Importance of impulse response

[fineMen]

Yeah? What happens when real instruments, their recordings playing through a speaker, start sounding more real?

Stopped reading. Why? Because I told you, a recording is not abour 'real'. I've got reference from so many artists in the recording business. I know them, I talk to them at the fire place at home, we love each other, go figure!

As I said, and I mean it, who is talking about the infamous glorious "real" is either praising a rarely successful recording artwork in its own right, or is deliberately lying about stereo equipment. Most of the time its the latter.

The naive audiophile is thinking of ... nothing actually ... but we already know that only ... leave it alone. How far can you separate from reality? And you still do and over again, hoping for the best equipment, instead of questioning the process from he start up, the recording. Sanitize your expectations!!

 

[dualazmak]

As the me I am wrote afterwards, please do not fall for people who infect you with an everlasting doubt regarding the capabilities of your stereo. Group delay, or whatever they name it, is actually of NO concern. (Except the infamous real big Klipsch horn.)

It is a marketing hoax and nothing else. I bet my life for it. (I trust technology when I step into a commercial jet. The typical audiophile is the infamous 'nervous flyer' who takes your hand on a cross-wind landing. I won't marry her ... . These anxious, even leaking contemporaries just don't understand anything.)

I'm writing this whilst listening to F/ Zappa's 'We're Only In It For The Money'. As time coherence is requested. "Psychedelic Dungeons", you name it ...

Have you carefully read my post #327 and also the linked posts?

You would please calm down, and take your enough time reading through those...

 

[dualazmak]

Hello @fineMen,

Also this post sharing nice discussion between @gnarly and myself would be of your interests and reference, I assume.

 

[thewas]

https://www.focal.com/sites/www.focal.fr/files/shared/catalog/document/notice_scalautopiaevo_web.pdf

https://wilsonaudio.com/media/858/Sasha-DAW-Manual.pdf

Some has. I never take notice for all the model.

No, those connection terminals adjust the level but not the phase.

 

[kongwee]

No, those connection terminals adjust the level but not the phase.

Not just level, electrical phase changes too, -1.5db, +1.5db slope. There are phase shifting, voltage and current not in sync. Some designer do take note about that and place the driver relative to each other. and you do not have the option to meddle the resistance. Of course the jumper is used to due to tonal control like that studio monitor does.

 

[thewas]

Not just level, electrical phase changes too, -1.5db, +1.5db slope. There are phase shifting, voltage and current not in sync. Some designer do take note about that and place the driver relative to each other. and you do not have the option to meddle the resistance. Of course the jumper is used to due to tonal control like that studio monitor does.

As said those electrical connections mainly change the level and are not used to align the phase, if you read the above manuals they writes to adjust the level and nothing about phase.

 

[kongwee]

As said those electrical connections mainly change the level and are not used to align the phase, if you read the above manuals they writes to adjust the level and nothing about phase.

Electrically these is phase shift. Crossover is using this relationship to cut the frequency off beside level. Of course you will not heard the phase shift and not suppose to heard it if done it correctly. Add or subtract the resistance will the phase angle change that is science. Unless you do in DSP form, you can set zero phase that impossible in LCR circuit.

 

[kimmosto]

No, reiterated no. To trust a blatant claim is not scientific.

The detail makes it.

What do you mean? That two signals with exactly (bit-perfect) equal magnitude response but different phase response and group delay should not be generated by time-reversing impulse response, or comparing original or all-passed signal to it's time-reversed version is not proper science, or something else? Proper answer in clear English please. Not any extended fuzzy audiophool BS.

Those guys were studying is there audible difference between two impulse-like signals with exactly the same magnitude response, but different phase response i.e. group delay. That is simple audibility test, and time-reversing is bulletproof method for generating (especially short) test signals without any difference in magnitude response. Nothing more - nothing less. Minimum-phase "original" was also included in the study so guys were using also test signals which were not time-reversed.

I'm more interested in comparison of minimum phase version with very low normal GD at LF and version with excess group delay especially with transient signals and loudspeakers without significant dynamic problems which could hide effect of group delay. Time-reversing method could provide some help also with that, but test signals should be generated so that one version produces output equal to speaker with minimum phase response. Otherwise advantage (minimal pressure loss with transients) of close to ideal timing cannot be detected.

 

[thewas]

Electrically these is phase shift. Crossover is using this relationship to cut the frequency off beside level. Of course you will not heard the phase shift and not suppose to heard it if done it correctly. Add or subtract the resistance will the phase angle change that is science. Unless you do in DSP form, you can set zero phase that impossible in LCR circuit.

Once again, the phase angle change when changing the series resistor value to those typical +-1 or 2 dB is by far not significant enough to be used for phase alignment and thus is used just to change the tonality, also it wouldn't make sense as you cannot change tonality and phase independently.

Here I have just run a simulation such to show you how insignificant it is with such a typical tweeter level change

and the corresponding phase plot of the tweeter

 

[kongwee]

Once again, the phase angle change when changing the series resistor value to those typical +-1 or 2 dB is by far not significant enough to be used for phase alignment and thus is used just to change the tonality, also it wouldn't make sense as you cannot change tonality and phase independently.

Here I have just run a simulation such to show you how insignificant it is with such a typical tweeter level change

View attachment 248079


and the corresponding phase plot of the tweeter

View attachment 248080

2000-1000hz, is significant . 1db different. 50hz shift. My Yamaha MSP7 monitor has +1.5db,-1.5db switches. I do switch them. This is real life, not graphs.

 

[thewas]

2000-1000hz, is significant . 1db different. 50hz shift.

Wrong interpretation again, the 50 Hz is just an artefact of the +-180° wrapping of the plot, what is important and you were talking about is the phase change and that is less than 5°. And of course it is 1dB different as I chose the resistor value accordingly.

My Yamaha MSP7 monitor has +1.5db,-1.5db switches. I do switch them. This is real life, not graphs.

And you mainly change the tonality, do some measurements and you will see that the simulations work for such perfectly.

 

[fineMen]

Not any extended fuzzy audiophool BS.

Those guys were studying is there audible difference between two impulse-like signals with exactly the same magnitude response, but different phase response i.e. group delay. That is simple audibility test, and time-reversing is bulletproof method for generating (especially short) test signals without any difference in magnitude response. Nothing more - nothing less. Minimum-phase "original" was also included in the study so guys were using also test signals which were not time-reversed.

Should I actually respond, how?

The 'white paper' tells that the few people in the test panel had to decide, if there was a difference between two signals:
x) an original monophonic testsignal with especially a bit more low frequency content, played over headphones simultaneously into both ears
y) the same testsignal, but ... (please explain in detail)

What does the 'time-reversal' do?

 

[Wesayso]

So glad we get to make up our own mind on this subject. No matter what "the others" say, judge for yourself. It's pretty obvious when you think about it. We expect the amplifier to respect the input and only add gain. We expect no less from our DAC too. Yet when it comes to speakers, anything goes...
The 'Audiophools' as @fineMen puts it, or 'Golden Ears' could never be bothered by things like the phase output of a speaker, they're too busy with the 8 watt set amp to care about stuff like that. Introducing FIR filters would mean adding a 'digital device', heck no! that can't be added to their sacred all analog setup! Sacrilege!

So it's only those that are interested in how this stuff really works scientifically, that give a damn and want to know what it does for perception.
A good start in finding out why it "could" be an advancement would be to start looking at some research on perception, Mr. Toole's colleague David Griesinger has a lot to say about that. Start with an easy to follow video and let it sink in, and ask yourself some questions: why do we want to let the music achieve our ears mostly in tact? Does it have something to do with our survival instincts? How can we get there? What did Griesinger use in this video to demonstrate and why?

 

[Wesayso]

Should I actually respond, how?

You don't have to, you're not adding anything useful. You're just in denial.

 

[fineMen]

The 'Audiophools' as @fineMen puts it, or 'Golden Ears' could never be bothered by things like the phase output of a speaker, they're too busy with the 8 watt set amp to care about stuff like that. Introducing FIR filters would mean adding a 'digital device', heck no! that can't be added to their sacred all analog setup! Sacrilege!

So it's only those that are interested in how this stuff really works scientifically, that give a damn and want to know what it does for perception.
...

Still my question isn't answered, namely what the 'time-reversal' does in the above mentioned 'white paper'. Thanks for the link anyway. It reiterates the audiophiles' perspective:

Listening is an automatic process, that is performed by "the brain". And it signals "danger".

I would say, listening to a stereo is an active process (because of the artificial origin of the recording), performed by the mind (because it is music). It signals pleasure (should do so).

Might be, that this difference makes the misunderstanding.

 

[kongwee]

Wrong interpretation again, the 50 Hz is just an artefact of the +-180° wrapping of the plot, what is important and you were talking about is the phase change and that is less than 5°. And of course it is 1dB different as I chose the resistor value accordingly.

And you mainly change the tonality, do some measurements and you will see that the simulations work for such perfectly.

You just use graph not ears. You don't know 50hz and 1.5db changes in listening. I do a lot DAW work. It is not uncommon to work in that "insufficient" window range. You can even extract the frequency respond and impose on any song wave file, you will hear the difference.

 

[kimmosto]

What does the 'time-reversal' do?

1st sample in test signal "1Wa" is around 300 ms, without time-reversing:

2nd sample in test signal "1Wa" is around 1300 ms, with time-reversing:

Time scale of the second sample is reversed to make difference to phase response and group delay. Amplitude responses are exactly equal. Got it?

 

[Wesayso]

There is a big difference between a system set up for listening pleasure and a system that's meant to provide pleasing background.
If you want the first, read Mr. Griesinger's papers. That video is just a start.

By the way, the Pro industry, including the recording business is quite aware of the perception of phase. It's where I got my information from to be able to do the same.

 

[thewas]

You just use graph not ears. You don't know 50hz and 1.5db changes in listening. I do a lot DAW work. It is not uncommon to work in that "insufficient" window range. You can even extract the frequency respond and impose on any song wave file, you will hear the difference.

It seems you still don't understand what I write, I never claimed that you don't hear 1.5 dB change which is of course easily heard, only that you probably won't hear less than 5° tweeter phase change.

 

[kongwee]

It seems you still don't understand what I write, I never claimed that you don't hear 1.5 dB change which is of course easily heard, only that you probably won't hear less than 5° tweeter phase change.

Because it is nicely crossed. That is vocal range, you can dim the vocal and vocal perceived distance won't change. It good for monitoring at that range.