[FFmpeg-devel] ZMBV Improved Motion Estimation

[Michael Niedermayer]

On Sun, May 11, 2008 at 11:41:56PM +0100, M?ns Rullg?rd wrote:
> Michael Niedermayer <michaelni at gmx.at> writes:
> 
> > On Sun, May 11, 2008 at 05:21:57AM -0700, Jason Garrett-Glaser wrote:
> >> >
> >> > I call and raise.
> >> >
> >> > 16765242 May 11 05:03 ffmpeg2.CAPCOM.FLC.avi
> >> > 792548 May 11 05:03 ffmpeg2.CRUSHER.FLC.avi
> >> > 2285332 May 11 05:03 ffmpeg2.FLOAT.FLC.avi
> >> 
> >> Wow, this is what I get from coding at 4 in the morning.  I just
> >> looked back at the code I wrote and realized that under your original
> >> code, "histogram[0]" already tracks the zero count, making almost all
> >> of my patch except the last line completely unnecessary.
> >> 
> >> It should work with something like this:
> >> 
> >> return sum + 256*bw*bh - histogram[0]*256;
> >> 
> >> But it doesn't.  I suspect I'm missing the obvious due to lack of
> >> sleep.  I'll look back on this in the morning.
> >
> > probably its that histogram is uint8_t and thus 256==0
> 
> That's not the problem, or at least not all of it.  All operands are

histogram is a uint8_t array the [0] element is increased 256 times thats not
distinguishable from 0 times.
This has nothing to do with the *256 which of course is not overflowing
either way ...

[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Old school: Use the lowest level language in which you can solve the problem
            conveniently.
New school: Use the highest level language in which the latest supercomputer
            can solve the problem without the user falling asleep waiting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20080512/1c79dc31/attachment.pgp>