[NUT-devel] Split index per stream?

Oded Shimon ods15 at ods15.dyndns.org
Sat Mar 11 17:51:44 CET 2006


On Sat, Mar 11, 2006 at 03:29:46PM +0100, Michael Niedermayer wrote:
> Hi
> 
> On Fri, Mar 10, 2006 at 11:28:38PM +0200, Oded Shimon wrote:
> > On Fri, Mar 10, 2006 at 02:42:00PM +0100, Michael Niedermayer wrote:
> > > Hi
> > > 
> > > one index for syncpoint pos, one for each stream?
> > 
> > Hmm, not sure it is worth it.. How do you even tell between them (I guess a 
> > stream_id_plus1 in the begginning). Doesn't add much for robustness (the 
> > stream stuff is tiny, 1-4kb of 80kb), and brings back the old index_ptr and 
> > several index complexity issues.. I'm against.
> 
> ok
> 
> btw, how much space would we gain/loose if we use
> - sp[i-2] + 2*sp[i-1] as sycpoint position predictor instead of sp[i-1]?
> or even
> sp[i-1] + median(sp[i-1] - sp[i-2], sp[i-2] - sp[i-3], sp[i-3] - sp[i-4])

Sorry, nothing:

current: 53686 bytes
method1: 52586
method2: 52823

Also, I was wrong, of the 83kb index, 53.7kb is syncpoint positions, 3kb is 
video, and 27kb is audio.
The 27kb of audio is all the pts differences, which are all very tiny (but 
not similar) numbers (1 to 5)

Hmm, on a different file...
65kb index, 39kb syncpoint positions, 7kb video, 20kb audio

current: 38821 bytes
method1: 20082 bytes
method2: 20414 bytes

I don't think it's worth it though, this file had 2-3mb overhead anyway 
(low bitrate, 620mb file), these 18kb aren't worth the complexity :)

- ods15




More information about the NUT-devel mailing list