Path: utzoo!attcan!uunet!lll-winken!lll-tis!helios.ee.lbl.gov!pasteur!ucbvax!hplabs!pyramid!cbmvax!jesup
From: jesup@cbmvax.UUCP (Randell Jesup)
Newsgroups: comp.sys.amiga.tech
Subject: Re: Message from designer of FlickerFixer
Message-ID: <4611@cbmvax.UUCP>
Date: 31 Aug 88 02:56:25 GMT
References: <3379@crash.cts.com>
Reply-To: jesup@cbmvax.UUCP (Randell Jesup)
Organization: Commodore Technology, West Chester, PA
Lines: 52

In article <3379@crash.cts.com> haitex@pnet01.cts.com (Wade Bickel) writes:
>I don't know what you guys are talking about.  I have a minor in Video and
>I've spent a lot of time doing frame editing.  On a video display frames are
>NEVER mismatched the way the are by the FlickerFixer.

	Because you always deal with fields, not frames, I suspect.
Also, even if you deal in 2 interlaced fields that make up a frame at
a time, camera-shot video is much more blurred than computer generated
video, which can hide much of the effect.  Try it with a baseball game
some time, if it is well focused (AND moving across the screen quickly).
Note that to see this, you may need to de-interlace the signal (the first
field has faded before the second shows, and your brain interpolates for
you).

	Trying to make non-interlaced versions of interlaced video is why
the TV manufacturers are playing with interpolation chips to try to
avoid to motion problems of de-interlacing video.

>What is particularely disturbing is when the second feild of a frame is
>overlayed with the first feild of the next frame to form a non interlaced
>screen.  This never happens with video tape or transmitted signals.  In fact,
>if it did, it would be very disturbing.

	Because "normal" video is interlaced.  You get a succession of
_fields_, not frames, on the monitor.  Each is displayed for ~ 1/60 sec.
Each is of a different "time".

>Perhaps someone can give some example of how they think this happens.  Lets
>consider a basketball moveing diagnally accross the screen.  When are scan
>lines drawn with the basketball appearing in two different positions at once?
>(ie: in the same pass of the beam).

	No.  At time T, the camera circuitry reads a field of an image
(I know it isn't instantaneous, but lets pretend).  It outputs this to
somehting (VCR, TV, whathaveyou).  At time T+1/60sec, it grabs and outputs
the next filed (that makes up a frame (more or less, there are some argue-
ments that a frame is really 4 fields, but that doesn't matter here)).

	Note that the ball has moved some distance in that 1/60th of a
second.  The distance may well be small, if the ball is at a distance.
If you were to deinterlace these two fields, it would be obvious (given
fast enough movement) that the ball is being displayed in two different
places.  In interlaced video, unless it is moving VERY fast, you won't
notice this, since it appears the ball is moving smoothly (which is why
we display so fast, so that the brain interprets a series of pictures
as a continuos stream.)  Even if it were moving very fast in interlace, it
wouldn't be noticable unless the frame (NOT field) were isolated and
repeated, so the brain isn't tricked into thinking it's in motion.

	Does that (finally) explain it?

-- 
Randell Jesup, Commodore Engineering {uunet|rutgers|allegra}!cbmvax!jesup