Commit Graph

181 Commits

Author SHA1 Message Date
Juanjo 43f1708f8b - Added PSNR feature to libavcodec and ffmpeg. By now just Y PSNR until I'm
sure it works ok. Also it's slow, so use it only when you _really_ need to
measure quality.
- Fix libavcodec Makefile to enable profiling.

Originally committed as revision 314 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-26 22:14:27 +00:00
Michael Niedermayer 2b9ab1d54a fixing msmpeg4 decoding if fps < 16 (i thought it was a indicator for the ext header, its the fps indeed)
Originally committed as revision 313 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-22 19:19:01 +00:00
Juanjo e7774f11ac - Bug fix on RTYPE (rounding type) not being honoured by H.263+ decoder.
Originally committed as revision 312 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-22 03:35:41 +00:00
Juanjo 9e15ad28bd - Bug fixes in H.263+ Advanced INTRA Coding decoder.
- H.263+ should be able to decode streams with AIC now :)

Originally committed as revision 311 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-19 19:23:34 +00:00
Juanjo d140623fc4 - H.263+ decoder support for Advanded INTRA Coding (buggy)
Originally committed as revision 310 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-18 19:33:27 +00:00
Zdenek Kabelac d6a4c0b111 * some modifications to allow gcc to compile same code for -fPIC
- introduced MOVQ_WONE, MOVQ_WTWO
* unrolled put_pixels_clamped_mmx
* added .baling where the jump address would be compiled on odd address
* added some testing code to disable all these routines (for time measurement)

Originally committed as revision 308 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-18 09:40:05 +00:00
Zdenek Kabelac 38d171ebbe * simplified indexing
* ignoring error from msmpeg4_decode_block

Originally committed as revision 307 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-18 09:34:54 +00:00
Zdenek Kabelac af8793ae75 * simplified indexing
Originally committed as revision 306 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-18 09:32:11 +00:00
Zdenek Kabelac 8257bf0534 * eliminating duplicated code from MPV_common_end
* simplified indexing

Originally committed as revision 305 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-18 09:31:37 +00:00
Zdenek Kabelac 3232bc885d * using static instead of extern
Originally committed as revision 304 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-18 09:29:32 +00:00
Zdenek Kabelac b4cc15e75e * continue after error in msmpeg4_decode_mb - helps for some movie samples
Originally committed as revision 303 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-18 09:28:05 +00:00
Michael Niedermayer bb6f51aeab avoid overflow of picturenumber*fps*10000
bug found by Lennert Buytenhek <buytenh@gnu.org>

Originally committed as revision 302 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-18 01:58:00 +00:00
Michael Niedermayer 3696d2bb83 block_last_index was too large (in mpeg1 decoding)
Originally committed as revision 301 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-18 00:31:16 +00:00
Juanjo b4dbd87c04 - Minor changes on bitrate control
Originally committed as revision 300 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-15 20:46:37 +00:00
Michael Niedermayer d8cf5aea49 optimized the normal bitstream writer, its faster than the alternative one on p3 now ... lets hope its at least not slower on p4 & k7
Originally committed as revision 299 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-15 00:14:01 +00:00
Michael Niedermayer f004ca1c1b optimized encode_dc() (+2% speed on P3 for mpeg1 intra only encodings)
Originally committed as revision 298 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-13 15:26:28 +00:00
Michael Niedermayer 80feb2a23e optimize mpeg1_encode_block()
(makeing some constant arrays global, merging put_bits(), ...)

Originally committed as revision 297 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-13 04:19:10 +00:00
Michael Niedermayer 4f0f8bfce3 aligned bitstream writer (1% slower on p3 but perhaps its faster on p4?)
Originally committed as revision 296 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-12 22:43:26 +00:00
Michael Niedermayer 17592475b3 alternative bitstream writer (disabled by default, uncomment #define ALT_BISTREAM_WRITER in common.h if u want to try it)
Originally committed as revision 295 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-12 15:43:16 +00:00
Juanjo e03c341ef3 - More work on preliminary bit rate control, just to be able to get an
average variance for picture's MBs so we can adjust qscale on the MB layer.

Originally committed as revision 294 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-10 06:10:50 +00:00
Juanjo 37fbfd0a91 - Now the ME is done for the entire picture when enconding, the
DCT/Quantization is done after we have calculated all the MV of the picture.
- This is the preamble for a better bit rate control.

Originally committed as revision 293 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-10 01:56:50 +00:00
Juanjo 81401c1fbc - Fixes on RTP and GOB headers for H.263.
- RTP callback system implemented for H.263/H.263+.
- Bug fix on DC coefficients of H.263.
- H.263 decoder now returns PTS on picture number.

Originally committed as revision 292 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-09 16:59:56 +00:00
Arpi 3cb32e3d1f PATCH by Rik Snel <rsnel@cube.dyndns.org>
this includes the range for quantized dct coefficients in dct_quantize()  (-1023...1023)

Originally committed as revision 291 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-09 01:25:06 +00:00
Arpi 80e103d04c PATCH by Rik Snel <rsnel@cube.dyndns.org>
this patch enhances the jpeg header writer. It can be asked to omit
quantisation and huffman tables and it can write different horizontal and
vertical sampling factors. (the last thing is useless for libavcodec
itself (because libavcodec only handles YUV420P at ecoder level), but the
values are initialized so that operation of libavcodec is not impaired)

Originally committed as revision 290 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-09 01:23:41 +00:00
Michael Niedermayer fa778d53ad workaround gcc 2.95.2 bug
Originally committed as revision 289 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-09 00:38:44 +00:00
Michael Niedermayer 0b2540a0c1 avoid copying input when encoding non intra stuff too
Originally committed as revision 288 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-07 23:48:24 +00:00
Michael Niedermayer d8c2efac4b width or height %16 != 0 bugfix
Originally committed as revision 287 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-07 02:08:36 +00:00
Michael Niedermayer 6fe84b4327 use multiply instead of divides for DC prediction on X86
Originally committed as revision 286 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-02-05 22:51:23 +00:00
Arpi 5dfd24ac98 workaround gcc bug "m" (0)
avoiding useage of ebx

Originally committed as revision 283 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-28 19:34:16 +00:00
Zdenek Kabelac 320680d420 * temporal solution for shared lib compilation
* using ALPHA code optimalization for non ARCH_I386

Originally committed as revision 282 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-28 18:06:28 +00:00
Michael Niedermayer 2f349de286 (commit by michael)
mmx & mmx2 quantizer
c dct permutation bugfix
dont copy input on intra only encodings if it can be avoided
dont draw edges on intra only stuff

Originally committed as revision 281 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-27 13:30:18 +00:00
Arpi 580b82fa24 (m)jpeg pad/flush with 1 instead of 0, fix by Rik Snel <rsnel@cube.dyndns.org>
Originally committed as revision 280 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-26 00:22:25 +00:00
Michael Niedermayer 2ddeed44a5 (commit by michael)
motion-estimation on width/height not divisable through 16 files bugfix

Originally committed as revision 279 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-24 19:06:47 +00:00
Michael Niedermayer d2b3c3d7bb (commit by michael)
dc coeff isnt dequantized (bug reported by falk hueffner)

Originally committed as revision 278 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-24 15:25:11 +00:00
Michael Niedermayer 4e3269e87f (commit by michael)
quantizer range bugfix (mpeg1 & mpeg4 encoding)

Originally committed as revision 277 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-24 14:34:03 +00:00
Nick Kurshev c7997007be Small simple idct improvement for Alpha by Falk Hueffner <falk.hueffner@student.uni-tuebingen.de>
Originally committed as revision 276 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-21 17:46:42 +00:00
Nick Kurshev 1e98dffb7a Alpha optimizations by Falk Hueffner <falk.hueffner@student.uni-tuebingen.de>
Originally committed as revision 274 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-20 14:48:02 +00:00
Nick Kurshev 4bdd9157cc Cygwin's mangling by Felix Buenemann <atmosfear@users.sourceforge.net>
Originally committed as revision 273 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-20 14:30:34 +00:00
Michael Niedermayer 0a8d8945fd (commit by michael)
faster simple idct in MMX

Originally committed as revision 269 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-17 20:00:41 +00:00
Michael Niedermayer ae40484c1c (commit by michael)
bye bye weird al rounding bug ;)

Originally committed as revision 268 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-15 22:22:41 +00:00
Michael Niedermayer 5975626d6f (commit by michael)
emms

Originally committed as revision 267 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-15 05:02:26 +00:00
Michael Niedermayer 3d9fccbf0f (commit by michael)
draw_edges_mmx()

Originally committed as revision 266 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-15 04:07:29 +00:00
Michael Niedermayer 4973971e37 (commit by michael)
faster simple_idct in C

Originally committed as revision 265 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-14 04:39:59 +00:00
Michael Niedermayer 6f903d8e27 (commit by michael)
16-bit divide instead of 32-bit on x86 in msmpeg_pred_dc()

Originally committed as revision 264 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-14 04:34:52 +00:00
Michael Niedermayer 8bdbae01fe (commit by michael)
memset() in MMX

Originally committed as revision 263 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-14 04:29:19 +00:00
Michael Niedermayer 99180fe39d (commit by michael)
new msmpeg4 dequantizer did dequantize too early

Originally committed as revision 261 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-13 06:12:35 +00:00
Michael Niedermayer badaf88eeb (commit by michael)
dequantizers skip trailing zeros
msmpeg4 non-intra decoder has its dequantizer "build in" now

Originally committed as revision 260 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-13 04:59:37 +00:00
Michael Niedermayer 05f361f0c7 (commited by michael / arpi was crazy enough to give me his password)
fixed a bug in get_vlc() when using the ALT_BITSTREAM_READER (rv10 & mpeg12 decoders where not working)
some optimizations
 replaced a if() with a >>
 changed get_bits1() a bit so it hopefully is faster on p4

Originally committed as revision 259 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-12 20:21:51 +00:00
Arpi eede215681 rv10 sig11 fix by Michael Niedermayer <michaelni@gmx.at>
Originally committed as revision 258 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-12 14:36:40 +00:00
Michael Niedermayer 5b3438c630 different fix, s->mbintra_table used only if h263_pred set. - patch by Michael Niedermayer <michaelni@gmx.at>
Originally committed as revision 257 to svn://svn.ffmpeg.org/ffmpeg/trunk
2002-01-11 22:07:12 +00:00