[XviD-devel] [CVS commit] devapi4 -- Speed optims
Edouard Gomez
ed.gomez at free.fr
Wed Nov 5 17:23:24 CET 2003
Following the previous discussion about setedges usage... here's the
patch that saves a fair amount of calls so we don't waste CPU anymore
(if we all agree using xvid is not a waste of CPU time ;-).
I don't have exact numbers anymore, but on the 1000 frame sequence i
tested for the memcpy optim, setdges was called ~1400 times, with this
patch it's called ~470 times. Interpolation calls might be saved as
well if the same rounding is required.
2003-11-05 16:05:44 GMT patch-90
Summary:
Speed improvement not wasting setedges and interpolate calls.
Revision:
xvidcore--devapi4--1.0--patch-90
Patch from syskin.
* This patch avoids calling setedges and interpolate for uneeded
cases:
- setedges is only called once per frame.
- interpolate is called only when the previous rounding
was different from the one needed.
* Interpolation has been optimized a bit for qpel case, we do the
hv pass down to top to use the cache more efficiently (hope so).
modified files:
src/encoder.c src/encoder.h src/image/image.c
--
Edouard Gomez
More information about the XviD-devel
mailing list