[XviD-devel] Changes to get_pmv2

Tom Jacobs T.R.Jacobs at lboro.ac.uk
Tue Oct 5 12:38:21 CEST 2004


I'm back to threading xvid after a week or so writing reports and have a few
(probably basic) questions about the variables and bitstream.

I'm kinda confused by mv16, mvs[] and pmvs[]. which one of these are the
differential prediction that is written to the bitstream and thus the one i
will need to recalculate? are all 4 values of the array written or just a
specific one?

i know these are probably simple questions but they are kinda confusing me.

also has anyone got a url for the actually iso mpeg4 standard? this would
help me with some of these questions i would guess.



Quoting Christoph Lampert <chl at math.uni-bonn.de>:

> On Wed, 22 Sep 2004, Tom Jacobs wrote:
> > hi
> >
> > my work on mpeg2 has only been simulated in an ideal SimpleScalar
> simulator.
> The software from ftp://ftp.cs.wisc.edu/sohi/Code/simplescalar ?
> > so no realworld results. but it did show a 96% complexity reduction for
> 20
> > processors on a 320x240 image.
> Hm, can you explain that? 96% complexity reduction means what?
> That 20 CPUs in parallel need only 4% of the time that 1 CPU needs?
> That would be a speedup factor of 25, which I supposed is impossible
> (well, at least very unlikely).
> And how did you parallelization look like? Just ME? Or everything?
> I'm really curious, because I did some experiments with parallel video
> encoding a while ago myself. If you stuff is "private", maybe we can
> discuss via private mail: gruel at gmx.de
> gruel
> _______________________________________________
> XviD-devel mailing list
> XviD-devel at xvid.org
> http://list.xvid.org/mailman/listinfo/xvid-devel

More information about the XviD-devel mailing list