Theories on Sampling Rates

Locked
User avatar
Brad
Joined: Wed Dec 20, 2000 9:32 am
Location: Chicago, IL
Contact:
Org Profile

Theories on Sampling Rates

Post by Brad » Fri May 02, 2003 1:12 am

This isnt anything I need help with persay, just random thoughts that I'd like peoples input on.

Now, I am by no means any kind of expert on audio. But I do have a basic understanding of how it works. So if what I say is just utterly retarded, please forgive me ^_^;;

Now, high end audio sample rate is around 96 khz, or 96000 samples per second. Now, what if there was some way to record audio at 1 mhz (or roughly 1 sample per nanosecond). First off, you'd have an obsenely large file on your hands. Most modern equipment probably couldnt write the data that fast anyway, but for the sake of theory, lets say it can. Now, if you were to stretch out that waveform, and essentially make it equivalent to around 44 khz, what kind of things do you think you'd be able to hear? Things that are simply too fast for our ears to pick up? Or maybe, sound waves dont even move that fast, and being able to record at that speed would be useless anyway.

Or maybe you'd hear ghosts........ woooooo......


:P

Anyways. Just some silly theorizing on my part. Don't mind me!
Image
Ask Brad an After Effects Question! - Forum - Site

User avatar
RadicalEd0
Joined: Mon Jun 24, 2002 2:58 pm
Org Profile

Post by RadicalEd0 » Fri May 02, 2003 2:11 am

atom is ok: I'd like to study audio/waveforms sometime.. Like, ok.. what's the "resolution" of audio? meaning, how far would i have to zoom in to a wave form before it would be only straight horizontal lines? and how much time
atom is ok: does that line represent?
atom is ok: 1 millisecond?
Taberis 17: that would be the sampling rate
Taberis 17: well
Taberis 17: no
Taberis 17: the
Taberis 17: well yeah
Taberis 17: the sampling rate would be like the resolution
Taberis 17: more samples per second
Taberis 17: = higher 'resolution'
atom is ok: if so, could it be possible to create a recording that sampled in nanoseconds?
Taberis 17: well
Taberis 17: the best
atom is ok: to create an even richer, more accurate sound?
Taberis 17: audio nowadays
Taberis 17: is 96khz
atom is ok: even though our ears probably wouldnt comprehend it
Taberis 17: so 96000 samples per second
NMEAMV: I thought you could go up to like 148
Taberis 17: oh
Taberis 17: theres probably
Taberis 17: 192 khz
Professor Pity has left the room.
Taberis 17: but the ear stops at like
Taberis 17: 48
atom is ok: so each sample is about 96 milliseconds?
atom is ok: er
atom is ok: no
Taberis 17: yeah that's right
NMEAMV: dear sir strong bad
NMEAMV: how do you operate the telegramaphone
NMEAMV: whilst wearing gentlemen's sport gloves
NMEAMV: sincerely
NMEAMV: Lord Elzingstond tingdingdingdingworth
NeoEVA05: heh heh... strong bad...
atom is ok: i wonder if you could like, record at 1 mhz, and if you slowed that down enough and stretched it out, what little things you could begin to hear
NMEAMV: try not to think of that
NMEAMV: thats like
atom is ok: i bet you'd hear ghosts
NMEAMV: trying to compress things differently
pspfarlo: lmfao ghosts
NMEAMV: and wondering if they compress in different ways
Taberis 17: hey
NMEAMV: each time
Taberis 17: I saw a show once
Taberis 17: where these guys
Taberis 17: went to a cemetary
NMEAMV: ghosts show up on the
NMEAMV: like
Taberis 17: with this l33t recording equipment
Taberis 17: and yeah
NMEAMV: -40kHz
Taberis 17: er..
NMEAMV: or something
Taberis 17: you cant have - hz
NMEAMV: YES
NMEAMV: YOU CAN
NMEAMV: WITH
NMEAMV: UBER
NMEAMV: GHOSTS
pspfarlo: i saw that on scifi
Taberis 17: lolo
NMEAMV: well it was some fucking wiggy number
Taberis 17: dude thats impossible
Taberis 17: a hz is a cycle per second
NMEAMV: at any rate.
Taberis 17: anyway
Taberis 17: its
Taberis 17: I believe
Taberis 17: 0 - 400
Taberis 17: hz
Taberis 17: not sure tho
Taberis 17: but yeah
atom is ok: im going to break the hz barrier
Taberis 17: theres like voices
atom is ok: someday
Taberis 17: at that frequency
pspfarlo: the voices ere discovered to be chipmunks =\
NeoEVA05: must... deinterlace...... video...
NMEAMV: NO
NMEAMV: IVTC
NMEAMV: YOU BASTARD
Taberis 17: chipmunks would be on the higher end of the scale
pspfarlo: chipmunks fucking =\
Taberis 17: this is
NeoEVA05: how do you deinterlace in premiere....?
Taberis 17: low
metroamv: eww
atom is ok: maybe if you could record at that many samples, and then stretched that out, to where it'd be comparable to like 44 khz, and listened, maybe you'd hear an echo of a conversation that happened like an hour ago


:O
NMEAMV: PENIS
NMEAMV: IN
NMEAMV: YO
NMEAMV: MIXED
NMEAMV: DRINK

User avatar
Farlo
expectations of deliberate annihilation
Joined: Tue Apr 02, 2002 8:04 am
Status: The Dark Host
Location: Fort Smith, Arkansas
Contact:
Org Profile

Post by Farlo » Fri May 02, 2003 2:51 am

RadicalEd0 wrote: Taberis 17: theres like voices
pspfarlo: the voices ere discovered to be chipmunks =\
pspfarlo: chipmunks fucking =\
bow down before my non-sensical uselessness

User avatar
Bushido Philosopher
Joined: Mon Jun 25, 2001 7:19 pm
Location: California
Org Profile

Post by Bushido Philosopher » Wed May 14, 2003 12:41 am

(I apologize ahead of time if I'm getting this wrong.)

The sampling rate is really just how often the sound is shapshotted, not the speed it goes. Actually the speed a sound goes won't even matter in this case. Actually, putting audio to 1 MHz is like putting a video to 100 frames per second or something. It's just simply not needed and won't make that much difference. You're just really downsizing how often it snapshots when "stretching out the waveform". You're not changing what sounds are being heard. Eventually I think that we will be able to do 1 MHz (or maybe we can already) but I think that there's really no need since it's good enough of a sample rate. Hec, I can barely tell the diference between 48 KHz and 44.1 KHz.

If I know what you're asking then I'd have to say that you will not have anything sounding different.
"Many people want to change the world, but very few even consider changing themselves."
<<A>My Member Profile</a>> <<A>Read my Xanga</a>>

User avatar
klinky
Joined: Mon Jul 23, 2001 12:23 am
Location: Cookie College...
Contact:
Org Profile

Post by klinky » Wed May 14, 2003 1:44 am

BTW @ 96Khz, each sample would be a 96,000th of a second. Not 96milliseconds. >_<

BTW, that's 1 sample every 10.41 nanoseconds.

1000Khz would be 1sample per nanosecond. Which would be a MAJOR waste of space.



~klinky

User avatar
madmallard
Joined: Sat Aug 04, 2001 6:07 pm
Status: Cracked up quacker, quacked up cracker
Location: Atlanta, GA
Contact:
Org Profile

Post by madmallard » Wed May 14, 2003 9:10 pm

actually, thats how SACD super audio cds work

They're only 1 bit wide at a huge sampling frequency.

Completely different method compared to PCM

User avatar
madmallard
Joined: Sat Aug 04, 2001 6:07 pm
Status: Cracked up quacker, quacked up cracker
Location: Atlanta, GA
Contact:
Org Profile

Post by madmallard » Wed May 14, 2003 9:20 pm

Its actually really cool. Instead of Pulse code, it uses pulse density. Instead of frame capturing, prese, it caputres the waveform amplitidue in binary in relative to the last bit.

SACD works at 2.8224 MHZ sampling rate. Expected frequency respones can reach from 0~100khz

Locked

Return to “Video & Audio Help”