mirror-github-bspeice/cuburn
1
0
Fork 0
mirror of https://github.com/stevenrobertson/cuburn.git synced 2025-02-05 11:40:04 -05:00
Code Issues Actions Packages Projects Releases Wiki Activity

Fix some color foibles (more yet remain)

Browse Source
This commit is contained in:
Steven Robertson 2011-11-12 10:42:02 -05:00
parent 9ef5363652
commit 24c0c8ee56
2 changed files with 6 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
cuburn/code/iter.py
View File

@ -351,7 +351,7 @@ void iter(
*accbuf = pix; *accbuf = pix;
{{elif info.acc_mode == 'deferred'}} {{elif info.acc_mode == 'deferred'}}
// 'color' gets the top 9 bits. TODO: add dithering via precalc. // 'color' gets the top 9 bits. TODO: add dithering via precalc.
uint32_t icolor = cc * 512.0f; uint32_t icolor = fminf(1.0f, cc) * 511.0f;
asm("bfi.b32 %0, %1, %0, 23, 9;" : "+r"(i) : "r"(icolor)); asm("bfi.b32 %0, %1, %0, 23, 9;" : "+r"(i) : "r"(icolor));
*log = i; *log = i;
{{endif}} {{endif}}
@ -377,8 +377,8 @@ __device__
void write_shmem_helper( void write_shmem_helper(
float4 *acc, float4 *acc,
const int glo_idx, const int glo_idx,
const int dr, const uint32_t dr,
const int gb const uint32_t gb
) { ) {
float4 pix = acc[glo_idx]; float4 pix = acc[glo_idx];
pix.x += (dr & 0xffff) / 255.0f; pix.x += (dr & 0xffff) / 255.0f;
@ -461,7 +461,7 @@ write_shmem(
bfe_decl(shr_addr, entry, 0, SHAB); bfe_decl(shr_addr, entry, 0, SHAB);
bfe_decl(color, entry, 23, 9); bfe_decl(color, entry, 23, 9);
float colorf = color / 512.0f; float colorf = color / 511.0f;
float4 outcol = tex2D(palTex, colorf, time); float4 outcol = tex2D(palTex, colorf, time);
// TODO: change texture sampler to return shorts and avoid this // TODO: change texture sampler to return shorts and avoid this

5
cuburn/render.py
View File

@ -246,15 +246,14 @@ class Renderer(object):
for i in range(nrounds): for i in range(nrounds):
iter_fun(np.uint64(d_log), d_seeds, d_points, d_infos, iter_fun(np.uint64(d_log), d_seeds, d_points, d_infos,
block=(32, self._iter.NTHREADS/32, 1), block=(32, self._iter.NTHREADS/32, 1),
grid=(ntemporal_samples, 1), grid=(ntemporal_samples, 1), stream=iter_stream)
texrefs=[tref], stream=iter_stream)
_sync_stream(write_stream, iter_stream) _sync_stream(write_stream, iter_stream)
sorter.sort(d_log_sorted, d_log, log_size, start_bit, True, sorter.sort(d_log_sorted, d_log, log_size, start_bit, True,
stream=write_stream) stream=write_stream)
_sync_stream(iter_stream, write_stream) _sync_stream(iter_stream, write_stream)
write_fun(d_accum, d_log_sorted, sorter.dglobal, log_shift, write_fun(d_accum, d_log_sorted, sorter.dglobal, log_shift,
block=(1024, 1, 1), grid=(nwriteblocks, 1), block=(1024, 1, 1), grid=(nwriteblocks, 1),
stream=write_stream) texrefs=[tref], stream=write_stream)
else: else:
iter_fun(np.uint64(d_accum), d_seeds, d_points, d_infos, iter_fun(np.uint64(d_accum), d_seeds, d_points, d_infos,
block=(32, self._iter.NTHREADS/32, 1), block=(32, self._iter.NTHREADS/32, 1),