Halve the number of motion search iterations, to eight.

The DIS code claims this is allowed after they changed their Sobel code;
for us, seemingly SOR was the breaking point. EPE is hardly moving
(<1% for most Sintel tests I've run), but speed goes up markedly.

diff --git a/motion_search.frag b/motion_search.frag
index f67cbba51cc3cb06869bbebc2cb037189404d04d..cd251f4f15a7613487aa2d7b5fc23730c3eeded3 100644 (file)
--- a/motion_search.frag
+++ b/motion_search.frag
@@ -36,7 +36,7 @@
  */
 
 const uint patch_size = 12;
-const uint num_iterations = 16;
+const uint num_iterations = 8;
 
 in vec2 flow_tc;
 in vec2 patch_center;