[M4IF Technotes] BVOPs question

[Zhen-Zhong Chen]

For B-frame, the compression rate depends on the efficiency of motion estimation and the temporal distance between two reference frames. I notice that the sequence Forman has a rapid modtion and changing background. The compression rate decreases in this case.
Regards,
Zhenzhong
----- Original Message ----- 
From: Klaus Leopold <klaus   itec.uni-klu.ac.at>
To: list-m4if <technotes   lists.m4if.org>
Sent: Monday, September 16, 2023 4:21 PM
Subject: [M4IF Technotes] BVOPs question
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Hi *,
> 
> I'm using the Microsoft-FPDAM MPEG-4 codec for encoding videos and have a
> question about BVOPs:
> 
> I encoded different streams (ISO conformance - foreman, silent, etc.) with
> differnt numbers of B-VOPs. I noticed that on adding B-Frames the overall
> file size of the video INCREASES. The overhead for the silent stream is for
> example 10 % if I encode it with 2 B_between_P and 9 P_between_I to 0
> B_between_P and 24 P_between_I.
> 
> The theory about B-Frames says that they provide the highest level of 
> compression. So my logical implication is that videos with B-Frames should be 
> smaller than videos without B-Frames. Why does the Why does the FPDAM codec 
> bahave different??
> 
> Klaus
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.0.6 (GNU/Linux)
> Comment: For info see http://www.gnupg.org
> 
> iD8DBQE9hZSILsrg1vEBM0IRAtYSAJ0XmUjyXS2jnoIc5MuQbw2A0nafiACdFFfe
> d+IOfROBUVjjTLOWOpG8W4k=
> =4V+q
> -----END PGP SIGNATURE-----
> 
> _______________________________________________
> Technotes mailing list
> Technotes   lists.m4if.org
> http://lists.m4if.org/mailman/listinfo/technotes
>