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 |
|
|