Annotation of /xvidcore/ChangeLog-1.0

Revision 1.4 - (view) (download)

1 :	edgomez	1.1	# Ed.Gomez: This ChangeLog is generated from a personal tree maintained
2 :			# under the arch revision control tool. That's why dates may be skewed. I
3 :			# also removed all my email adresses from the output because they are not
4 :			# relevant.
5 :
6 :	edgomez	1.2
7 :	edgomez	1.4	#########################################################################
8 :			# 1.0.2 (Bitstream Version 36)
9 :			#########################################################################
10 :
11 :			2004-08-29 11:24:26 GMT patch-55
12 :
13 :			Summary:
14 :			Merged one important forgotten bugfix from head
15 :			Revision:
16 :			xvidcore--stable--1.0--patch-55
17 :
18 :			Merged one important forgotten bugfix from head
19 :
20 :			Patches applied:
21 :
22 :			* ed.gomez@free.fr--2004-1/xvidcore--head--0.0--patch-70
23 :			Out of bounds MVs clipping
24 :
25 :			* ed.gomez@free.fr--2004-1/xvidcore--head--0.0--patch-71
26 :			Decoder optimization (fixing regression)
27 :
28 :
29 :			modified files:
30 :			src/decoder.c
31 :
32 :			new patches:
33 :			ed.gomez@free.fr--2004-1/xvidcore--head--0.0--patch-70
34 :			ed.gomez@free.fr--2004-1/xvidcore--head--0.0--patch-71
35 :
36 :
37 :			2004-08-29 10:51:58 GMT patch-54
38 :
39 :			Summary:
40 :			Marking 1.0.2
41 :			Revision:
42 :			xvidcore--stable--1.0--patch-54
43 :
44 :			From ed.gomez:
45 :			* Marking 1.0.2
46 :
47 :			modified files:
48 :			ChangeLog build/generic/configure.in src/xvid.c src/xvid.h
49 :
50 :
51 :			2004-08-22 13:08:44 GMT patch-53
52 :
53 :			Summary:
54 :			Thread safety problem in idct C version
55 :			Revision:
56 :			xvidcore--stable--1.0--patch-53
57 :
58 :			From ed.gomez:
59 :			* Fixed a thread safety problem in C version of the
60 :			idct function. Added some comments on some static
61 :			data not marked as RO.
62 :
63 :			modified files:
64 :			src/bitstream/mbcoding.c src/dct/idct.c
65 :
66 :
67 :			2004-08-21 11:45:55 GMT patch-52
68 :
69 :			Summary:
70 :			Stupid typo+error in fdct_xxx_skal macro generator.
71 :			Revision:
72 :			xvidcore--stable--1.0--patch-52
73 :
74 :			From Nicolas Boulay:
75 :			* Found a typo mistake (ecx->eax) and an error in the same line
76 :			But as we're lucky, the unrolled version was bugfree, and that
77 :			is that one which is used.
78 :
79 :			modified files:
80 :			src/dct/x86_asm/fdct_mmx_skal.asm
81 :
82 :
83 :			2004-07-26 20:21:24 GMT patch-51
84 :
85 :			Summary:
86 :			ChangeLog Update
87 :			Revision:
88 :			xvidcore--stable--1.0--patch-51
89 :
90 :			ChangeLog Update
91 :
92 :
93 :			modified files:
94 :			ChangeLog
95 :
96 :
97 :	edgomez	1.3	2004-07-24 11:33:57 GMT patch-50
98 :
99 :			Summary:
100 :			BVOP direct/interpolated ref block rounding fix.
101 :			Revision:
102 :			xvidcore--stable--1.0--patch-50
103 :
104 :			From ed.gomez:
105 :			* BVOP direct/interpolated ref block rounding fix.
106 :			It's been using rounding=1 for averaging stage since ever.
107 :			The standard says it's rounding=0.
108 :
109 :			See standard clause 7.6.9.4 for explicit code and Section 6.3.5
110 :			that says "rounding=0" in bframes as they don't set the
111 :			vop_rounding_type in VOP header. Both sections match, xvid was
112 :			wrong
113 :
114 :			modified files:
115 :			src/decoder.c
116 :
117 :
118 :			2004-07-23 20:37:09 GMT patch-49
119 :
120 :			Summary:
121 :			Removed data qualifer in .rodata
122 :			Revision:
123 :			xvidcore--stable--1.0--patch-49
124 :
125 :			From ed.gomez:
126 :			* long standing warning by yasm... data isn't a keyword for
127 :			.(ro)data sections.
128 :
129 :			modified files:
130 :			src/bitstream/x86_asm/cbp_mmx.asm
131 :			src/bitstream/x86_asm/cbp_sse2.asm
132 :			src/dct/x86_asm/fdct_mmx_ffmpeg.asm
133 :			src/dct/x86_asm/fdct_mmx_skal.asm
134 :			src/dct/x86_asm/fdct_sse2_skal.asm
135 :			src/dct/x86_asm/idct_3dne.asm src/dct/x86_asm/idct_mmx.asm
136 :			src/dct/x86_asm/idct_sse2_dmitry.asm
137 :			src/dct/x86_asm/simple_idct_mmx.asm
138 :			src/image/x86_asm/colorspace_rgb_mmx.asm
139 :			src/image/x86_asm/colorspace_yuyv_mmx.asm
140 :			src/image/x86_asm/interpolate8x8_3dn.asm
141 :			src/image/x86_asm/interpolate8x8_3dne.asm
142 :			src/image/x86_asm/interpolate8x8_mmx.asm
143 :			src/image/x86_asm/interpolate8x8_xmm.asm
144 :			src/image/x86_asm/qpel_mmx.asm
145 :			src/image/x86_asm/reduced_mmx.asm
146 :			src/motion/x86_asm/sad_3dn.asm src/motion/x86_asm/sad_3dne.asm
147 :			src/motion/x86_asm/sad_mmx.asm src/motion/x86_asm/sad_sse2.asm
148 :			src/motion/x86_asm/sad_xmm.asm
149 :			src/quant/x86_asm/quantize_h263_3dne.asm
150 :			src/quant/x86_asm/quantize_h263_mmx.asm
151 :			src/quant/x86_asm/quantize_mpeg_mmx.asm
152 :			src/quant/x86_asm/quantize_mpeg_xmm.asm
153 :			src/utils/x86_asm/cpuid.asm
154 :			src/utils/x86_asm/interlacing_mmx.asm
155 :			src/utils/x86_asm/mem_transfer_3dne.asm
156 :			src/utils/x86_asm/mem_transfer_mmx.asm
157 :
158 :
159 :			2004-07-19 18:45:14 GMT patch-48
160 :
161 :			Summary:
162 :			Complete previous xvid_decraw patch
163 :			Revision:
164 :			xvidcore--stable--1.0--patch-48
165 :
166 :			Complete previous xvid_decraw patch
167 :
168 :
169 :			modified files:
170 :			examples/xvid_decraw.c
171 :
172 :
173 :			2004-07-18 11:58:48 GMT patch-47
174 :
175 :			Summary:
176 :			ISO C99'ism fix
177 :			Revision:
178 :			xvidcore--stable--1.0--patch-47
179 :
180 :			ISO C99'ism fix
181 :
182 :
183 :			modified files:
184 :			src/encoder.c
185 :
186 :
187 :			2004-07-17 11:32:42 GMT patch-46
188 :
189 :			Summary:
190 :			Make sure time incr is never larger than 16bit.
191 :			Revision:
192 :			xvidcore--stable--1.0--patch-46
193 :
194 :			From ed.gomez:
195 :			* Keep both fbase and fincr under 16bit limit.
196 :
197 :			modified files:
198 :			src/encoder.c
199 :
200 :
201 :			2004-07-17 10:00:42 GMT patch-45
202 :
203 :			Summary:
204 :			Future version interoperability
205 :			Revision:
206 :			xvidcore--stable--1.0--patch-45
207 :
208 :			From ed.gomez:
209 :			* Zeroing the structures is the best way not to pass
210 :			wrong data when dealing with slightly new XviD (like
211 :			head). If we don't zero here, then brightness is not
212 :			initialized in HEAD leading to crash.
213 :
214 :			modified files:
215 :			examples/xvid_decraw.c
216 :
217 :
218 :			2004-07-10 17:30:40 GMT patch-44
219 :
220 :			Summary:
221 :			ChangeLog update
222 :			Revision:
223 :			xvidcore--stable--1.0--patch-44
224 :
225 :			ChangeLog update
226 :
227 :
228 :			modified files:
229 :			ChangeLog
230 :
231 :
232 :	edgomez	1.2	2004-07-10 17:27:06 GMT patch-43
233 :
234 :			Summary:
235 :			Small mem leak in vfw.
236 :			Revision:
237 :			xvidcore--stable--1.0--patch-43
238 :
239 :			From sysKin:
240 :			* Free zones.
241 :
242 :			modified files:
243 :			vfw/src/codec.c
244 :
245 :
246 :			2004-07-10 16:55:53 GMT patch-42
247 :
248 :			Summary:
249 :			Fix wrong matrix reading logic.
250 :			Revision:
251 :			xvidcore--stable--1.0--patch-42
252 :
253 :			From sysKin:
254 :			* Fix a problem reading quantization matrix.
255 :
256 :			modified files:
257 :			src/bitstream/bitstream.c
258 :
259 :
260 :			2004-06-26 21:26:35 GMT patch-41
261 :
262 :			Summary:
263 :			low delay guessing (il)logic fix.
264 :			Revision:
265 :			xvidcore--stable--1.0--patch-41
266 :
267 :			From sysKin:
268 :			* bframes were decoded wrong when trying to guess low_delay flag
269 :			as specified in the standard when vol_control_parameters aren't
270 :			available.
271 :
272 :			modified files:
273 :			src/decoder.c
274 :
275 :
276 :			2004-06-13 19:15:05 GMT patch-40
277 :
278 :			Summary:
279 :			Small memory error in ia32 cpuid function.
280 :			Revision:
281 :			xvidcore--stable--1.0--patch-40
282 :
283 :			From ed.gomez:
284 :			* Valgrind detected a write to suspicious stack space. To avoid
285 :			any false reporting, added an explicit stack space allocation.
286 :
287 :			modified files:
288 :			src/utils/x86_asm/cpuid.asm
289 :
290 :	edgomez	1.1	#########################################################################
291 :			# 1.0.1 (Bitstream Version 35)
292 :			#########################################################################
293 :
294 :	edgomez	1.2	2004-06-05 22:55:56 GMT patch-39
295 :
296 :			Summary:
297 :			Marking 1.0.1 release
298 :			Revision:
299 :			xvidcore--stable--1.0--patch-39
300 :
301 :			Marking 1.0.1 release
302 :
303 :
304 :			modified files:
305 :			ChangeLog TODO build/generic/configure.in src/xvid.c
306 :			src/xvid.h
307 :
308 :
309 :	edgomez	1.1	2004-06-02 20:58:38 GMT patch-38
310 :
311 :			Summary:
312 :			DC clipping bug for real
313 :			Revision:
314 :			xvidcore--stable--1.0--patch-38
315 :
316 :			From ed.gomez:
317 :			* patch-25 was supposed to fix a DC clipping bug. However i
318 :			added the additional clipping code in the wrong place. But
319 :			at least, my fix didn't cause any trouble, it was just noop.
320 :			This patch should really fix this very "unlikely bug" (i just
321 :			want to remind the reader that this bug isn't easy to trigger,
322 :			and eg: my test sequences don't trigger it at all)
323 :
324 :			BS version incremented: 35
325 :
326 :			modified files:
327 :			src/decoder.c src/motion/estimation_rd_based.c
328 :			src/prediction/mbprediction.c src/prediction/mbprediction.h
329 :			src/xvid.h
330 :
331 :
332 :			2004-05-31 21:11:49 GMT patch-37
333 :
334 :			Summary:
335 :			time fixes to decoder.
336 :			Revision:
337 :			xvidcore--stable--1.0--patch-37
338 :
339 :			From ed.gomez:
340 :			* timestamps were badly computed by teh decoder in some corner cases
341 :			(1fps). This bug revealed that, timestamps were indeed wrong as
342 :			expected, but that bvop blocks in direct mode (vectors interpolated)
343 :			were somewhat compensated with wrong vectors in these same corner
344 :			cases.
345 :
346 :			modified files:
347 :			src/bitstream/bitstream.c src/decoder.c src/decoder.h
348 :
349 :
350 :			2004-05-30 09:36:13 GMT patch-36
351 :
352 :			Summary:
353 :			Wrong license header.
354 :			Revision:
355 :			xvidcore--stable--1.0--patch-36
356 :
357 :			From ed.gomez:
358 :			* Pascal did agree a plain GPL migration long ago, but this file
359 :			remained GPL+location restriction.
360 :
361 :			modified files:
362 :			src/image/reduced.c
363 :
364 :
365 :			2004-05-29 09:02:25 GMT patch-35
366 :
367 :			Summary:
368 :			More missing va_end() calls.
369 :			Revision:
370 :			xvidcore--stable--1.0--patch-35
371 :
372 :			From pete:
373 :			* portab.h is plenty of missing calls to va_end().
374 :
375 :			modified files:
376 :			src/portab.h
377 :
378 :
379 :			2004-05-28 21:28:21 GMT patch-34
380 :
381 :			Summary:
382 :			FPS=1 problem in decoder.
383 :			Revision:
384 :			xvidcore--stable--1.0--patch-34
385 :
386 :			From ed.gomez:
387 :			* patch-24 did fix bad behavior in encoder, so at least, compliant
388 :			streams were generated but the decoder was still doing the maths
389 :			a wrong way. Apply same logic to decoder.
390 :
391 :			Thanks to the patch-24 bug reporter for this followup.
392 :
393 :			modified files:
394 :			src/bitstream/bitstream.c
395 :
396 :
397 :			2004-05-27 20:04:01 GMT patch-33
398 :
399 :			Summary:
400 :			Nasty typo in pvop vector lambdas.
401 :			Revision:
402 :			xvidcore--stable--1.0--patch-33
403 :
404 :			From sysKin:
405 :			* s/+/*/ in the lambda value array for vectors in the pvop
406 :			estimation module.
407 :
408 :			modified files:
409 :			src/motion/estimation_pvop.c
410 :
411 :
412 :			2004-05-26 13:23:38 GMT patch-32
413 :
414 :			Summary:
415 :			Bits/Bytes confusion in the VFW frontend.
416 :			Revision:
417 :			xvidcore--stable--1.0--patch-32
418 :
419 :			From sysKin:
420 :			* confusion between the kilo, in kilobits (1000) and
421 :			the kilo in kilobytes (1024, should be named KiB anyway)
422 :			* biSizeImage is in bytes, not bits according to the Win32
423 :			API.
424 :
425 :			modified files:
426 :			vfw/src/codec.c vfw/src/config.c
427 :
428 :
429 :			2004-05-26 09:28:31 GMT patch-31
430 :
431 :			Summary:
432 :			Close variable argument list.
433 :			Revision:
434 :			xvidcore--stable--1.0--patch-31
435 :
436 :			From ed.gomez:
437 :			* Close the variable argument list as specified by the ANSI C
438 :			standard. Reported by Carsten <list4darav at gmx dot de> on
439 :			xvid-devel.
440 :
441 :			modified files:
442 :			src/image/font.c
443 :
444 :
445 :			2004-05-26 09:00:26 GMT patch-30
446 :
447 :			Summary:
448 :			ICM compatibility for VFW
449 :			Revision:
450 :			xvidcore--stable--1.0--patch-30
451 :
452 :			From sysKin:
453 :			* Makes the VFW frontend compatible with ICM applications
454 :			(Ooo, MS Office... etc). Reported on IRC.
455 :
456 :			modified files:
457 :			vfw/src/config.c vfw/src/driverproc.c
458 :
459 :
460 :			2004-05-26 08:58:56 GMT patch-29
461 :
462 :			Summary:
463 :			Small trellis bug
464 :			Revision:
465 :			xvidcore--stable--1.0--patch-29
466 :
467 :			From sysKin:
468 :			* Last coeff wasn't summed.
469 :
470 :			Reported by Jean Marc.
471 :
472 :			modified files:
473 :			src/utils/mbtransquant.c
474 :
475 :
476 :			2004-05-26 08:46:45 GMT patch-28
477 :
478 :			Summary:
479 :			Small bug in bframe ME.
480 :			Revision:
481 :			xvidcore--stable--1.0--patch-28
482 :
483 :			From sysKin:
484 :			* Small bug in bframe ME.
485 :
486 :			modified files:
487 :			src/motion/estimation_bvop.c
488 :
489 :			#########################################################################
490 :			# 1.0.0 final (Bitstream Version 34)
491 :			#########################################################################
492 :
493 :			2004-05-08 22:26:06 GMT patch-27
494 :
495 :			Summary:
496 :			Marking 1.0.0 final
497 :			Revision:
498 :			xvidcore--stable--1.0--patch-27
499 :
500 :			From ed.gomez:
501 :			* Marking 1.0.0 final \o/
502 :
503 :			modified files:
504 :			ChangeLog build/generic/configure.in src/xvid.h
505 :
506 :
507 :			2004-05-06 17:56:52 GMT patch-26
508 :
509 :			Summary:
510 :			Small mismatch in hint<->widget in VFW
511 :			Revision:
512 :			xvidcore--stable--1.0--patch-26
513 :
514 :			From sysKin:
515 :			* Small mismatch in hint<->widget.
516 :
517 :			modified files:
518 :			vfw/src/resource.rc
519 :
520 :
521 :			2004-05-02 22:40:50 GMT patch-25
522 :
523 :			Summary:
524 :			DC prediction fix.
525 :			Revision:
526 :			xvidcore--stable--1.0--patch-25
527 :
528 :			From ed.gomez:
529 :			* DC predictors weren't clipped to the [-2048, 2047] range.
530 :
531 :			BS version increased to 33
532 :
533 :			Thanks to jnorish on our forums to point out the problem.
534 :
535 :			modified files:
536 :			src/bitstream/bitstream.c src/decoder.c
537 :			src/motion/estimation_rd_based.c src/prediction/mbprediction.c
538 :			src/prediction/mbprediction.h src/xvid.h
539 :
540 :
541 :			2004-05-02 10:30:29 GMT patch-24
542 :
543 :			Summary:
544 :			Possible VOL header corruption.
545 :			Revision:
546 :			xvidcore--stable--1.0--patch-24
547 :
548 :			From ed.gomez:
549 :			* The VOL header could be corrupted when passing fincr=fbase=1 which
550 :			happens for fps=1 sequences.
551 :
552 :			BS version bumped up to 32
553 :
554 :			Original report:
555 :			http://www.xvid.org/modules.php?op=modload&name=phpBB2&file=viewtopic&t=2026&highlight=
556 :
557 :			modified files:
558 :			src/bitstream/bitstream.c src/xvid.h
559 :
560 :
561 :			2004-04-30 23:10:19 GMT patch-23
562 :
563 :			Summary:
564 :			Some very light Unix build system changes
565 :			Revision:
566 :			xvidcore--stable--1.0--patch-23
567 :
568 :			To prepare testing framework merging.
569 :
570 :			From ed.gomez:
571 :			* Some typos
572 :			* Copyright updates (it's 2004 since a few months ;-)
573 :			* Added some checking to bootstrap.sh
574 :			* Added m4 AC_PREREQ macro to configure.in
575 :
576 :			modified files:
577 :			build/generic/Makefile build/generic/bootstrap.sh
578 :			build/generic/configure.in
579 :
580 :
581 :			2004-04-20 19:40:29 GMT patch-22
582 :
583 :			Summary:
584 :			Small visual fix.
585 :			Revision:
586 :			xvidcore--stable--1.0--patch-22
587 :
588 :			From sysKin:
589 :			* Small visual fix
590 :
591 :			modified files:
592 :			vfw/src/config.c
593 :
594 :
595 :			2004-04-20 19:38:24 GMT patch-21
596 :
597 :			Summary:
598 :			Fix crash in decoder for non IFrame 1st frame.
599 :			Revision:
600 :			xvidcore--stable--1.0--patch-21
601 :
602 :			From sysKin:
603 :			* Fixed the crash caused by non IFrame 1st frame.
604 :
605 :			modified files:
606 :			src/decoder.c
607 :
608 :
609 :			2004-04-18 16:21:50 GMT patch-20
610 :
611 :			Summary:
612 :			Typo
613 :			Revision:
614 :			xvidcore--stable--1.0--patch-20
615 :
616 :			Typo
617 :
618 :
619 :			modified files:
620 :			vfw/src/resource.rc
621 :
622 :
623 :			2004-04-17 17:04:20 GMT patch-19
624 :
625 :			Summary:
626 :			vfw opens audio file in shared access mode
627 :			Revision:
628 :			xvidcore--stable--1.0--patch-19
629 :
630 :			vfw opens audio file in shared access mode
631 :
632 :
633 :			modified files:
634 :			vfw/src/config.c
635 :
636 :
637 :			2004-04-15 22:39:12 GMT patch-18
638 :
639 :			Summary:
640 :			Tiny xvid_decraw cleaning
641 :			Revision:
642 :			xvidcore--stable--1.0--patch-18
643 :
644 :			Tiny xvid_decraw cleaning
645 :
646 :
647 :			modified files:
648 :			examples/xvid_decraw.c
649 :
650 :
651 :			2004-04-15 19:14:31 GMT patch-17
652 :
653 :			Summary:
654 :			Tiny minor fixes for msvc.
655 :			Revision:
656 :			xvidcore--stable--1.0--patch-17
657 :
658 :			From pete:
659 :			* Missing arch endianness define in project files.
660 :			* Add a textual warning about win32 console EOF misreading.
661 :			* Prevent a SIGFPE when no frames were decoded.
662 :
663 :			modified files:
664 :			build/win32/xvid_decraw.dsp build/win32/xvid_encraw.dsp
665 :			examples/xvid_decraw.c
666 :
667 :
668 :			2004-04-14 22:41:07 GMT patch-16
669 :
670 :			Summary:
671 :			Fixed missing 1st frame in dshow output.
672 :			Revision:
673 :			xvidcore--stable--1.0--patch-16
674 :
675 :			From sysKin:
676 :			* decoder flags were overwritten, this was preventing from outputing
677 :			the first frame immediatly.
678 :
679 :			modified files:
680 :			dshow/src/CXvidDecoder.cpp
681 :
682 :
683 :			2004-04-14 22:39:17 GMT patch-15
684 :
685 :			Summary:
686 :			Ressource leaking in dshow.
687 :			Revision:
688 :			xvidcore--stable--1.0--patch-15
689 :
690 :			From sysKin:
691 :			* Same kind of ressource leaking as in vfw. Same cure.
692 :
693 :			modified files:
694 :			dshow/src/CXvidDecoder.cpp dshow/src/CXvidDecoder.h
695 :
696 :
697 :			2004-04-14 19:45:23 GMT patch-14
698 :
699 :			Summary:
700 :			Fixed small bug in trellis code.
701 :			Revision:
702 :			xvidcore--stable--1.0--patch-14
703 :
704 :			From ed.gomez (spotted by jean marc):
705 :			* Trellis optimization was computing the sum \|coeffs\| wrongly
706 :			because the Compute_sum function wasn't doing zigzag scanning
707 :			and stopped at last non zero coeffs in linear scanning...
708 :			The trivial fix could have been to pass the zigzag to the compute
709 :			sum function so it could have done its job right.
710 :			But... Why computing stuff that is already known in the upper layer ?
711 :			I removed the compute sum function, and just pass the original sum
712 :			value in trellis function parameters, just in case we have to return
713 :			it again because trellis failed optimizing the block coeffs.
714 :
715 :			This fix the bug and saves cycles :-) (but should not be noticeable)
716 :
717 :			modified files:
718 :			src/utils/mbtransquant.c src/xvid.h
719 :
720 :
721 :			2004-04-12 12:06:12 GMT patch-13
722 :
723 :			Summary:
724 :			Don't do SAD and RD based searches for qp.
725 :			Revision:
726 :			xvidcore--stable--1.0--patch-13
727 :
728 :			From sysKin:
729 :			* MakeGoodFlags function wasn't disabling SAD based search when
730 :			using RD. This was causing slowdown for no gain at all. This
731 :			patch should speed up encoding in qp mode.
732 :
733 :
734 :			modified files:
735 :			src/motion/estimation_pvop.c
736 :
737 :
738 :			2004-04-12 12:01:19 GMT patch-12
739 :
740 :			Summary:
741 :			3dnow functions proper separation.
742 :			Revision:
743 :			xvidcore--stable--1.0--patch-12
744 :
745 :			From pete (thx to a forum report):
746 :			* Separate correctly pure 3dnow functions and 3dnow+mmxext functions.
747 :			This fix "Illegal instruction" crash on old k6-2 CPUs.
748 :
749 :			modified files:
750 :			src/xvid.c
751 :
752 :
753 :			2004-04-12 11:57:20 GMT patch-11
754 :
755 :			Summary:
756 :			Better MV clipping code.
757 :			Revision:
758 :			xvidcore--stable--1.0--patch-11
759 :
760 :			From sysKin:
761 :			* Better MV clipping. DivX 5 generates out of range vectors and clipping
762 :			them directly borks the decoding of other MVs for which out of range
763 :			MVs were predictors. So it's just better to clip them for the block
764 :			decoding and keep an unclipped version for predictions.
765 :
766 :			modified files:
767 :			src/decoder.c
768 :
769 :
770 :			2004-04-08 20:34:54 GMT patch-10
771 :
772 :			Summary:
773 :			PGM support back in xvid_decraw.
774 :			Revision:
775 :			xvidcore--stable--1.0--patch-10
776 :
777 :			From ed.gomez:
778 :			* pgm/pnm format added back to xvid_decraw
779 :			+ pgm/pnm formats are now default for yv12/i420/rgb24 pixel format.
780 :			+ tga is default for rgb16/32
781 :			+ use option -f to choose tga, or pnm/pgm
782 :
783 :			modified files:
784 :			examples/xvid_decraw.c
785 :
786 :
787 :			2004-04-07 22:30:15 GMT patch-9
788 :
789 :			Summary:
790 :			3DNow Ext functions use MMXEXT opcodes.
791 :			Revision:
792 :			xvidcore--stable--1.0--patch-9
793 :
794 :			From Soltius (XviD Forum):
795 :			* Most of 3dnow extension functions do use MMXEXT opcodes, so
796 :			classify these functions as 3dnowext+mmxext. Avoids K6-2
797 :			boxes to crash with an invalid instruction error reported
798 :			by the host OS.
799 :
800 :			PS: original bug report
801 :			http://www.xvid.org/modules.php?op=modload&name=phpBB2&file=viewtopic&t=1656
802 :
803 :			modified files:
804 :			src/xvid.c
805 :
806 :
807 :			2004-04-07 22:01:54 GMT patch-8
808 :
809 :			Summary:
810 :			RGB 16bit output fix.
811 :			Revision:
812 :			xvidcore--stable--1.0--patch-8
813 :
814 :			From ScarletteTout (XviD Forum):
815 :			* Fix RGB 16bit output in C functions.
816 :
817 :			From ed.gomez:
818 :			* Replaced PGM output by TGA output so it's easy to implement
819 :			RGB 16/24/32 and greyscale bitmaps support in a single format.
820 :			(pgm could have supported RGB 24 and Greyscale only)
821 :			* Added colorspace choice to xvid_decraw
822 :			Use option -c csp, where csp is either rgb16, rgb24, rgb32, yv12 or i420
823 :			Defaults to i420.
824 :
825 :			PS: original bug report
826 :			http://www.xvid.org/modules.php?op=modload&name=phpBB2&file=viewtopic&t=1960&highlight=
827 :
828 :			modified files:
829 :			examples/xvid_decraw.c src/image/colorspace.c
830 :
831 :
832 :			#########################################################################
833 :			# 1.0.0 RC4 (Bitstream Version 30)
834 :			#########################################################################
835 :
836 :			2004-04-04 20:21:38 GMT patch-7
837 :
838 :			Summary:
839 :			DShow widget hiding.
840 :			Revision:
841 :			xvidcore--stable--1.0--patch-7
842 :
843 :			From Michael:
844 :			* No need to keep widget visibles if they won't be in 1.0.0.
845 :
846 :			modified files:
847 :			dshow/src/xvid.ax.rc
848 :
849 :
850 :			2004-04-04 20:17:52 GMT patch-6
851 :
852 :			Summary:
853 :			Compiler quirk in portab.h
854 :			Revision:
855 :			xvidcore--stable--1.0--patch-6
856 :
857 :			From Michael:
858 :			* The VC.NET workaround was causing trouble. Inversed the test.
859 :
860 :			modified files:
861 :			src/portab.h
862 :
863 :
864 :			2004-04-04 14:19:10 GMT patch-5
865 :
866 :			Summary:
867 :			Marking RC4
868 :			Revision:
869 :			xvidcore--stable--1.0--patch-5
870 :
871 :			Marking RC4
872 :
873 :
874 :			modified files:
875 :			ChangeLog build/generic/configure.in src/xvid.h
876 :
877 :
878 :			2004-04-04 14:07:00 GMT patch-4
879 :
880 :			Summary:
881 :			Frame dropping disabling for bframes.
882 :			Revision:
883 :			xvidcore--stable--1.0--patch-4
884 :
885 :			From sysKin & Pete:
886 :			* Disable frame dropping with bframes enabled. These two options
887 :			do not play fine together.
888 :
889 :			modified files:
890 :			src/encoder.c
891 :
892 :
893 :			2004-04-04 14:05:50 GMT patch-3
894 :
895 :			Summary:
896 :			Dead code removal.
897 :			Revision:
898 :			xvidcore--stable--1.0--patch-3
899 :
900 :			From sysKin:
901 :			* FrameCodeP was always called with contanst parameters. Removed
902 :			these parameters and associated dead code.
903 :
904 :			modified files:
905 :			src/encoder.c
906 :
907 :
908 :			2004-04-04 14:03:42 GMT patch-2
909 :
910 :			Summary:
911 :			Typo in ME fast comparison.
912 :			Revision:
913 :			xvidcore--stable--1.0--patch-2
914 :
915 :			From sysKin:
916 :			* Small typo in Fast ME code.
917 :
918 :			modified files:
919 :			src/motion/estimation_common.c
920 :
921 :
922 :			2004-04-02 23:58:19 GMT patch-1
923 :
924 :			Summary:
925 :			VFW Resource leak fix (try #2)
926 :			Revision:
927 :			xvidcore--stable--1.0--patch-1
928 :
929 :			From Suiryc on IRC:
930 :			* both encoder and decoder ending functions were calling
931 :			the dll freeing code. This was an error as the first function
932 :			called would unbind core function for the second called one.
933 :			Thus xvidcore could not release buffers.
934 :
935 :			modified files:
936 :			vfw/src/codec.c vfw/src/driverproc.c
937 :
938 :
939 :			2004-04-02 20:33:02 GMT base-0
940 :
941 :			Summary:
942 :			tag of ed.gomez@free.fr--2004-1/xvidcore--devapi4--1.0--patch-53
943 :			Revision:
944 :			xvidcore--stable--1.0--base-0
945 :
946 :			(automatically generated log message)
947 :
948 :
949 :			2004-03-31 19:32:47 GMT patch-53
950 :
951 :			Summary:
952 :			Ressources leaking in VFW.
953 :			Revision:
954 :			xvidcore--devapi4--1.0--patch-53
955 :
956 :			From sysKin:
957 :			* Storing ressources in global vars is making multithreaded/instanced
958 :			apps leaking lot of memory. Moved these vars to codec struct.
959 :
960 :			Thanks to dalox <fdalleau at hymatom d0t fr> to spot and fix the bug.
961 :
962 :			modified files:
963 :			vfw/src/codec.c vfw/src/codec.h vfw/src/driverproc.c
964 :
965 :
966 :			2004-03-31 19:28:51 GMT patch-52
967 :
968 :			Summary:
969 :			Fix to bad NVOP+bframe interaction.
970 :			Revision:
971 :			xvidcore--devapi4--1.0--patch-52
972 :
973 :			From syskin:
974 :			* When generating a NVOP, it interferes with bframe flushing and
975 :			packing.
976 :
977 :			modified files:
978 :			src/encoder.c
979 :
980 :
981 :			2004-03-31 19:24:42 GMT patch-51
982 :
983 :			Summary:
984 :			Fix GMC 2 warp point.
985 :			Revision:
986 :			xvidcore--devapi4--1.0--patch-51
987 :
988 :			From Skal:
989 :			* Fix 2 warp points GMC.
990 :
991 :			modified files:
992 :			src/motion/gmc.c
993 :
994 :
995 :			2004-03-31 19:18:46 GMT patch-50
996 :
997 :			Summary:
998 :			Added intra quant testing.
999 :			Revision:
1000 :			xvidcore--devapi4--1.0--patch-50
1001 :
1002 :			From Skal:
1003 :			* Added intra quant testing to xvid_bench
1004 :
1005 :			modified files:
1006 :			examples/xvid_bench.c
1007 :
1008 :
1009 :			2004-03-31 19:07:55 GMT patch-49
1010 :
1011 :			Summary:
1012 :			input width/height check
1013 :			Revision:
1014 :			xvidcore--devapi4--1.0--patch-49
1015 :
1016 :			From Skal:
1017 :			* Input width and height aren't checked and this can cause crash.
1018 :
1019 :			modified files:
1020 :			src/encoder.c
1021 :
1022 :
1023 :			2004-03-28 01:02:21 GMT patch-48
1024 :
1025 :			Summary:
1026 :			$ CVS expansion removed
1027 :			Revision:
1028 :			xvidcore--devapi4--1.0--patch-48
1029 :
1030 :			$ CVS expansion removed
1031 :
1032 :
1033 :			modified files:
1034 :			dshow/src/Configure.cpp dshow/src/config.c
1035 :			examples/xvid_encraw.c
1036 :
1037 :
1038 :			2004-03-28 00:45:23 GMT patch-47
1039 :
1040 :			Summary:
1041 :			Allow bigger frames
1042 :			Revision:
1043 :			xvidcore--devapi4--1.0--patch-47
1044 :
1045 :			Allow bigger frames
1046 :
1047 :
1048 :			modified files:
1049 :			examples/xvid_encraw.c
1050 :
1051 :
1052 :			2004-03-28 00:41:54 GMT patch-46
1053 :
1054 :			Summary:
1055 :			Fix in postproc header
1056 :			Revision:
1057 :			xvidcore--devapi4--1.0--patch-46
1058 :
1059 :			Fix in postproc header
1060 :
1061 :
1062 :			modified files:
1063 :			src/image/postprocessing.h
1064 :
1065 :
1066 :			2004-03-28 00:33:02 GMT patch-45
1067 :
1068 :			Summary:
1069 :			Fixed xvidvfw build on real mingw+msys systems
1070 :			Revision:
1071 :			xvidcore--devapi4--1.0--patch-45
1072 :
1073 :			Fixed xvidvfw build on real mingw+msys systems
1074 :
1075 :
1076 :			modified files:
1077 :			vfw/bin/Makefile
1078 :
1079 :
1080 :			2004-03-15 21:48:48 GMT patch-44
1081 :
1082 :			Summary:
1083 :			VFW updates.
1084 :			Revision:
1085 :			xvidcore--devapi4--1.0--patch-44
1086 :
1087 :			VFW updates.
1088 :
1089 :			modified files:
1090 :			vfw/src/config.c vfw/src/resource.rc
1091 :
1092 :
1093 :			2004-03-15 21:44:17 GMT patch-43
1094 :
1095 :			Summary:
1096 :			GMC bugfix.
1097 :			Revision:
1098 :			xvidcore--devapi4--1.0--patch-43
1099 :
1100 :			From skal:
1101 :			* GMC bugfixes.
1102 :
1103 :			modified files:
1104 :			src/motion/gmc.c
1105 :
1106 :
1107 :			2004-03-15 21:41:18 GMT patch-42
1108 :
1109 :			Summary:
1110 :			Buffer overrun fix in post proc.
1111 :			Revision:
1112 :			xvidcore--devapi4--1.0--patch-42
1113 :
1114 :			From sysKin:
1115 :			* Fixed buffer overrun in postproc code.
1116 :
1117 :			modified files:
1118 :			src/image/postprocessing.c src/image/postprocessing.h
1119 :
1120 :
1121 :			2004-03-15 21:33:22 GMT patch-41
1122 :
1123 :			Summary:
1124 :			Forgotten files.
1125 :			Revision:
1126 :			xvidcore--devapi4--1.0--patch-41
1127 :
1128 :			Me:
1129 :			* Sorry i missed these file additions. Important for the packages
1130 :			as i build them from the tla archive.
1131 :
1132 :			new files:
1133 :			dshow/src/.arch-ids/Configure.cpp.id
1134 :			dshow/src/.arch-ids/config.c.id
1135 :			dshow/src/.arch-ids/config.h.id dshow/src/.arch-ids/debug.h.id
1136 :			dshow/src/Configure.cpp dshow/src/config.c dshow/src/config.h
1137 :			dshow/src/debug.h
1138 :
1139 :
1140 :			2004-03-03 21:01:09 GMT patch-40
1141 :
1142 :			Summary:
1143 :			MV clipping in decoder.
1144 :			Revision:
1145 :			xvidcore--devapi4--1.0--patch-40
1146 :
1147 :			From sysKin:
1148 :			* clip MVs to valid ranges.
1149 :
1150 :			modified files:
1151 :			src/decoder.c src/decoder.h
1152 :
1153 :
1154 :			#########################################################################
1155 :			# 1.0.0 RC3 (Bistream Version 29)
1156 :			#########################################################################
1157 :
1158 :			2004-02-29 13:17:10 GMT patch-39
1159 :
1160 :			Summary:
1161 :			Marking 1.0.0 RC3
1162 :			Revision:
1163 :			xvidcore--devapi4--1.0--patch-39
1164 :
1165 :			* Marking RC3
1166 :
1167 :			modified files:
1168 :			ChangeLog build/generic/configure.in src/xvid.h
1169 :
1170 :
1171 :			2004-02-29 13:08:38 GMT patch-38
1172 :
1173 :			Summary:
1174 :			Win32 project fixes for xvid_encraw and xvid_decraw.
1175 :			Revision:
1176 :			xvidcore--devapi4--1.0--patch-38
1177 :
1178 :			From sysKin:
1179 :			* xvid_dec/encraw were linking against libxvidcore.lib. Now they link
1180 :			against xvidcore.dll.a
1181 :
1182 :			NB: xvid_bench is left as is though it doesn't link. This program
1183 :			requires access to internal functions which aren't available
1184 :			through the dll link lib. No good solution exists for MSVC to
1185 :			build both a dll+its link lib+static lib.
1186 :
1187 :			modified files:
1188 :			build/win32/xvid_decraw.dsp build/win32/xvid_encraw.dsp
1189 :
1190 :
1191 :			2004-02-29 12:56:36 GMT patch-37
1192 :
1193 :			Summary:
1194 :			DShow updates
1195 :			Revision:
1196 :			xvidcore--devapi4--1.0--patch-37
1197 :
1198 :			Bunch of DShow updates
1199 :
1200 :			modified files:
1201 :			dshow/src/CXvidDecoder.cpp dshow/src/resource.h
1202 :			dshow/src/xvid.ax.rc
1203 :
1204 :
1205 :			2004-02-29 12:55:41 GMT patch-36
1206 :
1207 :			Summary:
1208 :			VFW updates.
1209 :			Revision:
1210 :			xvidcore--devapi4--1.0--patch-36
1211 :
1212 :			* Bunch of VFW frontends updates.
1213 :
1214 :			modified files:
1215 :			vfw/src/codec.c vfw/src/codec.h vfw/src/config.c
1216 :			vfw/src/resource.rc vfw/src/status.c vfw/vfw.dsp
1217 :
1218 :
1219 :			2004-02-29 12:49:24 GMT patch-35
1220 :
1221 :			Summary:
1222 :			Artefact workaround in bframes.
1223 :			Revision:
1224 :			xvidcore--devapi4--1.0--patch-35
1225 :
1226 :			From sysKin:
1227 :			* Workaround for some artefacts appearing in bframes.
1228 :
1229 :			modified files:
1230 :			src/motion/estimation_bvop.c
1231 :
1232 :
1233 :			2004-02-29 12:46:13 GMT patch-34
1234 :
1235 :			Summary:
1236 :			Decoder buffer overflow fix.
1237 :			Revision:
1238 :			xvidcore--devapi4--1.0--patch-34
1239 :
1240 :			From sysKin:
1241 :			* Avoids buffer overflow when reading the last align byte. Our
1242 :			bitstream do 4 byte reading because of some platform contraints
1243 :			(ARM) and can cause buffer overflow reads.
1244 :
1245 :			modified files:
1246 :			src/decoder.c
1247 :
1248 :
1249 :			2004-02-29 11:53:47 GMT patch-33
1250 :
1251 :			Summary:
1252 :			Compatibility decoding for old bitstreams.
1253 :			Revision:
1254 :			xvidcore--devapi4--1.0--patch-33
1255 :
1256 :			From syskin:
1257 :			* old core versions used in dev-api-3 distributed by nearly all win32 bin
1258 :			builders used to have a edging bug. So when this information is known,
1259 :			workaround the bug.
1260 :
1261 :			modified files:
1262 :			src/decoder.c src/encoder.c src/image/image.c
1263 :			src/image/image.h
1264 :
1265 :			#########################################################################
1266 :			# 1.0.0 RC2 (Bistream Version 28)
1267 :			#########################################################################
1268 :
1269 :			2004-02-08 01:06:40 GMT patch-32
1270 :
1271 :			Summary:
1272 :			Marking RC2
1273 :			Revision:
1274 :			xvidcore--devapi4--1.0--patch-32
1275 :
1276 :			Marking RC2
1277 :
1278 :
1279 :			modified files:
1280 :			ChangeLog build/generic/configure.in
1281 :
1282 :
1283 :			2004-02-07 13:54:24 GMT patch-31
1284 :
1285 :			Summary:
1286 :			Win32 project outputs dll lib for linking.
1287 :			Revision:
1288 :			xvidcore--devapi4--1.0--patch-31
1289 :
1290 :			From pete:
1291 :			* output a lib to link against the dll (xvidcore.dll.a).
1292 :
1293 :			modified files:
1294 :			build/win32/libxvidcore.dsp
1295 :
1296 :
1297 :			2004-02-07 13:51:01 GMT patch-30
1298 :
1299 :			Summary:
1300 :			DShow update.
1301 :			Revision:
1302 :			xvidcore--devapi4--1.0--patch-30
1303 :
1304 :			From pete:
1305 :			* cmd line driving
1306 :
1307 :			From sysKin(?):
1308 :			* bugfixes related to video flipping
1309 :			* bugfix for the 'crash at the end" bug
1310 :
1311 :			modified files:
1312 :			TODO dshow/dshow.dsp dshow/src/CAbout.cpp dshow/src/CAbout.h
1313 :			dshow/src/CXvidDecoder.cpp dshow/src/CXvidDecoder.h
1314 :			dshow/src/xvid.ax.def
1315 :
1316 :
1317 :			2004-02-07 13:47:45 GMT patch-29
1318 :
1319 :			Summary:
1320 :			VFW updates
1321 :			Revision:
1322 :			xvidcore--devapi4--1.0--patch-29
1323 :
1324 :			Sorry feeling lazy about splitting this patch...
1325 :
1326 :			From peter:
1327 :			* bitrate calculator
1328 :
1329 :			From sysKin:
1330 :			* WMP9 bugfix
1331 :
1332 :			modified files:
1333 :			TODO vfw/src/codec.c vfw/src/config.c vfw/src/config.h
1334 :			vfw/src/resource.h vfw/src/resource.rc
1335 :
1336 :
1337 :			2004-02-07 13:43:26 GMT patch-28
1338 :
1339 :			Summary:
1340 :			Implicit overflow tuning for 2nd pass.
1341 :			Revision:
1342 :			xvidcore--devapi4--1.0--patch-28
1343 :
1344 :			From sysKin:
1345 :			* When doing a bigger 2nd pass, the overflow loop must be more
1346 :			aggressive else no bonus bits are reinjected. So we can auto
1347 :			tune the overlow values in that case.
1348 :
1349 :			modified files:
1350 :			src/plugins/plugin_2pass2.c
1351 :
1352 :
1353 :			2004-02-07 13:38:33 GMT patch-27
1354 :
1355 :			Summary:
1356 :			GMC+interlaced bugfix in decoder.
1357 :			Revision:
1358 :			xvidcore--devapi4--1.0--patch-27
1359 :
1360 :			From sysKin:
1361 :			* GMC+interlaced bugfix in decoder.
1362 :
1363 :			modified files:
1364 :			src/decoder.c
1365 :
1366 :
1367 :			2004-02-07 13:35:16 GMT patch-26
1368 :
1369 :			Summary:
1370 :			Reverted patch-23
1371 :			Revision:
1372 :			xvidcore--devapi4--1.0--patch-26
1373 :
1374 :			From christoph:
1375 :			* reverted patch-23, old code was right.
1376 :			* Important typo for the YVYU csp (passing the y plane instead of u).
1377 :
1378 :			modified files:
1379 :			src/image/image.c src/xvid.h
1380 :
1381 :
1382 :			2004-01-31 11:20:36 GMT patch-25
1383 :
1384 :			Summary:
1385 :			DShow support for more mpeg4 fourccs.
1386 :			Revision:
1387 :			xvidcore--devapi4--1.0--patch-25
1388 :
1389 :			From sysKin:
1390 :			* Added support for the MP4V fourcc.
1391 :
1392 :			modified files:
1393 :			dshow/src/CAbout.cpp dshow/src/CAbout.h
1394 :			dshow/src/CXvidDecoder.cpp dshow/src/CXvidDecoder.h
1395 :			dshow/src/resource.h dshow/src/xvid.ax.rc
1396 :
1397 :
1398 :			2004-01-31 11:12:38 GMT patch-24
1399 :
1400 :			Summary:
1401 :			DivX decoder compatibility
1402 :			Revision:
1403 :			xvidcore--devapi4--1.0--patch-24
1404 :
1405 :			From sysKin:
1406 :			* DivX decoder compatibility improved for packed bitstreams.
1407 :			It should now detect them and play them fine.
1408 :
1409 :			modified files:
1410 :			src/bitstream/bitstream.c
1411 :
1412 :
1413 :			2004-01-31 11:10:26 GMT patch-23
1414 :
1415 :			Summary:
1416 :			YV12/I420 confusion fixed.
1417 :			Revision:
1418 :			xvidcore--devapi4--1.0--patch-23
1419 :
1420 :			From christoph:
1421 :			* I420/YV12 were swapped since ... ages.
1422 :			* CSP_USER renamed to CSP_PLANAR
1423 :
1424 :			modified files:
1425 :			src/encoder.c src/image/image.c src/xvid.h
1426 :
1427 :
1428 :			2004-01-31 10:53:20 GMT patch-22
1429 :
1430 :			Summary:
1431 :			Arch separation for mem transfer functions
1432 :			Revision:
1433 :			xvidcore--devapi4--1.0--patch-22
1434 :
1435 :			Arch separation for mem transfer functions
1436 :
1437 :
1438 :			modified files:
1439 :			src/utils/mem_transfer.h
1440 :
1441 :
1442 :			2004-01-27 14:47:08 GMT patch-21
1443 :
1444 :			Summary:
1445 :			Write to registry Flip video flag in dshow
1446 :			Revision:
1447 :			xvidcore--devapi4--1.0--patch-21
1448 :
1449 :			From sysKin:
1450 :			* The flip video flag is now saved in registry.
1451 :			* Changed internal flags name convention (use n prefix for all now)
1452 :
1453 :			modified files:
1454 :			dshow/src/CAbout.cpp dshow/src/CAbout.h
1455 :			dshow/src/CXvidDecoder.cpp
1456 :
1457 :
1458 :			2004-01-27 14:42:52 GMT patch-20
1459 :
1460 :			Summary:
1461 :			Added bitrate calc to VFW
1462 :			Revision:
1463 :			xvidcore--devapi4--1.0--patch-20
1464 :
1465 :			From Pete:
1466 :			* Added bitrate calculator.
1467 :			* Changed up a few function calls to static type.
1468 :
1469 :			modified files:
1470 :			vfw/src/codec.c vfw/src/codec.h vfw/src/config.c
1471 :			vfw/src/config.h vfw/src/resource.h vfw/src/resource.rc
1472 :			vfw/vfw.dsp
1473 :
1474 :
1475 :			2004-01-27 14:29:49 GMT patch-19
1476 :
1477 :			Summary:
1478 :			Bugfix in decoder
1479 :			Revision:
1480 :			xvidcore--devapi4--1.0--patch-19
1481 :
1482 :			From sysKin:
1483 :			* when stats are not used, don't write to the stats pointer.
1484 :
1485 :			modified files:
1486 :			src/decoder.c
1487 :
1488 :			#########################################################################
1489 :			# 1.0.0 RC1 (Bistream Version 26)
1490 :			#########################################################################
1491 :
1492 :			2004-01-25 16:01:06 GMT patch-18
1493 :
1494 :			Summary:
1495 :			Marking RC1
1496 :			Revision:
1497 :			xvidcore--devapi4--1.0--patch-18
1498 :
1499 :			Marking RC1
1500 :
1501 :			modified files:
1502 :			ChangeLog build/generic/configure.in src/xvid.h
1503 :
1504 :
1505 :			2004-01-25 15:37:57 GMT patch-17
1506 :
1507 :			Summary:
1508 :			VFW update (again�)
1509 :			Revision:
1510 :			xvidcore--devapi4--1.0--patch-17
1511 :
1512 :			From sysKin:
1513 :			* Added Constant Quant encoding.
1514 :
1515 :			modified files:
1516 :			vfw/src/codec.c vfw/src/config.c vfw/src/config.h
1517 :			vfw/src/resource.rc
1518 :
1519 :
1520 :			2004-01-25 15:35:38 GMT patch-16
1521 :
1522 :			Summary:
1523 :			Missing MB quants for PP.
1524 :			Revision:
1525 :			xvidcore--devapi4--1.0--patch-16
1526 :
1527 :			From sysKin:
1528 :			* MB Quants are used by the PP code, so don't forget to update
1529 :			them even if the block is skipped or not coded.
1530 :
1531 :			modified files:
1532 :			src/decoder.c
1533 :
1534 :
1535 :			2004-01-23 13:25:52 GMT patch-15
1536 :
1537 :			Summary:
1538 :			VFW update (again)
1539 :			Revision:
1540 :			xvidcore--devapi4--1.0--patch-15
1541 :
1542 :			From sysKin:
1543 :			* Status window updates.
1544 :			* Big resource.h cleanup, it seems msvc isn't able to do it automatically.
1545 :
1546 :			modified files:
1547 :			vfw/src/resource.h vfw/src/resource.rc vfw/src/status.c
1548 :
1549 :
1550 :			2004-01-23 11:17:20 GMT patch-14
1551 :
1552 :			Summary:
1553 :			VFW gcc warnings
1554 :			Revision:
1555 :			xvidcore--devapi4--1.0--patch-14
1556 :
1557 :			VFW gcc warnings
1558 :
1559 :
1560 :			modified files:
1561 :			vfw/src/codec.c vfw/src/config.c vfw/src/config.h
1562 :
1563 :
1564 :			2004-01-22 20:54:53 GMT patch-13
1565 :
1566 :			Summary:
1567 :			DShow updates.
1568 :			Revision:
1569 :			xvidcore--devapi4--1.0--patch-13
1570 :
1571 :			From sysKin:
1572 :			* Fixed registry params type. Bool cannot be used or something weird
1573 :			happens when writing to registry
1574 :			* Defaults set to what the Reset widget sets.
1575 :
1576 :			modified files:
1577 :			dshow/src/CAbout.h dshow/src/CXvidDecoder.cpp
1578 :
1579 :
1580 :			2004-01-22 20:35:27 GMT patch-12
1581 :
1582 :			Summary:
1583 :			VFW updates.
1584 :			Revision:
1585 :			xvidcore--devapi4--1.0--patch-12
1586 :
1587 :			From sysKin:
1588 :			* GUI improvements.
1589 :			* Stats fixing.
1590 :			* Automatic config clear upon installation.
1591 :			* Added postprocessing options in there too.
1592 :
1593 :			modified files:
1594 :			vfw/bin/xvid.inf vfw/src/codec.c vfw/src/codec.h
1595 :			vfw/src/config.c vfw/src/config.h vfw/src/resource.h
1596 :			vfw/src/resource.rc vfw/src/status.c vfw/src/status.h
1597 :
1598 :
1599 :			2004-01-22 20:28:54 GMT patch-11
1600 :
1601 :			Summary:
1602 :			Minor updates to text files
1603 :			Revision:
1604 :			xvidcore--devapi4--1.0--patch-11
1605 :
1606 :			Minor updates to text files
1607 :
1608 :
1609 :			modified files:
1610 :			AUTHORS TODO
1611 :
1612 :
1613 :			2004-01-22 20:27:10 GMT patch-10
1614 :
1615 :			Summary:
1616 :			2pass plugin changes.
1617 :			Revision:
1618 :			xvidcore--devapi4--1.0--patch-10
1619 :
1620 :			From sysKin:
1621 :			* Disabled QPel during first pass as well.
1622 :			* Fix a mistaking condition when enabling largers 2nd passes.
1623 :
1624 :			From ed.gomez:
1625 :			* Fix the fix logic. The previous fix enclosed a condition it should
1626 :			not have touched. So i removed the mistaking condition, which was
1627 :			wrong anyway as stated in the comment, and got back the sane condition
1628 :			test.
1629 :
1630 :			modified files:
1631 :			src/plugins/plugin_2pass1.c src/plugins/plugin_2pass2.c
1632 :
1633 :
1634 :			2004-01-17 13:03:11 GMT patch-9
1635 :
1636 :			Summary:
1637 :			Unitialized pointers during plugin creation.
1638 :			Revision:
1639 :			xvidcore--devapi4--1.0--patch-9
1640 :
1641 :			From sysKin:
1642 :			- plugins which do not require private data were leaving the param2
1643 :			unitialized. Just init it to NULL. This bug wasn't causing any
1644 :			trouble anyway...
1645 :
1646 :			modified files:
1647 :			src/plugins/plugin_dump.c src/plugins/plugin_psnr.c
1648 :			vfw/src/codec.c
1649 :
1650 :
1651 :			2004-01-17 01:09:01 GMT patch-8
1652 :
1653 :			Summary:
1654 :			DShow forwwards AR information.
1655 :			Revision:
1656 :			xvidcore--devapi4--1.0--patch-8
1657 :
1658 :			From syskin(?):
1659 :			- forward AR information to DShow framework.
1660 :			- grayed some widgets.
1661 :
1662 :			modified files:
1663 :			dshow/src/CXvidDecoder.cpp dshow/src/CXvidDecoder.h
1664 :			dshow/src/xvid.ax.rc
1665 :
1666 :
1667 :			2004-01-07 13:50:29 GMT patch-7
1668 :
1669 :			Summary:
1670 :			Scaled zones fix.
1671 :			Revision:
1672 :			xvidcore--devapi4--1.0--patch-7
1673 :
1674 :			From sysKin (ideas from Koepi iirc):
1675 :			* Fix scaled zones computing prescaled data and so on instead of
1676 :			guessing everything with a global zone weight factor.
1677 :
1678 :			modified files:
1679 :			src/plugins/plugin_2pass2.c
1680 :
1681 :
1682 :			2004-01-06 01:06:39 GMT patch-6
1683 :
1684 :			Summary:
1685 :			Tab->Spaces in header only
1686 :			Revision:
1687 :			xvidcore--devapi4--1.0--patch-6
1688 :
1689 :			Tab->Spaces in header only
1690 :
1691 :
1692 :			modified files:
1693 :			src/xvid.h
1694 :
1695 :
1696 :			2004-01-04 18:35:35 GMT patch-5
1697 :
1698 :			Summary:
1699 :			Typo fixed
1700 :			Revision:
1701 :			xvidcore--devapi4--1.0--patch-5
1702 :
1703 :			Typo fixed
1704 :
1705 :
1706 :			modified files:
1707 :			build/generic/configure.in
1708 :
1709 :
1710 :			2004-01-04 13:40:51 GMT patch-4
1711 :
1712 :			Summary:
1713 :			VFW safer code.
1714 :			Revision:
1715 :			xvidcore--devapi4--1.0--patch-4
1716 :
1717 :			From sysKin:
1718 :			* Protects some parts of the code depending on a previous
1719 :			xvidcore opening. Avoids resources leaking.
1720 :
1721 :			modified files:
1722 :			vfw/src/codec.c vfw/src/driverproc.c
1723 :
1724 :
1725 :			2004-01-04 13:33:28 GMT patch-3
1726 :
1727 :			Summary:
1728 :			Fixes VC debug target name
1729 :			Revision:
1730 :			xvidcore--devapi4--1.0--patch-3
1731 :
1732 :			Fixes VC debug target name
1733 :
1734 :
1735 :			modified files:
1736 :			vfw/vfw.dsp
1737 :
1738 :
1739 :			2004-01-02 23:10:56 GMT patch-2
1740 :
1741 :			Summary:
1742 :			Win32 linking policy revised.
1743 :			Revision:
1744 :			xvidcore--devapi4--1.0--patch-2
1745 :
1746 :			Finnaly Win32 linking policy is to separate all XviD components:
1747 :			- xvidcore.dll exports XviD API
1748 :			- xvidvfw.dll links against xvidcore DLL
1749 :			- xviddshow.dll links against xvidcore DLL
1750 :
1751 :			From sysKin:
1752 :			* Changed DShow linking policy in VS project file.
1753 :			* Changed VFW linking policy in VS project file.
1754 :			* Added runtime xvidcore.dll loading in DShow and VFW.
1755 :			* Installs xvidcore.dll along side with xvidvfw.dll.
1756 :
1757 :			From ed.gomez:
1758 :			* Changed libxvidcore.dll mingw32/cygwin target name to
1759 :			xvidcore.dll in the configure script.
1760 :			* Changed xvid.dll VFW target name to xvidvfw.dll in the
1761 :			generic Makefile.
1762 :
1763 :			PS: unlike CVS, i reverted back to MS build tools in VS project files
1764 :
1765 :			modified files:
1766 :			build/generic/configure.in build/win32/libxvidcore.dsp
1767 :			dshow/dshow.dsp dshow/src/CXvidDecoder.cpp
1768 :			dshow/src/CXvidDecoder.h vfw/bin/Makefile vfw/bin/sources.inc
1769 :			vfw/bin/xvid.inf vfw/src/codec.c vfw/src/codec.h
1770 :			vfw/src/config.c vfw/src/config.h vfw/vfw.dsp
1771 :
1772 :
1773 :			2004-01-02 22:02:07 GMT patch-1
1774 :
1775 :			Summary:
1776 :			Better seeking in dshow
1777 :			Revision:
1778 :			xvidcore--devapi4--1.0--patch-1
1779 :
1780 :			From Michael:
1781 :			* Better DShow seeking.
1782 :
1783 :			modified files:
1784 :			dshow/src/CXvidDecoder.cpp
1785 :
1786 :
1787 :			2004-01-02 12:28:39 GMT base-0
1788 :
1789 :			Summary:
1790 :			tag of ed.gomez@free.fr--2003-1/xvidcore--devapi4--1.0--patch-162
1791 :			Revision:
1792 :			xvidcore--devapi4--1.0--base-0
1793 :
1794 :			Archive cycling...
1795 :
1796 :			#########################################################################
1797 :			# 1.0.0 beta3 (Bistream Version 25)
1798 :			#########################################################################
1799 :
1800 :			2003-12-26 22:21:35 GMT patch-162
1801 :
1802 :			Summary:
1803 :			Marking 1.0.0 beta3
1804 :			Revision:
1805 :			xvidcore--devapi4--1.0--patch-162
1806 :
1807 :			Marking beta3
1808 :
1809 :			modified files:
1810 :			ChangeLog TODO build/generic/configure.in src/xvid.h
1811 :
1812 :
1813 :			2003-12-25 20:57:52 GMT patch-161
1814 :
1815 :			Summary:
1816 :			Thread safe PP.
1817 :			Revision:
1818 :			xvidcore--devapi4--1.0--patch-161
1819 :
1820 :			From Michael:
1821 :			* Thread safe PP, context is now stored in DECODER struct.
1822 :
1823 :			modified files:
1824 :			src/decoder.c src/decoder.h src/image/postprocessing.c
1825 :			src/image/postprocessing.h
1826 :
1827 :
1828 :			2003-12-25 20:49:36 GMT patch-160
1829 :
1830 :			Summary:
1831 :			Added Turbo option to VFW GUI
1832 :			Revision:
1833 :			xvidcore--devapi4--1.0--patch-160
1834 :
1835 :			From Michael:
1836 :			* Added turbo mode that enables all fast ME flags.
1837 :
1838 :			modified files:
1839 :			vfw/src/codec.c vfw/src/config.c vfw/src/config.h
1840 :			vfw/src/resource.h vfw/src/resource.rc
1841 :
1842 :
1843 :			2003-12-25 20:46:16 GMT patch-159
1844 :
1845 :			Summary:
1846 :			Cartoon mode usage written to stream.
1847 :			Revision:
1848 :			xvidcore--devapi4--1.0--patch-159
1849 :
1850 :			From Michael:
1851 :			* @encoding: write cartoon usage in the version user data.
1852 :			* @decoding: detect cartoon flag appended to version string
1853 :			and force FILM PP disabling.
1854 :
1855 :			modified files:
1856 :			src/bitstream/bitstream.c src/bitstream/bitstream.h
1857 :			src/decoder.c src/decoder.h src/encoder.c
1858 :
1859 :
1860 :			2003-12-21 13:34:03 GMT patch-158
1861 :
1862 :			Summary:
1863 :			Removed unused var in VFW GUI
1864 :			Revision:
1865 :			xvidcore--devapi4--1.0--patch-158
1866 :
1867 :			Removed unused var in VFW GUI
1868 :
1869 :
1870 :			modified files:
1871 :			vfw/src/config.c
1872 :
1873 :
1874 :			2003-12-21 13:32:52 GMT patch-157
1875 :
1876 :			Summary:
1877 :			Two pass small update
1878 :			Revision:
1879 :			xvidcore--devapi4--1.0--patch-157
1880 :
1881 :			from syskin:
1882 :			* allow second pass to be bigger than 1st one
1883 :			(not tested, the quant mapping formula may not be adapted for this
1884 :			usage, so take this change as experimental, and prefer doing
1885 :			second pass still smaller than 1st one)
1886 :			* let ivops benefit from positive overflow.
1887 :
1888 :			from me:
1889 :			* set frame type in quant zones (was a buglet)
1890 :
1891 :			modified files:
1892 :			src/plugins/plugin_2pass2.c
1893 :
1894 :
1895 :			2003-12-20 22:28:07 GMT patch-156
1896 :
1897 :			Summary:
1898 :			New VFW defaults
1899 :			Revision:
1900 :			xvidcore--devapi4--1.0--patch-156
1901 :
1902 :			From michael:
1903 :			* new VFW default values
1904 :
1905 :			modified files:
1906 :			vfw/src/config.c
1907 :
1908 :
1909 :			2003-12-20 22:12:38 GMT patch-155
1910 :
1911 :			Summary:
1912 :			Added ARGB colorspace.
1913 :			Revision:
1914 :			xvidcore--devapi4--1.0--patch-155
1915 :
1916 :			Christoph N�geli naegelic(at)ee{dot}ethz<dot>ch asked me to add C
1917 :			support for ARGB colorspace. He provided the encoding part, and i
1918 :			extended the original patch in order to have full support for ARGB
1919 :			both for encoding and decoding (though it's C only, read slow).
1920 :
1921 :			modified files:
1922 :			src/image/colorspace.c src/image/colorspace.h
1923 :			src/image/image.c src/xvid.c src/xvid.h
1924 :
1925 :
1926 :			2003-12-20 21:29:37 GMT patch-154
1927 :
1928 :			Summary:
1929 :			Added 2pass1 comment about fast 1st pass.
1930 :			Revision:
1931 :			xvidcore--devapi4--1.0--patch-154
1932 :
1933 :			Just added a comment on fast 1st pass, so it explains why
1934 :			we do it that way and why some things are left aside.
1935 :
1936 :			modified files:
1937 :			src/plugins/plugin_2pass1.c
1938 :
1939 :
1940 :			2003-12-20 20:03:51 GMT patch-153
1941 :
1942 :			Summary:
1943 :			Win32 VC6 wrong libc linking.
1944 :			Revision:
1945 :			xvidcore--devapi4--1.0--patch-153
1946 :
1947 :			From sysKin:
1948 :			* Changed single thread libc linking to multithreaded version.
1949 :
1950 :			modified files:
1951 :			vfw/vfw.dsp
1952 :
1953 :
1954 :			2003-12-20 15:28:53 GMT patch-152
1955 :
1956 :			Summary:
1957 :			VOL flags updating -- take #2
1958 :			Revision:
1959 :			xvidcore--devapi4--1.0--patch-152
1960 :
1961 :			From sysKin:
1962 :			* VOL flags updates fix take #2
1963 :
1964 :			modified files:
1965 :			src/encoder.c
1966 :
1967 :
1968 :			2003-12-20 15:10:30 GMT patch-151
1969 :
1970 :			Summary:
1971 :			Fast ME tunings.
1972 :			Revision:
1973 :			xvidcore--devapi4--1.0--patch-151
1974 :
1975 :			From michael:
1976 :			* fast refinement for 8x8 blocks
1977 :			* more reliable behavior for all fast ME decisions
1978 :
1979 :			modified files:
1980 :			src/motion/estimation_bvop.c src/motion/estimation_pvop.c
1981 :			src/xvid.h
1982 :
1983 :
1984 :			2003-12-20 14:59:58 GMT patch-150
1985 :
1986 :			Summary:
1987 :			VFW AR revamping -- take #2
1988 :			Revision:
1989 :			xvidcore--devapi4--1.0--patch-150
1990 :
1991 :			From sysKin:
1992 :			* more AR revamping
1993 :
1994 :			modified files:
1995 :			vfw/src/codec.c vfw/src/config.c vfw/src/config.h
1996 :			vfw/src/resource.h vfw/src/resource.rc
1997 :
1998 :
1999 :			2003-12-20 14:57:40 GMT patch-149
2000 :
2001 :			Summary:
2002 :			Unbuffured IO for 1st pass stat files
2003 :			Revision:
2004 :			xvidcore--devapi4--1.0--patch-149
2005 :
2006 :			Unbuffured IO for 1st pass stat files
2007 :
2008 :
2009 :			modified files:
2010 :			src/plugins/plugin_2pass1.c
2011 :
2012 :
2013 :			2003-12-18 17:44:07 GMT patch-148
2014 :
2015 :			Summary:
2016 :			Forgotten bit of patch-141
2017 :			Revision:
2018 :			xvidcore--devapi4--1.0--patch-148
2019 :
2020 :			Damn i forgot to merge the 1st pass changes... so lame, i tested the
2021 :			original patch but not the merged one.
2022 :
2023 :			modified files:
2024 :			src/plugins/plugin_2pass1.c
2025 :
2026 :
2027 :			2003-12-18 14:45:39 GMT patch-147
2028 :
2029 :			Summary:
2030 :			More postprocessing.
2031 :			Revision:
2032 :			xvidcore--devapi4--1.0--patch-147
2033 :
2034 :			From michael:
2035 :			* added film noise effect.
2036 :			* moved postproc initialization to decoder initialization.
2037 :			* added support for this postproc filter into DShow.
2038 :
2039 :			modified files:
2040 :			dshow/src/CAbout.cpp dshow/src/CAbout.h
2041 :			dshow/src/CXvidDecoder.cpp dshow/src/resource.h
2042 :			dshow/src/xvid.ax.rc src/decoder.c src/image/postprocessing.c
2043 :			src/image/postprocessing.h src/xvid.c src/xvid.h
2044 :
2045 :
2046 :			2003-12-18 14:38:19 GMT patch-146
2047 :
2048 :			Summary:
2049 :			BFrames ME speed up flags.
2050 :			Revision:
2051 :			xvidcore--devapi4--1.0--patch-146
2052 :
2053 :			From michael:
2054 :			* Added 3 ME flags to skip some bvop ME steps and thus speed up
2055 :			ME for bvops (at the expense of quality loss).
2056 :			- skip delta search
2057 :			- fast interpolate mode
2058 :			- early stop
2059 :
2060 :			modified files:
2061 :			src/motion/estimation.h src/motion/estimation_bvop.c
2062 :			src/motion/estimation_common.c src/motion/estimation_pvop.c
2063 :			src/xvid.h
2064 :
2065 :
2066 :			2003-12-18 14:11:01 GMT patch-145
2067 :
2068 :			Summary:
2069 :			VFW defaults changed
2070 :			Revision:
2071 :			xvidcore--devapi4--1.0--patch-145
2072 :
2073 :			From michael:
2074 :			* disabled DXN profiles. Better not say we are compatible to avoid
2075 :			problems with DXN. Mostly because of VBV lacking.
2076 :			* Changed overflow defaults to 5/5/5 with new 2pass code.
2077 :
2078 :			modified files:
2079 :			vfw/src/config.c
2080 :
2081 :
2082 :			2003-12-17 16:45:59 GMT patch-144
2083 :
2084 :			Summary:
2085 :			Forgotten bit for AR support in VFW
2086 :			Revision:
2087 :			xvidcore--devapi4--1.0--patch-144
2088 :
2089 :			Koepi might have forgotten to send me this change:
2090 :			* added resource id to the resource header
2091 :
2092 :			modified files:
2093 :			vfw/src/resource.h
2094 :
2095 :
2096 :			2003-12-17 15:11:37 GMT patch-143
2097 :
2098 :			Summary:
2099 :			Lower starting quantizer for CBR encoding
2100 :			Revision:
2101 :			xvidcore--devapi4--1.0--patch-143
2102 :
2103 :			From christoph:
2104 :			* Lower starting quantizer for CBR encodings.
2105 :
2106 :			modified files:
2107 :			src/plugins/plugin_single.c
2108 :
2109 :
2110 :			2003-12-17 15:04:33 GMT patch-142
2111 :
2112 :			Summary:
2113 :			Small xvid_encraw updates
2114 :			Revision:
2115 :			xvidcore--devapi4--1.0--patch-142
2116 :
2117 :			From christoph:
2118 :			* set upper frame size limit to 4096 pixels
2119 :			* GME refinement flag set where it belongs to.
2120 :
2121 :			modified files:
2122 :			examples/xvid_encraw.c
2123 :
2124 :
2125 :			2003-12-17 15:01:52 GMT patch-141
2126 :
2127 :			Summary:
2128 :			Two pass update.
2129 :			Revision:
2130 :			xvidcore--devapi4--1.0--patch-141
2131 :
2132 :			This patch improves the two pass code, quantizer distribution is
2133 :			smoother and results seem to be better.
2134 :
2135 :			* Two pass now scales only a specific part of the frame length.
2136 :			This required changing the stats file format (added a header+MV
2137 :			length field) and the xvid_plg_data_t structure (binary compatible).
2138 :			* Overflow improvement and degradation set to 10% instead of 60%
2139 :			Assymetric values may help...
2140 :			* Some cleanup work done on the encoder part of the API header.
2141 :
2142 :			NB: plg data struct will be cleaned up before 1.0 so if you rely on it
2143 :			please read the header file to know which part will disappear.
2144 :
2145 :			modified files:
2146 :			src/encoder.c src/plugins/plugin_2pass1.c
2147 :			src/plugins/plugin_2pass2.c src/xvid.h
2148 :
2149 :
2150 :			2003-12-17 13:53:34 GMT patch-140
2151 :
2152 :			Summary:
2153 :			VOL flags fix.
2154 :			Revision:
2155 :			xvidcore--devapi4--1.0--patch-140
2156 :
2157 :			From sysKin:
2158 :			* VOL flags updated as they should
2159 :
2160 :			modified files:
2161 :			src/encoder.c
2162 :
2163 :
2164 :			2003-12-17 11:07:15 GMT patch-139
2165 :
2166 :			Summary:
2167 :			VFW GUI Update.
2168 :			Revision:
2169 :			xvidcore--devapi4--1.0--patch-139
2170 :
2171 :			From Koepi:
2172 :			* typo in GMC description
2173 :			* Added AR widget (experimental)
2174 :
2175 :			modified files:
2176 :			vfw/src/codec.c vfw/src/config.c vfw/src/config.h
2177 :			vfw/src/resource.rc
2178 :
2179 :
2180 :			2003-12-14 12:44:36 GMT patch-138
2181 :
2182 :			Summary:
2183 :			Fix PP crash
2184 :			Revision:
2185 :			xvidcore--devapi4--1.0--patch-138
2186 :
2187 :			Fix PP crash
2188 :
2189 :
2190 :			modified files:
2191 :			src/decoder.c
2192 :
2193 :
2194 :			2003-12-14 12:43:21 GMT patch-137
2195 :
2196 :			Summary:
2197 :			Texture bit counting for bframes
2198 :			Revision:
2199 :			xvidcore--devapi4--1.0--patch-137
2200 :
2201 :			Texture bit counting for bframes
2202 :
2203 :
2204 :			modified files:
2205 :			src/bitstream/mbcoding.c
2206 :
2207 :
2208 :			2003-12-12 23:58:18 GMT patch-136
2209 :
2210 :			Summary:
2211 :			SSE2 code enabled.
2212 :			Revision:
2213 :			xvidcore--devapi4--1.0--patch-136
2214 :
2215 :			* SSE2 code enabled by default (only sane ones, _no_ idct)
2216 :			* Aligned data in xvid_bench to avoid crashes with SSE2 code
2217 :			because of unaligned read accesses.
2218 :
2219 :			modified files:
2220 :			examples/xvid_bench.c src/xvid.c
2221 :
2222 :
2223 :			2003-12-12 22:50:33 GMT patch-135
2224 :
2225 :			Summary:
2226 :			DShow update.
2227 :			Revision:
2228 :			xvidcore--devapi4--1.0--patch-135
2229 :
2230 :			From michael (from nic):
2231 :			* Dshow updates (colorspace etc...)
2232 :			* Deblocking option.
2233 :
2234 :			modified files:
2235 :			dshow/src/CAbout.cpp dshow/src/CAbout.h
2236 :			dshow/src/CXvidDecoder.cpp dshow/src/CXvidDecoder.h
2237 :			dshow/src/resource.h dshow/src/xvid.ax.rc
2238 :
2239 :
2240 :			2003-12-12 14:18:13 GMT patch-134
2241 :
2242 :			Summary:
2243 :			Added missing postproc init
2244 :			Revision:
2245 :			xvidcore--devapi4--1.0--patch-134
2246 :
2247 :			Added missing postproc init
2248 :
2249 :
2250 :			modified files:
2251 :			src/xvid.c
2252 :
2253 :
2254 :			2003-12-12 14:16:29 GMT patch-133
2255 :
2256 :			Summary:
2257 :			YUV space clarifications and fix.
2258 :			Revision:
2259 :			xvidcore--devapi4--1.0--patch-133
2260 :
2261 :			From christoph:
2262 :			* Put some comments
2263 :			* fixed UV swapping in USER case.
2264 :
2265 :			modified files:
2266 :			src/image/image.c
2267 :
2268 :
2269 :			2003-12-12 14:01:52 GMT patch-132
2270 :
2271 :			Summary:
2272 :			Default compilation flags change.
2273 :			Revision:
2274 :			xvidcore--devapi4--1.0--patch-132
2275 :
2276 :			from christoph:
2277 :			- -fgcse was causing trouble on Suse9 gcc
2278 :			- -01->-02
2279 :
2280 :			modified files:
2281 :			build/generic/configure.in
2282 :
2283 :
2284 :			2003-12-12 13:43:35 GMT patch-131
2285 :
2286 :			Summary:
2287 :			Pixel Aspect Ratio support improvement.
2288 :			Revision:
2289 :			xvidcore--devapi4--1.0--patch-131
2290 :
2291 :			* 1:1 VGA is default now (old behavior)
2292 :			* When passed EXT PAR type, we now (try to) sanityse the
2293 :			par_width/par_height value:
2294 :			- make it positive
2295 :			- using 0 (typical memset reseting) defaults to 1
2296 :			- simplify the PAR (using gcd)
2297 :			- then range it in [1..255] (can be lossy)
2298 :			* Specify valid range in API header
2299 :
2300 :			Bitstream version bumped to 24.
2301 :
2302 :			PS: this patch supersedes sysKin's one in CVS.
2303 :
2304 :			modified files:
2305 :			src/encoder.c src/xvid.h
2306 :
2307 :
2308 :			2003-12-10 22:57:50 GMT patch-130
2309 :
2310 :			Summary:
2311 :			xvid_decraw cmdline changes.
2312 :			Revision:
2313 :			xvidcore--devapi4--1.0--patch-130
2314 :
2315 :			* -nframes -> -frames (why did i put a 'n' there ?)
2316 :			* -save changed its meaning, it now controls per frame Elementary
2317 :			Stream saving.
2318 :			* -o string is now independent from -save.
2319 :
2320 :			So now it's possible to save both a ES file per frame + a ES file for the
2321 :			sequence.
2322 :
2323 :			modified files:
2324 :			examples/xvid_encraw.c
2325 :
2326 :
2327 :			2003-12-10 15:08:20 GMT patch-129
2328 :
2329 :			Summary:
2330 :			Decoder bugfixes.
2331 :			Revision:
2332 :			xvidcore--devapi4--1.0--patch-129
2333 :
2334 :			From syskin:
2335 :			* bvops MBs were going banana from time to time because they were
2336 :			referencing wrong future ref MBs.
2337 :			* decoder now informs the client app about bvop lag, returning
2338 :			XVID_TYPE_NOTHING, up to the client app to display (or not)
2339 :			the bvop lag frame (black with error message).
2340 :
2341 :			Fixing previous patch a bit:
2342 :			* added $Id$ fields
2343 :			* Fixed copyright
2344 :
2345 :			modified files:
2346 :			dshow/src/CXvidDecoder.cpp src/decoder.c
2347 :			src/image/postprocessing.c src/image/postprocessing.h
2348 :
2349 :
2350 :			2003-12-10 14:53:58 GMT patch-128
2351 :
2352 :			Summary:
2353 :			Deblocking code.
2354 :			Revision:
2355 :			xvidcore--devapi4--1.0--patch-128
2356 :
2357 :			Patch from michael:
2358 :			* added deblocking code
2359 :
2360 :			Merge work:
2361 :			* Added postprocessing.[ch] to project files
2362 :			* added #include "image/postprocessing.h" directive in decoder.c
2363 :			* new lines missing (gcc is so pedantic)
2364 :
2365 :			NB: slice rendering + postprocessing is impossible. Slice rendering is
2366 :			somewhat abandoned.
2367 :
2368 :			new files:
2369 :			src/image/.arch-ids/postprocessing.c.id
2370 :			src/image/.arch-ids/postprocessing.h.id
2371 :			src/image/postprocessing.c src/image/postprocessing.h
2372 :
2373 :			modified files:
2374 :			build/generic/sources.inc build/win32/libxvidcore.dsp
2375 :			src/decoder.c src/xvid.h
2376 :
2377 :
2378 :			2003-12-08 18:33:26 GMT patch-127
2379 :
2380 :			Summary:
2381 :			Don't read out of bounds
2382 :			Revision:
2383 :			xvidcore--devapi4--1.0--patch-127
2384 :
2385 :			Don't read out of bounds
2386 :
2387 :
2388 :			modified files:
2389 :			src/plugins/plugin_2pass2.c
2390 :
2391 :
2392 :			2003-12-08 18:31:41 GMT patch-126
2393 :
2394 :			Summary:
2395 :			Macroblock structure cleanup
2396 :			Revision:
2397 :			xvidcore--devapi4--1.0--patch-126
2398 :
2399 :			Macroblock structure cleanup
2400 :
2401 :
2402 :			modified files:
2403 :			src/global.h
2404 :
2405 :
2406 :			2003-12-07 15:09:41 GMT patch-125
2407 :
2408 :			Summary:
2409 :			Small fixes.
2410 :			Revision:
2411 :			xvidcore--devapi4--1.0--patch-125
2412 :
2413 :			From gruel:
2414 :			* xvid.h: Minor color space correction.
2415 :
2416 :			From sysKin:
2417 :			* codec.c: Zones fix
2418 :
2419 :			modified files:
2420 :			src/xvid.h vfw/src/codec.c
2421 :
2422 :
2423 :			2003-12-07 14:57:14 GMT patch-124
2424 :
2425 :			Summary:
2426 :			HUGE file handling in twopass.
2427 :			Revision:
2428 :			xvidcore--devapi4--1.0--patch-124
2429 :
2430 :			Because of a lacking cast, two pass did not handle well some very
2431 :			large target size (bitrate mode is not affected).
2432 :
2433 :			It should now be safe specifying target sizes up to 2^31kB which
2434 :			represents 2TB. Someone using XviD in studios ? ;-)
2435 :
2436 :			modified files:
2437 :			src/plugins/plugin_2pass2.c
2438 :
2439 :			#########################################################################
2440 :			# 1.0.0 beta2 (Bitstream Version 23)
2441 :			#########################################################################
2442 :
2443 :			2003-12-05 14:43:53 GMT patch-123
2444 :
2445 :			Summary:
2446 :			Marking 1.0.0 Beta2
2447 :			Revision:
2448 :			xvidcore--devapi4--1.0--patch-123
2449 :
2450 :			Marking 1.0.0 Beta2
2451 :
2452 :
2453 :			modified files:
2454 :			ChangeLog TODO build/generic/configure.in
2455 :
2456 :
2457 :			2003-12-05 14:35:22 GMT patch-122
2458 :
2459 :			Summary:
2460 :			Cap quants correctly (the best we can at least)
2461 :			Revision:
2462 :			xvidcore--devapi4--1.0--patch-122
2463 :
2464 :			Cap quants correctly (the best we can at least)
2465 :
2466 :
2467 :			modified files:
2468 :			src/plugins/plugin_single.c
2469 :
2470 :
2471 :			2003-12-05 14:33:48 GMT patch-121
2472 :
2473 :			Summary:
2474 :			Small glitch
2475 :			Revision:
2476 :			xvidcore--devapi4--1.0--patch-121
2477 :
2478 :			Small glitch
2479 :
2480 :
2481 :			modified files:
2482 :			src/motion/vop_type_decision.c
2483 :
2484 :
2485 :			2003-12-05 14:06:19 GMT patch-120
2486 :
2487 :			Summary:
2488 :			KFthresholding changes.
2489 :			Revision:
2490 :			xvidcore--devapi4--1.0--patch-120
2491 :
2492 :			As user reports proved, the logic behind the min_key_interval was
2493 :			1/ misleading because the parameter is kfthreshold indeed and not
2494 :			a minimum keyframe interval
2495 :			2/ the formula was a bit too aggressive (removing 20% of bitrate
2496 :			per frame until distance to next iframe was 1)
2497 :
2498 :			I posted a RFC to try to settle a decision on what behavior this
2499 :			setting should have. We have still have no clear answer so i prefer
2500 :			just fixing the misleading name right now and wait for a common
2501 :			position about its behavior later.
2502 :
2503 :			Libraries are binary compatible, but source code compatibility
2504 :			is broken (rename rc_2pass2_t->min_key_interval to kfthreshold).
2505 :			This is probably the last API change.
2506 :
2507 :			NB: fixes a type problem during scaling parameter computing which
2508 :			was causing insane pb_iboost_tax_ratio values.
2509 :
2510 :			modified files:
2511 :			src/plugins/plugin_2pass2.c src/xvid.h vfw/src/codec.c
2512 :			vfw/src/config.c vfw/src/config.h
2513 :
2514 :
2515 :			2003-12-05 00:20:28 GMT patch-119
2516 :
2517 :			Summary:
2518 :			ivop decision tuning.
2519 :			Revision:
2520 :			xvidcore--devapi4--1.0--patch-119
2521 :
2522 :			ivop decision tuning from sysKin.
2523 :
2524 :			modified files:
2525 :			src/motion/vop_type_decision.c
2526 :
2527 :
2528 :			2003-12-03 18:55:29 GMT patch-118
2529 :
2530 :			Summary:
2531 :			VOSH header always written.
2532 :			Revision:
2533 :			xvidcore--devapi4--1.0--patch-118
2534 :
2535 :			* profile is set to sane default value in BitstreamWriteVolHeaders
2536 :			* VOSH is now always written (note that the ending code is never
2537 :			written)
2538 :			* doubled variable removed from bvop estimation file.
2539 :
2540 :			bitstream version set to 23
2541 :			Version 22 was used in CVS by a fix from michael for VOSH, it was just
2542 :			lacking the sane default value setting when profile is 0x00 from user.
2543 :			0x00 is a reserved profile ID in the spec so it wasn't making much sense
2544 :			to write 0x00.
2545 :
2546 :			modified files:
2547 :			src/bitstream/bitstream.c src/motion/estimation_bvop.c
2548 :			src/xvid.h
2549 :
2550 :
2551 :			2003-12-03 15:29:30 GMT patch-117
2552 :
2553 :			Summary:
2554 :			VFW GUI fixes.
2555 :			Revision:
2556 :			xvidcore--devapi4--1.0--patch-117
2557 :
2558 :			min key was misleading because it was legacy code from dev-api-3.
2559 :
2560 :			modified files:
2561 :			vfw/src/codec.c vfw/src/config.c vfw/src/resource.rc
2562 :
2563 :
2564 :			2003-12-03 15:22:25 GMT patch-116
2565 :
2566 :			Summary:
2567 :			ME fixes.
2568 :			Revision:
2569 :			xvidcore--devapi4--1.0--patch-116
2570 :
2571 :			From syskin:
2572 :			- small typo in chroma sad reset
2573 :			- code tweaking
2574 :			+ adv diamond search used intead of mainsearch
2575 :			+ and some other stuff
2576 :			- thresholds tuned.
2577 :
2578 :			modified files:
2579 :			src/motion/estimation.h src/motion/estimation_bvop.c
2580 :			src/motion/estimation_pvop.c
2581 :
2582 :
2583 :			2003-12-01 11:17:20 GMT patch-115
2584 :
2585 :			Summary:
2586 :			Small changes and interlacing bugfix.
2587 :			Revision:
2588 :			xvidcore--devapi4--1.0--patch-115
2589 :
2590 :			* Interlacing bugfix, code got swapped in a very old patch
2591 :			(back in March)
2592 :			* xvid_encraw forces VOSH writing.
2593 :			* VFW GUI fixes.
2594 :
2595 :			modified files:
2596 :			examples/Makefile examples/xvid_encraw.c
2597 :			src/utils/mbtransquant.c vfw/src/config.c vfw/src/resource.rc
2598 :
2599 :
2600 :			2003-11-30 15:47:41 GMT patch-114
2601 :
2602 :			Summary:
2603 :			Thread safe MPEG4 quantization functions + xvid_bench update
2604 :			Revision:
2605 :			xvidcore--devapi4--1.0--patch-114
2606 :
2607 :			* Thread safe MPEG4 quantization functions.
2608 :			Cleaned up version of patch provided by Michael
2609 :			- fixed compiling problems on gcc
2610 :			- added const qualifiers every where it was possible to help C compiler
2611 :			optimization.
2612 :			- added the mpeg_quant_matrices param to all ASM function prototype
2613 :			in comments (even if it's not used, that shows we do it deliberatly)
2614 :			- forces m[intra][0][0] = 8, otherwise XviD could write invalid streams.
2615 :			* Added real CRC computing in xvid_bench.c
2616 :
2617 :			modified files:
2618 :			TODO examples/xvid_bench.c src/bitstream/bitstream.c
2619 :			src/decoder.c src/decoder.h src/encoder.c src/encoder.h
2620 :			src/motion/estimation.h src/motion/estimation_pvop.c
2621 :			src/motion/estimation_rd_based.c src/quant/quant.h
2622 :			src/quant/quant_h263.c src/quant/quant_matrix.c
2623 :			src/quant/quant_matrix.h src/quant/quant_mpeg.c
2624 :			src/quant/x86_asm/quantize_h263_3dne.asm
2625 :			src/quant/x86_asm/quantize_h263_mmx.asm
2626 :			src/quant/x86_asm/quantize_mpeg_mmx.asm
2627 :			src/quant/x86_asm/quantize_mpeg_xmm.asm
2628 :			src/utils/mbtransquant.c src/xvid.c
2629 :
2630 :
2631 :			2003-11-29 18:10:25 GMT patch-113
2632 :
2633 :			Summary:
2634 :			Fixed csp asm rules for real?
2635 :			Revision:
2636 :			xvidcore--devapi4--1.0--patch-113
2637 :
2638 :			Fixed csp asm rules for real?
2639 :
2640 :
2641 :			modified files:
2642 :			build/win32/libxvidcore.dsp
2643 :
2644 :
2645 :			2003-11-29 17:58:09 GMT patch-112
2646 :
2647 :			Summary:
2648 :			TODO/ChangeLog updated
2649 :			Revision:
2650 :			xvidcore--devapi4--1.0--patch-112
2651 :
2652 :			TODO/ChangeLog updated
2653 :
2654 :
2655 :			modified files:
2656 :			ChangeLog TODO
2657 :
2658 :
2659 :			#########################################################################
2660 :			# 1.0.0 beta1 (Bitstream Version 21)
2661 :			#########################################################################
2662 :
2663 :			2003-11-29 17:21:08 GMT patch-111
2664 :
2665 :			Summary:
2666 :			First beta marking
2667 :			Revision:
2668 :			xvidcore--devapi4--1.0--patch-111
2669 :
2670 :			First beta marking
2671 :
2672 :
2673 :			modified files:
2674 :			build/generic/configure.in src/xvid.c src/xvid.h
2675 :
2676 :
2677 :			2003-11-29 16:59:14 GMT patch-110
2678 :
2679 :			Summary:
2680 :			Catching up with CVS.
2681 :			Revision:
2682 :			xvidcore--devapi4--1.0--patch-110
2683 :
2684 :			Changes from sysKin:
2685 :			* dquant optimization.
2686 :			* CBR fix
2687 :
2688 :			modified files:
2689 :			src/encoder.c src/plugins/plugin_single.c
2690 :
2691 :
2692 :			2003-11-24 22:05:38 GMT patch-109
2693 :
2694 :			Summary:
2695 :			Big level handling in trellis.
2696 :			Revision:
2697 :			xvidcore--devapi4--1.0--patch-109
2698 :
2699 :			Trellis was treating big levels exactly the same way as lower ones.
2700 :			In some cases, trellis was doing wild optimizations favoring a 0
2701 :			because the distortion introduced by that big coeff change was acceptable.
2702 :			But visually this could result in some nasty blocks with wrong chroma
2703 :			information or similar brutal changes in other planes as well.
2704 :
2705 :			Skal added big levels handling where trellis just tries to minimize
2706 :			the cost varying the run value only. No level modification is done
2707 :			anymore.
2708 :
2709 :			modified files:
2710 :			TODO src/utils/mbtransquant.c
2711 :
2712 :
2713 :			2003-11-23 16:42:55 GMT patch-108
2714 :
2715 :			Summary:
2716 :			Trellis for MPEG.
2717 :			Revision:
2718 :			xvidcore--devapi4--1.0--patch-108
2719 :
2720 :			* Added trellis support for MPEg quantization type.
2721 :			* Changed RD fixed point precision, should help avoiding overflow
2722 :			(see the constant TL_SHIFT)
2723 :
2724 :			NB: we still have some problems when trellis optimizes DC for big DC
2725 :			values.
2726 :
2727 :			modified files:
2728 :			src/utils/mbtransquant.c
2729 :
2730 :
2731 :			2003-11-22 00:53:59 GMT patch-107
2732 :
2733 :			Summary:
2734 :			Win32 lib project fix (bis)
2735 :			Revision:
2736 :			xvidcore--devapi4--1.0--patch-107
2737 :
2738 :			* nasm >= 0.98.37 support in project file got reverted in a previous
2739 :			patch, push it back. Nota bene /O3 changed to /O2 for proper
2740 :			compilation with msvc compiler (everyone is not supposed to compile
2741 :			stuff with icc)
2742 :			* TODO update.
2743 :
2744 :			modified files:
2745 :			TODO build/win32/libxvidcore.dsp
2746 :
2747 :
2748 :			2003-11-19 21:26:34 GMT patch-106
2749 :
2750 :			Summary:
2751 :			updated bench crc
2752 :			Revision:
2753 :			xvidcore--devapi4--1.0--patch-106
2754 :
2755 :			updated bench crc
2756 :
2757 :
2758 :			modified files:
2759 :			examples/xvid_bench.c
2760 :
2761 :
2762 :			2003-11-19 16:00:00 GMT patch-105
2763 :
2764 :			Summary:
2765 :			Lumimasking fixes.
2766 :			Revision:
2767 :			xvidcore--devapi4--1.0--patch-105
2768 :
2769 :			from sysKin:
2770 :			* New plugin hook entry XVID_PLG_FRAME that happens inside FrameCodeIPB
2771 :			when both type and quant are known. Added hook handling in all plugins.
2772 :			* Fixed lumimasking.
2773 :
2774 :			from me:
2775 :			* small reverse commit in pvop estimation fixed.
2776 :
2777 :			modified files:
2778 :			src/bitstream/bitstream.c src/bitstream/bitstream.h
2779 :			src/encoder.c src/motion/estimation_pvop.c
2780 :			src/plugins/plugin_2pass1.c src/plugins/plugin_2pass2.c
2781 :			src/plugins/plugin_dump.c src/plugins/plugin_lumimasking.c
2782 :			src/plugins/plugin_psnr.c src/plugins/plugin_single.c
2783 :			src/xvid.h
2784 :
2785 :
2786 :			2003-11-19 15:37:16 GMT patch-104
2787 :
2788 :			Summary:
2789 :			Removed indirections from SearchData structure.
2790 :			Revision:
2791 :			xvidcore--devapi4--1.0--patch-104
2792 :
2793 :			Patch from sysKin:
2794 :			* removed indirections in SearchData structure. CheckCandidate functions
2795 :			don't use a const SearchData pointer anymore, but they should be a bit
2796 :			faster because of the less numerous indirections.
2797 :
2798 :			modified files:
2799 :			src/motion/estimation.h src/motion/estimation_bvop.c
2800 :			src/motion/estimation_common.c src/motion/estimation_gmc.c
2801 :			src/motion/estimation_pvop.c src/motion/estimation_rd_based.c
2802 :			src/motion/vop_type_decision.c
2803 :
2804 :
2805 :			2003-11-19 15:33:55 GMT patch-103
2806 :
2807 :			Summary:
2808 :			Formula error in twopass code.
2809 :			Revision:
2810 :			xvidcore--devapi4--1.0--patch-103
2811 :
2812 :			The reversing bframe formula in 2pass 2 was not right. This was in fact
2813 :			a test code i used when i was maintaining the code on its own branch.
2814 :			I should not have commited it :\
2815 :
2816 :			Fixed :-)
2817 :
2818 :			modified files:
2819 :			src/plugins/plugin_2pass2.c
2820 :
2821 :
2822 :			2003-11-18 21:41:08 GMT patch-102
2823 :
2824 :			Summary:
2825 :			Another problem with mis/unitialized reads.
2826 :			Revision:
2827 :			xvidcore--devapi4--1.0--patch-102
2828 :
2829 :			Michael introduced a fast subpel refine that uses a (iMinSAD2, currentQMV2)
2830 :			couple of data. The problem is that he plugged this in CheckCandidate16_qpel
2831 :			that is used outside this context, thus a if statement was traversed with
2832 :			garbage data in the standard subpel case. For perfection sake, using a
2833 :			iMinSAD=256*4096 value collects correct data even if it will not be used in
2834 :			the normal subpel case.
2835 :
2836 :			modified files:
2837 :			src/motion/estimation_pvop.c
2838 :
2839 :
2840 :			2003-11-16 17:29:39 GMT patch-101
2841 :
2842 :			Summary:
2843 :			The MEanalysis patch assumed bvops were always used
2844 :			Revision:
2845 :			xvidcore--devapi4--1.0--patch-101
2846 :
2847 :			The MEanalysis patch assumed bvops were always used
2848 :
2849 :
2850 :			modified files:
2851 :			src/encoder.c
2852 :
2853 :
2854 :			2003-11-16 15:12:15 GMT patch-100
2855 :
2856 :			Summary:
2857 :			MEanalysis using wrong mvs + bframe search using wrong mvs
2858 :			Revision:
2859 :			xvidcore--devapi4--1.0--patch-100
2860 :
2861 :			Valgrind reported lot of unitialized reads. These unitialized reads
2862 :			helped sysKin finding three bugs:
2863 :			- ZeroMacroblock did not reset the cbp field. So for some skipped
2864 :			blocks, a test was done on the cbp value...
2865 :			- MEanalysis was using wrong mvs from the current bvop (unitialized
2866 :			or just wrong in current context). That's because in devapi3, bframes
2867 :			used to share the same mvs array wheras now, it's one array a bvop.
2868 :			- Collocated skipped MBs for a bvop didn't reset mvs[0] and b_mvs[0].
2869 :
2870 :			modified files:
2871 :			src/encoder.c src/motion/estimation_bvop.c src/motion/motion.h
2872 :			src/motion/motion_inlines.h src/motion/vop_type_decision.c
2873 :
2874 :
2875 :			2003-11-15 15:21:09 GMT patch-99
2876 :
2877 :			Summary:
2878 :			Small fixes
2879 :			Revision:
2880 :			xvidcore--devapi4--1.0--patch-99
2881 :
2882 :			Small fixes
2883 :
2884 :
2885 :			modified files:
2886 :			src/encoder.c src/motion/vop_type_decision.c
2887 :
2888 :
2889 :			2003-11-15 15:02:47 GMT patch-98
2890 :
2891 :			Summary:
2892 :			DShow update + libxvidcore project file update
2893 :			Revision:
2894 :			xvidcore--devapi4--1.0--patch-98
2895 :
2896 :			From peter:
2897 :			* DShow now links against libxvidcore.lib
2898 :			* Some updates to the libxvidcore project file
2899 :
2900 :			modified files:
2901 :			TODO build/win32/libxvidcore.dsp dshow/dshow.dsp
2902 :			dshow/src/CXvidDecoder.cpp dshow/src/CXvidDecoder.h
2903 :
2904 :
2905 :			2003-11-15 01:51:28 GMT patch-97
2906 :
2907 :			Summary:
2908 :			Small updates to doc files
2909 :			Revision:
2910 :			xvidcore--devapi4--1.0--patch-97
2911 :
2912 :			Small updates to doc files
2913 :
2914 :
2915 :			modified files:
2916 :			AUTHORS CodingStyle TODO doc/INSTALL doc/README
2917 :
2918 :			renamed files:
2919 :			.arch-ids/authors.txt.id
2920 :			==> .arch-ids/AUTHORS.id
2921 :			.arch-ids/todo.txt.id
2922 :			==> .arch-ids/TODO.id
2923 :			authors.txt
2924 :			==> AUTHORS
2925 :			todo.txt
2926 :			==> TODO
2927 :
2928 :
2929 :			2003-11-14 11:23:55 GMT patch-96
2930 :
2931 :			Summary:
2932 :			Updated ChangeLog
2933 :			Revision:
2934 :			xvidcore--devapi4--1.0--patch-96
2935 :
2936 :			Updated ChangeLog
2937 :
2938 :
2939 :			modified files:
2940 :			ChangeLog
2941 :
2942 :
2943 :			2003-11-13 23:09:34 GMT patch-95
2944 :
2945 :			Summary:
2946 :			8x8 16bit Block SSE optimization.
2947 :			Revision:
2948 :			xvidcore--devapi4--1.0--patch-95
2949 :
2950 :			MMXed the calculation of SSE for 8x8 16bit blocks. This helps quite
2951 :			a lot VHQ=4 mode.
2952 :
2953 :			My tests show with trellis:chroma_me:
2954 :			- ~20% speed improvement for vhq=4.
2955 :			- at least 5% when using vhq=1.
2956 :
2957 :			Of course this speedup vanishes if more CPU intensive features are used.
2958 :			CruNcher who used gmc/qpel, noticed "only" a ~5% speed improvement.
2959 :
2960 :			NB: i'm of course talking about overall speed improvement. Such a small
2961 :			patch for such a big improvement :-)
2962 :
2963 :			modified files:
2964 :			src/motion/estimation_rd_based.c src/motion/sad.c
2965 :			src/motion/sad.h src/motion/x86_asm/sad_mmx.asm src/xvid.c
2966 :
2967 :
2968 :			2003-11-13 22:34:33 GMT patch-94
2969 :
2970 :			Summary:
2971 :			Various small bug fixes.
2972 :			Revision:
2973 :			xvidcore--devapi4--1.0--patch-94
2974 :
2975 :			* encoder.c: GMC code fix in encoder.c. Now gmcval is initialized correctly
2976 :			when using GME.
2977 :			* xvid_decraw.c: Fix elementary stream output.
2978 :			* plugin_2pass2.c: Small parsing bug in stats reading in 2pass2.
2979 :			* decoder.c: Read resync markers in bframes.
2980 :
2981 :			modified files:
2982 :			examples/xvid_decraw.c src/decoder.c src/encoder.c
2983 :			src/plugins/plugin_2pass2.c
2984 :
2985 :
2986 :			2003-11-11 16:24:05 GMT patch-93
2987 :
2988 :			Summary:
2989 :			VFW update for overflow control
2990 :			Revision:
2991 :			xvidcore--devapi4--1.0--patch-93
2992 :
2993 :			From Koepi.
2994 :			* Added widget and code for overflow control strength.
2995 :			* Removed widgets for payback options and kfthresholds.
2996 :			* Activated frame stats in DebugOutputView all the time.
2997 :
2998 :			From me:
2999 :			* Activated static motion detection in cartoon mode.
3000 :
3001 :			modified files:
3002 :			vfw/src/codec.c vfw/src/config.c vfw/src/resource.h
3003 :			vfw/src/resource.rc
3004 :
3005 :
3006 :			2003-11-09 20:47:47 GMT patch-92
3007 :
3008 :			Summary:
3009 :			New two pass code.
3010 :			Revision:
3011 :			xvidcore--devapi4--1.0--patch-92
3012 :
3013 :			New two pass code. I may say it's just a fixed version, though it looks
3014 :			more like a "take all the ideas and write it again" version. It performs
3015 :			better with all natural sequences i have and a bit worse with anime.
3016 :
3017 :			Including it now, allow me improving the code during the beta releases.
3018 :
3019 :			modified files:
3020 :			src/encoder.c src/plugins/plugin_2pass1.c
3021 :			src/plugins/plugin_2pass2.c src/xvid.h vfw/src/codec.c
3022 :			vfw/src/config.c vfw/src/config.h
3023 :
3024 :
3025 :			2003-11-09 17:07:16 GMT patch-91
3026 :
3027 :			Summary:
3028 :			Fixes for bframe compensation (used in psnr tests).
3029 :			Revision:
3030 :			xvidcore--devapi4--1.0--patch-91
3031 :
3032 :			* transfer_8to16_sub2_(c\|mmx\|xmm\|3dne) write back the compensated
3033 :			result to current frame pointer.
3034 :			* transfer_8to16_sub2_mmx uses proper rounding (a+b+1)/2. The +1
3035 :			operation was missing.
3036 :			* Blocks skipped in bframes must be compensated for psnr computing.
3037 :
3038 :			modified files:
3039 :			src/encoder.c src/motion/estimation_bvop.c
3040 :			src/utils/mem_transfer.c
3041 :			src/utils/x86_asm/mem_transfer_3dne.asm
3042 :			src/utils/x86_asm/mem_transfer_mmx.asm
3043 :
3044 :
3045 :			2003-11-05 16:05:44 GMT patch-90
3046 :
3047 :			Summary:
3048 :			Speed improvement not wasting setedges and interpolate calls.
3049 :			Revision:
3050 :			xvidcore--devapi4--1.0--patch-90
3051 :
3052 :			Patch from syskin.
3053 :
3054 :			* This patch avoids calling setedges and interpolate for uneeded cases:
3055 :			- setedges is only called once per frame.
3056 :			- interpolate is called only when the previous rounding
3057 :			was different from the one needed.
3058 :			* Interpolation has been optimized a bit for qpel case, we do the
3059 :			hv pass down to top to use the cache more efficiently (hope so).
3060 :
3061 :			modified files:
3062 :			src/encoder.c src/encoder.h src/image/image.c
3063 :
3064 :
3065 :			2003-11-03 19:51:12 GMT patch-89
3066 :
3067 :			Summary:
3068 :			SSE2 dev16 fix + xvid_bench DCT block alignments.
3069 :			Revision:
3070 :			xvidcore--devapi4--1.0--patch-89
3071 :
3072 :			* Small error fixed by Skal in his dev16 code (missing pshufd).
3073 :			* Blocks used by DCT tests are now aligned with DECLARE_ALIGNED_MATRIX
3074 :			this avoids the well know segfaults when using SSE2 instructions that
3075 :			suppose data alignment.
3076 :
3077 :			modified files:
3078 :			examples/xvid_bench.c src/motion/x86_asm/sad_sse2.asm
3079 :
3080 :
3081 :			2003-11-03 15:42:23 GMT patch-88
3082 :
3083 :			Summary:
3084 :			Align .rodata section for non coff objects
3085 :			Revision:
3086 :			xvidcore--devapi4--1.0--patch-88
3087 :
3088 :			Align .rodata section for non coff objects
3089 :
3090 :
3091 :			modified files:
3092 :			src/bitstream/x86_asm/cbp_mmx.asm
3093 :			src/bitstream/x86_asm/cbp_sse2.asm
3094 :			src/dct/x86_asm/fdct_mmx_ffmpeg.asm
3095 :			src/dct/x86_asm/fdct_mmx_skal.asm
3096 :			src/dct/x86_asm/fdct_sse2_skal.asm
3097 :			src/dct/x86_asm/idct_3dne.asm src/dct/x86_asm/idct_mmx.asm
3098 :			src/dct/x86_asm/idct_sse2_dmitry.asm
3099 :			src/dct/x86_asm/simple_idct_mmx.asm
3100 :			src/image/x86_asm/colorspace_rgb_mmx.asm
3101 :			src/image/x86_asm/colorspace_yuyv_mmx.asm
3102 :			src/image/x86_asm/interpolate8x8_3dn.asm
3103 :			src/image/x86_asm/interpolate8x8_3dne.asm
3104 :			src/image/x86_asm/interpolate8x8_mmx.asm
3105 :			src/image/x86_asm/interpolate8x8_xmm.asm
3106 :			src/image/x86_asm/qpel_mmx.asm
3107 :			src/image/x86_asm/reduced_mmx.asm
3108 :			src/motion/x86_asm/sad_3dn.asm src/motion/x86_asm/sad_3dne.asm
3109 :			src/motion/x86_asm/sad_mmx.asm src/motion/x86_asm/sad_sse2.asm
3110 :			src/motion/x86_asm/sad_xmm.asm
3111 :			src/quant/x86_asm/quantize_h263_3dne.asm
3112 :			src/quant/x86_asm/quantize_h263_mmx.asm
3113 :			src/quant/x86_asm/quantize_mpeg_mmx.asm
3114 :			src/quant/x86_asm/quantize_mpeg_xmm.asm
3115 :			src/utils/x86_asm/cpuid.asm
3116 :			src/utils/x86_asm/interlacing_mmx.asm
3117 :			src/utils/x86_asm/mem_transfer_3dne.asm
3118 :
3119 :
3120 :			2003-11-02 23:01:43 GMT patch-87
3121 :
3122 :			Summary:
3123 :			SSE2 update
3124 :			Revision:
3125 :			xvidcore--devapi4--1.0--patch-87
3126 :
3127 :			* Added Dmitry SSE2 iDCT code back.
3128 :			* Plugged Dmitry iDCT as default for SSE2
3129 :			* Fixed a bug in xvid_bench that was making it would test some CPU
3130 :			instruction set w/o host CPU support. xvidcore init was simply
3131 :			discarding irrelevant cpu flags.
3132 :
3133 :			new files:
3134 :			src/dct/x86_asm/.arch-ids/idct_sse2_dmitry.asm.id
3135 :			src/dct/x86_asm/idct_sse2_dmitry.asm
3136 :
3137 :			modified files:
3138 :			build/generic/sources.inc build/win32/libxvidcore.dsp
3139 :			examples/xvid_bench.c src/dct/fdct.h src/dct/idct.h src/xvid.c
3140 :
3141 :
3142 :			2003-10-31 14:53:26 GMT patch-86
3143 :
3144 :			Summary:
3145 :			Better handling of old windres versions + GNU make dependency.
3146 :			Revision:
3147 :			xvidcore--devapi4--1.0--patch-86
3148 :
3149 :			Old versions of GNU windres (<2.14) don't have the same short options.
3150 :			But long options remain the same so it's better to use long option names
3151 :			to have ful compatibility with older versions.
3152 :
3153 :			The Makefile appears to be dependent on GNU make because shell expansion
3154 :			for retrieving the path of the Makefile is wrong when using `` even with
3155 :			a single expansion assignment :=. It keeps being expanded when used.
3156 :
3157 :			modified files:
3158 :			doc/INSTALL vfw/bin/Makefile
3159 :
3160 :
3161 :			2003-10-29 11:31:28 GMT patch-85
3162 :
3163 :			Summary:
3164 :			Added sse2 f/iDCT code from skal
3165 :			Revision:
3166 :			xvidcore--devapi4--1.0--patch-85
3167 :
3168 :			* Added sse2 f/iDCT code from skal
3169 :			* Added hooking in xvid.c
3170 :
3171 :			new files:
3172 :			src/dct/x86_asm/.arch-ids/fdct_sse2_skal.asm.id
3173 :			src/dct/x86_asm/fdct_sse2_skal.asm
3174 :
3175 :			modified files:
3176 :			build/generic/sources.inc build/win32/libxvidcore.dsp
3177 :			src/xvid.c
3178 :
3179 :
3180 :			2003-10-29 00:19:10 GMT patch-84
3181 :
3182 :			Summary:
3183 :			Fix the static motion detection
3184 :			Revision:
3185 :			xvidcore--devapi4--1.0--patch-84
3186 :
3187 :			Fix the static motion detection
3188 :
3189 :
3190 :			modified files:
3191 :			src/motion/estimation_pvop.c
3192 :
3193 :
3194 :			2003-10-28 23:39:46 GMT patch-83
3195 :
3196 :			Summary:
3197 :			Added cartoon option handling.
3198 :			Revision:
3199 :			xvidcore--devapi4--1.0--patch-83
3200 :
3201 :			Added cartoon widgets + handling code.
3202 :
3203 :			NB: static motion detection is disabled because of crashes on P4
3204 :			cpus.
3205 :
3206 :			modified files:
3207 :			vfw/src/codec.c vfw/src/config.c vfw/src/config.h
3208 :			vfw/src/resource.h vfw/src/resource.rc
3209 :
3210 :
3211 :			2003-10-28 17:44:09 GMT patch-82
3212 :
3213 :			Summary:
3214 :			ASM cleanups;
3215 :			Revision:
3216 :			xvidcore--devapi4--1.0--patch-82
3217 :
3218 :			* Applied same style to all asm files
3219 :			* Replaced current sad sse2 operators with skal's ones
3220 :			* Removed old and unused colorspace asm files
3221 :
3222 :			removed files:
3223 :			src/image/x86_asm/.arch-ids/rgb_to_yv12_mmx.asm.id
3224 :			src/image/x86_asm/.arch-ids/yuv_to_yv12_mmx.asm.id
3225 :			src/image/x86_asm/.arch-ids/yuyv_to_yv12_mmx.asm.id
3226 :			src/image/x86_asm/.arch-ids/yv12_to_rgb24_mmx.asm.id
3227 :			src/image/x86_asm/.arch-ids/yv12_to_rgb32_mmx.asm.id
3228 :			src/image/x86_asm/.arch-ids/yv12_to_yuyv_mmx.asm.id
3229 :			src/image/x86_asm/rgb_to_yv12_mmx.asm
3230 :			src/image/x86_asm/yuv_to_yv12_mmx.asm
3231 :			src/image/x86_asm/yuyv_to_yv12_mmx.asm
3232 :			src/image/x86_asm/yv12_to_rgb24_mmx.asm
3233 :			src/image/x86_asm/yv12_to_rgb32_mmx.asm
3234 :			src/image/x86_asm/yv12_to_yuyv_mmx.asm
3235 :
3236 :			modified files:
3237 :			build/generic/sources.inc src/bitstream/x86_asm/cbp_3dne.asm
3238 :			src/bitstream/x86_asm/cbp_mmx.asm
3239 :			src/bitstream/x86_asm/cbp_sse2.asm
3240 :			src/dct/x86_asm/fdct_mmx_ffmpeg.asm
3241 :			src/dct/x86_asm/fdct_mmx_skal.asm
3242 :			src/dct/x86_asm/idct_3dne.asm src/dct/x86_asm/idct_mmx.asm
3243 :			src/dct/x86_asm/simple_idct_mmx.asm
3244 :			src/image/x86_asm/colorspace_mmx.inc
3245 :			src/image/x86_asm/colorspace_rgb_mmx.asm
3246 :			src/image/x86_asm/colorspace_yuv_mmx.asm
3247 :			src/image/x86_asm/colorspace_yuyv_mmx.asm
3248 :			src/image/x86_asm/interpolate8x8_3dn.asm
3249 :			src/image/x86_asm/interpolate8x8_3dne.asm
3250 :			src/image/x86_asm/interpolate8x8_mmx.asm
3251 :			src/image/x86_asm/interpolate8x8_xmm.asm
3252 :			src/image/x86_asm/qpel_mmx.asm
3253 :			src/image/x86_asm/reduced_mmx.asm
3254 :			src/motion/x86_asm/sad_3dn.asm src/motion/x86_asm/sad_3dne.asm
3255 :			src/motion/x86_asm/sad_mmx.asm src/motion/x86_asm/sad_sse2.asm
3256 :			src/motion/x86_asm/sad_xmm.asm
3257 :			src/quant/x86_asm/quantize_h263_3dne.asm
3258 :			src/quant/x86_asm/quantize_h263_mmx.asm
3259 :			src/quant/x86_asm/quantize_mpeg_mmx.asm
3260 :			src/quant/x86_asm/quantize_mpeg_xmm.asm
3261 :			src/utils/x86_asm/cpuid.asm
3262 :			src/utils/x86_asm/interlacing_mmx.asm
3263 :			src/utils/x86_asm/mem_transfer_3dne.asm
3264 :			src/utils/x86_asm/mem_transfer_mmx.asm
3265 :
3266 :
3267 :			2003-10-27 01:13:47 GMT patch-81
3268 :
3269 :			Summary:
3270 :			d_mv_bits speedup from sysKin
3271 :			Revision:
3272 :			xvidcore--devapi4--1.0--patch-81
3273 :
3274 :			d_mv_bits speedup from sysKin
3275 :
3276 :
3277 :			modified files:
3278 :			src/motion/motion_inlines.h
3279 :
3280 :
3281 :			2003-10-27 00:55:51 GMT patch-80
3282 :
3283 :			Summary:
3284 :			fDCT changes, new asm CodingStyle applied to dct dir
3285 :			Revision:
3286 :			xvidcore--devapi4--1.0--patch-80
3287 :
3288 :			* Ported the ffmpeg fDCT functions (mmx and xmm).
3289 :			* Modified the skal's versions a bit to allow rolling loops.
3290 :			* Activated Skal's fDCTs (unrolled versions) for mmx _and_ xmm
3291 :			(old code was ignoring xmm versions)
3292 :			* Removed the SSE2 versions (they'll be back later)
3293 :			* .data -> .rodata
3294 :			* Applied announced asm CodingStyle to the dct dir
3295 :			(I'll have to add a section with the said CodingStyle)
3296 :
3297 :			modified files:
3298 :			build/generic/sources.inc build/win32/libxvidcore.dsp
3299 :			src/dct/fdct.h src/dct/idct.h
3300 :			src/dct/x86_asm/fdct_mmx_ffmpeg.asm
3301 :			src/dct/x86_asm/fdct_mmx_skal.asm
3302 :			src/dct/x86_asm/idct_3dne.asm src/dct/x86_asm/idct_mmx.asm
3303 :			src/dct/x86_asm/simple_idct_mmx.asm src/xvid.c
3304 :
3305 :			renamed files:
3306 :			src/dct/x86_asm/.arch-ids/fdct_mmx.asm.id
3307 :			==> src/dct/x86_asm/.arch-ids/fdct_mmx_ffmpeg.asm.id
3308 :			src/dct/x86_asm/.arch-ids/fdct_xmm.asm.id
3309 :			==> src/dct/x86_asm/.arch-ids/fdct_mmx_skal.asm.id
3310 :			src/dct/x86_asm/fdct_mmx.asm
3311 :			==> src/dct/x86_asm/fdct_mmx_ffmpeg.asm
3312 :			src/dct/x86_asm/fdct_xmm.asm
3313 :			==> src/dct/x86_asm/fdct_mmx_skal.asm
3314 :
3315 :
3316 :			2003-10-25 13:48:42 GMT patch-79
3317 :
3318 :			Summary:
3319 :			BQuant->PQuant fix.
3320 :			Revision:
3321 :			xvidcore--devapi4--1.0--patch-79
3322 :
3323 :			When using closed_gop, a BFrame before an IFrame is turned into a PFrame.
3324 :			Thus is original quant has to be computed back, a rounding was causing these
3325 :			frames to be orig_quant-1. As a consequence we had very big frames before
3326 :			the IFrame, loosing many bits for nearly no visual benefit.
3327 :
3328 :			modified files:
3329 :			src/encoder.c
3330 :
3331 :
3332 :			2003-10-25 10:26:48 GMT patch-78
3333 :
3334 :			Summary:
3335 :			Added closed gop option to xvid_encraw
3336 :			Revision:
3337 :			xvidcore--devapi4--1.0--patch-78
3338 :
3339 :			Added closed gop option to xvid_encraw
3340 :
3341 :
3342 :			modified files:
3343 :			examples/xvid_encraw.c
3344 :
3345 :
3346 :			2003-10-24 17:39:53 GMT patch-77
3347 :
3348 :			Summary:
3349 :			RD fixes.
3350 :			Revision:
3351 :			xvidcore--devapi4--1.0--patch-77
3352 :
3353 :			Inter RD optimization relied on buggy functions to predict bitsize.
3354 :
3355 :			modified files:
3356 :			src/motion/estimation_pvop.c src/motion/motion_inlines.h
3357 :
3358 :
3359 :			2003-10-22 15:48:01 GMT patch-76
3360 :
3361 :			Summary:
3362 :			Small INSTALL update due to previous build patches.
3363 :			Revision:
3364 :			xvidcore--devapi4--1.0--patch-76
3365 :
3366 :			Small INSTALL update due to previous build patches.
3367 :
3368 :			modified files:
3369 :			doc/INSTALL
3370 :
3371 :
3372 :			2003-10-21 21:27:46 GMT patch-75
3373 :
3374 :			Summary:
3375 :			Removed unused next_block vars.
3376 :			Revision:
3377 :			xvidcore--devapi4--1.0--patch-75
3378 :
3379 :			Removed unused next_block vars.
3380 :
3381 :			modified files:
3382 :			src/decoder.c
3383 :
3384 :
3385 :			2003-10-21 21:24:15 GMT patch-74
3386 :
3387 :			Summary:
3388 :			VFW build changes.
3389 :			Revision:
3390 :			xvidcore--devapi4--1.0--patch-74
3391 :
3392 :			The build system has been modified to look like the core lib one
3393 :			minus the configure system.
3394 :
3395 :			modified files:
3396 :			vfw/bin/Makefile vfw/bin/sources.inc vfw/src/config.c
3397 :			vfw/vfw.dsp
3398 :
3399 :			renamed files:
3400 :			vfw/bin/.arch-ids/Makefile.cygwin.id
3401 :			==> vfw/bin/.arch-ids/Makefile.id
3402 :			vfw/bin/.arch-ids/Makefile.inc.id
3403 :			==> vfw/bin/.arch-ids/sources.inc.id
3404 :			vfw/bin/Makefile.cygwin
3405 :			==> vfw/bin/Makefile
3406 :			vfw/bin/Makefile.inc
3407 :			==> vfw/bin/sources.inc
3408 :			vfw/src/.arch-ids/config.rc.id
3409 :			==> vfw/src/.arch-ids/resource.rc.id
3410 :			vfw/src/config.rc
3411 :			==> vfw/src/resource.rc
3412 :
3413 :
3414 :			2003-10-21 17:00:09 GMT patch-73
3415 :
3416 :			Summary:
3417 :			Decoder cleanups and speedup
3418 :			Revision:
3419 :			xvidcore--devapi4--1.0--patch-73
3420 :
3421 :			cleanups, speedups from sysKin
3422 :
3423 :			modified files:
3424 :			src/decoder.c
3425 :
3426 :
3427 :			2003-10-21 16:22:15 GMT patch-72
3428 :
3429 :			Summary:
3430 :			Build fixes for newer nasm versions.
3431 :			Revision:
3432 :			xvidcore--devapi4--1.0--patch-72
3433 :
3434 :			nasm does not take care of adding trailing slashes to include paths.
3435 :			A patch to upstream authors has been refused because "the backslash()
3436 :			feature has been abandoned to get back to old nasm behavior"
3437 :
3438 :			Their choice is kinda stupid as nasm is now open to user mistakes... :\
3439 :			So we fix that on ou side.
3440 :
3441 :			modified files:
3442 :			build/generic/configure.in build/win32/libxvidcore.dsp
3443 :
3444 :
3445 :			2003-10-17 15:13:12 GMT patch-71
3446 :
3447 :			Summary:
3448 :			Updated docs.
3449 :			Revision:
3450 :			xvidcore--devapi4--1.0--patch-71
3451 :
3452 :			The doc of devapi4 is mostly out dated, i t is much better not to
3453 :			keep it in the repository at the moment. We'll add new docs later.
3454 :
3455 :			Added a INSTALL doc that explains the build/install process for
3456 :			supported platforms. It's a first try, things may be added later.
3457 :
3458 :			new files:
3459 :			doc/.arch-ids/INSTALL.id doc/INSTALL
3460 :
3461 :			removed files:
3462 :			doc/.arch-ids/API.dox.id doc/.arch-ids/Makefile.id
3463 :			doc/.arch-ids/foot.inc.in.id doc/.arch-ids/header.tex.in.id
3464 :			doc/.arch-ids/xvid-decoding.txt.id
3465 :			doc/.arch-ids/xvid-encoder.txt.id doc/API.dox doc/Makefile
3466 :			doc/foot.inc.in doc/header.tex.in doc/xvid-decoding.txt
3467 :			doc/xvid-encoder.txt
3468 :
3469 :			modified files:
3470 :			CodingStyle README doc/README
3471 :
3472 :			renamed files:
3473 :			.arch-ids/README.txt.id
3474 :			==> .arch-ids/README.id
3475 :			.arch-ids/changelog.txt.id
3476 :			==> .arch-ids/ChangeLog.id
3477 :			README.txt
3478 :			==> README
3479 :			changelog.txt
3480 :			==> ChangeLog
3481 :
3482 :
3483 :			2003-10-15 13:53:11 GMT patch-70
3484 :
3485 :			Summary:
3486 :			Better cross compilation handling.
3487 :			Revision:
3488 :			xvidcore--devapi4--1.0--patch-70
3489 :
3490 :			With this patch it is now possible to cross compile xvid
3491 :			quite easily for win32 platform on a build linux host.
3492 :
3493 :			Recipe for debian system:
3494 :			$ apt-get install mingw32
3495 :			(or create your own cross compiler/binutils suite and install mingw32
3496 :			header files -- sorry i don't have a recipe for this, this is let as
3497 :			an exercice for the reader)
3498 :			$ cd ${xvidcore}
3499 :			$ cd build/generic
3500 :			$ ./bootstrap.sh
3501 :			$ ./configure --host=i586-mingw32msvc
3502 :			(all occurences of i586-mingw32msvc may be replaced with the right
3503 :			prefix you've choosen for your cross compiler and cross binutils)
3504 :			$ make
3505 :			$ cd ../../vfw/bin
3506 :			$ make -f Makefile.cygwin \
3507 :			CC=i586-mingw32msvc-gcc WINDRES=i586-mingw32msvc-windres
3508 :
3509 :			Enjoy your win32 xvid.dll build by free software, on a free OS, for a devil
3510 :			OS target.
3511 :
3512 :			modified files:
3513 :			build/generic/configure.in vfw/bin/Makefile.cygwin
3514 :
3515 :
3516 :			2003-10-14 15:17:28 GMT patch-69
3517 :
3518 :			Summary:
3519 :			Fixed Qpel+Interpolation decoding. Cleaned up mb->mode usage.
3520 :			Revision:
3521 :			xvidcore--devapi4--1.0--patch-69
3522 :
3523 :			* Fixed interpolate mode + qpel decoding.
3524 :			* MB->mb_type completely replaced by MB->mode
3525 :
3526 :			modified files:
3527 :			src/decoder.c
3528 :
3529 :
3530 :			2003-10-12 21:57:24 GMT patch-68
3531 :
3532 :			Summary:
3533 :			ac/dc prediction for intra RD search.
3534 :			Revision:
3535 :			xvidcore--devapi4--1.0--patch-68
3536 :
3537 :			From syskin, added real ac/dc prediction for INTRA's bitcount.
3538 :
3539 :			modified files:
3540 :			src/motion/estimation_rd_based.c
3541 :
3542 :
3543 :			2003-10-09 18:15:50 GMT patch-67
3544 :
3545 :			Summary:
3546 :			Pigrated asm code to new quant API.
3547 :			Revision:
3548 :			xvidcore--devapi4--1.0--patch-67
3549 :
3550 :			Many changes that are mostly cosmetic in the asm files.
3551 :			* indent
3552 :			* added xor eax, eax in quant_(h263\|mpeg)_intra_.* functions
3553 :			(just to make sure the returned value isn't random)
3554 :			* added xor eax, eax in dequant_(h263\|mpeg)_.* functions
3555 :			(just to make sure the returned value isn't random)
3556 :			* synced cpuid.asm XVID_CPU_feature constants with the one
3557 :			defined in the C code (xvid.h)
3558 :			* enabled all cpu tests in xvid_bench.c
3559 :
3560 :			modified files:
3561 :			examples/xvid_bench.c src/quant/quant_h263.c
3562 :			src/quant/quant_mpeg.c
3563 :			src/quant/x86_asm/quantize_h263_3dne.asm
3564 :			src/quant/x86_asm/quantize_h263_mmx.asm
3565 :			src/quant/x86_asm/quantize_mpeg_mmx.asm
3566 :			src/quant/x86_asm/quantize_mpeg_xmm.asm
3567 :			src/utils/x86_asm/cpuid.asm src/xvid.h
3568 :
3569 :
3570 :			2003-10-08 21:05:47 GMT patch-66
3571 :
3572 :			Summary:
3573 :			Updated xvid_bench for quant API changes
3574 :			Revision:
3575 :			xvidcore--devapi4--1.0--patch-66
3576 :
3577 :			Updated xvid_bench for quant API changes
3578 :
3579 :
3580 :			modified files:
3581 :			examples/xvid_bench.c
3582 :
3583 :
3584 :			2003-10-07 13:03:51 GMT patch-65
3585 :
3586 :			Summary:
3587 :			Quant functions API changes (first step)
3588 :			Revision:
3589 :			xvidcore--devapi4--1.0--patch-65
3590 :
3591 :			In the road to instance safe mpeg quantization, a small cleanup
3592 :			to the quant API was needed. It consists in changing the way we
3593 :			name the functions quant_{mpeg\|h263}_{inter\|intra}_{arch} and in
3594 :			a move to a more unified API (even intra functions return the sum
3595 :			of coefficients, it can be used as a complexity measure at a later
3596 :			time).
3597 :
3598 :			This patch touch lot of files, but all changes are trivial.
3599 :
3600 :			NB: we should check the IA64 asm validity, i changed things but
3601 :			i can't test them.
3602 :
3603 :			new files:
3604 :			src/quant/.arch-ids/quant.h.id src/quant/quant.h
3605 :
3606 :			removed files:
3607 :			src/quant/.arch-ids/quant_h263.h.id
3608 :			src/quant/.arch-ids/quant_mpeg4.h.id src/quant/quant_h263.h
3609 :			src/quant/quant_mpeg4.h
3610 :
3611 :			modified files:
3612 :			build/generic/sources.inc build/win32/libxvidcore.dsp
3613 :			src/decoder.c src/encoder.c src/image/qpel.c
3614 :			src/motion/estimation_rd_based.c
3615 :			src/quant/ia64_asm/quant_h263_ia64.s src/quant/quant_h263.c
3616 :			src/quant/quant_matrix.c src/quant/quant_matrix.h
3617 :			src/quant/quant_mpeg.c
3618 :			src/quant/x86_asm/quantize_h263_3dne.asm
3619 :			src/quant/x86_asm/quantize_h263_mmx.asm
3620 :			src/quant/x86_asm/quantize_mpeg_mmx.asm
3621 :			src/quant/x86_asm/quantize_mpeg_xmm.asm
3622 :			src/utils/mbtransquant.c src/xvid.c src/xvid.h
3623 :
3624 :			renamed files:
3625 :			src/quant/.arch-ids/quant_mpeg4.c.id
3626 :			==> src/quant/.arch-ids/quant_mpeg.c.id
3627 :			src/quant/quant_mpeg4.c
3628 :			==> src/quant/quant_mpeg.c
3629 :			src/quant/x86_asm/.arch-ids/quantize4_mmx.asm.id
3630 :			==> src/quant/x86_asm/.arch-ids/quantize_mpeg_mmx.asm.id
3631 :			src/quant/x86_asm/.arch-ids/quantize4_xmm.asm.id
3632 :			==> src/quant/x86_asm/.arch-ids/quantize_mpeg_xmm.asm.id
3633 :			src/quant/x86_asm/.arch-ids/quantize_3dne.asm.id
3634 :			==> src/quant/x86_asm/.arch-ids/quantize_h263_3dne.asm.id
3635 :			src/quant/x86_asm/.arch-ids/quantize_mmx.asm.id
3636 :			==> src/quant/x86_asm/.arch-ids/quantize_h263_mmx.asm.id
3637 :			src/quant/x86_asm/quantize4_mmx.asm
3638 :			==> src/quant/x86_asm/quantize_mpeg_mmx.asm
3639 :			src/quant/x86_asm/quantize4_xmm.asm
3640 :			==> src/quant/x86_asm/quantize_mpeg_xmm.asm
3641 :			src/quant/x86_asm/quantize_3dne.asm
3642 :			==> src/quant/x86_asm/quantize_h263_3dne.asm
3643 :			src/quant/x86_asm/quantize_mmx.asm
3644 :			==> src/quant/x86_asm/quantize_h263_mmx.asm
3645 :
3646 :
3647 :			2003-10-05 00:15:15 GMT patch-64
3648 :
3649 :			Summary:
3650 :			Updated ChangeLog
3651 :			Revision:
3652 :			xvidcore--devapi4--1.0--patch-64
3653 :
3654 :			Updated ChangeLog
3655 :
3656 :
3657 :			modified files:
3658 :			changelog.txt
3659 :
3660 :
3661 :			2003-10-04 16:04:30 GMT patch-63
3662 :
3663 :			Summary:
3664 :			Removed legacy 2pass code from vfw
3665 :			Revision:
3666 :			xvidcore--devapi4--1.0--patch-63
3667 :
3668 :			Removed legacy 2pass code from vfw
3669 :
3670 :
3671 :			removed files:
3672 :			vfw/src/.arch-ids/2pass.c.id vfw/src/.arch-ids/2pass.h.id
3673 :			vfw/src/2pass.c vfw/src/2pass.h
3674 :
3675 :
3676 :			2003-10-04 00:41:38 GMT patch-62
3677 :
3678 :			Summary:
3679 :			Working VFW mingw/cygwin build system.
3680 :			Revision:
3681 :			xvidcore--devapi4--1.0--patch-62
3682 :
3683 :			This patch fixes the VFW building process. Now it should work out
3684 :			of the box using these steps:
3685 :			# cd ${xvidcore}
3686 :			# cd build/generic
3687 :			# ./bootstrap.sh <-- only needed for CVS checkouts.
3688 :			# ./configure
3689 :			# make
3690 :			# cd ../../vfw/bin
3691 :			# make -f Makefile.cygwin
3692 :
3693 :			Then install as usual clicking on the inf file or "make install"
3694 :			in the vfw/bin dir.
3695 :
3696 :			modified files:
3697 :			vfw/bin/Makefile.cygwin vfw/src/config.rc vfw/src/debug.h
3698 :			vfw/src/driverproc.c
3699 :
3700 :
3701 :			2003-10-03 17:00:53 GMT patch-61
3702 :
3703 :			Summary:
3704 :			Fixes for alternate scan and interlacing support.
3705 :			Revision:
3706 :			xvidcore--devapi4--1.0--patch-61
3707 :
3708 :			Fixes from CVS (by sysKin) for:
3709 :			- added alternate scan support with VHQ
3710 :			- fixed interlacing support in s/b-frames. May fix a potential problem
3711 :			as field_pred struct field seemed not to be initialized anywhere.
3712 :			As it's not supported yet, write an hardcoded 0 bit.
3713 :
3714 :			Fixes from me for the fixes from syKin:
3715 :			- scan_table effectively used in MBCodingBVOP
3716 :			- Block_CalcBits(Intra) fixes to data->scan_table (implies prototype
3717 :			change and code modification every where the functions were used)
3718 :
3719 :			I also increased BS version as it might result in different bitstreams
3720 :			It's now at version 20.
3721 :
3722 :			modified files:
3723 :			src/bitstream/mbcoding.c src/bitstream/mbcoding.h
3724 :			src/decoder.c src/encoder.c src/motion/estimation.h
3725 :			src/motion/estimation_rd_based.c src/xvid.h
3726 :
3727 :
3728 :			2003-10-03 15:41:37 GMT patch-60
3729 :
3730 :			Summary:
3731 :			Removed BIGLUT support.
3732 :			Revision:
3733 :			xvidcore--devapi4--1.0--patch-60
3734 :
3735 :			Remobed legacy code for BIGLUT support. It was unused and RD based
3736 :			Motion Estimation was not even compatible with this type of VLC coding.
3737 :
3738 :			modified files:
3739 :			build/generic/configure.in src/bitstream/mbcoding.c
3740 :			src/bitstream/mbcoding.h src/prediction/mbprediction.c
3741 :
3742 :
3743 :			2003-10-03 13:25:17 GMT patch-59
3744 :
3745 :			Summary:
3746 :			Bugfix for PFrames+ Ext Search.
3747 :			Revision:
3748 :			xvidcore--devapi4--1.0--patch-59
3749 :
3750 :			In Qpel mode, the code was doing a diamond search for wrong predictors.
3751 :			This resulted in poor performance as the diamond search was sitting there
3752 :			for sometime.
3753 :
3754 :			modified files:
3755 :			src/motion/estimation.h src/motion/estimation_pvop.c
3756 :
3757 :
3758 :			2003-10-02 16:50:51 GMT patch-58
3759 :
3760 :			Summary:
3761 :			Added VFW makefile for cygwin/minsys
3762 :			Revision:
3763 :			xvidcore--devapi4--1.0--patch-58
3764 :
3765 :			Added VFW makefile for cygwin/minsys. I can't test it so it is probably not
3766 :			right out of the box. Waiting for feedback in order to fix it.
3767 :
3768 :			new files:
3769 :			vfw/bin/.arch-ids/Makefile.cygwin.id
3770 :			vfw/bin/.arch-ids/Makefile.inc.id vfw/bin/Makefile.cygwin
3771 :			vfw/bin/Makefile.inc vfw/src/w32api/.arch-ids/=id
3772 :			vfw/src/w32api/.arch-ids/vfw.h.id vfw/src/w32api/vfw.h
3773 :
3774 :			new directories:
3775 :			vfw/src/w32api vfw/src/w32api/.arch-ids
3776 :
3777 :
3778 :			2003-10-02 13:35:15 GMT patch-57
3779 :
3780 :			Summary:
3781 :			Cleaned up the lumimasking code.
3782 :			Revision:
3783 :			xvidcore--devapi4--1.0--patch-57
3784 :
3785 :			The lumimasking code was not very plugin oriented as it has been ported
3786 :			from old XviD versions. This patch cleans up the code and integrates it
3787 :			better with plugin design. No changes done in teh functionnal code.
3788 :
3789 :			modified files:
3790 :			src/plugins/plugin_lumimasking.c
3791 :
3792 :
3793 :			2003-10-01 23:07:07 GMT patch-56
3794 :
3795 :			Summary:
3796 :			Cleaned up trailing space chars.
3797 :			Revision:
3798 :			xvidcore--devapi4--1.0--patch-56
3799 :
3800 :			The kind of patch we would love to avoid as they make merging
3801 :			a nightmare while they're kind of useless patches.
3802 :
3803 :			Applied sed 's/[ \t]*$//' to all c/h files.
3804 :
3805 :			modified files:
3806 :			src/bitstream/bitstream.c src/bitstream/bitstream.h
3807 :			src/bitstream/cbp.c src/bitstream/mbcoding.c src/dct/idct.c
3808 :			src/dct/simple_idct.c src/decoder.c src/decoder.h
3809 :			src/encoder.c src/encoder.h src/global.h
3810 :			src/image/colorspace.c src/image/colorspace.h src/image/font.c
3811 :			src/image/image.c src/image/image.h src/image/interpolate8x8.c
3812 :			src/image/interpolate8x8.h src/image/qpel.c src/image/qpel.h
3813 :			src/image/reduced.c src/motion/estimation_rd_based.c
3814 :			src/motion/gmc.c src/motion/gmc.h src/motion/motion.h
3815 :			src/motion/sad.c src/plugins/plugin_2pass1.c
3816 :			src/plugins/plugin_2pass2.c src/plugins/plugin_fixed.c
3817 :			src/plugins/plugin_psnr.c src/plugins/plugin_single.c
3818 :			src/portab.h src/prediction/mbprediction.c
3819 :			src/utils/mbtransquant.c src/utils/timer.c src/xvid.c
3820 :			src/xvid.h
3821 :
3822 :
3823 :			2003-09-30 18:10:18 GMT patch-55
3824 :
3825 :			Summary:
3826 :			Code cleanups.
3827 :			Revision:
3828 :			xvidcore--devapi4--1.0--patch-55
3829 :
3830 :			It's been a while since the last ISOC89 conformance cleanup.
3831 :
3832 :			Using the following switches help a lot :-)
3833 :			-Wall -Wsign-compare -Wredundant-decls -Wunreachable-code -Wnested-externs \
3834 :			-ansi
3835 :
3836 :			Result: 0 warning/0 error
3837 :
3838 :			modified files:
3839 :			src/bitstream/vlc_codes.h src/global.h src/image/qpel.h
3840 :			src/motion/estimation.h src/motion/estimation_bvop.c
3841 :			src/motion/estimation_gmc.c src/motion/estimation_pvop.c
3842 :			src/motion/estimation_rd_based.c src/motion/gmc.c
3843 :			src/motion/gmc.h src/motion/motion_inlines.h
3844 :			src/motion/vop_type_decision.c
3845 :
3846 :
3847 :			2003-09-29 00:31:32 GMT patch-54
3848 :
3849 :			Summary:
3850 :			Memory leakage fixes.
3851 :			Revision:
3852 :			xvidcore--devapi4--1.0--patch-54
3853 :
3854 :			The pEnc->queue was allocated but not freed when bframes == 0.
3855 :			And queue images were not freed as well.
3856 :
3857 :			modified files:
3858 :			examples/Makefile src/encoder.c src/image/image.c
3859 :			src/utils/mem_align.c
3860 :
3861 :
3862 :			2003-09-28 16:45:02 GMT patch-53
3863 :
3864 :			Summary:
3865 :			Fixes the unitialized mcsel bit in RD based ME
3866 :			Revision:
3867 :			xvidcore--devapi4--1.0--patch-53
3868 :
3869 :			Fixes the unitialized mcsel bit in RD based ME
3870 :
3871 :
3872 :			modified files:
3873 :			src/motion/estimation_rd_based.c
3874 :
3875 :
3876 :			2003-09-28 01:00:06 GMT patch-52
3877 :
3878 :			Summary:
3879 :			Fix the XviD constant version initialization
3880 :			Revision:
3881 :			xvidcore--devapi4--1.0--patch-52
3882 :
3883 :			Fix the XviD constant version initialization
3884 :
3885 :
3886 :			modified files:
3887 :			src/xvid.h
3888 :
3889 :
3890 :			2003-09-28 00:47:05 GMT patch-51
3891 :
3892 :			Summary:
3893 :			Fix to the build system (the come back).
3894 :			Revision:
3895 :			xvidcore--devapi4--1.0--patch-51
3896 :
3897 :			Ok this one fixes the way we build the targets. The VPATH thingy really
3898 :			works with the library targets. My understanding of the VPATH mechanism
3899 :			was wrong. Now it should be ok (i promess).
3900 :
3901 :			modified files:
3902 :			build/generic/Makefile
3903 :
3904 :
3905 :			2003-09-27 11:45:18 GMT patch-50
3906 :
3907 :			Summary:
3908 :			Small fix to previous patch.
3909 :			Revision:
3910 :			xvidcore--devapi4--1.0--patch-50
3911 :
3912 :			A pair ofdouble quotes prevented "make" to sort out the VPATH
3913 :			dependences in "=build". This resulted in compiling all the
3914 :			sources, all the time, whether a file changed or not.
3915 :
3916 :			modified files:
3917 :			build/generic/Makefile
3918 :
3919 :
3920 :			2003-09-26 22:39:44 GMT patch-49
3921 :
3922 :			Summary:
3923 :			Updated the build files for *nix.
3924 :			Revision:
3925 :			xvidcore--devapi4--1.0--patch-49
3926 :
3927 :			- Fixed MacOSX build (w/o module option). The subversion was not right
3928 :			it was just minor version though it has to be major.minor.
3929 :			- Fixed bootstrap.sh for MacOSX environment, it now looks for
3930 :			glibtoolize if libtoolize is not found.
3931 :			- The unified Makefile now builds XviD out of source tree in the directory
3932 :			=build. It's cleaner, and clashes much less w/ arch/tla source linting.
3933 :			- Tuned the tagging regexps so:
3934 :			+ autoconf files are ignored (considered backup => not erased, not copied
3935 :			and not commited/imported)
3936 :			+ =build is also considered backup.
3937 :
3938 :			modified files:
3939 :			build/generic/Makefile build/generic/bootstrap.sh
3940 :			build/generic/configure.in build/generic/platform.inc.in
3941 :			build/generic/sources.inc examples/Makefile
3942 :			{arch}/=tagging-method
3943 :
3944 :
3945 :			2003-09-24 01:38:03 GMT patch-48
3946 :
3947 :			Summary:
3948 :			Bug fix to decoder (mcsel/acpred bits swapped)
3949 :			Revision:
3950 :			xvidcore--devapi4--1.0--patch-48
3951 :
3952 :			As reported here:
3953 :			http://www.xvid.org/modules.php?op=modload&name=phpBB2&file=viewtopic&t=1513&highlight=
3954 :
3955 :			<quote>
3956 :			in the spec, 'there is 'mcsel' before 'ac_pred_flag'.
3957 :			however, if you see the code, it's changed.
3958 :			</quote>
3959 :
3960 :			We were doing the opposite, ac_pred before mcsel.
3961 :
3962 :			modified files:
3963 :			src/decoder.c
3964 :
3965 :
3966 :			2003-09-11 17:19:35 GMT patch-47
3967 :
3968 :			Summary:
3969 :			Small fix to GMC+QuarterPel -- BS version bumped to 19
3970 :			Revision:
3971 :			xvidcore--devapi4--1.0--patch-47
3972 :
3973 :			During the split up, a line has been disabled that prevented good
3974 :			quartelpel+GMC. Fixed.
3975 :
3976 :			This fix + patch-43 are a good reason to bump up the bitstream
3977 :			version to 19.
3978 :
3979 :			modified files:
3980 :			src/motion/gmc.h src/xvid.h
3981 :
3982 :
3983 :			2003-09-11 17:11:28 GMT patch-46
3984 :
3985 :			Summary:
3986 :			Build process fix for MacOSX+module option
3987 :			Revision:
3988 :			xvidcore--devapi4--1.0--patch-46
3989 :
3990 :			This patch fixes the install rule of the MacOSX module style library.
3991 :			It adds a PRE_SHARED_LIB == SHARED_LIB for all platforms except
3992 :			MacOSX that uses it a different way, and exploits the fact
3993 :			$(SPECIFIC_LDFLAGS) is the last var of the build line to insert a
3994 :			command for post linking the right .so file.
3995 :
3996 :			modified files:
3997 :			build/generic/Makefile build/generic/configure.in
3998 :			build/generic/platform.inc.in
3999 :
4000 :
4001 :			2003-09-11 13:56:40 GMT patch-45
4002 :
4003 :			Summary:
4004 :			Tree cleanup and build files updated.
4005 :			Revision:
4006 :			xvidcore--devapi4--1.0--patch-45
4007 :
4008 :			* Win32 files switched to Unix format.
4009 :			Ok, on IRC, we sorted out what was b0rking the project files each
4010 :			time I commit them in DOS format. In fact the Unix CVS does upload
4011 :			DOS format files if i commit them, and then Win32 CVS users get
4012 :			double \r\n files. So the best is to use unix format in CVS. This
4013 :			way, the checked out versions are right for Win32 users.
4014 :
4015 :			NB: this imply a unix2dos conversion when doing a release from a
4016 :			Unix box. I think it's better than current policy, but automatic
4017 :			tarballs/zip of the tree should take care of that.
4018 :
4019 :			* odivx and xvid_stat examples removed. Associated project files have
4020 :			been removed as well.
4021 :
4022 :			* The divx4 compatibility layer has been removed. Associated options
4023 :			in the configure script removed. libxvidcore.def has no more need
4024 :			to be an autoconf generated file.
4025 :
4026 :			* rawdec removed. It has never been used and will never be.
4027 :
4028 :			* Added major api appending to the macosx module build.
4029 :
4030 :			removed files:
4031 :			build/win32/.arch-ids/odivx_enc_dec.dsp.id
4032 :			build/win32/.arch-ids/xvid_stat.dsp.id
4033 :			build/win32/odivx_enc_dec.dsp build/win32/xvid_stat.dsp
4034 :			examples/.arch-ids/odivx_enc_dec.c.id
4035 :			examples/.arch-ids/xvid_stat.c.id examples/odivx_enc_dec.c
4036 :			examples/xvid_stat.c rawdec/.arch-ids/=id
4037 :			rawdec/.arch-ids/rawdec.c.id rawdec/.arch-ids/rawdec.dsp.id
4038 :			rawdec/rawdec.c rawdec/rawdec.dsp src/.arch-ids/divx4.c.id
4039 :			src/.arch-ids/divx4.h.id src/divx4.c src/divx4.h
4040 :
4041 :			modified files:
4042 :			build/generic/Makefile build/generic/configure.in
4043 :			build/generic/libxvidcore.def build/win32/libxvidcore.dsp
4044 :			build/win32/xvid_decraw.dsp build/win32/xvid_encraw.dsp
4045 :			build/win32/xvidcore.dsw
4046 :
4047 :			renamed files:
4048 :			build/generic/.arch-ids/libxvidcore.def.in.id
4049 :			==> build/generic/.arch-ids/libxvidcore.def.id
4050 :			build/generic/libxvidcore.def.in
4051 :			==> build/generic/libxvidcore.def
4052 :
4053 :			removed directories:
4054 :			rawdec rawdec/.arch-ids
4055 :
4056 :
4057 :			2003-09-11 12:59:19 GMT patch-44
4058 :
4059 :			Summary:
4060 :			Replaced malloc.h header file with stdlib.h
4061 :			Revision:
4062 :			xvidcore--devapi4--1.0--patch-44
4063 :
4064 :			Replaced malloc.h header file with stdlib.h
4065 :
4066 :
4067 :			modified files:
4068 :			src/plugins/plugin_lumimasking.c
4069 :
4070 :
4071 :			2003-09-10 22:33:04 GMT patch-43
4072 :
4073 :			Summary:
4074 :			Fixed a problem for 'power of 2' framerates
4075 :			Revision:
4076 :			xvidcore--devapi4--1.0--patch-43
4077 :
4078 :			Fixed a problem for 'power of 2' framerates
4079 :
4080 :
4081 :			modified files:
4082 :			src/bitstream/bitstream.c
4083 :
4084 :
4085 :			2003-09-10 21:57:12 GMT patch-42
4086 :
4087 :			Summary:
4088 :			Motion Estimation module splitting.
4089 :			Revision:
4090 :			xvidcore--devapi4--1.0--patch-42
4091 :
4092 :			The motion estimation module was the biggest file of the source
4093 :			tree. After some previous attempts, sysKin decided to split it
4094 :			up again. This time he's done it right.
4095 :
4096 :			This split up just changes the organization of ME functions
4097 :			inside different files. This should help a bit in keeping the
4098 :			motion estimation being manageable by a normal human ;-)
4099 :
4100 :			Here is the splitting logic quoted from sysKin's email to
4101 :			xvid-devel:
4102 :			- estimation.h: header files #included in all ME modules:
4103 :			+ deftypes,
4104 :			+ macros
4105 :			+ constants
4106 :			NB: no code.
4107 :			- estimation_bvop.c: motion estimation for b-vops.
4108 :			everything in it :)
4109 :			- estimation_common.c: functions shared among all ME
4110 :			modules:
4111 :			+ diamonds
4112 :			+ subpel
4113 :			+ refinement
4114 :			+ picture
4115 :			+ manipulation
4116 :			+ tables
4117 :			+ ... etc.
4118 :			- estimation_gmc.c: gruel's GME code
4119 :			- estimation_pvop.c: ME for p-vops. Also SAD-based mode
4120 :			decision
4121 :			- estimation_rd_based.c: everything R-D-based: mode
4122 :			decison (including _Fast) and ME.
4123 :			- gmc.c, gmc.h: no change.
4124 :
4125 :			new files:
4126 :			src/motion/.arch-ids/estimation.h.id
4127 :			src/motion/.arch-ids/estimation_bvop.c.id
4128 :			src/motion/.arch-ids/estimation_common.c.id
4129 :			src/motion/.arch-ids/estimation_gmc.c.id
4130 :			src/motion/.arch-ids/estimation_pvop.c.id
4131 :			src/motion/.arch-ids/estimation_rd_based.c.id
4132 :			src/motion/.arch-ids/motion_inlines.h.id
4133 :			src/motion/.arch-ids/vop_type_decision.c.id
4134 :			src/motion/estimation.h src/motion/estimation_bvop.c
4135 :			src/motion/estimation_common.c src/motion/estimation_gmc.c
4136 :			src/motion/estimation_pvop.c src/motion/estimation_rd_based.c
4137 :			src/motion/motion_inlines.h src/motion/vop_type_decision.c
4138 :
4139 :			removed files:
4140 :			src/motion/.arch-ids/motion_est.c.id
4141 :			src/motion/.arch-ids/motion_est.h.id
4142 :			src/motion/.arch-ids/smp_motion_est.c.id
4143 :			src/motion/.arch-ids/smp_motion_est.h.id
4144 :			src/motion/motion_est.c src/motion/motion_est.h
4145 :			src/motion/smp_motion_est.c src/motion/smp_motion_est.h
4146 :
4147 :			modified files:
4148 :			build/generic/sources.inc build/win32/libxvidcore.dsp
4149 :			src/bitstream/mbcoding.h src/motion/gmc.c src/motion/gmc.h
4150 :			src/motion/motion.h src/motion/motion_comp.c src/motion/sad.c
4151 :			src/motion/sad.h src/prediction/mbprediction.c
4152 :			src/prediction/mbprediction.h src/utils/mbfunctions.h
4153 :
4154 :
4155 :			2003-09-10 00:40:44 GMT patch-41
4156 :
4157 :			Summary:
4158 :			Cleanups and fix to (trellis+thresholding) logic
4159 :			Revision:
4160 :			xvidcore--devapi4--1.0--patch-41
4161 :
4162 :			Cleanups to some functions (loop unrolling, call to functions
4163 :			through function array pointers)...
4164 :
4165 :			Fix to the trellis+thresholding logic. It was comparing the
4166 :			return value of trellis with a threshold but the trellis function
4167 :			returns the last non zero coeff index... this was basically
4168 :			comparing apples with oranges... funny but wrong. Trellis now
4169 :			returns the sum of absolute coeffs, so the comparison is logical.
4170 :			Btw, as discussed on the devel ML, this is probably uneeded as
4171 :			trellis does an RD optimized coeff distribution.
4172 :
4173 :			modified files:
4174 :			src/utils/mbtransquant.c
4175 :
4176 :
4177 :			2003-09-09 13:13:58 GMT patch-40
4178 :
4179 :			Summary:
4180 :			Missing ressource for dshow frontend
4181 :			Revision:
4182 :			xvidcore--devapi4--1.0--patch-40
4183 :
4184 :			Missing ressource for dshow frontend
4185 :
4186 :
4187 :			new files:
4188 :			dshow/src/.arch-ids/XviD_logo.bmp.id dshow/src/XviD_logo.bmp
4189 :
4190 :
4191 :			2003-09-08 11:02:10 GMT patch-39
4192 :
4193 :			Summary:
4194 :			Small fixes for fast mode decision
4195 :			Revision:
4196 :			xvidcore--devapi4--1.0--patch-39
4197 :
4198 :			Small fixes for fast mode decision
4199 :
4200 :
4201 :			modified files:
4202 :			src/motion/motion_est.c src/xvid.h
4203 :
4204 :
4205 :			2003-09-05 23:45:48 GMT patch-38
4206 :
4207 :			Summary:
4208 :			New RD mode decision and subpel refinement.
4209 :			Revision:
4210 :			xvidcore--devapi4--1.0--patch-38
4211 :
4212 :			New stuff from michael. It deals with mode decision and subpel
4213 :			refinement. Integration of these new flags are not settled.
4214 :
4215 :			Wait and see. Further testing is needed.
4216 :
4217 :			modified files:
4218 :			src/motion/motion_est.c src/motion/motion_est.h src/xvid.h
4219 :
4220 :
4221 :			2003-08-29 13:56:30 GMT patch-37
4222 :
4223 :			Summary:
4224 :			Still more ME tuning
4225 :			Revision:
4226 :			xvidcore--devapi4--1.0--patch-37
4227 :
4228 :			Still more ME tuning
4229 :
4230 :
4231 :			modified files:
4232 :			src/motion/motion_est.c
4233 :
4234 :
4235 :			2003-08-28 12:43:22 GMT patch-36
4236 :
4237 :			Summary:
4238 :			Removed expanded the cvs Id field
4239 :			Revision:
4240 :			xvidcore--devapi4--1.0--patch-36
4241 :
4242 :			Removed expanded the cvs Id field
4243 :
4244 :
4245 :			modified files:
4246 :			src/image/x86_asm/qpel_mmx.asm
4247 :
4248 :
4249 :			2003-08-28 12:39:44 GMT patch-35
4250 :
4251 :			Summary:
4252 :			More motion est cleanup and bugfixes.
4253 :			Revision:
4254 :			xvidcore--devapi4--1.0--patch-35
4255 :
4256 :			Still more bugfixes, cleanups and improvements to the Motion Est
4257 :			by sysKin
4258 :
4259 :			modified files:
4260 :			src/motion/motion_est.c src/motion/motion_est.h
4261 :
4262 :
4263 :			2003-08-26 13:57:39 GMT patch-34
4264 :
4265 :			Summary:
4266 :			Added final bits of Aspect Ratio flag.
4267 :			Revision:
4268 :			xvidcore--devapi4--1.0--patch-34
4269 :
4270 :			Peter did think of the AR flag since the very beginning of devapi4.
4271 :			He just forgot to code the final bits to effectively write it to
4272 :			the bitstream. This patch adds these missing final bits so XviD now
4273 :			reads and writes AR flags.
4274 :
4275 :			modified files:
4276 :			./src/bitstream/bitstream.c ./src/encoder.c ./src/encoder.h
4277 :			./src/xvid.h
4278 :
4279 :
4280 :			2003-08-25 16:41:09 GMT patch-33
4281 :
4282 :			Summary:
4283 :			Small motion estimation cleanup.
4284 :			Revision:
4285 :			xvidcore--devapi4--1.0--patch-33
4286 :
4287 :			Cleanups from sysKin.
4288 :
4289 :			modified files:
4290 :			./src/motion/motion_est.c ./src/motion/motion_est.h
4291 :
4292 :
4293 :			2003-08-25 14:59:28 GMT patch-32
4294 :
4295 :			Summary:
4296 :			Frame padding bug.
4297 :			Revision:
4298 :			xvidcore--devapi4--1.0--patch-32
4299 :
4300 :			We were edging the image repeating pixels from the image directly,
4301 :			but the standard says we must repeat from a 16 pixel boundary.
4302 :
4303 :			See Chapter 7.6.4 of the standard.
4304 :
4305 :			modified files:
4306 :			./src/image/image.c
4307 :
4308 :
4309 :			2003-08-23 15:07:44 GMT patch-31
4310 :
4311 :			Summary:
4312 :			New Qpel code.
4313 :			Revision:
4314 :			xvidcore--devapi4--1.0--patch-31
4315 :
4316 :			Isibaar commited a new piece of QPel code that seems to be optimized
4317 :			for ia32(mmx) architectures. I had to clean it up a bit to make it
4318 :			respectful of architecture separations. This code is disabled for
4319 :			non ia32 arch, a comment mention it's only faster on ia32... i wonder
4320 :			if it's true, some tests have to be done on sourceforge compile farm
4321 :			in order to confirm that.
4322 :
4323 :			Compared to the bared CVS commit, this patch includes:
4324 :			- fixes the unix build.
4325 :			- better architecture separation.
4326 :			- CodingStyle respected.
4327 :
4328 :			new files:
4329 :			./src/image/.arch-ids/qpel.c.id
4330 :			./src/image/.arch-ids/qpel.h.id ./src/image/qpel.c
4331 :			./src/image/qpel.h
4332 :			./src/image/x86_asm/.arch-ids/qpel_mmx.asm.id
4333 :			./src/image/x86_asm/qpel_mmx.asm
4334 :
4335 :			modified files:
4336 :			./build/generic/sources.inc ./build/win32/libxvidcore.dsp
4337 :			./src/encoder.c ./src/motion/motion_comp.c ./src/xvid.c
4338 :
4339 :
4340 :			2003-08-22 13:20:36 GMT patch-30
4341 :
4342 :			Summary:
4343 :			sad32v does really what it's expected (ie 32x32 SAD :-)
4344 :			Revision:
4345 :			xvidcore--devapi4--1.0--patch-30
4346 :
4347 :			sad32v does really what it's expected (ie 32x32 SAD :-)
4348 :
4349 :
4350 :			modified files:
4351 :			./src/motion/sad.c
4352 :
4353 :
4354 :			2003-08-18 19:00:49 GMT patch-29
4355 :
4356 :			Summary:
4357 :			64bit fix.
4358 :			Revision:
4359 :			xvidcore--devapi4--1.0--patch-29
4360 :
4361 :			The interpolation code was unsafe on 64bit platforms, the offset was
4362 :			badly sized, resulting in segfaults.
4363 :
4364 :			modified files:
4365 :			./src/image/interpolate8x8.h
4366 :
4367 :
4368 :			2003-08-17 14:08:48 GMT patch-28
4369 :
4370 :			Summary:
4371 :			Greyscale mode fixes.
4372 :			Revision:
4373 :			xvidcore--devapi4--1.0--patch-28
4374 :
4375 :			We were missing some greyscale tests in the encoder loop... noticeably
4376 :			in the PVOP function when coding an intra block, and in BVOP function
4377 :			when coding all types.
4378 :
4379 :			I added the cbp trick in the cases discussed above.
4380 :
4381 :			modified files:
4382 :			./src/encoder.c
4383 :
4384 :
4385 :			2003-08-13 11:47:33 GMT patch-27
4386 :
4387 :			Summary:
4388 :			Forgotten bit for IA64 separation
4389 :			Revision:
4390 :			xvidcore--devapi4--1.0--patch-27
4391 :
4392 :			Forgotten bit for IA64 separation
4393 :
4394 :
4395 :			modified files:
4396 :			./src/image/interpolate8x8.h
4397 :
4398 :
4399 :			2003-08-11 15:42:30 GMT patch-26
4400 :
4401 :			Summary:
4402 :			Some qpel changes (sync with Isibaar)
4403 :			Revision:
4404 :			xvidcore--devapi4--1.0--patch-26
4405 :
4406 :			Some qpel changes (sync with Isibaar)
4407 :
4408 :
4409 :			modified files:
4410 :			./examples/xvid_encraw.c
4411 :
4412 :
4413 :			2003-08-11 15:30:04 GMT patch-25
4414 :
4415 :			Summary:
4416 :			Better architecture separation.
4417 :			Revision:
4418 :			xvidcore--devapi4--1.0--patch-25
4419 :
4420 :			Architecture depending functions were declared whatever arch you
4421 :			were compiling. This patch fixes that.
4422 :
4423 :			I also removed the simple_idct hack in decoder.c as it was simply
4424 :			not used. Better not have ugly code in there.
4425 :
4426 :			xvid_bench, should now compile and run on all archs. However I did
4427 :			not put the cpu definitions for each arch, i just separated
4428 :			ARCH_IS_IA32 so even ARCH_IS_GENERIC can compile
4429 :
4430 :			modified files:
4431 :			./examples/xvid_bench.c ./src/bitstream/cbp.h ./src/dct/fdct.h
4432 :			./src/dct/idct.h ./src/decoder.c ./src/image/colorspace.h
4433 :			./src/image/interpolate8x8.h ./src/image/reduced.h
4434 :			./src/motion/sad.h ./src/quant/quant_h263.h
4435 :			./src/quant/quant_mpeg4.h ./src/utils/emms.h
4436 :
4437 :
4438 :			2003-08-09 20:47:42 GMT patch-24
4439 :
4440 :			Summary:
4441 :			Updated changelog
4442 :			Revision:
4443 :			xvidcore--devapi4--1.0--patch-24
4444 :
4445 :			Updated changelog
4446 :
4447 :
4448 :			modified files:
4449 :			./changelog.txt
4450 :
4451 :
4452 :			2003-08-09 20:31:17 GMT patch-23
4453 :
4454 :			Summary:
4455 :			Workaround to a GMC bug due to a MS compiler bug.
4456 :			Revision:
4457 :			xvidcore--devapi4--1.0--patch-23
4458 :
4459 :			Christoph did use a trick to speed up code that resulted in badly
4460 :			optimized code (teh compiler was missing a cast)
4461 :
4462 :			modified files:
4463 :			./src/motion/gmc.c
4464 :
4465 :
4466 :			2003-08-09 17:09:00 GMT patch-22
4467 :
4468 :			Summary:
4469 :			Fixes to xvid_decraw
4470 :			Revision:
4471 :			xvidcore--devapi4--1.0--patch-22
4472 :
4473 :			xvid_decraw has always been used on not so high bitrate sequences
4474 :			and not so big sequences neither.
4475 :
4476 :			I've been doing lot of tests on the Matrix 2 trailer
4477 :			(1000x540 25fps ~5Mbits/s) and xvid_decraw was not able to handle that
4478 :			because of bugs in the buffer filling algorithm.
4479 :
4480 :			This patch fixes the buffer filling and catches up with christoph changes
4481 :			in CVS (wrong help message and option parsing).
4482 :
4483 :			modified files:
4484 :			./examples/xvid_decraw.c
4485 :
4486 :
4487 :			2003-08-08 21:31:59 GMT patch-21
4488 :
4489 :			Summary:
4490 :			Added QPel and GMC options.
4491 :			Revision:
4492 :			xvidcore--devapi4--1.0--patch-21
4493 :
4494 :			XviD has so many options that we forget to propose them all on the
4495 :			CLI, here are two more: GMC and Qpel.
4496 :
4497 :			modified files:
4498 :			./examples/xvid_encraw.c
4499 :
4500 :
4501 :			2003-08-07 19:26:28 GMT patch-20
4502 :
4503 :			Summary:
4504 :			SVOP handling in status window
4505 :			Revision:
4506 :			xvidcore--devapi4--1.0--patch-20
4507 :
4508 :			SVOP handling in status window
4509 :
4510 :
4511 :			modified files:
4512 :			./vfw/src/status.c
4513 :
4514 :
4515 :			2003-08-07 19:25:03 GMT patch-19
4516 :
4517 :			Summary:
4518 :			Warning cleanups by chl
4519 :			Revision:
4520 :			xvidcore--devapi4--1.0--patch-19
4521 :
4522 :			Warning cleanups by chl
4523 :
4524 :
4525 :			modified files:
4526 :			./src/encoder.c ./src/encoder.h ./src/image/interpolate8x8.c
4527 :			./src/motion/motion_est.c ./src/xvid.h
4528 :
4529 :
4530 :			2003-08-06 21:13:35 GMT patch-18
4531 :
4532 :			Summary:
4533 :			Fix to GMC sprite trajectory code
4534 :			Revision:
4535 :			xvidcore--devapi4--1.0--patch-18
4536 :
4537 :			Fix to GMC sprite trajectory code
4538 :
4539 :
4540 :			modified files:
4541 :			./src/bitstream/mbcoding.c ./src/xvid.h
4542 :
4543 :
4544 :			2003-08-06 10:57:25 GMT patch-17
4545 :
4546 :			Summary:
4547 :			Fixes a bug in BVOP block skipping thresholding
4548 :			Revision:
4549 :			xvidcore--devapi4--1.0--patch-17
4550 :
4551 :			Fixes a bug in BVOP block skipping thresholding
4552 :
4553 :
4554 :			modified files:
4555 :			./src/motion/motion_est.c
4556 :
4557 :
4558 :			2003-08-03 14:57:32 GMT patch-16
4559 :
4560 :			Summary:
4561 :			Functions renaming + motion fixes.
4562 :			Revision:
4563 :			xvidcore--devapi4--1.0--patch-16
4564 :
4565 :			BITS flags have been renamed to RD (Rate Distorsion) flags... however
4566 :			function names were still xxxBitsxxx.
4567 :
4568 :			Improved frame type decision
4569 :
4570 :			Fix for DQUANTS plugins, their quant was never checked against valid
4571 :			[1..31] range.
4572 :
4573 :			modified files:
4574 :			./src/encoder.c ./src/motion/motion_est.c
4575 :			./src/motion/motion_est.h
4576 :
4577 :
4578 :			2003-08-02 15:00:49 GMT patch-15
4579 :
4580 :			Summary:
4581 :			API cleanup.
4582 :			Revision:
4583 :			xvidcore--devapi4--1.0--patch-15
4584 :
4585 :			Since we started devapi3 and then devapi4, feature names did not change
4586 :			because it was just convenient to keep them to minimize the change impact.
4587 :			But most of the flags were now not even suggesting what they do. So this
4588 :			patch cleans the API.
4589 :
4590 :			This patch also change the way we describe flags, it's more compact and
4591 :			shows better flags are bit sets that must not overlap. This change fixes
4592 :			a plugin flag overlapping problem as well.
4593 :
4594 :			modified files:
4595 :			./examples/xvid_encraw.c ./src/decoder.c ./src/encoder.c
4596 :			./src/encoder.h ./src/motion/motion_est.c
4597 :			./src/motion/motion_est.h ./src/xvid.c ./src/xvid.h
4598 :			./vfw/src/codec.c ./vfw/src/config.c
4599 :
4600 :
4601 :			2003-07-29 22:25:12 GMT patch-14
4602 :
4603 :			Summary:
4604 :			Fixed bogus memory accesses
4605 :			Revision:
4606 :			xvidcore--devapi4--1.0--patch-14
4607 :
4608 :			Fixed bogus memory accesses
4609 :
4610 :
4611 :			modified files:
4612 :			./src/encoder.c ./src/plugins/plugin_2pass1.c
4613 :
4614 :
4615 :			2003-07-28 12:22:33 GMT patch-13
4616 :
4617 :			Summary:
4618 :			Bitstream version increased to 16
4619 :			Revision:
4620 :			xvidcore--devapi4--1.0--patch-13
4621 :
4622 :			Bitstream version increased to 16
4623 :
4624 :
4625 :			modified files:
4626 :			./src/xvid.h
4627 :
4628 :
4629 :			2003-07-25 12:01:51 GMT patch-12
4630 :
4631 :			Summary:
4632 :			Added gmc files to teh windows project file
4633 :			Revision:
4634 :			xvidcore--devapi4--1.0--patch-12
4635 :
4636 :			Added gmc files to teh windows project file
4637 :
4638 :
4639 :			modified files:
4640 :			./build/win32/libxvidcore.dsp ./vfw/src/codec.c
4641 :
4642 :
4643 :			2003-07-25 12:00:31 GMT patch-11
4644 :
4645 :			Summary:
4646 :			Added cartoon mode from Isibaar
4647 :			Revision:
4648 :			xvidcore--devapi4--1.0--patch-11
4649 :
4650 :			Added cartoon mode from Isibaar
4651 :
4652 :
4653 :			modified files:
4654 :			./src/motion/motion_est.c ./src/plugins/plugin_single.c
4655 :			./src/utils/mbtransquant.c ./src/xvid.h
4656 :
4657 :
4658 :			2003-07-25 10:30:41 GMT patch-10
4659 :
4660 :			Summary:
4661 :			Bitstream syntax comments.
4662 :			Revision:
4663 :			xvidcore--devapi4--1.0--patch-10
4664 :
4665 :			This patch does not change the bitstream but adds some comments that
4666 :			can help in order to understand (lack of) calls to BitstreamPadAlways.
4667 :
4668 :			modified files:
4669 :			./src/bitstream/bitstream.c ./src/encoder.c
4670 :
4671 :
4672 :			2003-07-22 16:34:25 GMT patch-9
4673 :
4674 :			Summary:
4675 :			Fixes Bistream errors in VOL (+ forced stuffing)
4676 :			Revision:
4677 :			xvidcore--devapi4--1.0--patch-9
4678 :
4679 :			After a detailed bugreport at:
4680 :			http://www.xvid.org/modules.php?op=modload&name=phpBB2&file=viewtopic&t=1387&highlight=
4681 :
4682 :			I discovered that:
4683 :
4684 :			1/ we did not write video_signal_type, but we were padding to the next
4685 :			byte, that's why we had video_signal_type=0 and then only 1s until the
4686 :			next byte boundary. This explains the 11 next_start_code();
4687 :
4688 :			2/ video_object_type_indication = Reserved is right on my machine,
4689 :			please check again, but i doubt there is a bug there, we use 3 hard
4690 :			wired values and none of them is zero.
4691 :
4692 :			3/ 01 : next_start_code() *** Was wrong in 24.02.2003; is correct in
4693 :			dev-api-4!!! *** was a bug in fact... when we write user data, we pad to
4694 :			the next byte boundary (if needed) like the standard says... by
4695 :			chancepadding was almost always done, thus the next_start_code() was
4696 :			respected.
4697 :
4698 :			4/ The extra stuffing bits were caused by a forced padding between our
4699 :			VOL function writer and VOP header function writer.
4700 :
4701 :			modified files:
4702 :			./src/bitstream/bitstream.c ./src/encoder.c
4703 :
4704 :
4705 :			2003-07-16 22:57:44 GMT patch-8
4706 :
4707 :			Summary:
4708 :			Fixed quant4_intra_xmm and quant_intra_3dne bug for DC<0.
4709 :			Revision:
4710 :			xvidcore--devapi4--1.0--patch-8
4711 :
4712 :			These two functions were suffering the same error that consists in
4713 :			emulating idiv with some an inversed divisor array and an imul
4714 :			instruction followed by a right shift...
4715 :
4716 :			That was always decreasing the ressult by 1 for negative DC values.
4717 :			A not so bad solution is simply to use a cmov instruction and choose
4718 :			the right value according to the DC value. As these function were
4719 :			for PIII and Athlon, we are sure we can use the cmov instruction.
4720 :
4721 :			PS: the fix is somewhere in cosmetic changes... sorry but the code was
4722 :			too ugly to fix it like it was.
4723 :
4724 :			modified files:
4725 :			./src/quant/x86_asm/quantize4_xmm.asm
4726 :			./src/quant/x86_asm/quantize_3dne.asm
4727 :
4728 :
4729 :			2003-07-16 12:58:21 GMT patch-7
4730 :
4731 :			Summary:
4732 :			Fixed the build system (error caused by patch-5)
4733 :			Revision:
4734 :			xvidcore--devapi4--1.0--patch-7
4735 :
4736 :			IA64 cahnges were wrong in the build system, they make all platforms
4737 :			try to compile a directory... That patch should fix the IA64 target
4738 :			build and get back to previous behavior for other architectures.
4739 :
4740 :			modified files:
4741 :			./build/generic/configure.in ./build/generic/platform.inc.in
4742 :			./build/generic/sources.inc
4743 :
4744 :
4745 :			2003-07-13 12:16:55 GMT patch-6
4746 :
4747 :			Summary:
4748 :			Updates for GME and some cleanups.
4749 :			Revision:
4750 :			xvidcore--devapi4--1.0--patch-6
4751 :
4752 :			This is a all in one patch from syskin:
4753 :			* mcsel decision moved to ModeDecision() function. That makes motion
4754 :			loop completely aware of macroblock mode and vectors (amv in that case).
4755 :			A simple copy&paste was needed to make the the decision R-D based, to be
4756 :			compatible with MODEDECISION_BITS.
4757 :
4758 :			* many bugs fixed. Most of them very small, the only big one was that
4759 :			BITS was misunderstanding a flag and was thinking that mpeg quant is
4760 :			used when h263 quant is used and vice versa :( Also, correct cbp with
4761 :			inter4v mode makes mode decision better. Two speedups - for BITS (no
4762 :			more dequantization when sum == 0) and for ChromaME (chroma sad not
4763 :			computed if total sad too big before that). Some GMC compiler warnings
4764 :			removed. Probably more, I don't remember ;) I haven't touched P/B/I
4765 :			decision for once.
4766 :
4767 :			* compiler warnings removed, mostly "const mismatch" in get_amv() <-- or
4768 :			what was his name.
4769 :
4770 :			* two functions made 2x smaller, shorter and faster.
4771 :
4772 :			modified files:
4773 :			./src/encoder.c ./src/global.h ./src/image/interpolate8x8.h
4774 :			./src/motion/gmc.c ./src/motion/gmc.h
4775 :			./src/motion/motion_est.c ./src/motion/motion_est.h
4776 :			./vfw/src/codec.c
4777 :
4778 :
4779 :			2003-07-10 17:35:59 GMT patch-5
4780 :
4781 :			Summary:
4782 :			IA64 updates.
4783 :