Commit 3fa13569 authored by Jim Bankoski's avatar Jim Bankoski

remove complex rd modeling.

The affect on quality was minimal.  Less than .1%, various sets
yt ( +.15%), derf (-.1%), hd ( -.1% ), std hd(-.15%)...

The affect on speed of encode at speed -5 was substantial ( ~3% ).

Change-Id: I8903346fbae0c35f5b9ea20f81fdd239ae81247d
parent 466f42aa
......@@ -419,18 +419,12 @@ static void model_rd_for_sb(VP9_COMP *cpi, BLOCK_SIZE bsize,
struct macroblock_plane *const p = &x->plane[i];
struct macroblockd_plane *const pd = &xd->plane[i];
const BLOCK_SIZE bs = get_plane_block_size(bsize, pd);
int rate;
int64_t dist;
(void) cpi->fn_ptr[bs].vf(p->src.buf, p->src.stride,
pd->dst.buf, pd->dst.stride, &sse);
if (i == 0)
x->pred_sse[ref] = sse;
// sse works better than var, since there is no dc prediction used
model_rd_from_var_lapndz(sse, 1 << num_pels_log2_lookup[bs],
pd->dequant[1] >> 3, &rate, &dist);
rate_sum += rate;
dist_sum += (int)dist;
dist_sum += (int)sse;
}
*out_rate_sum = rate_sum;
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment