3 @c man begin CODEC OPTIONS
5 libavcodec provides some generic global options, which can be set on
6 all the encoders and decoders. In addition each codec may support
7 so-called private options, which are specific for a given codec.
9 Sometimes, a global option may only affect a specific kind of codec,
10 and may be unsensical or ignored by another, so you need to be aware
11 of the meaning of the specified options. Also some options are
12 meant only for decoding or encoding.
14 Options may be set by specifying -@var{option} @var{value} in the
15 FFmpeg tools, or by setting the value explicitly in the
16 @code{AVCodecContext} options or using the @file{libavutil/opt.h} API
19 The list of supported options follow:
22 @item b @var{integer} (@emph{encoding,audio,video})
23 Set bitrate in bits/s. Default value is 200K.
25 @item ab @var{integer} (@emph{encoding,audio})
26 Set audio bitrate (in bits/s). Default value is 128K.
28 @item bt @var{integer} (@emph{encoding,video})
29 Set video bitrate tolerance (in bits/s). In 1-pass mode, bitrate
30 tolerance specifies how far ratecontrol is willing to deviate from the
31 target average bitrate value. This is not related to min/max
32 bitrate. Lowering tolerance too much has an adverse effect on quality.
34 @item flags @var{flags} (@emph{decoding/encoding,audio,video,subtitles})
40 Use four motion vector by macroblock (mpeg4).
42 Use 1/4 pel motion compensation.
50 Always try a mb with mv=<0,0>.
54 Use internal 2pass ratecontrol in first pass mode.
56 Use internal 2pass ratecontrol in second pass mode.
58 Only decode/encode grayscale.
62 Set error[?] variables during encoding.
66 Normalize adaptive quantization.
72 Place global headers in extradata instead of every keyframe.
74 Only write platform-, build- and time-independent data. (except (I)DCT).
75 This ensures that file and data checksums are reproducible and match between
76 platforms. Its primary use is for regression testing.
78 Apply H263 advanced intra coding / mpeg4 ac prediction.
80 Deprecated, use mpegvideo private options instead.
82 Deprecated, use mpegvideo private options instead.
84 Apply interlaced motion estimation.
89 @item me_method @var{integer} (@emph{encoding,video})
90 Set motion estimation method.
95 zero motion estimation (fastest)
97 full motion estimation (slowest)
99 EPZS motion estimation (default)
101 esa motion estimation (alias for full)
103 tesa motion estimation
105 dia motion estimation (alias for epzs)
107 log motion estimation
109 phods motion estimation
113 hex motion estimation
115 umh motion estimation
117 iter motion estimation
120 @item extradata_size @var{integer}
123 @item time_base @var{rational number}
126 It is the fundamental unit of time (in seconds) in terms of which
127 frame timestamps are represented. For fixed-fps content, timebase
128 should be @code{1 / frame_rate} and timestamp increments should be
131 @item g @var{integer} (@emph{encoding,video})
132 Set the group of picture size. Default value is 12.
134 @item ar @var{integer} (@emph{decoding/encoding,audio})
135 Set audio sampling rate (in Hz).
137 @item ac @var{integer} (@emph{decoding/encoding,audio})
138 Set number of audio channels.
140 @item cutoff @var{integer} (@emph{encoding,audio})
141 Set cutoff bandwidth.
143 @item frame_size @var{integer} (@emph{encoding,audio})
144 Set audio frame size.
146 Each submitted frame except the last must contain exactly frame_size
147 samples per channel. May be 0 when the codec has
148 CODEC_CAP_VARIABLE_FRAME_SIZE set, in that case the frame size is not
149 restricted. It is set by some decoders to indicate constant frame
152 @item frame_number @var{integer}
153 Set the frame number.
155 @item delay @var{integer}
157 @item qcomp @var{float} (@emph{encoding,video})
158 Set video quantizer scale compression (VBR). It is used as a constant
159 in the ratecontrol equation. Recommended range for default rc_eq:
162 @item qblur @var{float} (@emph{encoding,video})
163 Set video quantizer scale blur (VBR).
165 @item qmin @var{integer} (@emph{encoding,video})
166 Set min video quantizer scale (VBR). Must be included between -1 and
167 69, default value is 2.
169 @item qmax @var{integer} (@emph{encoding,video})
170 Set max video quantizer scale (VBR). Must be included between -1 and
171 1024, default value is 31.
173 @item qdiff @var{integer} (@emph{encoding,video})
174 Set max difference between the quantizer scale (VBR).
176 @item bf @var{integer} (@emph{encoding,video})
177 Set max number of B frames between non-B-frames.
179 Must be an integer between -1 and 16. 0 means that B-frames are
180 disabled. If a value of -1 is used, it will choose an automatic value
181 depending on the encoder.
185 @item b_qfactor @var{float} (@emph{encoding,video})
186 Set qp factor between P and B frames.
188 @item rc_strategy @var{integer} (@emph{encoding,video})
189 Set ratecontrol method.
191 @item b_strategy @var{integer} (@emph{encoding,video})
192 Set strategy to choose between I/P/B-frames.
194 @item ps @var{integer} (@emph{encoding,video})
195 Set RTP payload size in bytes.
197 @item mv_bits @var{integer}
198 @item header_bits @var{integer}
199 @item i_tex_bits @var{integer}
200 @item p_tex_bits @var{integer}
201 @item i_count @var{integer}
202 @item p_count @var{integer}
203 @item skip_count @var{integer}
204 @item misc_bits @var{integer}
205 @item frame_bits @var{integer}
206 @item codec_tag @var{integer}
207 @item bug @var{flags} (@emph{decoding,video})
208 Workaround not auto detected encoder bugs.
215 some old lavc generated msmpeg4v3 files (no autodetection)
217 Xvid interlacing bug (autodetected if fourcc==XVIX)
219 (autodetected if fourcc==UMP4)
221 padding bug (autodetected)
225 illegal vlc bug (autodetected per fourcc)
229 old standard qpel (autodetected per fourcc/version)
232 @item direct_blocksize
233 direct-qpel-blocksize bug (autodetected per fourcc/version)
235 edge padding bug (autodetected per fourcc/version)
241 Workaround various bugs in microsoft broken decoders.
246 @item lelim @var{integer} (@emph{encoding,video})
247 Set single coefficient elimination threshold for luminance (negative
248 values also consider DC coefficient).
250 @item celim @var{integer} (@emph{encoding,video})
251 Set single coefficient elimination threshold for chrominance (negative
252 values also consider dc coefficient)
254 @item strict @var{integer} (@emph{decoding/encoding,audio,video})
255 Specify how strictly to follow the standards.
260 strictly conform to a older more strict version of the spec or reference software
262 strictly conform to all the things in the spec no matter what consequences
266 allow unofficial extensions
268 allow non standardized experimental things, experimental
269 (unfinished/work in progress/not well tested) decoders and encoders.
270 Note: experimental decoders can pose a security risk, do not use this for
271 decoding untrusted input.
274 @item b_qoffset @var{float} (@emph{encoding,video})
275 Set QP offset between P and B frames.
277 @item err_detect @var{flags} (@emph{decoding,audio,video})
278 Set error detection flags.
285 detect bitstream specification deviations
287 detect improper bitstream length
289 abort decoding on minor error detection
291 ignore decoding errors, and continue decoding.
292 This is useful if you want to analyze the content of a video and thus want
293 everything to be decoded no matter what. This option will not result in a video
294 that is pleasing to watch in case of errors.
296 consider things that violate the spec and have not been seen in the wild as errors
298 consider all spec non compliancies as errors
300 consider things that a sane encoder should not do as an error
303 @item has_b_frames @var{integer}
305 @item block_align @var{integer}
307 @item mpeg_quant @var{integer} (@emph{encoding,video})
308 Use MPEG quantizers instead of H.263.
310 @item qsquish @var{float} (@emph{encoding,video})
311 How to keep quantizer between qmin and qmax (0 = clip, 1 = use
312 differentiable function).
314 @item rc_qmod_amp @var{float} (@emph{encoding,video})
315 Set experimental quantizer modulation.
317 @item rc_qmod_freq @var{integer} (@emph{encoding,video})
318 Set experimental quantizer modulation.
320 @item rc_override_count @var{integer}
322 @item rc_eq @var{string} (@emph{encoding,video})
323 Set rate control equation. When computing the expression, besides the
324 standard functions defined in the section 'Expression Evaluation', the
325 following functions are available: bits2qp(bits), qp2bits(qp). Also
326 the following constants are available: iTex pTex tex mv fCode iCount
327 mcVar var isI isP isB avgQP qComp avgIITex avgPITex avgPPTex avgBPTex
330 @item maxrate @var{integer} (@emph{encoding,audio,video})
331 Set max bitrate tolerance (in bits/s). Requires bufsize to be set.
333 @item minrate @var{integer} (@emph{encoding,audio,video})
334 Set min bitrate tolerance (in bits/s). Most useful in setting up a CBR
335 encode. It is of little use elsewise.
337 @item bufsize @var{integer} (@emph{encoding,audio,video})
338 Set ratecontrol buffer size (in bits).
340 @item rc_buf_aggressivity @var{float} (@emph{encoding,video})
343 @item i_qfactor @var{float} (@emph{encoding,video})
344 Set QP factor between P and I frames.
346 @item i_qoffset @var{float} (@emph{encoding,video})
347 Set QP offset between P and I frames.
349 @item rc_init_cplx @var{float} (@emph{encoding,video})
350 Set initial complexity for 1-pass encoding.
352 @item dct @var{integer} (@emph{encoding,video})
358 autoselect a good one (default)
368 floating point AAN DCT
371 @item lumi_mask @var{float} (@emph{encoding,video})
372 Compress bright areas stronger than medium ones.
374 @item tcplx_mask @var{float} (@emph{encoding,video})
375 Set temporal complexity masking.
377 @item scplx_mask @var{float} (@emph{encoding,video})
378 Set spatial complexity masking.
380 @item p_mask @var{float} (@emph{encoding,video})
383 @item dark_mask @var{float} (@emph{encoding,video})
384 Compress dark areas stronger than medium ones.
386 @item idct @var{integer} (@emph{decoding/encoding,video})
387 Select IDCT implementation.
400 Automatically pick a IDCT compatible with the simple one
423 floating point AAN IDCT
426 @item slice_count @var{integer}
428 @item ec @var{flags} (@emph{decoding,video})
429 Set error concealment strategy.
434 iterative motion vector (MV) search (slow)
436 use strong deblock filter for damaged MBs
438 favor predicting from the previous frame instead of the current
441 @item bits_per_coded_sample @var{integer}
443 @item pred @var{integer} (@emph{encoding,video})
444 Set prediction method.
456 @item aspect @var{rational number} (@emph{encoding,video})
457 Set sample aspect ratio.
459 @item debug @var{flags} (@emph{decoding/encoding,audio,video,subtitles})
460 Print specific debug info.
473 per-block quantization parameter (QP)
487 memory management control operations (H.264)
491 visualize quantization parameter (QP), lower QP are tinted greener
493 visualize block types
495 picture buffer allocations
500 @item vismv @var{integer} (@emph{decoding,video})
501 Visualize motion vectors (MVs).
503 This option is deprecated, see the codecview filter instead.
508 forward predicted MVs of P-frames
510 forward predicted MVs of B-frames
512 backward predicted MVs of B-frames
515 @item cmp @var{integer} (@emph{encoding,video})
516 Set full pel me compare function.
521 sum of absolute differences, fast (default)
523 sum of squared errors
525 sum of absolute Hadamard transformed differences
527 sum of absolute DCT transformed differences
529 sum of squared quantization errors (avoid, low quality)
531 number of bits needed for the block
533 rate distortion optimal, slow
537 sum of absolute vertical differences
539 sum of squared vertical differences
541 noise preserving sum of squared differences
543 5/3 wavelet, only used in snow
545 9/7 wavelet, only used in snow
552 @item subcmp @var{integer} (@emph{encoding,video})
553 Set sub pel me compare function.
558 sum of absolute differences, fast (default)
560 sum of squared errors
562 sum of absolute Hadamard transformed differences
564 sum of absolute DCT transformed differences
566 sum of squared quantization errors (avoid, low quality)
568 number of bits needed for the block
570 rate distortion optimal, slow
574 sum of absolute vertical differences
576 sum of squared vertical differences
578 noise preserving sum of squared differences
580 5/3 wavelet, only used in snow
582 9/7 wavelet, only used in snow
589 @item mbcmp @var{integer} (@emph{encoding,video})
590 Set macroblock compare function.
595 sum of absolute differences, fast (default)
597 sum of squared errors
599 sum of absolute Hadamard transformed differences
601 sum of absolute DCT transformed differences
603 sum of squared quantization errors (avoid, low quality)
605 number of bits needed for the block
607 rate distortion optimal, slow
611 sum of absolute vertical differences
613 sum of squared vertical differences
615 noise preserving sum of squared differences
617 5/3 wavelet, only used in snow
619 9/7 wavelet, only used in snow
626 @item ildctcmp @var{integer} (@emph{encoding,video})
627 Set interlaced dct compare function.
632 sum of absolute differences, fast (default)
634 sum of squared errors
636 sum of absolute Hadamard transformed differences
638 sum of absolute DCT transformed differences
640 sum of squared quantization errors (avoid, low quality)
642 number of bits needed for the block
644 rate distortion optimal, slow
648 sum of absolute vertical differences
650 sum of squared vertical differences
652 noise preserving sum of squared differences
654 5/3 wavelet, only used in snow
656 9/7 wavelet, only used in snow
663 @item dia_size @var{integer} (@emph{encoding,video})
664 Set diamond type & size for motion estimation.
666 @item last_pred @var{integer} (@emph{encoding,video})
667 Set amount of motion predictors from the previous frame.
669 @item preme @var{integer} (@emph{encoding,video})
670 Set pre motion estimation.
672 @item precmp @var{integer} (@emph{encoding,video})
673 Set pre motion estimation compare function.
678 sum of absolute differences, fast (default)
680 sum of squared errors
682 sum of absolute Hadamard transformed differences
684 sum of absolute DCT transformed differences
686 sum of squared quantization errors (avoid, low quality)
688 number of bits needed for the block
690 rate distortion optimal, slow
694 sum of absolute vertical differences
696 sum of squared vertical differences
698 noise preserving sum of squared differences
700 5/3 wavelet, only used in snow
702 9/7 wavelet, only used in snow
709 @item pre_dia_size @var{integer} (@emph{encoding,video})
710 Set diamond type & size for motion estimation pre-pass.
712 @item subq @var{integer} (@emph{encoding,video})
713 Set sub pel motion estimation quality.
715 @item dtg_active_format @var{integer}
717 @item me_range @var{integer} (@emph{encoding,video})
718 Set limit motion vectors range (1023 for DivX player).
720 @item ibias @var{integer} (@emph{encoding,video})
721 Set intra quant bias.
723 @item pbias @var{integer} (@emph{encoding,video})
724 Set inter quant bias.
726 @item color_table_id @var{integer}
728 @item global_quality @var{integer} (@emph{encoding,audio,video})
730 @item coder @var{integer} (@emph{encoding,video})
735 variable length coder / huffman coder
746 @item context @var{integer} (@emph{encoding,video})
749 @item slice_flags @var{integer}
751 @item xvmc_acceleration @var{integer}
753 @item mbd @var{integer} (@emph{encoding,video})
754 Set macroblock decision algorithm (high quality mode).
763 use best rate distortion
766 @item stream_codec_tag @var{integer}
768 @item sc_threshold @var{integer} (@emph{encoding,video})
769 Set scene change threshold.
771 @item lmin @var{integer} (@emph{encoding,video})
772 Set min lagrange factor (VBR).
774 @item lmax @var{integer} (@emph{encoding,video})
775 Set max lagrange factor (VBR).
777 @item nr @var{integer} (@emph{encoding,video})
780 @item rc_init_occupancy @var{integer} (@emph{encoding,video})
781 Set number of bits which should be loaded into the rc buffer before
784 @item flags2 @var{flags} (@emph{decoding/encoding,audio,video})
789 Allow non spec compliant speedup tricks.
791 Deprecated, use mpegvideo private options instead.
793 Skip bitstream encoding.
795 Ignore cropping information from sps.
797 Place global headers at every keyframe instead of in extradata.
799 Frame data might be split into multiple chunks.
801 Show all frames before the first keyframe.
803 Deprecated, use mpegvideo private options instead.
805 Export motion vectors into frame side-data (see @code{AV_FRAME_DATA_MOTION_VECTORS})
806 for codecs that support it. See also @file{doc/examples/export_mvs.c}.
809 @item error @var{integer} (@emph{encoding,video})
811 @item qns @var{integer} (@emph{encoding,video})
812 Deprecated, use mpegvideo private options instead.
814 @item threads @var{integer} (@emph{decoding/encoding,video})
819 detect a good number of threads
822 @item me_threshold @var{integer} (@emph{encoding,video})
823 Set motion estimation threshold.
825 @item mb_threshold @var{integer} (@emph{encoding,video})
826 Set macroblock threshold.
828 @item dc @var{integer} (@emph{encoding,video})
829 Set intra_dc_precision.
831 @item nssew @var{integer} (@emph{encoding,video})
834 @item skip_top @var{integer} (@emph{decoding,video})
835 Set number of macroblock rows at the top which are skipped.
837 @item skip_bottom @var{integer} (@emph{decoding,video})
838 Set number of macroblock rows at the bottom which are skipped.
840 @item profile @var{integer} (@emph{encoding,audio,video})
878 @item level @var{integer} (@emph{encoding,audio,video})
886 @item lowres @var{integer} (@emph{decoding,audio,video})
887 Decode at 1= 1/2, 2=1/4, 3=1/8 resolutions.
889 @item skip_threshold @var{integer} (@emph{encoding,video})
890 Set frame skip threshold.
892 @item skip_factor @var{integer} (@emph{encoding,video})
893 Set frame skip factor.
895 @item skip_exp @var{integer} (@emph{encoding,video})
896 Set frame skip exponent.
897 Negative values behave identical to the corresponding positive ones, except
898 that the score is normalized.
899 Positive values exist primarily for compatibility reasons and are not so useful.
901 @item skipcmp @var{integer} (@emph{encoding,video})
902 Set frame skip compare function.
907 sum of absolute differences, fast (default)
909 sum of squared errors
911 sum of absolute Hadamard transformed differences
913 sum of absolute DCT transformed differences
915 sum of squared quantization errors (avoid, low quality)
917 number of bits needed for the block
919 rate distortion optimal, slow
923 sum of absolute vertical differences
925 sum of squared vertical differences
927 noise preserving sum of squared differences
929 5/3 wavelet, only used in snow
931 9/7 wavelet, only used in snow
938 @item border_mask @var{float} (@emph{encoding,video})
939 Increase the quantizer for macroblocks close to borders.
941 @item mblmin @var{integer} (@emph{encoding,video})
942 Set min macroblock lagrange factor (VBR).
944 @item mblmax @var{integer} (@emph{encoding,video})
945 Set max macroblock lagrange factor (VBR).
947 @item mepc @var{integer} (@emph{encoding,video})
948 Set motion estimation bitrate penalty compensation (1.0 = 256).
950 @item skip_loop_filter @var{integer} (@emph{decoding,video})
951 @item skip_idct @var{integer} (@emph{decoding,video})
952 @item skip_frame @var{integer} (@emph{decoding,video})
954 Make decoder discard processing depending on the frame type selected
957 @option{skip_loop_filter} skips frame loop filtering, @option{skip_idct}
958 skips frame IDCT/dequantization, @option{skip_frame} skips decoding.
966 Discard useless frames like 0-sized frames.
969 Discard all non-reference frames.
972 Discard all bidirectional frames.
975 Discard all frames excepts keyframes.
981 Default value is @samp{default}.
983 @item bidir_refine @var{integer} (@emph{encoding,video})
984 Refine the two motion vectors used in bidirectional macroblocks.
986 @item brd_scale @var{integer} (@emph{encoding,video})
987 Downscale frames for dynamic B-frame decision.
989 @item keyint_min @var{integer} (@emph{encoding,video})
990 Set minimum interval between IDR-frames.
992 @item refs @var{integer} (@emph{encoding,video})
993 Set reference frames to consider for motion compensation.
995 @item chromaoffset @var{integer} (@emph{encoding,video})
996 Set chroma qp offset from luma.
998 @item trellis @var{integer} (@emph{encoding,audio,video})
999 Set rate-distortion optimal quantization.
1001 @item sc_factor @var{integer} (@emph{encoding,video})
1002 Set value multiplied by qscale for each frame and added to
1005 @item mv0_threshold @var{integer} (@emph{encoding,video})
1006 @item b_sensitivity @var{integer} (@emph{encoding,video})
1007 Adjust sensitivity of b_frame_strategy 1.
1009 @item compression_level @var{integer} (@emph{encoding,audio,video})
1010 @item min_prediction_order @var{integer} (@emph{encoding,audio})
1011 @item max_prediction_order @var{integer} (@emph{encoding,audio})
1012 @item timecode_frame_start @var{integer} (@emph{encoding,video})
1013 Set GOP timecode frame start number, in non drop frame format.
1015 @item request_channels @var{integer} (@emph{decoding,audio})
1016 Set desired number of audio channels.
1018 @item bits_per_raw_sample @var{integer}
1019 @item channel_layout @var{integer} (@emph{decoding/encoding,audio})
1024 @item request_channel_layout @var{integer} (@emph{decoding,audio})
1029 @item rc_max_vbv_use @var{float} (@emph{encoding,video})
1030 @item rc_min_vbv_use @var{float} (@emph{encoding,video})
1031 @item ticks_per_frame @var{integer} (@emph{decoding/encoding,audio,video})
1032 @item color_primaries @var{integer} (@emph{decoding/encoding,video})
1033 @item color_trc @var{integer} (@emph{decoding/encoding,video})
1034 @item colorspace @var{integer} (@emph{decoding/encoding,video})
1035 @item color_range @var{integer} (@emph{decoding/encoding,video})
1036 @item chroma_sample_location @var{integer} (@emph{decoding/encoding,video})
1038 @item log_level_offset @var{integer}
1039 Set the log level offset.
1041 @item slices @var{integer} (@emph{encoding,video})
1042 Number of slices, used in parallelized encoding.
1044 @item thread_type @var{flags} (@emph{decoding/encoding,video})
1045 Select which multithreading methods to use.
1047 Use of @samp{frame} will increase decoding delay by one frame per
1048 thread, so clients which cannot provide future frames should not use
1054 Decode more than one part of a single frame at once.
1056 Multithreading using slices works only when the video was encoded with
1060 Decode more than one frame at once.
1063 Default value is @samp{slice+frame}.
1065 @item audio_service_type @var{integer} (@emph{encoding,audio})
1066 Set audio service type.
1090 @item request_sample_fmt @var{sample_fmt} (@emph{decoding,audio})
1091 Set sample format audio decoders should prefer. Default value is
1094 @item pkt_timebase @var{rational number}
1096 @item sub_charenc @var{encoding} (@emph{decoding,subtitles})
1097 Set the input subtitles character encoding.
1099 @item field_order @var{field_order} (@emph{video})
1100 Set/override the field order of the video.
1106 Interlaced video, top field coded and displayed first
1108 Interlaced video, bottom field coded and displayed first
1110 Interlaced video, top coded first, bottom displayed first
1112 Interlaced video, bottom coded first, top displayed first
1115 @item skip_alpha @var{integer} (@emph{decoding,video})
1116 Set to 1 to disable processing alpha (transparency). This works like the
1117 @samp{gray} flag in the @option{flags} option which skips chroma information
1118 instead of alpha. Default is 0.
1120 @item codec_whitelist @var{list} (@emph{input})
1121 "," separated List of allowed decoders. By default all are allowed.
1123 @item dump_separator @var{string} (@emph{input})
1124 Separator used to separate the fields printed on the command line about the
1126 For example to separate the fields with newlines and indention:
1128 ffprobe -dump_separator "
1129 " -i ~/videos/matrixbench_mpeg2.mpg
1134 @c man end CODEC OPTIONS
1136 @ifclear config-writeonly
1137 @include decoders.texi
1139 @ifclear config-readonly
1140 @include encoders.texi