From: "Ivar" <avnet@online.no>
Newsgroups: alt.home-theater.marketplace,alt.home-theater.misc,rec.audio.marketplace
Subject: Re: Crystal Vision - Improve ALL composite sources!  Great S-Switcher too :-)
Date: Fri, 24 Jul 1998 10:56:41 +0200
Organization: Telenor Online Public Access
Lines: 58
Message-ID: <6p9i71$o5o$1@o.online.no>

Greg Rogers wrote in message ...
>In article <6p7uab$ef1$5@nntp2.ba.best.com>, thad@thadlabs.com (Thad
>Floryan) wrote:
>
>> I would still like to see an explanation of how it is expected to work
with
>> a player such as the CLD-99 which _must_ do internal Y/C separation in
order
>> for the CLD-99 to do its video tricks, then recombine the Y/C signals
back
>> into composite for additional processing by the Crystal Vision.  Seems to
me
>> there's too much Y/C separation/re-combination going-on with such a setup
but
>> I would like to be proven wrong.  :-)
>
>You are exactly correct. And the CLD-97 does the very same thing. It does
>internal Y/C separation prior to noise reduction and then recombines again
>to create the composite signal. The CLD-97 simply doesn't provide the user
>with access to the noise reduction circuit adjustments like some newer
>Pioneer players. The older CLD-95 also created its composite output after
>separating the Y/C signals. The same thing is done on most of Pioneer's
>better laserdisc players.


Except the HLD-X9 and the HLD-X0, the composite signal is not Y/C separated
and recombined in these two players.  After the field buffer, the signal
data path is divided into two, one is Y/C separated one is not.  The HLD-X9
even has a 3D motion adaptive Y/C separation filter with three different
selectable algorithms, for its Y/C data path.

>
>So the idea of running the composite video (from the re-combined Y/C
>video) through a second layer of comb filtering is ... well, an
>interesting technical exercise to say the least.


I agree, but that is exactly what Pioneer is doing in the CLD-79 and the
CLD-99. The digitized composite signal is first Y/C separated in by a 2D Y/C
separation ( I do not think it is adaptive, it is hard to find out from the
service manual ) filter in the video noise reduction section.  The Y leg and
C leg is then processed and recombined to composite and then again Y/C
separated, in the CLD-99 by a 3D motion adaptive filter, just before the
D/As.  This architecture is quite different from the CLD-97, HLD-X9 and
HLD-X0 where the "good" Y/C separation it done just after the field buffer
and no more Y/C separation or recombination is done. ( Except for the
composite output of the CLD-97 ).  I clearly prefer latter way of doing it.
This might some of the explanation why the CLD-97 is better than the CLD-99
in most video parameters.

Snip

Best regards,

Ivar Vikøren.