home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!pipex!bnr.co.uk!uknet!edcastle!dcs.ed.ac.uk!jxp
- From: jxp@dcs.ed.ac.uk (Joe Potter)
- Newsgroups: alt.sys.amiga.demos
- Subject: Re: Source of Problem (was Re: Criti!)
- Message-ID: <BzCoEx.K74@dcs.ed.ac.uk>
- Date: 16 Dec 92 11:40:08 GMT
- References: <doug.04ou@dsij.uucp> <1992Dec14.174633.25907@tdb.uu.se> <1992Dec15.054434.26886@u.washington.edu>
- Sender: cnews@dcs.ed.ac.uk (UseNet News Admin)
- Organization: Department of Computer Science, University of Edinburgh
- Lines: 36
-
- In article <1992Dec15.054434.26886@u.washington.edu>, foregone@stein.u.washington.edu (Carl Chavez) writes:
- > You know, I think the main problem is that some people don't understand
- > what a sample is. To the laypeople or hobby programmer (like me), a sample is
- > something recorded, and a simple electrical click or wave is a sound. To the
- > demo coders, a sample is *anything* generated by a computer.
- >
- > Could somebody give the *real* definition of "sample"?
-
- Deja vu, anyone?
-
- A single sample refers to an instantaneous capture of a waveform's
- amplitude at some specified instant. A table can be built up of all such
- captures over short time delays, and then can be replayed by changing the
- signal level to a loudspeaker to values obtained from the table at a rate
- identical to the "short time delays" referred to above. The table is referred
- to as a sampled sound, often abbreviated to just "sample."
-
- An example: waveform is 100*sin(t*2*pi)
-
- We set the "short time delays" to be 0.01, then record 100 samples of
- the function setting t to 0.00, 0.01, 0.02, ... , 0.98, 0.99. Now we've got a
- table of 100 values between 0 and 100. Then, we can send these table values to
- the D to A converter at any rate we choose and have a sine wave!
-
- Usually, sampled sounds are captured by using an A to D converter on an
- analogue input signal from an instrument. However, it's valid to sample any
- continuous function in this manner, generating a discontinuous "approximation"
- whose error (called aliasing distortion) has a frequency greater than audible
- sounds (~20kHz). In this way, a computed sound can be generated and replayed
- on the Amiga.
- And, contrary to my previous post which everyone ignored, computed
- sounds ARE sampled sounds.
-
- Joe.
-
- "What the f--k was all that about?"
-