Xref: utzoo sci.electronics:3715 rec.video:4116
Path: utzoo!attcan!uunet!lll-winken!lll-tis!ames!mailrus!purdue!decwrl!labrea!glacier!jbn
From: jbn@glacier.STANFORD.EDU (John B. Nagle)
Newsgroups: sci.electronics,rec.video
Subject: Re: VCR comercial eliminator
Keywords: VCR
Message-ID: <17659@glacier.STANFORD.EDU>
Date: 30 Aug 88 05:49:24 GMT
References: <257@ivucsb.UUCP> <16314@apple.Apple.COM>
Reply-To: jbn@glacier.UUCP (John B. Nagle)
Organization: Stanford University
Lines: 33


      About four years ago, I saw such a unit on sale at a video store
(since defunct) at the Stanford Shopping Center.  It was a very
sophisticated device.  It had some elaborate algorithm for determining
whether a program was a commercial or not; the algorithm was not disclosed.
It took it a few seconds to decide whether a new segment was a commercial
or not.  Upon detecting the start of a commercial, it would stop recording
and back up the VCR to the beginning of the commercial.  As soon as a
break in the program occured, it would start recording again, and if the
new segment also turned out to be a commercial, would again back over the
commercial.  

      The unit only worked with certain VCRs.  The backup and reposition
logic was very sophisticated; it backed up too far and then played forward
until detecting some kind of mark it was placing in the audio or video
track, then stopped.  The amount of time to rewind before reentering
play was determined adaptively, and the unit learned the dynamics of the
VCR involved over time.

      I tried this unit on a basketball game.  It recorded the game until
a commercial started.  A few seconds in, it detected the commercial, and
backed up, then waited.  Another commercial followed, and it started
recording again, but soon detected that this was a second message and
again repositioned.  A third commercial was treated the same way.
Finally, the game resumed, and this time, the recording process was
allowed to proceed.

      Playback showed a clean tape, with a clean cut from the game
to the game.  I was impressed.  But the price, about $350, was a bit high
at the time.

      I have no idea how it was done.

						John Nagle