Commit Graph

7441 Commits

Author SHA1 Message Date
Loren Merritt 45d9d61889 don't mark the delayed samples for upmixing if they haven't been downmixed
Originally committed as revision 14741 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-13 23:30:53 +00:00
Loren Merritt 3b6516f7e7 oops, non-normalized downmix to mono contaminated add_bias for non-simd float-to-int.
Originally committed as revision 14740 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-13 23:30:28 +00:00
Ramiro Polla 06cddea060 mlp: Cosmetics: Close comments in an extra line and
remove empty lines between doxy comments and function declarations.

Originally committed as revision 14735 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-13 19:43:18 +00:00
Ramiro Polla ce15710f55 mlp: Split common code from parser and decoder to be used by encoder.
Originally committed as revision 14733 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-13 18:47:03 +00:00
Diego Biurrun e7327dbba7 Remove unused variable.
Originally committed as revision 14732 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-13 18:33:25 +00:00
Ramiro Polla ee5b34d56e mlp_parser: Initialize crc data in its own function.
Originally committed as revision 14722 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-13 02:44:38 +00:00
Ramiro Polla 5dae0dd22c mlpdec: Put some doxy comments to the right of vars instead of interleaved.
Originally committed as revision 14721 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-13 01:41:06 +00:00
Ramiro Polla f53acb7b17 mlpdec: Split channel parameters from context into their own struct.
Originally committed as revision 14720 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-13 01:36:01 +00:00
Loren Merritt 557ac0c448 optimize ac3_downmix.
1.3x faster 5.1->stereo, 1.9x faster 5.1->mono.

Originally committed as revision 14719 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-12 23:45:46 +00:00
Ramiro Polla db01fa131b mlpdec: whitespace and line-breaking cosmetics.
Originally committed as revision 14717 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-12 18:05:58 +00:00
Ramiro Polla d6aa052dc5 mlpdec: Split filter parameters from context into their own struct.
Originally committed as revision 14716 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-12 17:53:59 +00:00
Peter Ross 3b4b113781 Fix CODEC_ID_PCM_U8 decoder output size calculation to support odd-number of samples.
Originally committed as revision 14712 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-12 12:32:40 +00:00
Loren Merritt 72745cff20 use float_to_int16_interleave in ac3
Originally committed as revision 14706 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-12 03:01:17 +00:00
Loren Merritt 916d5d6c32 use imdct_half in ac3
Originally committed as revision 14705 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-12 01:30:24 +00:00
Loren Merritt 862b98d42c cosmetics in dsp init
Originally committed as revision 14704 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-12 00:51:45 +00:00
Loren Merritt d46ac5bfde mdct wrapper function to match fft
Originally committed as revision 14703 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-12 00:38:30 +00:00
Loren Merritt 0a570e826d remove mdct tmp buffer
Originally committed as revision 14702 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-12 00:36:36 +00:00
Loren Merritt e8dd7b0c9a remove a float->double->float conversion.
1.8x faster imdct_c with fpmath=sse, no change with x87

Originally committed as revision 14701 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-12 00:34:08 +00:00
Loren Merritt 46803f4f67 optimize imdct_half:
remove tmp buffer.
skip fft reinterleave pass, leaving data in a format more convenient for simd.
merge post-rotate with post-reorder.

Originally committed as revision 14700 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-12 00:33:34 +00:00
Loren Merritt 49c0dd754c indent
Originally committed as revision 14699 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-12 00:27:21 +00:00
Loren Merritt 5d0ddd1a9f split-radix FFT
c is 1.9x faster than previous c (on various x86 cpus), sse is 1.6x faster than previous sse.

Originally committed as revision 14698 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-12 00:26:58 +00:00
Loren Merritt bafad220a7 import yasm macros from x264
Originally committed as revision 14697 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-11 23:54:09 +00:00
Robert Swain 9edae4ad81 Synchronise code with AAC decoder in SoC
Originally committed as revision 14695 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-11 11:22:48 +00:00
Robert Swain 9cc04edff9 More OKed hunks of the AAC decoder from SoC
Originally committed as revision 14694 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-11 11:16:06 +00:00
Roman Shaposhnik 3f66d16821 Fixing the units in DV50 tables (both coordinates are
supposed to be in units of 8 pixels) and simplifying
the calculations of y_ptr and c_offset.

Originally committed as revision 14693 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-11 10:14:46 +00:00
Peter Ross 5e8ecfde50 Apply PCM ENCODE/DECODE() macros to the S/U,8/24/32,LE/BE PCM codecs.
Originally committed as revision 14692 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-11 09:52:17 +00:00
Roman Shaposhnik d557838cfb Coalescing the un-weighting and de-quantization steps for faster
processing

Originally committed as revision 14691 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-11 09:35:58 +00:00
Kostya Shishkov 76c655fb8f Simplify colour conversion in Fraps as suggested by Michael
Originally committed as revision 14690 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-11 05:11:05 +00:00
Michael Niedermayer 2967101173 The deblock filter does not need all the intra availability stuff so do
not recalculate it.

Originally committed as revision 14689 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-11 03:08:02 +00:00
Michael Niedermayer faa7e39466 Simplify constrained_intra_pred code in fill_caches().
Originally committed as revision 14688 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-11 02:52:14 +00:00
Michael Niedermayer d1d10e9125 Support MBAFF + constrained intra prediction.
(no i would not have tried to implement this had i known what mess it is)
fixes at least:
CAMACI3_Sony_C

Originally committed as revision 14687 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-11 02:21:33 +00:00
Robert Swain 697c7cd44b Bump minor version as av_log_missing_feature() was added to public API in
r14685

Originally committed as revision 14686 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-10 22:18:46 +00:00
Justin Ruggles ea779d91ca Add a generic function to lavc to log messages about missing features.
Patch by Justin Ruggles (justin ruggles gmail com)

Originally committed as revision 14685 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-10 20:29:43 +00:00
Kostya Shishkov af1761915b R and B components are stored as a differences to G component in Fraps v5.
This fixes roundup issue 574.

Originally committed as revision 14684 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-10 18:15:38 +00:00
Michael Niedermayer beca9a286c Mask bits of reference indexes properly to avoid negative numbers entangling
each other.
Fixes at least:
CAMA1_TOSHIBA_B
cama1_vtc_c
CAMA3_Sand_E
cama3_vtc_b
CAMASL3_Sony_B
CVMA1_TOSHIBA_B
CVMAQP3_Sony_D
cvmp_mot_mbaff0_full_B
FRExt/HCAMFF1_HHI
FRExt/HCHP3_HHI_A
FRExt/HVLCMFF0_Sony_B

Originally committed as revision 14683 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-10 16:10:39 +00:00
Michael Niedermayer e248cb601e Remove a few more lines of unneeded code in fill_caches() for MBAFF.
Also document why the remaining related part is apparently needed.

Originally committed as revision 14682 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-10 15:39:46 +00:00
Michael Niedermayer d50cdd82de Use the correct reference tables for interlaced MBAFF MBs in the loop filter.
Fixes at least
CAMP_MOT_MBAFF_L30
CAMP_MOT_MBAFF_L31
FRExt/FREXT01_JVC_D

Originally committed as revision 14680 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-10 02:38:51 +00:00
Michael Niedermayer 0281d32550 Do not execute the part of fill_caches() for the loopfilter that overwrites
reference indexes and motion vectors.
Fixes at least:
CVMAQP2_Sony_G

Originally committed as revision 14678 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-10 00:38:41 +00:00
Michael Niedermayer 3d11b8ce13 Remove more code that has become useless after the pairless MBAFF loop
filter cleanup.

Originally committed as revision 14676 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-09 18:11:25 +00:00
Robert Swain cc0591dab0 Sync already committed code with that in SoC and commit more OKed hunks of code
Originally committed as revision 14674 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-09 10:46:27 +00:00
Michael Niedermayer 5f7f9719ac Remove odd pairwise deblock filtering applied to MBAFF MBs.
Originally committed as revision 14673 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-09 01:17:34 +00:00
Stefano Sabatini 96afa6be0b Doxygenate avcodec_version().
Originally committed as revision 14666 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-08 18:34:27 +00:00
Baptiste Coudurier ec7e2582ae fix 4:2:2 psnr
Originally committed as revision 14662 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-08 02:56:44 +00:00
Peter Ross 3ee573a300 Add simpler/cleaner/faster F32BE encoding/decoding.
Originally committed as revision 14660 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-07 09:32:10 +00:00
Peter Ross ff66caab40 Simplify PCM codec; change 'n' in pcm_decode_frame() to equal "total number of samples".
Originally committed as revision 14659 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-07 09:23:56 +00:00
Peter Ross 13dbd9b705 Prevent DAUD PCM encoder from fetching values outside of ff_reverse[] array bounds when input sample values are < 0.
Originally committed as revision 14658 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-07 08:16:42 +00:00
Vitor Sessak b888abe1be Cosmetics: rename loop counters in eval_refl()
Originally committed as revision 14653 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-06 21:19:30 +00:00
Vitor Sessak bdb34c9f5e Simplify eval_refl(): do not store a buffer value in a var when the buffer could be used directly
Originally committed as revision 14652 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-06 21:16:24 +00:00
Vitor Sessak aa425ada54 Simplify eval_refl(): s/b == 0/!b/
Originally committed as revision 14651 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-06 21:04:40 +00:00
Vitor Sessak 40812fdcf1 Simplify range checking in eval_refl(). Before the
condition b == 0 could never be true, because it would
have already been caught in the previous ifs. Also the
condition b == 0x1000 could never be true (it would
triggered the return).

Originally committed as revision 14650 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-06 21:02:56 +00:00