The importance of latency

I would like to point out something that not everybody is aware of:

As much as sight is processed immediately by our brain, making it our #1 sense, our hearing is not functioning the same way (alas!)

Our ear pavilions are mere receivers. Sound has then to be processed by our brain, and he's not quick at it. Many reasons for this: the intricacy of the structure of the ear first, but then a fact which is very subjective.

As weird as it may "sound", left-handed and ambidextrous people ex aequo are faster at processing the transformed wave into an electrical signal via the auditive nerve.
Right handed people are slower at processing it.

What does it mean?

Well, simply put, we are not equally apt facing latency.
I hope that everyone here knows how much latency is a key factor in our perception of both periodic and non periodic events.

Our perception of sound, globally, is a non-linear process.

When considering any tuning, interval, anything tangible and non purely theoretic, we face the issue of our own human built-in latency.

We often blame external devices for their excesive amount of latency, yet we forget or ignore our own.

Real latency is always depending on our brains.

So, when measuring consonances and hearing beating, we are always fooled by our own brain. The physical event of beating doesn't match how the brain works.

I could go on and on about fMRI results proving my statement, but I won't.

I had a pretty interesting conversation about latency yesterday. I thought I'd share it here.

/JB

Rowdy JB!

You should be aware that the level of psychoacoustic knowledge on this
list is very high, so you should always feel free to speak in-depth
about these things! I'm always down for a deep discussion about
psychoacoustics and would love to hear more about your ideas.

On Fri, Jul 8, 2011 at 11:50 PM, collinet <jbcollinet@...> wrote:
>
> Well, simply put, we are not equally apt facing latency.
> I hope that everyone here knows how much latency is a key factor in our perception of both periodic and non periodic events.

In what sense would you imagine that latency interferes with our
perception of periodic events? I would imagine the biggest impact
would come if group delay is non-uniform along the signal chain (which
it probably is), but how do you predict that'll make an impact?

> Our perception of sound, globally, is a non-linear process.

The addition of latency in and of itself would be a linear process, as
it can be represented by a convolution of the original signal with an
impulse response that represents the latency (perhaps just an impulse
a few milliseconds later). Do you claim that something about the
latency added is nonlinear, or that the auditory system just behaves
nonlinearly in general, outside of the latency issue?

> So, when measuring consonances and hearing beating, we are always fooled by our own brain. The physical event of beating doesn't match how the brain works.

What do you mean by this?

> I could go on and on about fMRI results proving my statement, but I won't.

Feel free to post them, I'd find it an interesting read.

-Mike

Mike,

Weeks ago I sent information to your personal address regarding the derived true value of the musical octave. There, I didn�t mention specifical values since until the conversion of a conventional piano to the new octave and its evaluation are done, it is better not to give out this information. We shall tune the piano to the new octave and according to the results I would send them to you and to the list.

The true musical octave was correctly derived.

The omission of your response is not what I had expected, I tell you this despite the respect and my appraisement for you.

Should you only respond to messages which are addressed to the list, I have explained why I didn�t send it to them. Thanks, Mario.

<<<<<<<<<<<<<<<<<<<<<<<<<<
----- Original Message ----- From: "Mike Battaglia" <battaglia01@...>
To: <tuning@yahoogroups.com>
Sent: Friday, July 08, 2011 11:02 PM
Subject: Re: [tuning] The importance of latency

> Rowdy JB!
>
> You should be aware that the level of psychoacoustic knowledge on this
> list is very high, so you should always feel free to speak in-depth
> about these things! I'm always down for a deep discussion about
> psychoacoustics and would love to hear more about your ideas.
>
> On Fri, Jul 8, 2011 at 11:50 PM, collinet <jbcollinet@...> wrote:
>>
>> Well, simply put, we are not equally apt facing latency.
>> I hope that everyone here knows how much latency is a key factor in our >> perception of both periodic and non periodic events.
>
> In what sense would you imagine that latency interferes with our
> perception of periodic events? I would imagine the biggest impact
> would come if group delay is non-uniform along the signal chain (which
> it probably is), but how do you predict that'll make an impact?
>
>> Our perception of sound, globally, is a non-linear process.
>
> The addition of latency in and of itself would be a linear process, as
> it can be represented by a convolution of the original signal with an
> impulse response that represents the latency (perhaps just an impulse
> a few milliseconds later). Do you claim that something about the
> latency added is nonlinear, or that the auditory system just behaves
> nonlinearly in general, outside of the latency issue?
>
>> So, when measuring consonances and hearing beating, we are always fooled >> by our own brain. The physical event of beating doesn't match how the >> brain works.
>
> What do you mean by this?
>
>> I could go on and on about fMRI results proving my statement, but I >> won't.
>
> Feel free to post them, I'd find it an interesting read.
>
> -Mike
>
>
> ------------------------------------
>
> You can configure your subscription by sending an empty email to one
> of these addresses (from the address at which you receive the list):
> tuning-subscribe@yahoogroups.com - join the tuning group.
> tuning-unsubscribe@yahoogroups.com - leave the group.
> tuning-nomail@yahoogroups.com - turn off mail from the group.
> tuning-digest@yahoogroups.com - set group to send daily digests.
> tuning-normal@yahoogroups.com - set group to send individual emails.
> tuning-help@yahoogroups.com - receive general help information.
> Yahoo! Groups Links
>
>
>
>

--- In tuning@yahoogroups.com, Mike Battaglia <battaglia01@...> wrote:
>
> Rowdy JB!
>
> You should be aware that the level of psychoacoustic knowledge on this
> list is very high, so you should always feel free to speak in-depth
> about these things! I'm always down for a deep discussion about
> psychoacoustics and would love to hear more about your ideas.

Great to know this group is not another kind of phony stuff, but something really serious and passionate!
Now I know how hot it is here. Thermometer test:check!
>
> On Fri, Jul 8, 2011 at 11:50 PM, collinet <jbcollinet@...> wrote:
> >
> > Well, simply put, we are not equally apt facing latency.
> > I hope that everyone here knows how much latency is a key factor in our perception of both periodic and non periodic events.
>
> In what sense would you imagine that latency interferes with our
> perception of periodic events? I would imagine the biggest impact
> would come if group delay is non-uniform along the signal chain (which
> it probably is), but how do you predict that'll make an impact?

I think that latency should be studied more thoroughly. So far, not much attention has been paid to brain processing. I'll volunteer for more EEGs focused on specific aspects.

I didn't mention that I'm deaf in one ear, and that even if there's no brain damage. It has a normal structure for an ambidextrous person.

Latency is most of the time very theoretic, it's a concept. Few people care to go beyond this (I'm glad this group does) and even fewer care to define in clear terms what periodicity and latency are. It's a kind of call to arms actually.

The biggest impact would, in my opinion, be what you're imagining. Now we'll need practical and comprehensive measurement tools.
>
> > Our perception of sound, globally, is a non-linear process.
>
> The addition of latency in and of itself would be a linear process, as
> it can be represented by a convolution of the original signal with an
> impulse response that represents the latency (perhaps just an impulse
> a few milliseconds later). Do you claim that something about the
> latency added is nonlinear, or that the auditory system just behaves
> nonlinearly in general, outside of the latency issue?

I'm not claiming, I only think. To me, the auditory system behaves non-linearly in general, but again, we'll need more EEGs and fMRIs. If you add the latency issue, it may well change the paradigm, also.
>
> > So, when measuring consonances and hearing beating, we are always fooled by our own brain. The physical event of beating doesn't match how the brain works.
>
> What do you mean by this?

The discussion from which stems this topic was a rough draft, I hadn't time to develop this much. I'll be curious to know which are your conceptions of the "soundwave moving at a speed of 343 m/s" (beating) working with "different types of brain wave patterns and their amount in Hz. (brain latency)"
>
> > I could go on and on about fMRI results proving my statement, but I won't.
>
> Feel free to post them, I'd find it an interesting read.
>
Will do! Actually I'm still searching for the most relevant published results on this topic.
What I really would like to clarify is the notion of "global latency" (applicable to human and non-human) and "brain latency". Are they the same thing with different names, or really two distinct types of latency, two "languages of latency"?

I'm just jotting down ideas and questioning. Can we enhance and broaden our understanding of latency (or latencies, if more than an universal phenomenon)?

To me, it's also extremely important to fully understand every aspect of latency when coming to code apps for iPad (or other devices) which have an important native latency that can be reduced only by reducing the buffer size and the weight of the UI.

/JB

> -Mike
>

--- In tuning@yahoogroups.com, "Mario Pizarro" <piagui@...> wrote:
>
> Mike,
>
> Weeks ago I sent information to your personal address regarding the derived
> true value of the musical octave. There, I didn´t mention specifical values
> since until the conversion of a conventional piano to the new octave and its
> evaluation are done, it is better not to give out this information. We shall
> tune the piano to the new octave and according to the results I would send
> them to you and to the list.
>
> The true musical octave was correctly derived.

I'm sort of intrigued by this. By "true value of the musical octave", do you mean some frequency ratio other than the conventional value of exactly two (2)? What makes yours the "true" one?

Keenan

On Sat, Jul 9, 2011 at 12:11 PM, Mario Pizarro <piagui@...> wrote:
>
> Mike,
>
> Weeks ago I sent information to your personal address regarding the derived
> true value of the musical octave. There, I didn´t mention specifical values
> since until the conversion of a conventional piano to the new octave and its
> evaluation are done, it is better not to give out this information. We shall
> tune the piano to the new octave and according to the results I would send
> them to you and to the list.
>
> The true musical octave was correctly derived.
>
> The omission of your response is not what I had expected, I tell you this
> despite the respect and my appraisement for you.
>
> Should you only respond to messages which are addressed to the list, I have
> explained why I didn´t send it to them. Thanks, Mario.

Apologies Mario, I must have missed it. Feel free to resend or to post
here as well.

-Mike

On Sat, Jul 9, 2011 at 2:19 PM, collinet <jbcollinet@...> wrote:
>
> --- In tuning@yahoogroups.com, Mike Battaglia <battaglia01@...> wrote:
> >
> > In what sense would you imagine that latency interferes with our
> > perception of periodic events? I would imagine the biggest impact
> > would come if group delay is non-uniform along the signal chain (which
> > it probably is), but how do you predict that'll make an impact?
>
> I think that latency should be studied more thoroughly. So far, not much attention has been paid to brain processing. I'll volunteer for more EEGs focused on specific aspects.

We've been pretty obsessed with brain processing over here, but I
haven't seen as of yet a specific role for something like latency to
play in fundamental frequency estimation.

> I didn't mention that I'm deaf in one ear, and that even if there's no brain damage. It has a normal structure for an ambidextrous person.
>
> Latency is most of the time very theoretic, it's a concept. Few people care to go beyond this (I'm glad this group does) and even fewer care to define in clear terms what periodicity and latency are. It's a kind of call to arms actually.
>
> The biggest impact would, in my opinion, be what you're imagining. Now we'll need practical and comprehensive measurement tools.

When you talk about latency, are you literally talking about a
straight linear delay on the signal? Or are you talking about some
kind of nonlinear hysteretic memory effect?

> > The addition of latency in and of itself would be a linear process, as
> > it can be represented by a convolution of the original signal with an
> > impulse response that represents the latency (perhaps just an impulse
> > a few milliseconds later). Do you claim that something about the
> > latency added is nonlinear, or that the auditory system just behaves
> > nonlinearly in general, outside of the latency issue?
>
> I'm not claiming, I only think. To me, the auditory system behaves non-linearly in general, but again, we'll need more EEGs and fMRIs. If you add the latency issue, it may well change the paradigm, also.

I'm sure it's nonlinear pretty much all over the place.

> > > So, when measuring consonances and hearing beating, we are always fooled by our own brain. The physical event of beating doesn't match how the brain works.
> >
> > What do you mean by this?
>
> The discussion from which stems this topic was a rough draft, I hadn't time to develop this much. I'll be curious to know which are your conceptions of the "soundwave moving at a speed of 343 m/s" (beating) working with "different types of brain wave patterns and their amount in Hz. (brain latency)"

I don't think that beating has anything to do with the speed of
sound... beating has to do with when two tones are close enough to get
caught in the same auditory filter in the cochlea.

> > > I could go on and on about fMRI results proving my statement, but I won't.
> >
> > Feel free to post them, I'd find it an interesting read.
> >
> Will do! Actually I'm still searching for the most relevant published results on this topic.
> What I really would like to clarify is the notion of "global latency" (applicable to human and non-human) and "brain latency". Are they the same thing with different names, or really two distinct types of latency, two "languages of latency"?

When you say latency, do you mean the delay in signal propagation from
the initial transduction of the signal to the auditory nerve to its
reception in the auditory midbrain? Or something else?

-Mike

Keenan,

In the past century some musicians by themselves or assisted by mathematicians, intended to demonstrate that the musical octave, that is, the true musical octave, slightly differs from the range 1 up to 2. One of them, the mexican Juli�n Carrillo has written the book titled "Sonido 13"; he was a distinguished person and notable musician who firmly declared that he can detect by ear that tiny discrepance asserting that the higher figure is slightly greater than 2. I had the comments written by Joaquin Zama�ois in his book titled "Teor�a de la M�sica" and recall that he commented that J. Carrillo directed an orchestra that played pieces using a slightly stretched octave. I also read that J. Z. wrote that Carrillo�s nephew was the mathematician who assisted to him. As far as I recall, the comments refer that J. Carrillo didn�t deduce the discrepant value.

I missed the group of photocopied pages that contain the comments.

Obviously, I didn�t try to detect by ear such a true musical octave. Recently, I asked to myself: � what indorses the 2 exactness of the octave?. There might be a scientific reason for asserting that it is really 2 exactly. Probably some institution in this planet has already measured this natural range. �Nobody proved that 2 is the exact musical octave or this range was measured with inapropriated instrumentation / procedures?.

I examined the progression of musical cells given in my book "The Piagui Musical Scale: Perfecting Harmony" (Author: C. Mario Pizarro). The progression contains 624 frequency steps comprised in the range C = 1 up to cell # 624 that equals to (9/8)^6 = 2.02728652954098. It is a geometrical progression since any cell value equals to the product of the preceding cell multiplied by one of the M, J, U commas (M = Smallest interval or schisma = (32805 / 32768)).

The first and last groups of 12 cells make a product of 1.01364326477... = Pythagorean comma. What a coincidence.

The analysis of the progression properties led to the obtainment of the true musical octave. In a few days, (July 16), a grand piano will be tuned and adjusted to this octave.

Thanks
Mario
Lima, July 09

<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
----- Original Message ----- From: "Keenan Pepper" <keenanpepper@...>
To: <tuning@yahoogroups.com>
Sent: Saturday, July 09, 2011 1:44 PM
Subject: [tuning] Re: The importance of latency

--- In tuning@yahoogroups.com, "Mario Pizarro" <piagui@...> wrote:
>
> Mike,
>
> Weeks ago I sent information to your personal address regarding the > derived
> true value of the musical octave. There, I didn�t mention specifical > values
> since until the conversion of a conventional piano to the new octave and > its
> evaluation are done, it is better not to give out this information. We > shall
> tune the piano to the new octave and according to the results I would send
> them to you and to the list.
>
> The true musical octave was correctly derived.

I'm sort of intrigued by this. By "true value of the musical octave", do you mean some frequency ratio other than the conventional value of exactly two (2)? What makes yours the "true" one?

Keenan

------------------------------------

You can configure your subscription by sending an empty email to one
of these addresses (from the address at which you receive the list):
tuning-subscribe@yahoogroups.com - join the tuning group.
tuning-unsubscribe@yahoogroups.com - leave the group.
tuning-nomail@yahoogroups.com - turn off mail from the group.
tuning-digest@yahoogroups.com - set group to send daily digests.
tuning-normal@yahoogroups.com - set group to send individual emails.
tuning-help@yahoogroups.com - receive general help information.
Yahoo! Groups Links

Hi Rowdy JB
 
Have you checked out Sir James Jean, "The Science of Sound"? it might have further insight for you.
Tim

--- On Sat, 7/9/11, collinet <jbcollinet@...> wrote:

From: collinet <jbcollinet@...>
Subject: [tuning] The importance of latency
To: tuning@yahoogroups.com
Date: Saturday, July 9, 2011, 3:50 AM

I would like to point out something that not everybody is aware of:

As much as sight is processed immediately by our brain, making it our #1 sense, our hearing is not functioning the same way (alas!)

Our ear pavilions are mere receivers. Sound has then to be processed by our brain, and he's not quick at it. Many reasons for this: the intricacy of the structure of the ear first, but then a fact which is very subjective.

As weird as it may "sound", left-handed and ambidextrous people ex aequo are faster at processing the transformed wave into an electrical signal via the auditive nerve.
Right handed people are slower at processing it.

What does it mean?

Well, simply put, we are not equally apt facing latency.
I hope that everyone here knows how much latency is a key factor in our perception of both periodic and non periodic events.

Our perception of sound, globally, is a non-linear process.

When considering any tuning, interval, anything tangible and non purely theoretic, we face the issue of our own human built-in latency.

We often blame external devices for their excesive amount of latency, yet we forget or ignore our own.

Real latency is always depending on our brains.

So, when measuring consonances and hearing beating, we are always fooled by our own brain. The physical event of beating doesn't match how the brain works.

I could go on and on about fMRI results proving my statement, but I won't.

I had a pretty interesting conversation about latency yesterday. I thought I'd share it here.

/JB

------------------------------------

You can configure your subscription by sending an empty email to one
of these addresses (from the address at which you receive the list):
  tuning-subscribe@yahoogroups.com - join the tuning group.
  tuning-unsubscribe@yahoogroups.com - leave the group.
  tuning-nomail@yahoogroups.com - turn off mail from the group.
  tuning-digest@yahoogroups.com - set group to send daily digests.
  tuning-normal@yahoogroups.com - set group to send individual emails.
  tuning-help@yahoogroups.com - receive general help information.
Yahoo! Groups Links

--- In tuning@yahoogroups.com, Mike Battaglia <battaglia01@...> wrote:
>
> On Sat, Jul 9, 2011 at 2:19 PM, collinet <jbcollinet@...> wrote:
> >
> > --- In tuning@yahoogroups.com, Mike Battaglia <battaglia01@> wrote:
> > >
> > > In what sense would you imagine that latency interferes with our
> > > perception of periodic events? I would imagine the biggest impact
> > > would come if group delay is non-uniform along the signal chain (which
> > > it probably is), but how do you predict that'll make an impact?
> >
> > I think that latency should be studied more thoroughly. So far, not much attention has been paid to brain processing. I'll volunteer for more EEGs focused on specific aspects.
>
> We've been pretty obsessed with brain processing over here, but I
> haven't seen as of yet a specific role for something like latency to
> play in fundamental frequency estimation.

I've found a study (non official) claiming that the brain already compensates for any delay even before any sound is heard... Meaning that when you play music, your auditory nerve already adds latency? Confusing study...
>
> > I didn't mention that I'm deaf in one ear, and that even if there's no brain damage. It has a normal structure for an ambidextrous person.
> >
> > Latency is most of the time very theoretic, it's a concept. Few people care to go beyond this (I'm glad this group does) and even fewer care to define in clear terms what periodicity and latency are. It's a kind of call to arms actually.
> >
> > The biggest impact would, in my opinion, be what you're imagining. Now we'll need practical and comprehensive measurement tools.
>
> When you talk about latency, are you literally talking about a
> straight linear delay on the signal? Or are you talking about some
> kind of nonlinear hysteretic memory effect?

Hybrid kind of. But more NHM.

>
> > > The addition of latency in and of itself would be a linear process, as
> > > it can be represented by a convolution of the original signal with an
> > > impulse response that represents the latency (perhaps just an impulse
> > > a few milliseconds later). Do you claim that something about the
> > > latency added is nonlinear, or that the auditory system just behaves
> > > nonlinearly in general, outside of the latency issue?
> >
> > I'm not claiming, I only think. To me, the auditory system behaves non-linearly in general, but again, we'll need more EEGs and fMRIs. If you add the latency issue, it may well change the paradigm, also.
>
> I'm sure it's nonlinear pretty much all over the place.
>
> > > > So, when measuring consonances and hearing beating, we are always fooled by our own brain. The physical event of beating doesn't match how the brain works.
> > >
> > > What do you mean by this?
> >
> > The discussion from which stems this topic was a rough draft, I hadn't time to develop this much. I'll be curious to know which are your conceptions of the "soundwave moving at a speed of 343 m/s" (beating) working with "different types of brain wave patterns and their amount in Hz. (brain latency)"
>
> I don't think that beating has anything to do with the speed of
> sound... beating has to do with when two tones are close enough to get
> caught in the same auditory filter in the cochlea.

Come with me at my next venue as FOH sound engineer. It's kind of hard to explain how confusing it is to have a signal moving at the speed of light in cables from source to speakers, and getting it back at 343 m/s.
>
> > > > I could go on and on about fMRI results proving my statement, but I won't.
> > >
> > > Feel free to post them, I'd find it an interesting read.
> > >
> > Will do! Actually I'm still searching for the most relevant published results on this topic.
> > What I really would like to clarify is the notion of "global latency" (applicable to human and non-human) and "brain latency". Are they the same thing with different names, or really two distinct types of latency, two "languages of latency"?
>
> When you say latency, do you mean the delay in signal propagation from
> the initial transduction of the signal to the auditory nerve to its
> reception in the auditory midbrain? Or something else?

First one!

I think I will just read and learn until I get a really good grasp of things here.
Learning is the main reason why I subscribed to this group... Usually, I'm not the kind of dude to wait for someone to come by and teach me. I ask questions. A lot of!

>
> -Mike
>

On Tue, Jul 12, 2011 at 12:48 AM, collinet <jbcollinet@...> wrote:
>
> --- In tuning@yahoogroups.com, Mike Battaglia <battaglia01@...> wrote:
> >
> > We've been pretty obsessed with brain processing over here, but I
> > haven't seen as of yet a specific role for something like latency to
> > play in fundamental frequency estimation.
>
> I've found a study (non official) claiming that the brain already compensates for any delay even before any sound is heard... Meaning that when you play music, your auditory nerve already adds latency? Confusing study...
//snip
> > When you talk about latency, are you literally talking about a
> > straight linear delay on the signal? Or are you talking about some
> > kind of nonlinear hysteretic memory effect?
>
> Hybrid kind of. But more NHM.

Can you define precisely what it is that you mean by latency?

> > I don't think that beating has anything to do with the speed of
> > sound... beating has to do with when two tones are close enough to get
> > caught in the same auditory filter in the cochlea.
>
> Come with me at my next venue as FOH sound engineer. It's kind of hard to explain how confusing it is to have a signal moving at the speed of light in cables from source to speakers, and getting it back at 343 m/s.

OK, but when you play something like 440 Hz and 441 Hz at the same
time, and you hear 1 Hz amplitude fluctuation, that doesn't have to do
with the speed of sound, but with the characteristics of the cochlear
auditory filter.

> > When you say latency, do you mean the delay in signal propagation from
> > the initial transduction of the signal to the auditory nerve to its
> > reception in the auditory midbrain? Or something else?
>
> First one!

So how should that affect pitch perception if everything hits your
brain a millisecond later?

-Mike

Can you quantify the latency?
I would have thought it would be (very) small compared to the latency introduced by distance?

Steve P.

On 9 Jul 2011, at 19:19, collinet wrote:

> I think that latency should be studied more thoroughly. So far, not > much attention has been paid to brain processing.