[vdr] Re: workaround for Rev 2.x cards

[Christian Berger <einStein@donau.de>]

Jarkko Santala schrieb:
> 
> On Fri, 21 Jun 2002, Carsten Koch wrote:
> 
> > I may be blind, but I must admit that even on my high-end beamer I
> > can hardly tell the difference between composite, Y/C (from my 1.3 DVB-S)
> > and RGB (from my graphics card). All three are very good.
> 
> I think there must be something terribly wrong with your setup if the
> differences are really so small. I tried different methods of connecting
> my DVD player to my TV  when I got it and IMHO S-Video was twice as good
> as Video and RGB was twice as good as S-Video.
> 
> This is all very device dependent however, I hear that on some TV sets
> using S-Video input might even produce a better picture than RGB, but
> surely one should be able to see the difference between using two pairs
> of wire over using just one pair even with one's eyes closed.

Well the picture isn't really "better" it's just "more pleasing". They
have stuff like noise filtering. Phillips is famous for it. If you
compare the Phillips image you will find it looks a lot different. Skin
looses all it's structure and simply looks plastic.

You really have less information on composite. In RGB you easily have
4Mhz of bandwidth for all 3 channels resulting in a theoretical total
bandwidth of 12 Mhz. On S-Video you have about 5Mhz (if you're lucky)
for luma (brightness) and 1Mhz for croma (color) resulting in a total
bandwith of theoretical 7 Mhz. Now for composite croma and luma are
mixed, so you need to filter out all the croma out of the luma, this
cannot be done that well, typically resulting in the luma bandwidth
beeing cut to maybe 3Mhz resulting in a complete bandwidth of 5 Mhz.

DVB usually has about 5.5 Mhz for luma and 2.8Mhz for croma.

> And no, I don't see the difference between copper and gold plated
> connectors, neither can I hear it.

Well _IF_ equipment manufacturers would care about impendance matching
and stuff, there _could_ be a different, but even then it could probably
easily be corrected by doing the "termination" with a variable resistor.
But since they don't do, a lot of equipment suffers from reflections on
the cable. Gold plated contacts won't make any sense till the rest is
well designed.

Servus
  Casandro
 
>         -jake
> 
> --
> Jarkko Santala <jake@iki.fi>            http://www.iki.fi/~jake/
> System Administrator                    2001:670:83:f08::/64