Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [off list] Making Movies for Motion Analysis



See answers below. --Pat Cooney

Note to others on the list: John's questions all concern converting analog
video to digital using a video-capture card in the computer. His questions
are less relevant if you use cameras that produce and record digital movies
internally. Then, the "capture" of the movie into your computer is merely
the transfer of digital data from the camera to the computer via Firewire or
some other means. See

http://muweb.millersville.edu/~pjcooney/making-movies/

for more information on all the various possibilities.

----- Original Message -----
From: "John S. Denker" <jsd@monmouth.com>
To: "Pat Cooney" <pjcooney@HOTMAIL.COM>
Sent: Wednesday, September 11, 2002 12:09 AM
Subject: [off list] Making Movies for Motion Analysis


Some questions:

Is the idea to compress the movies on-the-fly,
or to store them uncompressed and (maybe)
compress them later? (The JDR card you mention
apparently doesn't do hardware compression.)


You need to experiment with your hardware and software to see which choice
allow initial capture with no dropped frames. We capture "raw" uncompressed
video buffered in RAM and streamed to the HD, then compress the video after
editing out unwanted segments.


In either case, what's the longest movie that
can be handled? A few seconds? A few minutes?


Video clips for motion analysis are rarely over a few seconds long. Again,
if you need minutes of video, experiment.


If uncompressed, is the idea to
-- hold uncompressed movies in main memory?
-- stream uncompressed movies to disk?


All the analog video capture software I've used streams video to the
computer's hard disk.

In either case, what sort of computer is needed?
-- main memory size
-- main memory speed
-- disk size
-- disk speed


Any recent computer should be adequate. If you already have computers older
than, say, 3 years, you should try video capture on them before buying a new
computer.

==========

Reply to the list if you like.


---------------------------------------------------------------------
Patrick J. Cooney Ph.D.
Professor of Physics
Dept. of Physics, Millersville U.
Millersville PA 17551
717/872-3770 (voice)
717/872-3985 (fax)
www.millersville.edu/~pjcooney/

"Be well, do good work, and keep in touch."
-- Garrison Keillor
---------------------------------------------------------------------