[FFmpeg-devel] [RFC] Runtime-inited versus Hardcoded tables.

[Kostya]

On Thu, Jan 31, 2008 at 12:54:16PM +0100, Zdenek Kabelac wrote:
> 2008/1/31, Reimar D?ffinger <Reimar.Doeffinger at stud.uni-karlsruhe.de>:
> > Hello,
> > On Thu, Jan 31, 2008 at 11:47:33AM +0100, Zdenek Kabelac wrote:
> > > Yes - that's obvious - but I think distribution of some large binary
> > > tables (either in sources or executables) is by faaaaaar more painful
> > > then having them initialized on the runtime and only when needed.
> >
> > Well, that so far is an assertion without reasoning behind it though.
> > At least for all I know, the tables could easily be smaller than the
> > code to generate them after compression in many cases.
> > It also raises the question which scenario exactly you are thinking of
> > that you would consider distribution of (at my random guess) a few 100 kB of
> > extra data painful...
> 
> Well of course we are not talking about tables which take ~1kB here ;)
> 
> But I think that files like rv34vlc.h,indeo3data.h are already quite
> too much if there exists simple code to generate them at runtime I
> would prefer rather this version.

Bad example in case of rv34vlc.h - tables there contain only codeword
lengths and actual VLC tables are generated from that data.

And the real problem are vector codebooks (like in monster dcadata.h).

> The scenario is simple - you release binary/source package which is
> then downloaded by thousands of user - which spends time/money waiting
> for download :)
> 
> Zdenek