home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: rec.audio
- Path: sparky!uunet!think.com!ames!data.nas.nasa.gov!win31.nas.nasa.gov!fineberg
- From: fineberg@win31.nas.nasa.gov (Samuel A. Fineberg)
- Subject: Re: Why filter D/A output?
- References: <1992Nov8.222911.27702@doug.cae.wisc.edu> <1992Nov16.161335.4400@bnr.uk> <1992Nov20.080807.1@fencer.cis.dsto.gov.au> <1992Nov21.132015.8881@bilver.uucp>
- Sender: news@nas.nasa.gov (News Administrator)
- Organization: CSC, NASA Ames Research Center, NAS Division
- Date: Sat, 21 Nov 92 22:39:00 GMT
- Message-ID: <1992Nov21.223900.12289@nas.nasa.gov>
- Reply-To: fineberg@nas.nasa.gov
- Lines: 34
-
- In article <1992Nov21.132015.8881@bilver.uucp>, bill@bilver.uucp (Bill Vermillion) writes:
- |> In article <1992Nov20.080807.1@fencer.cis.dsto.gov.au> st_waldman@fencer.cis.dsto.gov.au writes:
- |> >> You have a good point regarding filtering ... The best that I can come up with in
- |> >> terms of explaning why filtering is needed is that the distortion products at
- |> >> higher frequencies have images that 'fold' back into the audio band. Therefore
- |> >> filtering is required to suppress noise in the audio band.
- |> >
- |> >When an output signal is generated by a DAC, it looks more like a
- |> >stair-case type of signal. eg. for a sine wave we might get:
- |> >
- |> > -- -- --
- |> > - - - - - -
- |> > - - - - - -
- |> > - - - - - -
- |> > - -- -- -
- |> >
- |>
- |>
- |>
- |> Wrong. Wrong. Wrong. Wrong. Wrong.
- |>
- |> It looks like the sine wave that went in!
- |> --
- |> Bill Vermillion - bill@bilver.oau.org bill.vermillion@oau.org
- |> - bill@bilver.uucp
- |> - ..!{peora|tous|tarpit}!bilver!bill
- |>
- He is right. Think about it this way, there is one level encoded on the
- CD for every 22.68 usec. What do you do with the output of the DAC during
- these intervals? Usually, the ouput level is held at the sample level, and
- then the signal is lowpass filtered to remove the stairsteps (which add high
- frequency distortion).
-
- Sam
-