I'm no fan of 4:2:2 (2:1 chroma subsampling), but the comparison looks a lot better for JPEG when you use 4:2:2 (which is the usual JPEG default). Using 4:4:4 i

I'm no fan of 4:2:2 (2:1 chroma subsampling), but the comparison looks a lot better for JPEG when you use 4:2:2 (which is the usual JPEG default). Using 4:4:4 instead drives the blocking artifacts way up, and isn't fair to JPEG.