diff options
| author | Yaowu Xu <yaowu@google.com> | 2013-02-22 16:18:08 -0800 |
|---|---|---|
| committer | Yaowu Xu <yaowu@google.com> | 2013-02-25 07:47:27 -0800 |
| commit | 499fe05dc030b31d15c2ad9717e6bcee839f7d12 (patch) | |
| tree | 0a1940649fca62295840ea1ccdecb40d95d743b8 /test/test.mk | |
| parent | 0c9e2e9a1ddca6e120c4afa7f225968a9ab989bf (diff) | |
| download | libvpx-499fe05dc030b31d15c2ad9717e6bcee839f7d12.tar libvpx-499fe05dc030b31d15c2ad9717e6bcee839f7d12.tar.gz libvpx-499fe05dc030b31d15c2ad9717e6bcee839f7d12.tar.bz2 libvpx-499fe05dc030b31d15c2ad9717e6bcee839f7d12.zip | |
optimize forward 16x16 DCT for accuracy
This commit added pre/post scaling for first half of fDCT16x16 to
reduce error, by simulation of 100,000 blocks for random inputs,
the average sse reduced from 2.1/block to 0.0498/block.
also enabled tests for 16x16 fDCT and iDCT
Change-Id: Id2a95f0464c6dd4118797d456237ae90274c0f02
Diffstat (limited to 'test/test.mk')
| -rw-r--r-- | test/test.mk | 3 |
1 files changed, 2 insertions, 1 deletions
diff --git a/test/test.mk b/test/test.mk index 29657625c..d879f9fb2 100644 --- a/test/test.mk +++ b/test/test.mk @@ -72,8 +72,9 @@ endif LIBVPX_TEST_SRCS-$(CONFIG_VP9) += convolve_test.cc LIBVPX_TEST_SRCS-$(CONFIG_VP9_ENCODER) += fdct4x4_test.cc + LIBVPX_TEST_SRCS-$(CONFIG_VP9_ENCODER) += fdct8x8_test.cc -#LIBVPX_TEST_SRCS-$(CONFIG_VP9_ENCODER) += dct16x16_test.cc +LIBVPX_TEST_SRCS-$(CONFIG_VP9_ENCODER) += dct16x16_test.cc LIBVPX_TEST_SRCS-$(CONFIG_VP9_ENCODER) += variance_test.cc #LIBVPX_TEST_SRCS-$(CONFIG_VP9_ENCODER) += dct32x32_test.cc |