"The secret of big speakers"

[oivavoi]

(underline mine). That might just be reason why it was not detectable. Use music and linear-phase loudspeakers (this is mandatory, of course) and it is easily detected in proper ABX blind test (as is a change of absolute polarity, btw). All you need is a few allpass kernels you convolve you test signals/music with and compare that to the original (convolved with a dummy kernel so as to avoid hints from any processing artifacts). It also is a matter of training, I found that skilled listeners are readily able to detect phase shifts that, say, our 4th order crossover @ 120Hz introduces. I did a lot of these experiments both for myself as for the companies I worked for, so I have lots evidence, sadly I can't disclose. And mind you, this isn't Q&D hobby stuff, I'm working on these topics seriously for more than a decade professionally. If you look harder you might find papers and reports of experiments that give more insight than what AES etc have to offer.

I'd be happy to upload a few allpass-processed audio samples if there is any interest, and normally even with standard non-linearphase loudspeaker one can detect the differences but of course lin-phase speakers are best. The other option is you measure the phase response of your multiway, preferably 3-way speakers (or obtain the info otherwise) and use a tool like Rephase to create a correction kernel so you can test yourself.
Personally I have little interest to engage in endless discussions where everything boils down to "do you have any peer-reviewed papers that prove your point?" "there is no evidence in the literature" etc when it is so easy to get to your own personal evidence as it is in this case. Remember, this isn't useless snake oil discussions like cable directionality and such. This is real, rewarding experiments where really something is happening (well, some people are actually "phase-deaf" even after a lot of training).

Interesting points!

Concerning peer review and stuff: I ageee that it’s difficult to just take something on the authority of somebody saying it. At the same time, I do think people often forget that the “applied psychoacoustic loudspeaker research” is extremely small as a research field, and that many of the findings probably are inconclusive. Just an example: at my social science department, there were around 20 people doing a PhD last year (including me). Give or take, that’s probably more than the total number of people doing PHDs last year on applied psychoacoustic loudspeaker research in the whole world.

I do think, therefore, that a lot of the knowledge in this field is (unfortunately) hidden in the industry, where experiments are sometimes done with commercialization in mind. Still, the more evidence that can be presented, the better it is.

 

[andreasmaaan]

I agree with you @oivavoi, but that shouldn’t lower the standard of evidence for new claims that that seem to contradict the existing evidence.

Published studies in this particular area have been quite consistent to date and don’t appear to have had obvious flaws. That doesn’t mean they’re right, but it does mean that their findings are persuasive in the absence of contradictory evidence.

If anyone who has read the benchmark studies on this have an opinion as to how they’re flawed, this too would be interesting.

 

[oivavoi]

I agree with you @oivavoi, but that shouldn’t lower the standard of evidence for new claims that that seem to contradict the existing evidence.

Published studies in this particular area have been quite consistent to date and don’t appear to have had obvious flaws. That doesn’t mean they’re right, but it does mean that their findings are persuasive in the absence of contradictory evidence.

If anyone who has read the benchmark studies on this have an opinion as to how they’re flawed, this too would be interesting.

True. As for me, I'm no expert at all on these issues. I just get naturally cautious when I hear claims about what 'science says'. In my own field of scientific work, where there are boatloads of studies done every year, results are generally not consistent, and few things are written in stone. There are many findings I have believed to be beyond reproach, which have shown themselves to be false. I'm no postmodernist at all, and my own passion for research is to separate truth from falsehoods. But in general I adhere to Popper's principle that it's much easier to show that a claim is false than to show that a claim is true.

On phase etc, it seems intuitively reasonable to me that we will be able to hear phase in the lower frequencies, where wavelengths are long, and not in the higher frequencies, where they are short. Possible shortcomings with the studies that have been done:
- were they done with phase linear transducers or speakers? It may be that we are able to identify phase correctness when we hear it, but that it can be difficult to hear impairments from a baseline that is already impaired
- program material: were the studies done with purist recordings where orignal soundfields were preserved, or with the artificial soup of sound that goes for studio recordings these days?
- mono vs stereo: one of the possible roles phase may play is about creating good imaging. It's possible that messed-up phase is more noticeable in stereo with regards to imaging, than in mono.
- EDIT: and an important caveat with all tests that are based on listeners reporting their experiences is that we may actually sense a lot of things which affects us, but doesn't make their way to our "conscious display"

But for all I know, these points may not matter, may have been answered already, etc. I have absolutely no stakes in this. I'm just interested in the issues. FWIW, the monitors I'm listening to just now are ported and not based around fir-filters (but supposedly, the 8th order crossover at 2600 hz is as correct phase-wise as the manufacturer could get them). They still sound absolutely awesome to my subjective ears

 

[Wombat]

True. As for me, I'm no expert at all on these issues. I just get naturally cautious when I hear claims about what 'science says'. In my own field of scientific work, where there are boatloads of studies done every year, results are generally not consistent, and few things are written in stone. There are many findings I have believed to be beyond reproach, which have shown themselves to be false. I'm no postmodernist at all, and my own passion for research is to separate truth from falsehoods. But in general I adhere to Popper's principle that it's much easier to show that a claim is false than to show that a claim is true.

On phase etc, it seems intuitively reasonable to me that we will be able to hear phase in the lower frequencies, where wavelengths are long, and not in the higher frequencies, where they are short. Possible shortcomings with the studies that have been done:
- were they done with phase linear transducers or speakers? It may be that we are able to identify phase correctness when we hear it, but that it can be difficult to hear impairments from a baseline that is already impaired
- program material: were the studies done with purist recordings where orignal soundfields were preserved, or with the artificial soup of sound that goes for studio recordings these days?
- mono vs stereo: one of the possible roles phase may play is about creating good imaging. It's possible that messed-up phase is more noticeable in stereo with regards to imaging, than in mono.

But for all I know, these points may not matter, may have been answered already, etc. I have absolutely no stakes in this. I'm just interested in the issues. FWIW, the monitors I'm listening to just now are ported and not based around fir-filters (but supposedly, the 8th order crossover at 2600 hz is as correct phase-wise as the manufacturer could get them). They still sound absolutely awesome to my subjective ears

Are you mainly active in physical sciences?

 

[oivavoi]

Are you mainly active in physical sciences?

Nope, social science all the way. But definitely in the non-postmodernist camp of social science. In general, I think results in the physical sciences can be much more objective and secure than results in the social sciences. Most of what goes for psychoacoustics is social science.

 

[Wombat]

Nope, social science all the way. But definitely in the non-postmodernist camp of social science. In general, I think results in the physical sciences can be much more objective and secure than results in the social sciences. Most of what goes for psychoacoustics is social science.

What we generally think of it may be social-science but the mechanics/confirmation of it are biological-science and neurological-science IMHO. The latter two come first and they are early in their development.

 

[andreasmaaan]

Nope, social science all the way. But definitely in the non-postmodernist camp of social science. In general, I think results in the physical sciences can be much more objective and secure than results in the social sciences. Most of what goes for psychoacoustics is social science.

I would have thought that psychoacoustics is the study of the sensory aspect of the hearing system, and is therefore a branch of psychophysics. This distinguishes it from the psychology of listening, which is very much a social science.

 

[oivavoi]

What we generally think of it may be social-science but the mechanics/confirmation of it are biological-science and neurological-science IMHO. The latter two come first and they are early in their development.

It also depends on the measurements used, and the outcome of interest. Asking listening panels to report on what they "like", and crunching the data statistically afterwards, is most definitely social science IMHO. Trying to find out whether something is audible or not, particularly if measuring it directly/physically through eye tracking or brain scans, becomes physical science.

 

[andreasmaaan]

It also depends on the measurements used, and the outcome of interest. Asking listening panels to report on what they "like", and crunching the data statistically afterwards, is most definitely social science IMHO. Trying to find out whether something is audible or not, particularly if measuring it directly/physically through eye tracking or brain scans, becomes physical science.

Totally agree about the preference studies. This phase audibility topic and other audibility related areas are what I had in mind with my post. Thinking about it now following what you just said, our approach here is actually quite old-fashioned (dependant on subjects' reports) compared to some other psychophysics areas (not that this prevents it being a physical science).

 

[FrantzM]

I am not sure that the condescension applies. Science is not perfect and Social Science as in the studies of things that relates toa society is/are Science. Applying statistics to people perception is Science and it works.
We have moved O.T. ..
I continue to contend that no small speakers I have yet heard have swayed into thinking i was am listening to the massive scale and stage thrown by a Genesis 1.2 .. IOW there is something going on with the realy big speakers ... I need to hear some of the best "small " speakers out there namely the Kii and especially the Dutch and Dutch 8c. these may change my opinion. I am not sure the Kii is right there since kii themselves came up with the BXT module ...

 

[oivavoi]

Totally agree about the preference studies. This phase audibility topic and other audibility related areas are what I had in mind with my post. Thinking about it now following what you just said, our approach here is actually quite old-fashioned (dependant on subjects' reports) compared to some other psychophysics areas (not that this prevents it being a physical science).

Very true. I just edited my post above and added this:
"and an important caveat with all tests that are based on listeners reporting their experiences is that we may actually sense a lot of things which affects us, but doesn't make their way to our 'conscious display'"

Wikipedia has a nice entry on it: https://en.wikipedia.org/wiki/Pre-attentive_processing

I hope to see more psychoacoustic studies employing brain scans, eye tracking and direct physical measurements!

But yeah, this is venturing off-topic.

 

[KSTR]

I did a web search on the audibility of phase distortions (so called 'excess phase response' and how designers deal with it) and the consensus seems to be:
- Effect of phase distortion is audible but quite an individual property, some are less sensitive to it than others,
- It seldom is a reall issue for the perceived audio quality. A bad speaker doesn't improve much with phase linearization while an excellent one might improve a bit but not "worlds apart".

In my personal experience, a linear phase speaker both sounds more "compact" in the time domain together with a more original timbre of instruments and with improved transparency of the "soundstage" ... and less listening fatigue long term overall. All off which are no major concerns for general reproduction quality but can be the 'icing on the cake'.

 

[Thomas savage]

If we can make a linear phase speaker , what’s the advantage in not doing so?

We already split hairs with our DAC reviews/requirements , wanting best theoretical performance so why should this be different in this instance ...

Make it the best possible and worry about if you can hear it ,,, later or ,,never.

 

[svart-hvitt]

If we can make a linear phase speaker , what’s the advantage in not doing so?

We already split hairs with our DAC reviews/requirements , wanting best theoretical performance so why should this be different in this instance ...

Make it the best possible and worry about if you can hear it ,,, later or ,,never.

Cost?

 

[Blumlein 88]

Very true. I just edited my post above and added this:
"and an important caveat with all tests that are based on listeners reporting their experiences is that we may actually sense a lot of things which affects us, but doesn't make their way to our 'conscious display'"

Wikipedia has a nice entry on it: https://en.wikipedia.org/wiki/Pre-attentive_processing

I hope to see more psychoacoustic studies employing brain scans, eye tracking and direct physical measurements!

But yeah, this is venturing off-topic.

Do you know it has been observed in some blind listening tests that participants believe they are not able to perceive differences and are guessing. Yet they are perceiving differences and their "guesses" are correct.

 

[Thomas savage]

Cost?

When has cost been a factor in hifi ..

There’s far more expensive examples that don’t adhere.

 

[Thomas savage]

Do you know it has been observed in some blind listening tests that participants believe they are not able to perceive differences and are guessing. Yet they are perceiving differences and their "guesses" are correct.

The rabbit hole , we should put a charge on the door lol

 

[maverickronin]

When has cost been a factor in hifi ..

There’s far more expensive examples that don’t adhere.

IME the cost is no object stuff tends more towards the cable riser types while the more objectively minded seem to be more interested in value for money.

 

[Juhazi]

I just had to register to this forum after reading through this thread! Nice to see familiar nicks here too.

I support KSTRs statements, phase rotations will harm the timbre of instruments! Most people seem to think only of sine wave and it's distortion products. But music instruments are more complex, often the fundamental note is weaker than first harmonic etc. and sound includes also nonharmonic components. If you listen to only modern pop/rock made in studios in several sessions and sound files gone through several cadgets, you don't learn to recognize natural timbre of upright bass, violin, piano etc.! Of course every instrument has it's own timbre and mic techiques are different too. I go to live concerts quite often and hear piano, violin, flute, guitar and saxophone at home too, played by my kids. I love to listen to 1940-60's classical, jazz and pop recordings from FM broadcasts or records/CDs, which use simple studio techniques and are often monophonic.

I have been playing with minidsp 4x10HD and 2x4HD for many years now, with 4-way, 3way and 2-way plus sub systems. Also tried to accommodate to passive LR4 xo but switching to LR2 makes all my speakers to sound more natural and "easy to ears". I have eq'd each way of multiways to be similar, but still I and my friends notice easily when XO type is switched - and LR2 is preferred! Same experience with one passive 2-way diy speaker.

After these experiences I have started to understand fullrange entusiasts better. But FR or first order xo have too many bad features and nonlinearities to stand in my living room as no1 speakers. I haven't tried FIR correction of phase or even heard such, but I believe that it is for good when done right - no room correction, only to direct sound.

ps. I hate bass reflex and other ported or back-loaded bass systems.

 

[KSTR]

For those of us that use some sort of computer based playback implemetation cost is very little if not zero. Most commercially available DRC ("digital room correction") software allows for phase linearization anyway so while you at it, that is linearising frequency response at the listening position you might (and will, typically) correct phase distortion as well.

All you actually need for a most basic linearization is the time-inverted excess phase response of your speakers in form of an impulse response that can be loaded into a convolver plugin, and for most players (Foobar, JRiver, etc) such plugins are available. You can either measure this excess phase (the phase coming from the overall allpass behavior of the crossover, the acoustic output being relevant here, or obtain it analytically if the XO freqs and characteristics are known. The result is that you have a speaker with the same phase response a single wideband driver with the identical frequency response would have.
On top of that, one can add a bit of phase de-rotation at the lower cutoff where the phase is determind by the natural phase of the system acoustic highpass. So if you know that your speaker is 6th order Butterworth @50Hz (ported small speaker with protective electric highpass) you can change phase response to that of a 2nd-oder system, undoing the effect of the remaining four orders. This is to be used careful as too much of it can introduce "pre-ringing" type of effects. If you play back a step function (1Hz squarewave) the slopes of the square will have leading exponential edges which sounds like a rising level of noise before the main "hump" sound. This is an excercise in compromise because "speeding up" the onset of a LF transient (say, a short 50Hz wavelet) so that it is "in line" with the rest also makes the transient a bit wider, sometimes "smeared" again.

To make life easy it would be very nice if manufacturers started to offer measured (or otherwise derived) excess phase correction data (IR) that can be fed directly into convolvers. Or offer some sort of plugin with "hardcoded" corrections for their speakers if they want the whole thing being self-contained and ready to use. This requires a changed mindset, though. Phase errors are nothing to be ashamed of so no need to hide the exact data needed for a digital correction to help those who think they want to try and use this sort of fine-tuning, even without doing full-blown DRC. Digital precorrection of phase can have close to zero negative side-effects and close to zero cost. Since most speakers are still passive multiway with 99% of them being non-linear phase many many people could try and judge the benefits with no expert knowledge required other than using a simple plugin in the digital playback chain... now, if we just had that IR database...