or Connect
Head-Fi.org › Forums › Equipment Forums › Sound Science › Skeptico Saloon: An Objectivist Joint
New Posts  All Forums:Forum Nav:

Skeptico Saloon: An Objectivist Joint - Page 52  

post #766 of 1671
Quote:
Originally Posted by swspiers View Post
 
Quote:
Originally Posted by ab initio View Post

auditory cortex

You could read a book on it.

Cheers

And another freshman answer. Correct but incomplete. Want to try again?

 I don't think disdain is the right tone to employ when asking a question.

 

it's a vast domain and you can't make one answer that could cover it all. I don't really see how having a described part of the brain would enlighten anyone about our limit of perception.

and about the "good enough for human ears" part, it depends on a lot of parameters, most of which you could google. sometimes a sound can be completely masked by another sound just 20 or 25db louder.

under other circumstances we can notice one sound, and later, another sound 100db louder without a problem. but as soon as music is playing, we lose that ability as the brain prioritize the data.

all in all I think it is widely accepted that while listening to music, anything manifesting at least 80db under the loudest played sound is considered unnoticed.

so that means noises and distortion at -80db (or 0.01%) are considered inaudible while music is playing.

 

as a joke, the industry counts headphones with under 1% of distortion as ok. (it show how bad headphones are compared to the rest of the audio system).

 

 

about objective arguments, you give 2 samples of music with a difference at -40db and see if people can accurately discriminate the samples. you do the same with differences at -50db etc.

it has been done. they tried to bait giant squids with glowing ping pong balls, so you can imagine that scientists and doctors were already long bored with what humans can and cannot do.

it was also demonstrated that some kind of noises or distortions are more noticeable than others, same for some frequencies over others.

all this is well documented, but that's medicine more than headfi.

post #767 of 1671
Quote:
Originally Posted by castleofargh View Post
 

 I don't think disdain is the right tone to employ when asking a question.

 

it's a vast domain and you can't make one answer that could cover it all. I don't really see how having a described part of the brain would enlighten anyone about our limit of perception.

and about the "good enough for human ears" part, it depends on a lot of parameters, most of which you could google. sometimes a sound can be completely masked by another sound just 20 or 25db louder.

under other circumstances we can notice one sound, and later, another sound 100db louder without a problem. but as soon as music is playing, we lose that ability as the brain prioritize the data.

all in all I think it is widely accepted that while listening to music, anything manifesting at least 80db under the loudest played sound is considered unnoticed.

so that means noises and distortion at -80db (or 0.01%) are considered inaudible while music is playing.

 

as a joke, the industry counts headphones with under 1% of distortion as ok. (it show how bad headphones are compared to the rest of the audio system).

 

 

about objective arguments, you give 2 samples of music with a difference at -40db and see if people can accurately discriminate the samples. you do the same with differences at -50db etc.

it has been done. they tried to bait giant squids with glowing ping pong balls, so you can imagine that scientists and doctors were already long bored with what humans can and cannot do.

it was also demonstrated that some kind of noises or distortions are more noticeable than others, same for some frequencies over others.

all this is well documented, but that's medicine more than headfi.


Disdain was totally appropriate considering the nature of the response.

post #768 of 1671

The parameters of human hearing have been studied for over a century and are well understood. Just detectable thresholds have been measured and established. It's safe to say if an amp does 20Hz to 20kHz stone flat with .1% distortion, as long as everything else is equal, it's going to sound EXACTLY the same as an amp that does 10Hz to 40kHz stone flat with .0001% distortion. Sure, one measures better than the other, but THEY ARE BOTH THE SAME TO HUMAN EARS.

post #769 of 1671

@swspiers If you do want to discuss the brain and audio perception, please, start a new thread and we can have a fruitful discussion there. I encourage you to do so because I'm sure most everybody here would welcome the opportunity to learn something new.

 

Cheers


Edited by ab initio - 7/1/14 at 10:44am
post #770 of 1671

does anyone have an example of any hearing related phenomena 1st being deduced from Physiology?

 

Psychoacoustics is the more useful field for audio equipment design engineers looking for human hearing description - as far as I know the tracing the Anatomy, Neurophysiology of Psychoacoustic phenomena mainly flows in practice from the Psychoacoustic clues/observations to the Physiology detective work

 

so the book I look at is Fastl Psychoacoustics http://www.amazon.com/Psychoacoustics-Springer-Series-Information-Sciences-ebook/dp/B00DZ0NP72/ref=dp_kinw_strp_1

 

Tonotopic mapping from the basilar membrane up to auditory cortex seems very well documented - and its Psychoacoustic manifestations, such as ERB, frequency masking is demonstratively useful as seen in lossy CODEC design

 

but as an evolved system there are other behaviors, no one easily examined/explained grand unified theory of human auditory sensing/processing exists or is expected

 

for instance Tonotopic mapping isn't the only way we process the frequency of sounds, auditory neuron discharge phase locking likely plays an important role in pitch perception too

 

somewhere where right and left auditory neurons get together we get a very high temporal resolution correlation - click stream difference times between the ears can be  resolved down to few microseconds - but most neurons firing rate saturates at mid kHz, few can manage 10 kHz - none have been observed in any mammal that can fire at 100s of kHz - none of acoustic signals have to have >20 kHz components

does it matter if the general location where this correlation is happening is called the Olivary Complex?

 

 

electronics, transducers, recording practice, audio reproduction all have limits, well described in Physics, EE/Signal and System Theory, these can be matched/compared with Psychoacoustics knowledge to give estimates of what parts of the chain have clearly audible errors, limitations, and the many on the electronics side nowadays that simply are way below so far known human Psychoacoustic thresholds


Edited by jcx - 7/1/14 at 8:59am
post #771 of 1671
Quote:
Originally Posted by swspiers View Post

Bigshot, I respect the heck out of you, but I strongly disagree with you. Yes, the ears are vital. But the processing of that information is far more complicated, and not anywhere near to being completely understood. There are countless theories of perception, but very few laws.

I find that very few objective arguments take that into account.


The human visual processing system is not fully understood, and the way we can interpret shapes, faces, and subtle visual details is extremely complicated. That doesn't mean that we need to include UV and IR in movies to ensure we get the complete experience though - even though we don't fully understand visual processing, we can say with quite a bit of confidence that all human visual perception is encompassed by the range of electromagnetic radiation from about 380nm to 750nm (give or take a bit on either end).

 

Similarly, for audio, you're right that there's a lot of subconscious processing that goes on, but that doesn't mean that we can't set bounds on what is and is not perceptible.

post #772 of 1671

I would think that psychoacoustics would more likely make us *not* hear something audible, rather than perceive something that is inaudible. Frequency masking and filtering out constant sounds, like tape hiss are examples of not hearing something audible. I can't think of an example of perceiving something that is inaudible. I guess the fact that super-audible frequencies at loud volumes can cause headaches... but that isn't exactly a desirable thing.

post #773 of 1671
The Stax sr-207 manages to have distortion figures around 0.1-3% from 20-20kHz if my memory serves me correct. Rest could probably be EQ'ed, then pair with a 323 and a DAC and you'd have an amazing combo for little money wrt measurements.
Almost, if not transparent.
post #774 of 1671
Quote:
Originally Posted by davidsh View Post

The Stax sr-207 manages to have distortion figures around 0.1-3% from 20-20kHz if my memory serves me correct. Rest could probably be EQ'ed, then pair with a 323 and a DAC and you'd have an amazing combo for little money wrt measurements.
Almost, if not transparent.
post #775 of 1671

If you figure for Fletcher Munson, that looks pretty much flat. My cans, the Oppo PM-1s are very similar. They're so close, I don't bother to EQ. For cans, if it's within 5dB from 40Hz to 10kHz, that is flat enough for government work.

post #776 of 1671
Quote:
Originally Posted by bigshot View Post
 

If you figure for Fletcher Munson, that looks pretty much flat. My cans, the Oppo PM-1s are very similar. They're so close, I don't bother to EQ. For cans, if it's within 5dB from 40Hz to 10kHz, that is flat enough for government work.

You like them 'O mighty one?  Currawong said they sound dark, but I guess it's preference though.  Graphzzz plzzz.  :tongue_smile:  

 

Also, can somebody post a graph of how a audibly flat phones will look like taking an account of Fletcher Munson?

post #777 of 1671

Besides, the F-M curve is made based on what? Diffused field or? Anyway, that doesn't necessarily translate directly into what sound good for a headphone. Is it even relevant to use that curve wrt headphones?

 

Linear speakers in a diffused field act rather different in a room I suppose, for example the treble generally rolls, which it doesn't with headphones with flat response.

 

It'd probably make sense to find out which compensation curve that works best with headphones first (sean olive among others has looked into that?), EQ them flat to the ear and then apply said compensation curve. I have a feeling F-M alone as a compensation wouldn't make for a good sound.

 

Can't really remember all the stuff and get my facts straight, sorry. Maybe you can?

post #778 of 1671
Quote:
Originally Posted by SilverEars View Post
 

...how a audibly flat phones will look like taking an account of Fletcher Munson?

 

 

:p

 

Cheers

post #779 of 1671
Quote:
Originally Posted by davidsh View Post
 

Besides, the F-M curve is made based on what? Diffused field or? Anyway, that doesn't necessarily translate directly into what sound good for a headphone. Is it even relevant to use that curve wrt headphones?

 

Linear speakers in a diffused field act rather different in a room I suppose, for example the treble generally rolls, which it doesn't with headphones with flat response.

 

It'd probably make sense to find out which compensation curve that works best with headphones first (sean olive among others has looked into that?), EQ them flat to the ear and then apply said compensation curve. I have a feeling F-M alone as a compensation wouldn't make for a good sound.

 

Can't really remember all the stuff and get my facts straight, sorry. Maybe you can?

Is there one collective info on different types of theories on how the compensation should be done?  I'me sure there are disagreements.  I would like to look into them, and possibly fine a favorite one.  I'm aware of the Olive, and diffuse field.  I believe the diffuse filed is used by Etymotic and Golden Ears.  Golden Ears apply more than one type of compensation from what I recall.  And then, there are Rin's graphs, his looks the wildest, more so than Tyll's.  Not sure who of these guys has the most accurate tools and philosophy.  

post #780 of 1671
Quote:
Originally Posted by ab initio View Post
 

 

 

:p

 

Cheers

NOoooooo.  Heaven's Gods no.  I will not.. Never.  It's seems so irrational to go with those.  But, it seems everybody on headfi are headed in that direction.  People will reach it and then they will finally forget about the equipment and realized it was the music after all.  :tongue:


Edited by SilverEars - 7/1/14 at 5:57pm
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
This thread is locked  
Head-Fi.org › Forums › Equipment Forums › Sound Science › Skeptico Saloon: An Objectivist Joint