[FFmpeg-devel] FFv1.3: full GOP 1/300, 2-pass statistics for Derf's collection

Pascal Massimino pascal.massimino at gmail.com
Wed Dec 5 10:26:01 CET 2012


Hi,

On Wed, Dec 5, 2012 at 1:01 AM, Peter B. <pb at das-werkstatt.com> wrote:

> On 12/05/2012 06:12 AM, Loren Merritt wrote:
> > On Wed, 5 Dec 2012, Peter B. wrote:
> >
> >> It's very interesting to see that 2-pass with GOP=1 reaches almost the
> >> same compression ratio as single-pass with GOP=300. In some cases, it's
> >> even small than GOP=300 (marked red).
> > Of course. Each of your test videos is a single scene. The probability
> > tables don't need to change much from one frame to the next, because the
> > content doesn't change much.
> >
> > The same result *might* also hold for a whole movie, but your experiment
> > doesn't predict that one way or the other.
> >
> Well, I totally agree.
> I wanted to use different, longer material from our archive collection,
> but Michael suggested to use Derf's collection - and added:
>
> "I really dont understand what is so hard on simply downloading 10 files
> from a page and not messing with them before encoding. no other tools,
> no capture, no A/D nothing."
>

suggestion: concatenate together the sources with same resolutions.
This will inject scene-changes artificially.


>
> So, I understood this as "use these official test videos".
>
> Pb
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>


More information about the ffmpeg-devel mailing list