home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: rec.video
- Path: sparky!uunet!think.com!ames!pacbell.com!unet!loren!larson
- From: larson@loren.net.com (Alan Larson)
- Subject: Re: Photo CD
- Message-ID: <1992Dec21.201831.14296@unet.net.com>
- Sender: news@unet.net.com
- Nntp-Posting-Host: loren
- Organization: me
- References: <PD.92Dec17104743@horus.sics.se> <BEN_WEN.92Dec19153613@w20-575-51.mit.edu> <1992Dec21.001452.19135@welchgate.welch.jhu.edu>
- Date: Mon, 21 Dec 1992 20:18:31 GMT
- Lines: 17
-
- In article <1992Dec21.001452.19135@welchgate.welch.jhu.edu> peterli@welchgate.welch.jhu.edu (Peter Li) writes:
-
- > Kodachrome has resolving power of 165 lines/mm, so it is equivalent to
- > 330 pixels/mm. However, what happens if a region of 165 lines/mm
- > picture start on the quarter cycle relative to the digitalization
- > position ? Since Kodachrome is analog, you still see the fine lines.
- > But if we only scan 330 pixels/mm at fixed positions, we end up with
- > gray ! So shouldn't the scan frequency bumped up to 660 (Nyquist
- > limit) pixels/mm ?
-
- As another poster noted, the Nyquist limit assumes that the signal
- is properly bandwidth limited before sampling.
-
- One other thing you forgot, though, was that the Nyquist limit requires
- two samples per cycle, not per half-cycle.
-
- Alan
-