on closer examination it doesn't look like interlace problems either.
I'm rolling back on that one. Here's what I think is happening:
The source material is probably interlaced, which means it is transmitted with a frame made up of two fields of alternating lines (all the odd numbered lines in one field, then all the even numbered lines in the next). This is done so that the data rate is halved but the overall resolution of the full frame remains the same - what is lost is temporal resolution.
With the rolling credits, the text has moved significantly in the interval between fields. Meanwhile, the de-interlacer in the receiver is trying to estimate what the image should be, and because the immediately adjacent lines are not comparable it comes up with a version that ghosts the texts.
So why does the Panasonic produce a better result than the Humax? There are several possibilities (and maybe more I haven't thought of):
- The Panasonic has a better de-interlacing engine than the Humax.
- The TV has a better de-interlacer than the Humax, and the Panasonic is feeding the TV at 1080i but the Humax is set for 1080p.
- The source material is p (progressive), and the Panasonic is feeding the the TV at 1080p but the Humax is set for 1080i.
- It may even be the scaler rather than the de-interlacer, and using 576i or 720i/p as the Humax output format (whatever matches the source material) fixes it - unlike the Humax, the Panasonic may be operating in an automatic mode where the HDMI link is run at the same setting as the source material and letting the TV's scaler/de-interlacer handle it.
Meanwhile, I have to say, I don't think this would bother me much. It is only an effect that will be seen on high-contrast sharp-edged objects moving rapidly on screen in a vertical direction - ie end credits.