Hi,
On Sun, May 3, 2015 at 2:15 AM, James Almer <jamr...@gmail.com> wrote:
> Signed-off-by: James Almer <jamr...@gmail.com>
> ---
> libavcodec/avcodec.h | 3 +++
> libavcodec/version.h | 4 ++--
> libavcodec/vp9.c | 22 ++++++++++++++--------
> 3 files changed, 19 insertions(+), 10 deletions(-)
>
> diff --git a/libavcodec/avcodec.h b/libavcodec/avcodec.h
> index 8c7c420..3d5e82d 100644
> --- a/libavcodec/avcodec.h
> +++ b/libavcodec/avcodec.h
> @@ -2912,6 +2912,9 @@ typedef struct AVCodecContext {
> #define FF_PROFILE_HEVC_MAIN_STILL_PICTURE 3
> #define FF_PROFILE_HEVC_REXT 4
>
> +#define FF_PROFILE_VP9_0 0
> +#define FF_PROFILE_VP9_1 1
> [..]
+static const AVProfile profiles[] = {
> + { FF_PROFILE_VP9_0, "Profile 0" },
> + { FF_PROFILE_VP9_1, "Profile 1" },
> + { FF_PROFILE_UNKNOWN },
> +};
Do people agree we should use real names here, even if they're made up?
I believe profile 0 can simply be called "main". The significant feature in
profile 1 is extended colorspaces (422, 440, 444), so we can probably call
it "extended colorspaces". Profile 2 would then be called "high bitdepth".
Profile 3 would then be kind of long, since it's "extended colorspaces +
high bitdepth", but I don't have better ideas right now...
Other than that (so if people disagree with the proposal above, as-is),
patch is fine.
Ronald
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
