home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.atari.st
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!sdd.hp.com!uakari.primate.wisc.edu!usenet.coe.montana.edu!news.u.washington.edu!milton.u.washington.edu!mpark
- From: mpark@milton.u.washington.edu (Michael Park)
- Subject: Re: (none)
- Message-ID: <1992Sep11.171848.29728@u.washington.edu>
- Sender: news@u.washington.edu (USENET News System)
- Organization: University of Washington, Seattle
- References: <2250597@overmind.citadel>
- Date: Fri, 11 Sep 1992 17:18:48 GMT
- Lines: 23
-
- In article <2250597@overmind.citadel> only_bbs!arthur_dent@overmind.mind.org writes:
- >Actually, it
- >depends on your definition of accuracy. If you use a sampling frequency
- >twice that of the sine wave's frequency, it would end up as a triangle wave.
- > At three times the frequency, you would get a square wave. Therefore, to
- >retain the shape of the sine wave, you would need a frequency of at least 5
- >times that of the sound itself.
-
- Although your idea of "retaining the shape of the sine wave" is
- intuitively appealing, the Sampling Theorem (tm) assures us that
- a sine wave can be completely represented even when sampled at
- "only" twice its frequency. To understand why, you must realize
- that a triangle wave of a given frequency contains components
- of much higher frequencies. If the analog signal has been properly
- sampled, then even though the digital samples may _look_ like a
- triangle wave, the only "legal" underlying analog signal is
- a sine wave.
-
-
- --
- Ciao-abunga! +-------------------------------------+
- Michael Park | This space intentionally left blank |
- mpark@u.washington.edu +-------------------------------------+
-