I understand that the narrowest hair follicle in the Coclea is 12 micro metres across. At 44.1Kps how far across that follicle does sound travel before the next reading? Is it all the way, 100th, 1000th or what?
Sorry but my math is not good enough to work it out.
Speed of sound is approx 343 m/s... so:
343 000 mm/s
We divide id with 44100 to get distance in mm which sound travels in one sample, and it's around 7,8 mm
So, it would need more than one sample decoded to push the air over hair.
There is only one problem here: sound is analogue, not digital, also, ears are in analogue domain, not digital. You can't decode one sample and get meaningful information, nor is sound decoded in 44100 incremental steps.
What are you trying to prove?
Not quite sure of the context of the question but samples do not arrive at the ear from a digital recording playback. After the train of samples has gone through the d to a there are no samples in the waveform. It is a continuous analogue waveform.
Hair follicles vibrate at a given frequency. The question as posited is nonsensical.
Someone I will meet up next Wednesday tried to tell me,last year, that he could hear the individual samples of digital as "graininess".
I could not think of a reply then. But have since thought that if I could show that each sound samples hit the hair follicle before the previous one has crossed the whole hair he could not hear individual sampled even as "graininess".
Knowing him he will not consider reading to correct his thoughts; just one simple idea may.
Obviously I am completely wrong in how I think. So my math and science are even weaker than I thought.
I will withdraw from this discuss with apologises for mis-understanding.
How fast the speed of sound is in the cochlea is an interesting question. I suspect it's quite fast because it's liquid filled. But it's not relevant to the detection of sound because the detection is by the resonance of the hairs, determined by their geometry. As has been pointed out, the individual samples aren't audible anyway. Not only are they at far too high a frequency to hear (44.1 KHz or more), the output filter of the DAC "smooths" them into an exact replica of the original analog signal that was digitised. I suggest you watch this video:
Come back and ask for explanations of anything you don't understand.
Rest assured, if your friend believes he hears "graininess", it's all in his head.
What happened last year was that I mentioned, in the same group I will meet on Wednesday, that my amp had broken. The LP enthusiest asked what TT I had. I said a Rega Planar 3, in the loft. He said I should bring it down and listen for the "musicality" in LPs. I asked what did he mean by musicality in LPs, that's not in digital. He said its the graininess in digital, from the sampling he objects to. Interesting but acquantance with a large LP collection also mentions musicality.
Since last yesr I have read a great deal, including viewing that video, all of which reinforces my bias that I should stick with digital; unless I can only get the music on LP, when I rip that to a digital file.
My musical tastes vere to classical and jazz. But real rock and roll still figures; after all I was 14 the Day the Music Died.
he could hear the individual samples of digital as "graininess".
Sure, people can hear Santa Claus too.
Those are mental health/intellectual issues, not electro-acoustic ones.
Your ear is a filter, so the speed at which information passes through it (group velocity) is a function of frequency. You cannot compute a single number that describes it.
Your friend is so profoundly confused about how hearing works that you are not going to be able to talk to him about the factual aspects of hearing. It would be like trying to describe traffic laws to someone who has convinced himself that cars only work underwater. He will not understand what you mean by roads and street signs.
But have since thought that if I could show that each sound samples hit the hair follicle before the previous one has crossed the whole hair he could not hear individual sampled even as "graininess".
my math and science are even weaker than I thought.
I can’t attest to your math and science skills, but it seems you need to study some anatomy...
...as well as sampling theory (see the above video).
As far as digital goes, if each sample were replaced with an *analog* sinc function using the amplitude of the sample and the sampling frequency, the sampled waveform will be reconstructed to within the toleralces defined by the bit depth and sample rate as well as the design of the ADC. The sinc fuction is created through the use of the reconstruction filter within the ADC. Again, the result is an analog signal.
The SNR of the restored analog signal will exceed that of vinyl assuming you’re using more than ~13 bits. Frequency response using 44.1 kHz sample rate will be less but it will extend beyond that which is necessary for adult hearing. Vinyl playback will exceed 22.5 kHz, but there will be distortion. In stark contrast there will be none from the conversion of a digital signal. Furthermore, any content above the Nyquist frequency from vinyl, assuming it was actually part of the recording, will almost certainly be masked by the energy in correlated frequencies that are lower.
He said I should bring it down and listen for the "musicality" in LPs. I asked what did he mean by musicality in LPs, that's not in digital. He said its the graininess in digital, from the sampling he objects to. Interesting but acquantance with a large LP collection also mentions musicality.Vinyl DOES sound DIFFERENT from digital
and some people prefer vinyl. You can't argue with someone's taste.However, vinyl is TECHNICALLY inferior
to digital (noise, distortion, and frequency response). Personally, it's the noise that bothers me the most, and it's the noise that limits the resolution & dynamic range. A record can actually have a wider
frequency response (on the high end) than a CD but the CD is flatter in the audible range where it counts. ...I can make an amplifier that goes up to 1MHz, but if it's not flat across the audio range it doesn't make a good audio amplifier.
A digital copy of a record can sound like the record. Here
(https://hydrogenaud.io/index.php/topic,117219.msg967843.html#msg967843) is a report of one-such demonstration.
But, a record can't sound like a digital original. That's not as easy to "demonstrate" since we can't make our own vinyl records but it's pretty obvious that records don't have a perfectly-silent background. You also can't argue with "musicality" because it's a MEANINGLESS word and musicality can't be measured.
The audiophile community uses lots of meaningless words and non-measurable qualities. Let's stick with noise, frequency response, distortion, etc... Things that engineers & scientists can define & measure.P.S.
I actually had a soundcard with an unfiltered stair-stepped output. But, I couldn't hear the "graininess" because the noise/harmonics are above the audio range (44.1kHz and above). Also, the speaker provides some mechanical filtering so the soundwaves are not stair-stepped. The only way I knew the soundcard was unfiltered was that I was doing some experiments with an oscilloscope attached. It was a shocking thing to see
, but like I said it sounded
I actually had a soundcard with an unfiltered stair-stepped output. But, I couldn't hear the "graininess" because the noise/harmonics are above the audio range (44.1kHz and above). Also, the speaker provides some mechanical filtering so the soundwaves are not stair-stepped. The only way I knew the soundcard was unfiltered was that I was doing some experiments with an oscilloscope attached. It was a shocking thing to see, but like I said it sounded fine.
Oh dear lord, I misread and instead of micrometers used milimeters :) (am a bit dyslexic)
Effin' eejit LOL
So, 343 000 000 μm/s divided by 44100 is around 7800 μm
Much more, but there is this thing that everyone else explained :)