On Tue, Sep 17, 2013 at 11:16 PM, Kieran Kunhya <[email protected]> wrote: >> I've never heard of NV20, but i know NV16 as 4:2:2 8-bit, similar to NV12 >> just for 4:2:2, not as 16bpp - so this would conflict with this defintion. >> >> > 8-bit 4:2:2 is 16 bits per sample (I don't know why bpp is used in that > file)
You mean 8-bit per sample, but 16-bit per pixel (due to sub-sampling)? So it matches my understanding of NV16? (just NV12 with less subsampling) If the samples are 8-bit, why a BE/LE variant? - Hendrik _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
