretty simple really...
There is the frequency domain, and there is the time domain......
The freq domain is a direct function of time itself; the time domain (phase) is the relative time between frequency components.
And all together it is called sound......again, how can we not hear it?
Oh, i find it is incorrect to say "if you can set up your system with RePhase perfectly, you better not move your head from that one exact spot because it will change the phases you just set."
I've found flat phase traces can hold up quite well for both on and off-axis measurements & listening.
Outdoors certainly, and even indoors some, with a high ratio of direct sound.
I find SQ to be much higher with flat phase than without...at numerous listening positions.
Outdoors really proves it....and for me largely negates Tooles' assertions that the brain doesn't seem to care about phase.
The first time I played around with dsp in my Yamaha receiver (around 3 yrs ago) I found it to be a powerful tool to focus the sound and define the soundstage. I have little understanding of what I was actually doing as Yamaha goes into very little detail, I was able to determine (?) from other reports/testing that it is a blend of FIR and IIR for both delay and phase relationship, working up to a 650hz limit and no higher.
I do know after running the test tone and letting it set itself it was not ‘better’ in my view, it wasn’t until I blindly set it up by ear when it finally gelled……I’m a firm believer now and have bought a minidsp SHD for sub integration in my new setup which will be going together soon. If that works well I may even try the 4x10 for the rest of the xo.
I do know after running the test tone and letting it set itself it was not ‘better’ in my view, it wasn’t until I blindly set it up by ear when it finally gelled……I’m a firm believer now and have bought a minidsp SHD for sub integration in my new setup which will be going together soon. If that works well I may even try the 4x10 for the rest of the xo.
Could be broken down to the phase that is common in our environment vs. the phase irregularities that a source might produce. Both are different I think.
I haven't done my own experiments here, so leaning on others views. Always dangerous I know.
From what I gather - phase is a derivation of frequency response. Alternatively, a change in relative phase causes a change in frequency response.
If our ears move in the listening space, the relative time distances change between non-co-incident sources. This changes phase and thus results in a frequency change.
I am isolating all other variables which may dominate the listening experience, such as baffle diffraction and room effects. Let's assume an anechoic listening experience.
More knowledgeable people than me have argued we cannot and do not hear phase. We hear the effect of phase relationships and what they do to the frequency response.
We hear frequency, we hear distortion of frequency, we hear delay. Each of these is above a threshold of hearing.
I've never heard these people say "we hear phase". We hear the effect of it.
From what I gather - phase is a derivation of frequency response. Alternatively, a change in relative phase causes a change in frequency response.
If our ears move in the listening space, the relative time distances change between non-co-incident sources. This changes phase and thus results in a frequency change.
I am isolating all other variables which may dominate the listening experience, such as baffle diffraction and room effects. Let's assume an anechoic listening experience.
More knowledgeable people than me have argued we cannot and do not hear phase. We hear the effect of phase relationships and what they do to the frequency response.
We hear frequency, we hear distortion of frequency, we hear delay. Each of these is above a threshold of hearing.
I've never heard these people say "we hear phase". We hear the effect of it.
I certainly am not qualified to argue the technical aspects but I can tell you once dialed the sound is improved everywhere in the house…..even listening from different rooms.
I think we are accustomed to "phase" changes in our environment. It's how are brain defines a space we are in, walls, floor, ceiling proximity. If we hear some "unexpected" phase change that's when confusion sets in. An extreme example might be an approaching train, blowing it's whistle, but you hearing the Doppler effect before the train passes you. That would certainly confuse your mind.
If you have another look at post #47, you'll see that engineers have been trying to show it is audible for decades. It is only barely discriminable using trick sounds and when there is no room which always scrambles the phase structure no matter how much RePhase you use (testing was in an anechoic chamber or trick tones in headphones).
Doppler is a change in frequency you are hearing, not phase. All drivers (even horns and ESLs to a far smaller amount*) necessarily produce some Doppler-like distortion (and of course, related phase weirdness) but nobody on DIYaudio seems to fret about that.
B.
* not sure about Ionovac - anybody know?
Doppler is a change in frequency you are hearing, not phase. All drivers (even horns and ESLs to a far smaller amount*) necessarily produce some Doppler-like distortion (and of course, related phase weirdness) but nobody on DIYaudio seems to fret about that.
B.
* not sure about Ionovac - anybody know?
Last edited:
This may be due to where the reflections come from, how late they are, whether they are quiet and how their response has changed compared to the main sound.
Right, and the phase shift that is associated with that response change. It's probably not a conscious realization unless the change isn't as expected.
AllenB -
Related to what you say, there definitely is a reality to "learning" the sound of a room before a new installation that was great in a previous room sounds OK in the new room. But I have no good idea what it is you learn.
And I think it is also related to what I've been calling (elsewhere), "the Toole Critique". Toole in recent years has been saying that EQ is to fix speakers and rooms are forever just what they are (I forget the term used by Immanuel Kant for that... sounds better in German). Once our hearing has a sense of the room, we overlook it, just as we heard Uncle Fred sounding the same in all locations.