Is transient response the most important thing for the perceived audio quality in a system ?

[dlaloum]

As much as this "looks bad" there really is no evidence (and there has been a fair amount of research) that this "sounds bad". Intuition can be useful but not always and I think phase and square waves are examples of things where it makes intuitive sense that good square wave reproduction and flat phase should sound better but there is little evidence this is the case and a lot of evidence that it doesn't make any audible difference. This is backed up by my experience with REW and Rephase to flatten phase.... I can measure and see the "success" of the phase correction but if I ABX with and without the phase correction I can not tell a difference. I think this is valuable information because it prevents me from wasting time and money on things that don't matter.

For years, I could walk into a listening room and tell whether it was a panel or a box playing - sight unseen.

Was it identifying box resonances? Or was it phase - which panels like the Quad ESL63 to current do so well? (or both?)

Whichever it is, "boxes" have seldom sounded "real" - although the best have sounded very good indeed.... souding good, and sounding "real" are two different things. I do wonder whether that ability to reproduce a decent square wave, even though they lacked both bottom and top extension, was the magic of electrostatic speakers?

 

[dlaloum]

It doesn’t exist, for me, because there is no formal proof. I can’t say if it exists for you.

Psycho Acoustics is still a nascent field.

There is far too little work being done in this area to identify what it is exactly that triggers/affects our preception.

Most speaker manufacturers are only marginally interested - it doesn't generate sales.

Even the Harman efforts, were focused not on how we perceive, but on consumer preference - ie: what it is that will sell more speakers

So there is nothing there that can be discounted - the work simply hasn't been done.

In science, it is never a matter of "proof" rather there is a hypothesis which science tries valiently to disprove... the more "substantial" tests that fail, the more likely it is that the hypothesis is true... sadly providing a "proof" is limited to very abstract mathematics, whereas in the more applied/physical world of acoustics, and perception.... there really is no proof.... that is just a pseudo-scientific myth.

 

[Galliardist]

Even the Harman efforts, were focused not on how we perceive, but on consumer preference - ie: what it is that will sell more speakers

The Harman efforts used preference as a shortcut to understanding, rather than just how to sell speakers. The idea that Toole and Olive were only speaker sellers is wrong.

The work you want, will likely be derived from other work into how we hear more generally, and on linking hearing into the wider concept of virtual reality, most likely. And it will take a long time.

However, work is ongoing into how we hear music. Recent stories in New Scientist (I don't have them to hand) reported that songs trigger the speech centre in the brain (surprise), and that listening to Bach triggered the parts of the brain used in pattern matching, if I remember correctly. There is evidence (again, I'm trying to recall what was said, I don't have references) that playing the violin in the way used by modern soloists with lots of vibrato can also stimulate the speech centre. And music is known to stimulate parts of the brain involved in movement (why we dance, tap our feet, etc). Simply showing that these things happen is progress, and tells us that a lot of the brain is used in music.
What we don't know is what you point to: we don't know how an audiophile judges sound quality. It seems likely to me that the brain has a problem with that activity, because for the biases idea to work that is referred to often on this forum, it is drawing on things other than just experience of hearing to make judgements.

The Harman work is the best we have, but it investigates hearing through controlled testing - which is not what we do in the prevalent mode of judging uncontrolled and sighted.

For my part, I believe there are two lines of research that would open up the field. These may be being done already, I've not looked. First is understanding what parts of the brain are active when we are trying to determine the quality of sounds. That I feel is important for virtual reality research, because making our hearing more satisfied is going to be part of making VR usable in long stints.

The second is actually understanding the electrical signal passed by the ear to the brain. A lot of what we know about sight comes from the fact that we know that the brain is doing things to the output of the eye (switching to right way up, masking the hole in our vision where the optic nerve joins the retina, etc) but I have never seen any similar description of the output of the ear. I expect that the information does exist somewhere, because of cochlear research and implants. I must have a look for it!

 

[levimax]

Was it identifying box resonances? Or was it phase - which panels like the Quad ESL63 to current do so well? (or both?)

There are so many differences between box speakers and panel speakers that it is hard to say what you were hearing. The reason I don't believe phase is important is because when I ABX the same speaker with and without phase correction I can not reliably tell them apart. I am not saying there is no difference but if there is it is extremely subtle where as the differece between panels and boxes is often much more apparent.

 

[ernestcarl]

It doesn’t exist, for me, because there is no formal proof. I can’t say if it exists for you.

No reason for you or anyone to have to believe my own anecdotal listening results. If you’ve done your own tests and weren’t able to hear any substantial difference, then good for you. One less thing to bother worrying about.

It is the potential for lobing or magnitude cancellation and comb filtering between none phase compliant LCR mains that’s more of a concern to me than phase linearization itself. The fact that I even hear a difference between my FIR DSP presets is only a small bonus. As long as the passive/active crossovers or all pass system is well designed… then there’s no reason why it should not sound good.

 

[gnarly]

I have not seen any experimments/ABX that show audibility of phase shift of eg LR4 @ 3 kHz. Can you link if so?

I'd be very surprised to see a valid study that gave an audibility difference with music for that specific example.
Maybe/probably heard with pink noise, but a single LR4 @ 3kHz isn't enough phase rotation to even bother with, ime.

The real benefit of flat phase ime, is down low.
Like all things in audio, the harder it is to achieve....like very low bass extension, high SPL that stays linear, etc etc.....the greater the potential gains.
Flat phase down low, aka greatly reduced group delay, gives music a sense of greater rhythm, and cleaner transients.
It kinda makes bass sound like it's less...until I measure and see it's all still there..only cleaner sounding.
Let's me turn up the bass a little to match tonality i was accustomed to, for a very pleasing tonality with newfound clarity.
like always...... ymmv

 

[leManu]

Here's a page regrouping a series of article by Soulnote Chief Designer. What he says is that transient is everything and static performance doesn't really matter.
I'm sure almost everyone here will disagree, but interesting read nonetheless.

DESIGN PHILOSOPHY

www.kcsr.co.jp

 

[levimax]

No reason for you or anyone to have to believe my own anecdotal listening results. If you’ve done your own tests and weren’t able to hear any substantial difference, then good for you. One less thing to bother worrying about.

It is the potential for lobing or magnitude cancellation and comb filtering between none phase compliant LCR mains that’s more of a concern to me than phase linearization itself. The fact that I even hear a difference between my FIR DSP presets is only a small bonus. As long as the passive/active crossovers or all pass system is well designed… then there’s no reason why it should not sound good.

From my limited experience I have not found that chasing around "flat" phase is worth the effort but your "well designed" crossover comment got me thinking. My system is a DIY 4 way (3 way plus sub) which uses all analog active LR 4 crossovers and DSP Fir Filters for FR adjustment and crossover linearization (basically all pass filters to reduce group delay and flatten phase). Since I don't have individual DSP on each driver and an active LR 4 crossover is inherently "phase coherent" (360 degrees out of phase) I actually don't think it is possible for me to break the "phase coherence" of the LR 4 crossovers with DSP and maybe that is why I don't hear any difference with "flat phase". I am not sure if research has looked at phase coherence vs group delay but maybe it is coherence that is more important i.e 360 degrees, 720 degrees, etc. rather than the total amount of delay and that is one of the reasons why LR 4 has proven to be a good crossover design. If anyone knows the research or can link to an article I would be interested to learn more.

 

[ernestcarl]

Since I don't have individual DSP on each driver and an active LR 4 crossover is inherently "phase coherent" (360 degrees out of phase) I actually don't think it is possible for me to break the "phase coherence" of the LR 4 crossovers with DSP and maybe that is why I don't hear any difference with "flat phase".

I would think the coherence stays the same… but, that’s probably not the main reason.

Test design matters here. I mentioned headphones or speakers “nearfield” in a “very dry room” and using foobar2000 ABX plugin where one can repeat particular sections of test tracks over and over ad infinitum… the telltale signs are very/extremely subtle for mid-HF xo, but once you’ve latched to an audible difference it becomes repeatable from there on out. Frankly, it’s tedious and no fun at all. And just to be clear, I haven’t tested this for anything higher than 2kHz.

The lower you go, though, the more audible it is… like stacking all pass filters in the bass region to phase match none phase compliant speakers, instead just of using linearizing (mixed) phase FIR EQ to avoid increasing overall processing time delay.

 

[Thomas_A]

I'd be very surprised to see a valid study that gave an audibility difference with music for that specific example.
Maybe/probably heard with pink noise, but a single LR4 @ 3kHz isn't enough phase rotation to even bother with, ime.

The real benefit of flat phase ime, is down low.
Like all things in audio, the harder it is to achieve....like very low bass extension, high SPL that stays linear, etc etc.....the greater the potential gains.
Flat phase down low, aka greatly reduced group delay, gives music a sense of greater rhythm, and cleaner transients.
It kinda makes bass sound like it's less...until I measure and see it's all still there..only cleaner sounding.
Let's me turn up the bass a little to match tonality i was accustomed to, for a very pleasing tonality with newfound clarity.
like always...... ymmv

What I know may be audible is polarity switch or phase shifts in the bass region. But not transientwise but for asymmetric waveforms you get a timbre shift.

 

[gnarly]

What I know may be audible is polarity switch or phase shifts in the bass region. But not transientwise but for asymmetric waveforms you get a timbre shift.

I'm not sure what you're saying, but i think we may be on the same page.

Sure, a polarity change (of say just a sub), will cause misalignment of asymmetric waveforms with the rest of drivers participating in the waveform.
I think that effects both timbre and transients. Both of those rely of relative phase between all frequencies (at least mathematically, whether or not our speakers are resolving enough to make it audible.
Personally, i think out ears are more resolving than past studies have concluded ...we haven't had flat phase speakers readily available until relatively recently.

Anyway, Flat phase takes correct polarity (and overlapping phase traces through the xover region) a step further.

 

[Thomas_A]

I'm not sure what you're saying, but i think we may be on the same page.

Sure, a polarity change (of say just a sub), will cause misalignment of asymmetric waveforms with the rest of drivers participating in the waveform.
I think that effects both timbre and transients. Both of those rely of relative phase between all frequencies (at least mathematically, whether or not our speakers are resolving enough to make it audible.
Personally, i think out ears are more resolving than past studies have concluded ...we haven't had flat phase speakers readily available until relatively recently.

Anyway, Flat phase takes correct polarity (and overlapping phase traces through the xover region) a step further.

I'm not sure what you're saying, but i think we may be on the same page.

Sure, a polarity change (of say just a sub), will cause misalignment of asymmetric waveforms with the rest of drivers participating in the waveform.
I think that effects both timbre and transients. Both of those rely of relative phase between all frequencies (at least mathematically, whether or not our speakers are resolving enough to make it audible.
Personally, i think out ears are more resolving than past studies have concluded ...we haven't had flat phase speakers readily available until relatively recently.

Anyway, Flat phase takes correct polarity (and overlapping phase traces through the xover region) a step further.

I am specifically talking about this:

Absolute Polarity - Myth or "Important"?

I'm really confident it's a matter of training and preconditioning. The polarity ABX test is my go-to warm-up test when doing any kind of analytical listening and testing. On some days it just doesn't work... Number #1 rule of ABX'ing: The moment you feel any, even subtle kind of stress or...

www.audiosciencereview.com

 

[gnarly]

I am specifically talking about this:

Absolute Polarity - Myth or "Important"?

I'm really confident it's a matter of training and preconditioning. The polarity ABX test is my go-to warm-up test when doing any kind of analytical listening and testing. On some days it just doesn't work... Number #1 rule of ABX'ing: The moment you feel any, even subtle kind of stress or...

www.audiosciencereview.com

Oh, isn't that thread about absolute polarity of the entire speaker?
Haven't waded through it, because seems low on the list to me.

Much high, and what i was referring to, is relative polarity of bass (sub) to rest of the speaker...
Just like phase....speaker relative to itself...

 

[audiofooled]

”Why is transient response important? Well, music is full of transients. If your audio system can’t reproduce them accurately, the music will sound muffled, lifeless, or distorted. On the other hand, a system with good transient response can make the music sound dynamic, detailed, and vibrant.”

1. No, not very important - explain why

 

[Thomas_A]

Oh, isn't that thread about absolute polarity of the entire speaker?
Haven't waded through it, because seems low on the list to me.

Much high, and what i was referring to, is relative polarity of bass (sub) to rest of the speaker...
Just like phase....speaker relative to itself...

Yes it is all polarity. But the issue is when phase shifts happen in the bass.

 

[audiofooled]

I'd be very surprised to see a valid study that gave an audibility difference with music for that specific example.
Maybe/probably heard with pink noise, but a single LR4 @ 3kHz isn't enough phase rotation to even bother with, ime.

The real benefit of flat phase ime, is down low.
Like all things in audio, the harder it is to achieve....like very low bass extension, high SPL that stays linear, etc etc.....the greater the potential gains.
Flat phase down low, aka greatly reduced group delay, gives music a sense of greater rhythm, and cleaner transients.
It kinda makes bass sound like it's less...until I measure and see it's all still there..only cleaner sounding.
Let's me turn up the bass a little to match tonality i was accustomed to, for a very pleasing tonality with newfound clarity.
like always...... ymmv

I agree.

 

[Tangband]

Is anything explained in detail yet about how perceptions are formed in the brain?

There are some books one can read with some facts, but noone has the proven explaination how its done.

1. No, not very important - explain why

Hmm…This video is about compression and loudness-war. He is right about that loudness-war kills the recorded playback ( also transients ) , but I dont listen much to a guy who has Yamaha ns10 as monitors….

 

[audiofooled]

There are some books one can read with some facts, but noone has the proven explaination how its done.


Hmm…This video is about compression and loudness-war. He is right about that loudness-war kills the recorded playback ( also transients ) , but I dont listen much to a guy who has Yamaha ns10 as monitors….

 

[IPunchCholla]

Psycho Acoustics is still a nascent field.

There is far too little work being done in this area to identify what it is exactly that triggers/affects our preception.

Most speaker manufacturers are only marginally interested - it doesn't generate sales.

Even the Harman efforts, were focused not on how we perceive, but on consumer preference - ie: what it is that will sell more speakers

So there is nothing there that can be discounted - the work simply hasn't been done.

In science, it is never a matter of "proof" rather there is a hypothesis which science tries valiently to disprove... the more "substantial" tests that fail, the more likely it is that the hypothesis is true... sadly providing a "proof" is limited to very abstract mathematics, whereas in the more applied/physical world of acoustics, and perception.... there really is no proof.... that is just a pseudo-scientific myth.

I don’t believe it as nascent as many here would like to believe, it is just that there has not been a compilation/distillation of the research for many aspects as Toole did for the goal of audio reproduction. I work at a university and have access to pretty much all publicly published research (Though oddly only 12 universities in the OCLC system stock AES, and only one in the US). Whenever I have gone looking for a particular topic, say the effect of training on being able to hear distortion, the issue hasn’t been that there is no research, even when I limit to full text sources online. The problem is that there is so much research that even just skimming the abstracts takes more time than I am willing to dedicate. I have this giant file that just keeps growing of research in pyschoacoustics.

So much of what we talk about here could probably be discounted. It just isn’t since it would require a Toole or Diamond to condense and distill.

I mean, apparently Neurolink just got permission to start human trials of brain implanted chips. They have been able to use fMRI to vocalize speech that subjects just thought about saying. They’ve even been able to ”see” letters that subjects visualize in their head, measure the time it takes for someone to recognize something and use it to tell if they have seen that thing before. So much is understood, it just isn’t getting communicated in a way that is relevant to this forum. But our ignorance here does not reflect the overall state of the knowledge.

And, yes. Science opporates by falsifiable hypothesis. “I heard a thing one time so you should try and disprove it.” does not fall into that realm.

 

[Cbdb2]

There are some books one can read with some facts, but noone has the proven explaination how its done.


Hmm…This video is about compression and loudness-war. He is right about that loudness-war kills the recorded playback ( also transients ) , but I dont listen much to a guy who has Yamaha ns10 as monitors….

Talk about Dunning-Kruger.

David Mellor worked as theatre sound engineer at London's Royal Opera House, with activities including sound design, front-of-house operation, stage monitoring and electronic design satisfying the likes of Luciano Pavarotti, Placido Domingo, Dame Kiri Te Kanawa and Karlheinz Stockhausen.

And I guess you think Abbey Roads is a lousy studio.

https://dt7v1i9vyp3mf.cloudfront.net/styles/news_large/s3/imagelibrary/S/SF_Nov_08C-B1QDKn4dpcyg1HpXLlsEQdWpwzbL606p.jpg