[mythtv-users] overscan problems with nvidia 8400 GS

[Brian Wood]

On Monday 06 July 2009 19:46:00 Phil Wild wrote:
> > I think its actually really quite simple, if the TV manufacturer promoted
> > a TV display resolution then the set should provide it. (With option of
> > no overscanning)  Manufacturers might be resistant to providing the first
> > few but once the precedent is confirmed they will likely become more
> > obliging.
>
> I agree with you.... BUT....
>
> I run my panel at 1080P with no overscan. I often get garbage
> appearing around the edges of the display and sometimes, there is also
> just black (maybe 20 pixels of black) around the outside of the video
> stream.
>
> So it would appear that the conversion of video from the original
> source material to 1920 x 1080 does not always produce an exact 1920 x
> 1080 video stream. I would expect that the everyday consumer would
> complain more about what they see on the screen rather than what they
> don't see (being a better end result).

The same conclusion as was made in the 1950s, and still probably correct.
>
> The fix needs to start with the broadcasters...

Or perhaps with the folks who provide the video material to them, though the 
broadcasters should be capable of correcting errors, it takes time and thus 
costs money. Picture information once removed can never be brought back 
again, usually.

It's just so funny to see the past repeat itself.

A problem TV broadcasters used to have in the NTSC days was too-wide 
horizontal blanking. In extreme cases this could result in a black vertical 
stripe to the left of the picture, usually it was not that extreme 
(especially since the home receiver's overscan hid it).

The problem was often the result of too many processing devices in the chain, 
time base correctors and frame syncs were offten to blame, especially when 3 
or 4 such devices, improperly set up, were cascaded.

The FCC used to regularly fine broadcasters for overly wide blanking, it was 
common and could be checked without having to visit the station, an easy 
write up.

I suspect directors will continue to observe "safe action" and "safe title" 
standards, and who can blame them? Running a title graphic to the very edge 
of the theoretical picture area is not a good way to make sure it will be 
seen properly.

So a problem from the old days now exists in the "improved" digital system, 
when we had the perfect chance to correct things. Oh well.

-- 
Brian Wood
beww at beww.org