[MPlayer-dev-eng] Lots of stuff for NUT

[Oded Shimon]

On Fri, Dec 30, 2005 at 11:19:07PM -0500, Rich Felker wrote:
> On Sat, Dec 31, 2005 at 04:32:49AM +0100, Michael Niedermayer wrote:
> > Hi
> > 
> > On Fri, Dec 30, 2005 at 10:23:55PM +0200, Oded Shimon wrote:
> > [...]
> > > > > 7. real back ptr is back ptr / (10 + number of streams)
> > > > > 
> > > > > The smallest amount of space a syncpoint can take:
> > > > > 
> > > > > <8 byte startcode> <1 byte pts> <1 byte per back ptr> <1 byte frame code>
> > > > > 
> > > > > That's (10+number of streams), back ptr can divide by that, it actually 
> > > > > does make the index about 1kb smaller or so...
> > > > 
> > > > Saving 1k out of 80k is rather pointless, and this is rather ugly.
> > > > It's nice to keep multiplication and division to a minimum anyway for
> > > > small implementations (even tho I know it's already needed for
> > > > timebase).
> > > 
> > > Ok, whatever. /8 is pretty in regards to >>3 anyway. :)
> > 
> > why exactly dont you remove more LSBs? for example if the back ptr only
> > specifies the position in 16k units you would have to search 16k in worst
> > case, this isnt much ...
> > so a single byte back_ptr could cover 16k*128= 2mb instead of 1k
> 
> I considered this once before but there seemed to be a problem. I
> can't remember what it was though. :(

Dude, you flamed about this like 200 times.
We wanted to know exactly which syncpoint a back ptr is pointing to, 
otherwise it's practically useless.

- ods15