Menu ▾ ▴

Svn Commit Log


Commit Date  
[r2615] by ulatec

The recent parallel-processing changes broke the -B switch. It's
fixed now.

Also fixed several problems related to thread termination.

2005-01-05 23:39:54 Tree
[r2614] by ulatec

Updated to explain the new "-p" option and to remove the "-s" option.

2005-01-05 06:02:18 Tree
[r2613] by ulatec

MotionSearcher.hh
SetRegion2D.hh
main.c
newdenoise.cc
newdenoise.hh

Implemented the first, easy version of parallelism. "-p 0" produces
the original behavior. "-p 1" moves the reading/writing of video
frames to separate threads. "-p 2" moves the denoising of color to a
separate thread. Some work was needed to separate their memory spaces
better, using private allocators for the set-region objects.

This does pretty well, but since color generally takes far less time to
denoise than intensity, it still doesn't fill up 2 processors. The
next step is to rewrite denoising so that 2 search-windows can process
different slices of the same frame at the same time, but that is a BIG
project.

Also, the "skip" option was removed.

2005-01-05 06:00:58 Tree
[r2612] by sms00

Slightly more conservative defaults for the bandwidth percentages.

2005-01-03 19:03:54 Tree
[r2611] by trckjunky

Add SSE code for y4mspatialfilter -- only 20% improvement :-(

2005-01-02 23:11:03 Tree
[r2610] by trckjunky

Convert frame_i2f and frame_f2i to SSE, yields 20% performance improvement

2005-01-02 22:07:27 Tree
[r2609] by trckjunky

use floats for intermediates (speeds things a little and changes 18% of the values!)

2005-01-02 20:55:46 Tree
[r2608] by trckjunky

code reorg'd so there is one easy to vectorize function (convolveLine)

2005-01-02 20:39:43 Tree
[r2607] by trckjunky

fix major bug in reversing vert/horiz filter processing order

2005-01-02 20:27:52 Tree
[r2606] by trckjunky

Process vertical filter first (changes results but enables future optimizations)

2005-01-02 20:24:33 Tree
Older >