[svn] / branches / dev-api-4 / xvidcore / changelog.txt Repository:
ViewVC logotype

Diff of /branches/dev-api-4/xvidcore/changelog.txt

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 1172, Sat Oct 4 16:01:18 2003 UTC revision 1173, Sun Oct 5 00:11:18 2003 UTC
# Line 5  Line 5 
5  # relevent  # relevent
6  #  #
7    
8    2003-10-04 16:04:30 GMT                                         patch-63
9    
10        Summary:
11          Removed legacy 2pass code from vfw
12        Revision:
13          xvidcore--devapi4--1.0--patch-63
14    
15        Removed legacy 2pass code from vfw
16    
17    
18        removed files:
19         vfw/src/.arch-ids/2pass.c.id vfw/src/.arch-ids/2pass.h.id
20         vfw/src/2pass.c vfw/src/2pass.h
21    
22    
23    2003-10-04 00:41:38 GMT                                         patch-62
24    
25        Summary:
26          Working VFW mingw/cygwin build system.
27        Revision:
28          xvidcore--devapi4--1.0--patch-62
29    
30        This patch fixes the VFW building process. Now it should work out
31        of the box using these steps:
32         # cd ${xvidcore}
33         # cd build/generic
34         # ./bootstrap.sh <-- only needed for CVS checkouts.
35         # ./configure
36         # make
37         # cd ../../vfw/bin
38         # make -f Makefile.cygwin
39    
40        Then install as usual clicking on the inf file or "make install"
41        in the vfw/bin dir.
42    
43        modified files:
44         vfw/bin/Makefile.cygwin vfw/src/config.rc vfw/src/debug.h
45         vfw/src/driverproc.c
46    
47    
48    2003-10-03 17:00:53 GMT                                         patch-61
49    
50        Summary:
51          Fixes for alternate scan and interlacing support.
52        Revision:
53          xvidcore--devapi4--1.0--patch-61
54    
55        Fixes from CVS (by sysKin) for:
56         - added alternate scan support with VHQ
57         - fixed interlacing support in s/b-frames. May fix a potential problem
58           as field_pred struct field seemed not to be initialized anywhere.
59           As it's not supported yet, write an hardcoded 0 bit.
60    
61        Fixes from me for the fixes from syKin:
62         - scan_table effectively used in MBCodingBVOP
63         - Block_CalcBits(Intra) fixes to data->scan_table (implies prototype
64           change and code modification every where the functions were used)
65    
66        I also increased BS version as it might result in different bitstreams
67        It's now at version 20.
68    
69        modified files:
70         src/bitstream/mbcoding.c src/bitstream/mbcoding.h
71         src/decoder.c src/encoder.c src/motion/estimation.h
72         src/motion/estimation_rd_based.c src/xvid.h
73    
74    
75    2003-10-03 15:41:37 GMT                                         patch-60
76    
77        Summary:
78          Removed BIGLUT support.
79        Revision:
80          xvidcore--devapi4--1.0--patch-60
81    
82        Remobed legacy code for BIGLUT support. It was unused and RD based
83        Motion Estimation was not even compatible with this type of VLC coding.
84    
85        modified files:
86         build/generic/configure.in src/bitstream/mbcoding.c
87         src/bitstream/mbcoding.h src/prediction/mbprediction.c
88    
89    
90    2003-10-03 13:25:17 GMT                                         patch-59
91    
92        Summary:
93          Bugfix for PFrames+ Ext Search.
94        Revision:
95          xvidcore--devapi4--1.0--patch-59
96    
97        In Qpel mode, the code was doing a diamond search for wrong predictors.
98        This resulted in poor performance as the diamond search was sitting there
99        for sometime.
100    
101        modified files:
102         src/motion/estimation.h src/motion/estimation_pvop.c
103    
104    
105    2003-10-02 16:50:51 GMT                                         patch-58
106    
107        Summary:
108          Added VFW makefile for cygwin/minsys
109        Revision:
110          xvidcore--devapi4--1.0--patch-58
111    
112        Added VFW makefile for cygwin/minsys. I can't test it so it is probably not
113        right out of the box. Waiting for feedback in order to fix it.
114    
115        new files:
116         vfw/bin/.arch-ids/Makefile.cygwin.id
117         vfw/bin/.arch-ids/Makefile.inc.id vfw/bin/Makefile.cygwin
118         vfw/bin/Makefile.inc vfw/src/w32api/.arch-ids/=id
119         vfw/src/w32api/.arch-ids/vfw.h.id vfw/src/w32api/vfw.h
120    
121        new directories:
122         vfw/src/w32api vfw/src/w32api/.arch-ids
123    
124    
125    2003-10-02 13:35:15 GMT                                         patch-57
126    
127        Summary:
128          Cleaned up the lumimasking code.
129        Revision:
130          xvidcore--devapi4--1.0--patch-57
131    
132        The lumimasking code was not very plugin oriented as it has been ported
133        from old XviD versions. This patch cleans up the code and integrates it
134        better with plugin design. No changes done in teh functionnal code.
135    
136        modified files:
137         src/plugins/plugin_lumimasking.c
138    
139    
140    2003-10-01 23:07:07 GMT                                         patch-56
141    
142        Summary:
143          Cleaned up trailing space chars.
144        Revision:
145          xvidcore--devapi4--1.0--patch-56
146    
147        The kind of patch we would love to avoid as they make merging
148        a nightmare while they're kind of useless patches.
149    
150        Applied sed 's/[ \t]*$//' to all c/h files.
151    
152        modified files:
153         src/bitstream/bitstream.c src/bitstream/bitstream.h
154         src/bitstream/cbp.c src/bitstream/mbcoding.c src/dct/idct.c
155         src/dct/simple_idct.c src/decoder.c src/decoder.h
156         src/encoder.c src/encoder.h src/global.h
157         src/image/colorspace.c src/image/colorspace.h src/image/font.c
158         src/image/image.c src/image/image.h src/image/interpolate8x8.c
159         src/image/interpolate8x8.h src/image/qpel.c src/image/qpel.h
160         src/image/reduced.c src/motion/estimation_rd_based.c
161         src/motion/gmc.c src/motion/gmc.h src/motion/motion.h
162         src/motion/sad.c src/plugins/plugin_2pass1.c
163         src/plugins/plugin_2pass2.c src/plugins/plugin_fixed.c
164         src/plugins/plugin_psnr.c src/plugins/plugin_single.c
165         src/portab.h src/prediction/mbprediction.c
166         src/utils/mbtransquant.c src/utils/timer.c src/xvid.c
167         src/xvid.h
168    
169    
170    2003-09-30 18:10:18 GMT                                         patch-55
171    
172        Summary:
173          Code cleanups.
174        Revision:
175          xvidcore--devapi4--1.0--patch-55
176    
177        It's been a while since the last ISOC89 conformance cleanup.
178    
179        Using the following switches help a lot :-)
180         -Wall -Wsign-compare -Wredundant-decls -Wunreachable-code -Wnested-externs \
181         -ansi
182    
183        Result: 0 warning/0 error
184    
185        modified files:
186         src/bitstream/vlc_codes.h src/global.h src/image/qpel.h
187         src/motion/estimation.h src/motion/estimation_bvop.c
188         src/motion/estimation_gmc.c src/motion/estimation_pvop.c
189         src/motion/estimation_rd_based.c src/motion/gmc.c
190         src/motion/gmc.h src/motion/motion_inlines.h
191         src/motion/vop_type_decision.c
192    
193    
194    2003-09-29 00:31:32 GMT                                         patch-54
195    
196        Summary:
197          Memory leakage fixes.
198        Revision:
199          xvidcore--devapi4--1.0--patch-54
200    
201        The pEnc->queue was allocated but not freed when bframes == 0.
202        And queue images were not freed as well.
203    
204        modified files:
205         examples/Makefile src/encoder.c src/image/image.c
206         src/utils/mem_align.c
207    
208    
209    2003-09-28 16:45:02 GMT                                         patch-53
210    
211        Summary:
212          Fixes the unitialized mcsel bit in RD based ME
213        Revision:
214          xvidcore--devapi4--1.0--patch-53
215    
216        Fixes the unitialized mcsel bit in RD based ME
217    
218    
219        modified files:
220         src/motion/estimation_rd_based.c
221    
222    
223    2003-09-28 01:00:06 GMT                                         patch-52
224    
225        Summary:
226          Fix the XviD constant version initialization
227        Revision:
228          xvidcore--devapi4--1.0--patch-52
229    
230        Fix the XviD constant version initialization
231    
232    
233        modified files:
234         src/xvid.h
235    
236    
237    2003-09-28 00:47:05 GMT                                         patch-51
238    
239        Summary:
240          Fix to the build system (the come back).
241        Revision:
242          xvidcore--devapi4--1.0--patch-51
243    
244        Ok this one fixes the way we build the targets. The VPATH thingy really
245        works with the library targets. My understanding of the VPATH mechanism
246        was wrong. Now it should be ok (i promess).
247    
248        modified files:
249         build/generic/Makefile
250    
251    
252    2003-09-27 11:45:18 GMT                                         patch-50
253    
254        Summary:
255          Small fix to previous patch.
256        Revision:
257          xvidcore--devapi4--1.0--patch-50
258    
259        A pair ofdouble quotes prevented "make" to sort out the VPATH
260        dependences in "=build". This resulted in compiling all the
261        sources, all the time, whether a file changed or not.
262    
263        modified files:
264         build/generic/Makefile
265    
266    
267    2003-09-26 22:39:44 GMT                                         patch-49
268    
269        Summary:
270          Updated the build files for *nix.
271        Revision:
272          xvidcore--devapi4--1.0--patch-49
273    
274        - Fixed MacOSX build (w/o module option). The subversion was not right
275          it was just minor version though it has to be major.minor.
276        - Fixed bootstrap.sh for MacOSX environment, it now looks for
277          glibtoolize if libtoolize is not found.
278        - The unified Makefile now builds XviD out of source tree in the directory
279          =build. It's cleaner, and clashes much less w/ arch/tla source linting.
280        - Tuned the tagging regexps so:
281          + autoconf files are ignored (considered backup => not erased, not copied
282            and not commited/imported)
283          + =build is also considered backup.
284    
285        modified files:
286         build/generic/Makefile build/generic/bootstrap.sh
287         build/generic/configure.in build/generic/platform.inc.in
288         build/generic/sources.inc examples/Makefile
289         {arch}/=tagging-method
290    
291    
292    2003-09-24 01:38:03 GMT                                         patch-48
293    
294        Summary:
295          Bug fix to decoder (mcsel/acpred bits swapped)
296        Revision:
297          xvidcore--devapi4--1.0--patch-48
298    
299        As reported here:
300        http://www.xvid.org/modules.php?op=modload&name=phpBB2&file=viewtopic&t=1513&highlight=
301    
302        <quote>
303        in the spec, 'there is 'mcsel' before 'ac_pred_flag'.
304        however, if you see the code, it's changed.
305        </quote>
306    
307        We were doing the opposite, ac_pred before mcsel.
308    
309        modified files:
310         src/decoder.c
311    
312    
313    2003-09-11 17:19:35 GMT                                         patch-47
314    
315        Summary:
316          Small fix to GMC+QuarterPel -- BS version bumped to 19
317        Revision:
318          xvidcore--devapi4--1.0--patch-47
319    
320        During the split up, a line has been disabled that prevented good
321        quartelpel+GMC. Fixed.
322    
323        This fix + patch-43 are a good reason to bump up the bitstream
324        version to 19.
325    
326        modified files:
327         src/motion/gmc.h src/xvid.h
328    
329    
330    2003-09-11 17:11:28 GMT                                         patch-46
331    
332        Summary:
333          Build process fix for MacOSX+module option
334        Revision:
335          xvidcore--devapi4--1.0--patch-46
336    
337        This patch fixes the install rule of the MacOSX module style library.
338        It adds a PRE_SHARED_LIB == SHARED_LIB for all platforms except
339        MacOSX that uses it a different way, and exploits the fact
340        $(SPECIFIC_LDFLAGS) is the last var of the build line to insert a
341        command for post linking the right .so file.
342    
343        modified files:
344         build/generic/Makefile build/generic/configure.in
345         build/generic/platform.inc.in
346    
347    
348    2003-09-11 13:56:40 GMT                                         patch-45
349    
350        Summary:
351          Tree cleanup and build files updated.
352        Revision:
353          xvidcore--devapi4--1.0--patch-45
354    
355        * Win32 files switched to Unix format.
356          Ok, on IRC, we sorted out what was b0rking the project files each
357          time I commit them in DOS format. In fact the Unix CVS does upload
358          DOS format files if i commit them, and then Win32 CVS users get
359          double \r\n files. So the best is to use unix format in CVS. This
360          way, the checked out versions are right for Win32 users.
361    
362          NB: this imply a unix2dos conversion when doing a release from a
363              Unix box. I think it's better than current policy, but automatic
364              tarballs/zip of the tree should take care of that.
365    
366        * odivx and xvid_stat examples removed. Associated project files have
367          been removed as well.
368    
369        * The divx4 compatibility layer has been removed. Associated options
370          in the configure script removed. libxvidcore.def has no more need
371          to be an autoconf generated file.
372    
373        * rawdec removed. It has never been used and will never be.
374    
375        * Added major api appending to the macosx module build.
376    
377        removed files:
378         build/win32/.arch-ids/odivx_enc_dec.dsp.id
379         build/win32/.arch-ids/xvid_stat.dsp.id
380         build/win32/odivx_enc_dec.dsp build/win32/xvid_stat.dsp
381         examples/.arch-ids/odivx_enc_dec.c.id
382         examples/.arch-ids/xvid_stat.c.id examples/odivx_enc_dec.c
383         examples/xvid_stat.c rawdec/.arch-ids/=id
384         rawdec/.arch-ids/rawdec.c.id rawdec/.arch-ids/rawdec.dsp.id
385         rawdec/rawdec.c rawdec/rawdec.dsp src/.arch-ids/divx4.c.id
386         src/.arch-ids/divx4.h.id src/divx4.c src/divx4.h
387    
388        modified files:
389         build/generic/Makefile build/generic/configure.in
390         build/generic/libxvidcore.def build/win32/libxvidcore.dsp
391         build/win32/xvid_decraw.dsp build/win32/xvid_encraw.dsp
392         build/win32/xvidcore.dsw
393    
394        renamed files:
395         build/generic/.arch-ids/libxvidcore.def.in.id
396           ==> build/generic/.arch-ids/libxvidcore.def.id
397         build/generic/libxvidcore.def.in
398           ==> build/generic/libxvidcore.def
399    
400        removed directories:
401         rawdec rawdec/.arch-ids
402    
403    
404    2003-09-11 12:59:19 GMT                                         patch-44
405    
406        Summary:
407          Replaced malloc.h header file with stdlib.h
408        Revision:
409          xvidcore--devapi4--1.0--patch-44
410    
411        Replaced malloc.h header file with stdlib.h
412    
413    
414        modified files:
415         src/plugins/plugin_lumimasking.c
416    
417    
418    2003-09-10 22:33:04 GMT                                         patch-43
419    
420        Summary:
421          Fixed a problem for 'power of 2' framerates
422        Revision:
423          xvidcore--devapi4--1.0--patch-43
424    
425        Fixed a problem for 'power of 2' framerates
426    
427    
428        modified files:
429         src/bitstream/bitstream.c
430    
431    
432    2003-09-10 21:57:12 GMT                                         patch-42
433    
434        Summary:
435          Motion Estimation module splitting.
436        Revision:
437          xvidcore--devapi4--1.0--patch-42
438    
439        The motion estimation module was the biggest file of the source
440        tree. After some previous attempts, sysKin decided to split it
441        up again. This time he's done it right.
442    
443        This split up just changes the organization of ME functions
444        inside different files. This should help a bit in keeping the
445        motion estimation being manageable by a normal human ;-)
446    
447        Here is the splitting logic quoted from sysKin's email to
448        xvid-devel:
449         - estimation.h: header files #included in all ME modules:
450                         + deftypes,
451                         + macros
452                         + constants
453                         NB: no code.
454         - estimation_bvop.c: motion estimation for b-vops.
455                              everything in it :)
456         - estimation_common.c: functions shared among all ME
457                                modules:
458                                + diamonds
459                                + subpel
460                                + refinement
461                                + picture
462                                + manipulation
463                                + tables
464                                + ...  etc.
465         - estimation_gmc.c: gruel's GME code
466         - estimation_pvop.c: ME for p-vops. Also SAD-based mode
467                              decision
468         - estimation_rd_based.c: everything R-D-based: mode
469                                  decison (including _Fast) and ME.
470         - gmc.c, gmc.h: no change.
471    
472        new files:
473         src/motion/.arch-ids/estimation.h.id
474         src/motion/.arch-ids/estimation_bvop.c.id
475         src/motion/.arch-ids/estimation_common.c.id
476         src/motion/.arch-ids/estimation_gmc.c.id
477         src/motion/.arch-ids/estimation_pvop.c.id
478         src/motion/.arch-ids/estimation_rd_based.c.id
479         src/motion/.arch-ids/motion_inlines.h.id
480         src/motion/.arch-ids/vop_type_decision.c.id
481         src/motion/estimation.h src/motion/estimation_bvop.c
482         src/motion/estimation_common.c src/motion/estimation_gmc.c
483         src/motion/estimation_pvop.c src/motion/estimation_rd_based.c
484         src/motion/motion_inlines.h src/motion/vop_type_decision.c
485    
486        removed files:
487         src/motion/.arch-ids/motion_est.c.id
488         src/motion/.arch-ids/motion_est.h.id
489         src/motion/.arch-ids/smp_motion_est.c.id
490         src/motion/.arch-ids/smp_motion_est.h.id
491         src/motion/motion_est.c src/motion/motion_est.h
492         src/motion/smp_motion_est.c src/motion/smp_motion_est.h
493    
494        modified files:
495         build/generic/sources.inc build/win32/libxvidcore.dsp
496         src/bitstream/mbcoding.h src/motion/gmc.c src/motion/gmc.h
497         src/motion/motion.h src/motion/motion_comp.c src/motion/sad.c
498         src/motion/sad.h src/prediction/mbprediction.c
499         src/prediction/mbprediction.h src/utils/mbfunctions.h
500    
501    
502    2003-09-10 00:40:44 GMT                                         patch-41
503    
504        Summary:
505          Cleanups and fix to (trellis+thresholding) logic
506        Revision:
507          xvidcore--devapi4--1.0--patch-41
508    
509        Cleanups to some functions (loop unrolling, call to functions
510        through function array pointers)...
511    
512        Fix to the trellis+thresholding logic. It was comparing the
513        return value of trellis with a threshold but the trellis function
514        returns the last non zero coeff index... this was basically
515        comparing apples with oranges... funny but wrong. Trellis now
516        returns the sum of absolute coeffs, so the comparison is logical.
517        Btw, as discussed on the devel ML, this is probably uneeded as
518        trellis does an RD optimized coeff distribution.
519    
520        modified files:
521         src/utils/mbtransquant.c
522    
523    
524    2003-09-09 13:13:58 GMT                                         patch-40
525    
526        Summary:
527          Missing ressource for dshow frontend
528        Revision:
529          xvidcore--devapi4--1.0--patch-40
530    
531        Missing ressource for dshow frontend
532    
533    
534        new files:
535         dshow/src/.arch-ids/XviD_logo.bmp.id dshow/src/XviD_logo.bmp
536    
537    
538    2003-09-08 11:02:10 GMT                                         patch-39
539    
540        Summary:
541          Small fixes for fast mode decision
542        Revision:
543          xvidcore--devapi4--1.0--patch-39
544    
545        Small fixes for fast mode decision
546    
547    
548        modified files:
549         src/motion/motion_est.c src/xvid.h
550    
551    
552    2003-09-05 23:45:48 GMT                                         patch-38
553    
554        Summary:
555          New RD mode decision and subpel refinement.
556        Revision:
557          xvidcore--devapi4--1.0--patch-38
558    
559        New stuff from michael. It deals with mode decision and subpel
560        refinement. Integration of these new flags are not settled.
561    
562        Wait and see. Further testing is needed.
563    
564        modified files:
565         src/motion/motion_est.c src/motion/motion_est.h src/xvid.h
566    
567    
568    2003-08-29 13:56:30 GMT                                         patch-37
569    
570        Summary:
571          Still more ME tuning
572        Revision:
573          xvidcore--devapi4--1.0--patch-37
574    
575        Still more ME tuning
576    
577    
578        modified files:
579         src/motion/motion_est.c
580    
581    
582    2003-08-28 12:43:22 GMT                                         patch-36
583    
584        Summary:
585          Removed expanded the cvs Id field
586        Revision:
587          xvidcore--devapi4--1.0--patch-36
588    
589        Removed expanded the cvs Id field
590    
591    
592        modified files:
593         src/image/x86_asm/qpel_mmx.asm
594    
595    
596    2003-08-28 12:39:44 GMT                                         patch-35
597    
598        Summary:
599          More motion est cleanup and bugfixes.
600        Revision:
601          xvidcore--devapi4--1.0--patch-35
602    
603        Still more bugfixes, cleanups and improvements to the Motion Est
604        by sysKin
605    
606        modified files:
607         src/motion/motion_est.c src/motion/motion_est.h
608    
609    
610    2003-08-26 13:57:39 GMT                                         patch-34
611    
612        Summary:
613          Added final bits of Aspect Ratio flag.
614        Revision:
615          xvidcore--devapi4--1.0--patch-34
616    
617        Peter did think of the AR flag since the very beginning of devapi4.
618        He just forgot to code the final bits to effectively write it to
619        the bitstream. This patch adds these missing final bits so XviD now
620        reads and writes AR flags.
621    
622        modified files:
623         ./src/bitstream/bitstream.c ./src/encoder.c ./src/encoder.h
624         ./src/xvid.h
625    
626    
627    2003-08-25 16:41:09 GMT                                         patch-33
628    
629        Summary:
630          Small motion estimation cleanup.
631        Revision:
632          xvidcore--devapi4--1.0--patch-33
633    
634        Cleanups from sysKin.
635    
636        modified files:
637         ./src/motion/motion_est.c ./src/motion/motion_est.h
638    
639    
640    2003-08-25 14:59:28 GMT                                         patch-32
641    
642        Summary:
643          Frame padding bug.
644        Revision:
645          xvidcore--devapi4--1.0--patch-32
646    
647        We were edging the image repeating pixels from the image directly,
648        but the standard says we must repeat from a 16 pixel boundary.
649    
650        See Chapter 7.6.4 of the standard.
651    
652        modified files:
653         ./src/image/image.c
654    
655    
656    2003-08-23 15:07:44 GMT                                         patch-31
657    
658        Summary:
659          New Qpel code.
660        Revision:
661          xvidcore--devapi4--1.0--patch-31
662    
663        Isibaar commited a new piece of QPel code that seems to be optimized
664        for ia32(mmx) architectures. I had to clean it up a bit to make it
665        respectful of architecture separations. This code is disabled for
666        non ia32 arch, a comment mention it's only faster on ia32... i wonder
667        if it's true, some tests have to be done on sourceforge compile farm
668        in order to confirm that.
669    
670        Compared to the bared CVS commit, this patch includes:
671         - fixes the unix build.
672         - better architecture separation.
673         - CodingStyle respected.
674    
675        new files:
676         ./src/image/.arch-ids/qpel.c.id
677         ./src/image/.arch-ids/qpel.h.id ./src/image/qpel.c
678         ./src/image/qpel.h
679         ./src/image/x86_asm/.arch-ids/qpel_mmx.asm.id
680         ./src/image/x86_asm/qpel_mmx.asm
681    
682        modified files:
683         ./build/generic/sources.inc ./build/win32/libxvidcore.dsp
684         ./src/encoder.c ./src/motion/motion_comp.c ./src/xvid.c
685    
686    
687    2003-08-22 13:20:36 GMT                                         patch-30
688    
689        Summary:
690          sad32v does really what it's expected (ie 32x32 SAD :-)
691        Revision:
692          xvidcore--devapi4--1.0--patch-30
693    
694        sad32v does really what it's expected (ie 32x32 SAD :-)
695    
696    
697        modified files:
698         ./src/motion/sad.c
699    
700    
701    2003-08-18 19:00:49 GMT                                         patch-29
702    
703        Summary:
704          64bit fix.
705        Revision:
706          xvidcore--devapi4--1.0--patch-29
707    
708        The interpolation code was unsafe on 64bit platforms, the offset was
709        badly sized, resulting in segfaults.
710    
711        modified files:
712         ./src/image/interpolate8x8.h
713    
714    
715    2003-08-17 14:08:48 GMT                                         patch-28
716    
717        Summary:
718          Greyscale mode fixes.
719        Revision:
720          xvidcore--devapi4--1.0--patch-28
721    
722        We were missing some greyscale tests in the encoder loop... noticeably
723        in the PVOP function when coding an intra block, and in BVOP function
724        when coding all types.
725    
726        I added the cbp trick in the cases discussed above.
727    
728        modified files:
729         ./src/encoder.c
730    
731    
732    2003-08-13 11:47:33 GMT                                         patch-27
733    
734        Summary:
735          Forgotten bit for IA64 separation
736        Revision:
737          xvidcore--devapi4--1.0--patch-27
738    
739        Forgotten bit for IA64 separation
740    
741    
742        modified files:
743         ./src/image/interpolate8x8.h
744    
745    
746    2003-08-11 15:42:30 GMT                                         patch-26
747    
748        Summary:
749          Some qpel changes (sync with Isibaar)
750        Revision:
751          xvidcore--devapi4--1.0--patch-26
752    
753        Some qpel changes (sync with Isibaar)
754    
755    
756        modified files:
757         ./examples/xvid_encraw.c
758    
759    
760    2003-08-11 15:30:04 GMT                                         patch-25
761    
762        Summary:
763          Better architecture separation.
764        Revision:
765          xvidcore--devapi4--1.0--patch-25
766    
767        Architecture depending functions were declared whatever arch you
768        were compiling. This patch fixes that.
769    
770        I also removed the simple_idct hack in decoder.c as it was simply
771        not used. Better not have ugly code in there.
772    
773        xvid_bench, should now compile and run on all archs. However I did
774        not put the cpu definitions for each arch, i just separated
775        ARCH_IS_IA32 so even ARCH_IS_GENERIC can compile
776    
777        modified files:
778         ./examples/xvid_bench.c ./src/bitstream/cbp.h ./src/dct/fdct.h
779         ./src/dct/idct.h ./src/decoder.c ./src/image/colorspace.h
780         ./src/image/interpolate8x8.h ./src/image/reduced.h
781         ./src/motion/sad.h ./src/quant/quant_h263.h
782         ./src/quant/quant_mpeg4.h ./src/utils/emms.h
783    
784    
785    2003-08-09 20:47:42 GMT                                         patch-24
786    
787        Summary:
788          Updated changelog
789        Revision:
790          xvidcore--devapi4--1.0--patch-24
791    
792        Updated changelog
793    
794    
795        modified files:
796         ./changelog.txt
797    
798    
799  2003-08-09 20:31:17 GMT                                         patch-23  2003-08-09 20:31:17 GMT                                         patch-23
800    

Legend:
Removed from v.1172  
changed lines
  Added in v.1173

No admin address has been configured
ViewVC Help
Powered by ViewVC 1.0.4