Originally posted by: Karyudo
Its colourspace and chroma subsampling. Not to mention MPEG artifacts...

NTSC DV is 4:2:0, which means one chroma channel is sampled at just half the resolution of the luma (i.e. 2 pixels share common colour info in this channel) and the other channel is sampled at half that of the first chroma channel (i.e. 4 pixels share common colour info in that channel). Compare with RGB—or even 4:4:4 planar formats—which stores 24 bits for each pixel.

Hi everybody!

This is my first post. I am very interested in all the ongoing projects here and asking myself if I should start one as well. Please correct me, if I am wrong but the NTSC Signal produced from the Laserdisc player, is always subsampled on 4:2:2. What you loose by using NTSC DV is all the color information on the even lines of the signal. I got that. My question is, why all the expensive and "for professional use" hardware is using DV to digitize an analogue signal like the Decklink Systems for example?

In which colorspace does MPEG2-Video work? YUV2 or RGB? I've read somewhere that all display devices on the world produce color from RGB.

What advantage has capturing in RGB colorspace compared to YUV2 colorspace, since the source (NTSC or PAL) is 4:2:2 subsampled anyway?

I guess avoiding colorspace conversion is good because you will loose when convert from RGB to YUV2 back to RGB.

Could anyone explain what advantages to expect when capturing with 10bit (like the Decklink) compared to 8bit? Will there be finer color levels (like 0-255 <> 0-1023)?

Thanks everyone!

R2D2 from Vienna, Austria