Quote:
|
Originally Posted by welwynnick
Madshi, do you think YCbCr looks significantly better than RGB over HDMI?
|
I think it depends very much. The source is encoded in YCbCr. The displays all show RGB. So someone somewhere in the video chain has to convert YCbCr to RGB. The big question is: Who does that the best way? If you tell the DVD player to do the conversion, there will be some (quite little, but still some) loss, because HDMI supports only 8 bits in RGB, as a result there will be some minor rounding losses. I'd guess that the Lumagen is better in color space conversion, cause that's one of the things it was designed for. But even if it's "only" as good as the Pioneer, you get a benefit by transporting YCbCr to the VP, because then there are no rounding losses. The Lumagen then takes 8bit YCbCr, does its calculations (deinterlacing + scaling) and finally converts things to RGB. Honestly, I'm not sure whether the Lumagen is doing the conversion to RGB first and then doing deinterlacing/scaling, or vice versa. But you'll have benefits either way, as the Lumagen with the newest firmware does all calculations in 10bit internally. Only the end result is then dithered down to 8bit RGB.
If the Lumagen deinterlacing/scaling algorithms only work on YCbCr data, then sending it RGB is double bad, because then the DVD player would do YCbCr->RGB, the Lumagen would do RGB->YCbCr, then do deinterlacing/scaling etc, then do YCbCr->RGB again. But I'm not sure how the Lumagen does things internally. I do know that the Mosquito is internally doing its calculations in YCbCr, though.
In any case sending YCbCr to the Lumagen can't be worse than sending it RGB (unless the Lumagen screws things up majorly, which I'm sure it does not). Sending it YCbCr is most probably better. Maybe the difference is minor, but why settle for a suboptimal connection?