[x264-devel] Question on keyint_min and open gop

[Li, Stephen]

It makes sense. Thanks for the explanation.

Steve

-----Original Message-----
From: x264-devel [mailto:x264-devel-bounces at videolan.org] On Behalf Of BugMaster
Sent: May-05-16 1:50 PM
To: Mailing list for x264 developers
Subject: Re: [x264-devel] Question on keyint_min and open gop

On Thu, 5 May 2016 15:40:13 +0000, Li, Stephen wrote:
>   
> Hi,
>
> I am wondering what is the effect of keyint_min when open gop is 
> enabled. From what I understand, keyint_min affects whether an I-frame 
> or IDR-frame should be used when a scene cut is detected.
> But in open gop, there are no IDR frames.  It seems to make sense that 
> keyint_min should have no effect. But based on my testing, the output 
> bitstream is quite different when different keyint_min is used.
>
> Any ideas?
>  

Use of open gop doesn't mean there is no more different between simple I-frames and I-frames that are key-frames (recovery points). There is still scenecut detection which decide which frame should be key-frame which is influenced by keyint_min. With simple I-frame *any* subsequent frame (be this P or B-frame) can reference frames before it while they still in DPB. With open gop key-frames (also I-frames) the only frames that can reference before it are B-frames which directly follow this key-frame in decoding order (i.e. have latter DTS) and which in presentation order precede it (i.e. have earlier PTS). Also open gop key-frames need write recovery point SEI which simple I-frames don't.

_______________________________________________
x264-devel mailing list
x264-devel at videolan.org
https://mailman.videolan.org/listinfo/x264-devel

______________________________________________________________________
This email has been scanned by the Symantec Email Security.cloud service.
For more information please visit http://www.symanteccloud.com ______________________________________________________________________

______________________________________________________________________
This email has been scanned by the Symantec Email Security.cloud service.
For more information please visit http://www.symanteccloud.com
______________________________________________________________________