Jingning Han [Fri, 13 Mar 2015 19:57:49 +0000 (12:57 -0700)]
Merge "Use sdx4df to do 1-step refinement"
Marco [Fri, 13 Mar 2015 17:31:35 +0000 (10:31 -0700)]
Merge "Lower bitrate threshold below which cyclic refresh is turned off."
Alex Converse [Fri, 13 Mar 2015 17:20:15 +0000 (10:20 -0700)]
Merge "Reconcile active_map and cyclic refresh"
Jingning Han [Fri, 13 Mar 2015 16:56:00 +0000 (09:56 -0700)]
Merge "Reset src buffer only once in vp9_int_pro_motion_estimation"
Jingning Han [Fri, 13 Mar 2015 16:55:52 +0000 (09:55 -0700)]
Merge "Reduce the number of full block SAD calls"
Jingning Han [Fri, 13 Mar 2015 16:53:15 +0000 (09:53 -0700)]
Use sdx4df to do 1-step refinement
Change-Id: Ie0c3ef3ae3aedf049b1a296de607730b79c12672
Marco [Fri, 13 Mar 2015 16:36:22 +0000 (09:36 -0700)]
Lower bitrate threshold below which cyclic refresh is turned off.
Change-Id: Ib54ab11adf8178eec74f65388a89c8f912c7869a
paulwilkins [Fri, 13 Mar 2015 15:45:52 +0000 (08:45 -0700)]
Merge "Shorten GF/arf interval in hard scenes."
Yaowu Xu [Fri, 13 Mar 2015 04:27:04 +0000 (21:27 -0700)]
Merge changes Ic54c88ec,I6f120b65
* changes:
Added comment for vp9 only codec controls
Rename a vp9 only control with vp9e prefix
Jingning Han [Fri, 13 Mar 2015 01:46:40 +0000 (18:46 -0700)]
Reset src buffer only once in vp9_int_pro_motion_estimation
Change-Id: I5c96b6a25f9df60da65b7af7c92a921b611746e3
Yaowu Xu [Fri, 13 Mar 2015 01:43:23 +0000 (18:43 -0700)]
Merge "vp9_pick_inter_mode(): Use single loop to evaluate inter modes"
Yaowu Xu [Fri, 13 Mar 2015 01:14:43 +0000 (18:14 -0700)]
Added comment for vp9 only codec controls
Comments are updated to reflect that these controls apply to VP9 only,
thereby, to insure the document produced by doxygent to reflect the
same fact too.
Change-Id: Ic54c88ec066aa0ec4552d43dd4a7016e1f810f42
Yaowu Xu [Fri, 13 Mar 2015 00:59:18 +0000 (17:59 -0700)]
Rename a vp9 only control with vp9e prefix
Also updated comments and doc to make it clear that the control is
supported in VP9 only.
Change-Id: I6f120b65135fc1185695e9e3dc29fbe0d430824b
Jingning Han [Thu, 12 Mar 2015 23:18:12 +0000 (16:18 -0700)]
Reduce the number of full block SAD calls
This commit uses a 6-point 1-step refine motion search in the
integral projection based full pixel motion estimation, to replace
the current 9-point search.
It reduces runtime cost of speed -6 on some noisy clips, e.g.,
dark720p single thread
33314 b/f, 40.076 dB, 18231 ms ->
33307 b/f, 40.067 dB, 17768 ms
The compression performance for rtc set remains unchanged.
Change-Id: I194ea5a9ce52e5a10baeee36338633adc22f764c
Yunqing Wang [Fri, 13 Mar 2015 00:16:48 +0000 (17:16 -0700)]
Merge "Minorly modify model_rd_for_sb_y function"
Jingning Han [Thu, 12 Mar 2015 23:43:09 +0000 (16:43 -0700)]
Merge "Fix fdct8x8_quant ssse3 overflow issue"
Alex Converse [Tue, 10 Mar 2015 23:08:17 +0000 (16:08 -0700)]
Reconcile active_map and cyclic refresh
Change-Id: Id7f8654aeeb20caa402bc822521b1d72c658f4f9
Yaowu Xu [Wed, 11 Mar 2015 17:16:39 +0000 (10:16 -0700)]
vp9_pick_inter_mode(): Use single loop to evaluate inter modes
This commit changes to use single loop to evaluate all inter modes.
There is no impact on compression quality and speed, but allow future
experiment with the order of modes evaluated.
Change-Id: I71696ce1014cbe127e25e98710d835987f5ecc09
Yunqing Wang [Wed, 11 Mar 2015 18:31:47 +0000 (11:31 -0700)]
Minorly modify model_rd_for_sb_y function
Added a skip_dc check. If skip_dc = 1, we could eliminate calling
of vp9_model_rd_from_var_lapndz(). This gave slight PSNR & SSIM
gain(<0.1%), and no speed change.
Change-Id: If5ca733366148c86b98e196a00cc890f50e9a3e5
Jingning Han [Thu, 12 Mar 2015 19:16:27 +0000 (12:16 -0700)]
Fix fdct8x8_quant ssse3 overflow issue
This resolves webm issue 968.
Change-Id: Ieb363129b1e135a561141c68211d413226aba754
Deb Mukherjee [Thu, 12 Mar 2015 18:10:59 +0000 (11:10 -0700)]
Merge "Some rate control adjustments to control overshoot"
Jingning Han [Thu, 12 Mar 2015 16:24:02 +0000 (09:24 -0700)]
Merge "Prevent integer overflow in choose_partitioning"
Jingning Han [Thu, 12 Mar 2015 16:23:51 +0000 (09:23 -0700)]
Merge "Remove unnecessary speed feature checking"
Jingning Han [Thu, 12 Mar 2015 16:23:41 +0000 (09:23 -0700)]
Merge "Apply fast motion search to golden reference frame"
Jingning Han [Thu, 12 Mar 2015 16:23:31 +0000 (09:23 -0700)]
Merge "Refactor to remove GLOBAL_MOTION"
Yaowu Xu [Thu, 12 Mar 2015 15:01:08 +0000 (08:01 -0700)]
Fix a typo introduced in #
94401aff
This fixes all test vector failures
Change-Id: Ie1a9fe0f023f7a0c7e89eb55df1b40ff65302adc
Marco [Thu, 12 Mar 2015 00:19:47 +0000 (17:19 -0700)]
Merge "Fix visual studio build failure."
Jingning Han [Wed, 11 Mar 2015 23:29:13 +0000 (16:29 -0700)]
Prevent integer overflow in choose_partitioning
Re-arrange the multiplication and right shift operations to avoid
integer overflow in choose_partitioning.
Change-Id: Ib4005cafb410a67c1960486471d75b6ebe38c4e0
hkuang [Wed, 11 Mar 2015 23:19:14 +0000 (16:19 -0700)]
Merge "Refactor the block decode code to make it simpler."
Marco [Wed, 11 Mar 2015 23:11:35 +0000 (16:11 -0700)]
Fix visual studio build failure.
Change-Id: Ifeb14f945d0f0300eb7b21b38e5720ac1c11a6cf
Jingning Han [Wed, 11 Mar 2015 23:08:59 +0000 (16:08 -0700)]
Remove unnecessary speed feature checking
This commit removes the pred_mv_sad comparison from rtc motion
search, given that a stronger comparison has been done at the
mode search level to eliminate unlikely selected reference frames.
Change-Id: I49b8d24b2174303066fd8eff2102c0648f2869df
Adrian Grange [Wed, 11 Mar 2015 23:09:49 +0000 (16:09 -0700)]
Merge "Clamp rate correction factor after scaling it"
Jingning Han [Tue, 10 Mar 2015 01:55:38 +0000 (18:55 -0700)]
Apply fast motion search to golden reference frame
This commit enables the rtc coding mode to run integral projection
based motion search for golden reference frame. It improves the
speed -6 compression performance by 1.1% on average, 3.46% for
jimred_vga, 6.46% for tacomascmvvga, and 0.5% for vidyo clips. The
speed -6 is about 6% slower.
Change-Id: I0fe402ad2edf0149d0349ad304ab9b2abdf0c804
Jingning Han [Wed, 11 Mar 2015 22:44:26 +0000 (15:44 -0700)]
Refactor to remove GLOBAL_MOTION
Make the vp9_int_pro_motion_estimation() function return zero
motion vector if high bit depth is turned on, instead of removing
it from compiled codes.
Change-Id: Ia48f010eb590b2d517d5678c394110b326a1a95e
Marco [Wed, 11 Mar 2015 21:07:34 +0000 (14:07 -0700)]
vpx_temporal_svc_encoder: Update some default settings.
Set flags=0 if layering_mode=0, and set static_thresh=1.
Change-Id: Id9874923b23966a2b1d0e83965add7d5c9d26d16
hkuang [Tue, 10 Mar 2015 22:37:12 +0000 (15:37 -0700)]
Refactor the block decode code to make it simpler.
Change-Id: I0f983cb821ad7ec6fbefe7895cb8124a8fa39df6
Yaowu Xu [Wed, 11 Mar 2015 17:41:20 +0000 (10:41 -0700)]
Merge "Separate rd_thresh adaption by ref_frame"
Adrian Grange [Tue, 10 Mar 2015 19:04:59 +0000 (12:04 -0700)]
Clamp rate correction factor after scaling it
Added clamp on the rate correction factor after
it has been scaled.
Change-Id: I5d4b46a101987b43c5bcfd7e0bd1b7b4d53640a4
paulwilkins [Fri, 6 Mar 2015 16:32:20 +0000 (16:32 +0000)]
Shorten GF/arf interval in hard scenes.
This patch accounts in the first pass stats for blocks that
while not coded as intra, are complex and have an intra error /
best error ratio below a threshold.
The modification shortens the GF arf interval for a particular
class of content that contains a lot of blocks matching the
above criteria. (In one short problem test sequence the average
interval dropped from about 14-15 to 10-11)
The change results in small net gains in metrics results for the
Yt(~0.2%) and yt-hd (~0.5%) sets and is approximately neutral
for the other test sets.
The change is currently shielded by a flag and off by default
pending verification that it does not cause other regressions
in tests on a wider YT test set.
Change-Id: I6b803daa6a4ac09a6f428fb3a18be1ecedd974b7
Yaowu Xu [Wed, 11 Mar 2015 02:06:52 +0000 (19:06 -0700)]
Separate rd_thresh adaption by ref_frame
Only update the rd_thresh factors for modes sharing same reference
frame. This helps overall compression of 6 and 7 by .13% and .19%
respectively without any noticeable speed difference.
Change-Id: Idb3a3879512c5d7d0880034516079949290690c5
Deb Mukherjee [Wed, 11 Mar 2015 00:25:10 +0000 (17:25 -0700)]
Some rate control adjustments to control overshoot
Some rate control adjustments to control overshoot
in the constrained quality mode.
Change-Id: I8907b9a883642d779009d0a138adfa6ba67e7f41
Marco [Tue, 10 Mar 2015 18:48:10 +0000 (11:48 -0700)]
Merge "Modify update golden reference update under aq-mode=3 mode."
Marco [Fri, 6 Mar 2015 01:26:52 +0000 (17:26 -0800)]
Modify update golden reference update under aq-mode=3 mode.
For non-SVC 1 pass CBR: make the GF update interval a multiple of the
cyclic refresh period, and use encoding stats to prevent GF update at certain times.
Change-Id: I4c44cacc2f70f1d27391a47644837e1eaa065017
Yunqing Wang [Tue, 10 Mar 2015 17:52:55 +0000 (10:52 -0700)]
Merge "Accumulate tx_totals counters in multi-threaded encoder"
Yaowu Xu [Tue, 10 Mar 2015 17:48:52 +0000 (10:48 -0700)]
Merge "Enable using Golden reference in choose_partition()"
paulwilkins [Tue, 10 Mar 2015 17:45:02 +0000 (10:45 -0700)]
Merge "Allow q adjustment for VPX_CQ and VPX_CBR."
Yunqing Wang [Tue, 10 Mar 2015 16:53:45 +0000 (09:53 -0700)]
Accumulate tx_totals counters in multi-threaded encoder
Tx_totals counters weren't handled correctly in multi-thread
case, which caused the mismatch while encoding using threads > 1.
This patch fixed that.
Change-Id: Ice9b0386f57175fb92a0bdcd5042686a3106246a
Adrian Grange [Tue, 10 Mar 2015 16:14:54 +0000 (09:14 -0700)]
Fix vp9_compute_qdelta_by_rate loop behavior
The return value from vp9_compute_qdelta_by_rate, which is
a delta value for the quantizer, could never be 0 if
(qindex == rc->worst_quality).
This occurs because target_index was setup unconditionally
in the loop and yet the loop counter stopped at
(rc->worst_quality - 1).
Change-Id: I6b59cd9b5811ff33357e71cd7d814c5e53d291f2
Yaowu Xu [Fri, 6 Mar 2015 21:45:39 +0000 (13:45 -0800)]
Enable using Golden reference in choose_partition()
Choose_partition uses only the last frame as reference frame in making
partition decision, this commit adds the check on how well Golden
frame with (0,0) predicts the current block, and uses GF(0,0) as
basis for partition decision if it produces better prediction.
The commit improves rtc speed 6 and 7 encoding by 0.14% and 0.19%
respectively.
Change-Id: I156acf925bd6e0b586d48155d1940d27270a3915
Frank Galligan [Mon, 9 Mar 2015 23:25:01 +0000 (16:25 -0700)]
Refactor read_intra_frame_mode_info
Change-Id: I56b0614154408e8ec613784b2007374df00fbf17
Alex Converse [Mon, 9 Mar 2015 20:02:16 +0000 (13:02 -0700)]
Merge "Don't waste time partitioning skip superblocks."
Jingning Han [Mon, 9 Mar 2015 19:27:19 +0000 (12:27 -0700)]
Merge "Skip golden ref frame check when it is same as last ref frame"
Jingning Han [Mon, 9 Mar 2015 17:59:29 +0000 (10:59 -0700)]
Skip golden ref frame check when it is same as last ref frame
When golden reference frame is refreshed, the next frame has both
its last and golden reference frames point to the same reference
frame in real-time coding mode. Experiments suggest that using
two separate reference frames for frames right after golden refresh
frame does not provide further compression performance advantage.
This commit hence retains the current encoder implementation and
shuts off the mode search over golden reference frame in this case.
It makes the encoder run slightly faster at no coding performance
change.
Change-Id: I1561f7799253a10e675d05c63c1749fe9e85b472
Alex Converse [Mon, 9 Mar 2015 17:45:38 +0000 (10:45 -0700)]
Don't waste time partitioning skip superblocks.
Force 64x64 partitioning when a whole superblock is SEGMENT_LVL_SKIP. This
drops encode times of screens mostly at rest by 20%.
Change-Id: Ieba554b0b8a0c1679aae784a8bd11f038ab942c3
paulwilkins [Mon, 9 Mar 2015 17:12:33 +0000 (17:12 +0000)]
Allow q adjustment for VPX_CQ and VPX_CBR.
Adjustment previously only enabled in VBR mode.
This patch allows adjustment of min and max q for CBR
and adjustment of max q only for CQ mode.
Change-Id: Id5e583f3d50453cd544fc57249acacd946457482
Marco [Fri, 6 Mar 2015 18:47:09 +0000 (10:47 -0800)]
Add aq-mode=3 to ethread test.
Change-Id: I619a2be502f1d2eb0ae746f7fcb465d19ddae048
Yunqing Wang [Mon, 9 Mar 2015 16:42:12 +0000 (09:42 -0700)]
Merge "vp9_ethread: fix me consts initialization to support aq_mode=3 encoding"
Jingning Han [Mon, 9 Mar 2015 16:34:26 +0000 (09:34 -0700)]
Merge "Move pred_mv assign outside integral projection motion search"
Deb Mukherjee [Mon, 9 Mar 2015 16:21:35 +0000 (09:21 -0700)]
Merge "Reduce default overshoot/undershoot params to 25"
Yunqing Wang [Mon, 9 Mar 2015 15:35:57 +0000 (08:35 -0700)]
Merge "Modify the setting of transform skip flags in non-rd mode"
Yunqing Wang [Fri, 6 Mar 2015 23:17:59 +0000 (15:17 -0800)]
vp9_ethread: fix me consts initialization to support aq_mode=3 encoding
While turning on "--aq_mode=3", the quantizers are updated by each
thread. Fixed the me consts initialization function to make sure
that the correct thread data are updated.
Change-Id: Ied27bb7bae76fc3fa2cda4f8c35ac0b46271bef4
Johann [Sat, 7 Mar 2015 00:04:59 +0000 (16:04 -0800)]
Merge "Cherry pick libyuv VS fix"
Johann [Fri, 6 Mar 2015 20:25:14 +0000 (12:25 -0800)]
Cherry pick libyuv VS fix
Change-Id: I1f2e077af87d29b07eaf7acaec8d955e00da9333
hkuang [Tue, 3 Mar 2015 01:15:00 +0000 (17:15 -0800)]
Remove some unnecessary code in thread context copy.
Change-Id: Iddf098e1bae9c10fc2f325f84156f50a0bd0055a
Hangyu Kuang [Fri, 6 Mar 2015 18:27:26 +0000 (10:27 -0800)]
Merge "Only wait for previous frame's motion vector if needed."
Marco [Fri, 6 Mar 2015 18:00:18 +0000 (10:00 -0800)]
Merge "Set the threads/tile_columns in vpx_temporal_svc_encoder."
Marco [Fri, 6 Mar 2015 17:51:21 +0000 (09:51 -0800)]
Set the threads/tile_columns in vpx_temporal_svc_encoder.
Change-Id: I794e8b64f26ebe68d7f18c3a11cb7c80e1cd3110
Yunqing Wang [Thu, 5 Mar 2015 17:44:18 +0000 (09:44 -0800)]
Modify the setting of transform skip flags in non-rd mode
While searching for the best mode in non-rd case, SSE of
a partition block is calculated and the transform size is set.
This patch rewrites the skip checking conditions based on
transform size instead of partition size to be more precise.
Small gains were seen in rtc set borg test (speed 6).
AVG PSNR: 0.087%, overall PSNR: 0.073%, SSIM: 0.146%.
No noticeable speed change.
Change-Id: I5603ca5339c784dfa02263f4005988ccd8c32f6e
Yaowu Xu [Fri, 6 Mar 2015 15:27:58 +0000 (07:27 -0800)]
Merge changes I1b972c94,I9c897d32
* changes:
Prevent invalid memory access
Use correct bsize for uv
Hangyu Kuang [Mon, 2 Mar 2015 21:59:03 +0000 (13:59 -0800)]
Only wait for previous frame's motion vector if needed.
Change-Id: Iecce685a33b64844446c0009f21bc85566d7469f
Yaowu Xu [Thu, 5 Mar 2015 22:57:11 +0000 (14:57 -0800)]
Prevent invalid memory access
Change-Id: I1b972c945274254d896d772d859840b2f8211b4f
hkuang [Thu, 5 Mar 2015 22:55:27 +0000 (14:55 -0800)]
Merge "Fix a tsan error bug in frame parallel decode."
Alex Converse [Thu, 5 Mar 2015 22:43:42 +0000 (14:43 -0800)]
Merge changes I219c287b,I6adee670
* changes:
Call encoder control before running ethread test.
Don't copy thread data for the main thread.
Alex Converse [Thu, 5 Mar 2015 22:43:38 +0000 (14:43 -0800)]
Merge "Fix misleading indentation."
Alex Converse [Thu, 5 Mar 2015 19:47:21 +0000 (11:47 -0800)]
Call encoder control before running ethread test.
The first value of video->frame() is 15 because of the source range used.
Change-Id: I219c287b500fa376160d6dcfb937092e577ec959
Deb Mukherjee [Thu, 5 Mar 2015 21:58:37 +0000 (13:58 -0800)]
Reduce default overshoot/undershoot params to 25
Reducing these values from 100 makes rate control tighter by
default.
Change-Id: I979409c93e6f19e128898f08ad581bf67eb0f824
Frank Galligan [Thu, 5 Mar 2015 21:58:26 +0000 (13:58 -0800)]
Merge "Use 8 threads by default for VP9 encodes"
Frank Galligan [Thu, 5 Mar 2015 21:58:21 +0000 (13:58 -0800)]
Merge "VP9: turn on tile-columns and frame-parallel-mode by default"
Alex Converse [Thu, 5 Mar 2015 21:54:44 +0000 (13:54 -0800)]
Merge "Don't inline cost_coeffs."
Adrian Grange [Thu, 5 Mar 2015 20:52:57 +0000 (12:52 -0800)]
Merge "Fix valgrind memcpy memory overlaps warning"
Alex Converse [Thu, 5 Mar 2015 20:39:02 +0000 (12:39 -0800)]
Don't inline cost_coeffs.
It was tiny when it was orginally marked INLINE. Forcing this function
to be inlined prevents the compiler from inlining its much smaller
callers.
No measurable speed impact, 28320 byte smaller libvpx.a
Change-Id: I6bf4c917157d15cbadb3cd3e20a9e82d35dc7d6f
Alex Converse [Thu, 5 Mar 2015 20:09:03 +0000 (12:09 -0800)]
Fix misleading indentation.
Change-Id: Ic82b039a3d42f9aa01b85a3a69facfaa84b43a53
Alex Converse [Thu, 5 Mar 2015 20:09:48 +0000 (12:09 -0800)]
Don't copy thread data for the main thread.
Change-Id: I6adee6704cacfeae0ed0b217a91095457d1be74a
Jingning Han [Thu, 5 Mar 2015 19:43:01 +0000 (11:43 -0800)]
Move pred_mv assign outside integral projection motion search
Change-Id: I040b066fdce08e2f05115a22ea808715aa147779
Jingning Han [Thu, 5 Mar 2015 17:25:16 +0000 (09:25 -0800)]
Merge "Move integral projection motion search to vp9_mcomp.c"
Yaowu Xu [Thu, 5 Mar 2015 16:20:35 +0000 (08:20 -0800)]
Use correct bsize for uv
Change-Id: I9c897d32af6c3a956bb6f424a74c12737727038a
Johann [Wed, 4 Mar 2015 23:34:55 +0000 (15:34 -0800)]
Declare function used by 'once' with 'void' parameters
Visual Studio is exceptionally picky about this:
vp9_reconintra.c(900): warning C4113: 'void (__cdecl *)()' differs in
parameter lists from 'void (__cdecl *)(void)'
[.build-x86_64-win64-vs10\vpx.vcxproj]
Change-Id: I564c7415f4608fd962be8c699d6133a996b545f7
hkuang [Wed, 4 Mar 2015 22:23:01 +0000 (14:23 -0800)]
Merge "Fix variable shadowing."
Adrian Grange [Wed, 4 Mar 2015 20:49:58 +0000 (12:49 -0800)]
Merge "Small rationalization of code in vp9_first_pass"
hkuang [Wed, 4 Mar 2015 19:40:20 +0000 (11:40 -0800)]
Fix variable shadowing.
Change-Id: I287c61ad4aa85654aef3c871a62131bcd0d891ac
hkuang [Wed, 4 Mar 2015 19:27:58 +0000 (11:27 -0800)]
Merge "Remove unnecessary loopfilter init."
Adrian Grange [Wed, 4 Mar 2015 18:54:10 +0000 (10:54 -0800)]
Merge "Make encoder buffer allocation dynamic"
Adrian Grange [Fri, 27 Feb 2015 23:52:58 +0000 (15:52 -0800)]
Small rationalization of code in vp9_first_pass
Change-Id: I87cc0e038171c60a957298827e312fead500f7fb
Jingning Han [Wed, 4 Mar 2015 18:47:56 +0000 (10:47 -0800)]
Merge "Use SAD value to set chroma cost flag"
Jingning Han [Wed, 4 Mar 2015 18:28:33 +0000 (10:28 -0800)]
Move integral projection motion search to vp9_mcomp.c
Make it a general purpose fast motion estimation function, to be
used in the mode search process.
Change-Id: Ib354cb0e664dc61c30c0b2314297835ee75b157a
Jingning Han [Wed, 4 Mar 2015 17:40:01 +0000 (09:40 -0800)]
Use SAD value to set chroma cost flag
This saves an extra 64x64 variance calculation and replaces two
32x32 variance functions with sad functions. The compression
performance change is unnoticeable.
Change-Id: I6d33868695664ec73b56c42945162ae61c484856
Jingning Han [Wed, 4 Mar 2015 17:01:33 +0000 (09:01 -0800)]
Merge "Properly handle the boundary blocks for integral projection search"
Adrian Grange [Wed, 18 Feb 2015 17:40:34 +0000 (09:40 -0800)]
Make encoder buffer allocation dynamic
Frame buffers are now allocated dynamically on-demand.
Entries in the reference frame map, cm->ref_frame_map,
may now be set to -1 (INVALID_IDX) to indicate that
there is not a valid reference buffer in that "slot".
All slots in the reference frame map are now initialized
to the empty state (-1) and each buffer is initialized
to have a reference count of 0.
Change-Id: Id1afe98de98db4ae8b2dfefed7889c3b28c68582
Deb Mukherjee [Wed, 4 Mar 2015 07:23:44 +0000 (23:23 -0800)]
Merge "dc quantizer fix for 32x32 transforms"
Jingning Han [Wed, 4 Mar 2015 03:04:34 +0000 (19:04 -0800)]
Merge "Scale the normalization factor depending on the block size"