Update of /cvsroot/mplayer/0_90/DOCS In directory mail:/var/tmp.root/cvs-serv25001 Modified Files: mplayer.1 Log Message: Finally merged libavc-options.txt into the man page. Index: mplayer.1 =================================================================== RCS file: /cvsroot/mplayer/0_90/DOCS/mplayer.1,v retrieving revision 1.336 retrieving revision 1.337 diff -u -r1.336 -r1.337 --- mplayer.1 17 Mar 2003 03:07:38 -0000 1.336 +++ mplayer.1 20 Mar 2003 03:36:55 -0000 1.337 @@ -1748,7 +1748,7 @@ Flip image upside\-down. .TP .B \-lavdopts <option1:option2:...> (DEBUG CODE) -If decoding with a codec from libavcodec, you can specify its parameters here. +If decoding with libavcodec, you can specify its parameters here. .I EXAMPLE: .PD 0 @@ -2179,7 +2179,7 @@ .REss .IPs "field[=n]" Extracts a single field from interlaced image using stride arithmetic -to avoid wasting cpu time. The optional argument n specifies whether +to avoid wasting CPU time. The optional argument n specifies whether to extract the even or the odd field (depending on whether n is even or odd). .IPs "detc[=var1=value2:var2=value2:...]" @@ -2526,7 +2526,7 @@ .br 1: mt .br -2: rh(default) +2: rh (default) .br 3: abr .br @@ -2604,7 +2604,7 @@ . .TP .B \-lavcopts <option1:option2:...> -If encoding with a codec from libavcodec, you can specify its parameters +If encoding with libavcodec, you can specify its parameters here. .I EXAMPLE: @@ -2637,11 +2637,25 @@ mpeg1video: MPEG1 video :) .REss .IPs vqmin=<1\-31> -minimum quantizer (pass\ 1/\:2) (default: 2) +minimum quantizer (pass\ 1/\:2) +.RSss +1: Not recommended (much larger file, little quality difference and weird side +effects: msmpeg4, h263 will be very low quality, ratecontrol will be confused +resulting in lower quality and some decoders will not be able to decode it). +.br +2: Recommended for normal mpeg4/mpeg1video encoding (default). +.br +3: Recommended for h263(p)/msmpeg4. The reason for preferring 3 over 2 is that +2 could lead to overflows (this will be fixed for h263(p) by changing the +quantizer per MB in the future, msmpeg4 cannot be fixed as it does not support +that) +.REss .IPs vqscale=<1\-31> -constant quantizer (selects fixed quantizer mode) (default: 0 (disabled)) +Constant quantizer /\: constant quality encoding (selects fixed quantizer mode). +A lower value means better quality but larger files (default: 0 (disabled)). +1 is not recommended (see \-vqmin for details). .IPs vqmax=<1\-31> -maximum quantizer (pass\ 1/\:2) (default: 31) +maximum quantizer (pass\ 1/\:2) 10\-31 should be a sane range (default: 31) .IPs mbqmin=<1\-31> minimum macroblock quantizer (pass\ 1/\:2) (default: 2) .IPs mbqmax=<1\-31> @@ -2653,50 +2667,64 @@ .RSss 0: no B frames (default) .br -0\-2: sane range +0\-2: sane range for MPEG4 .REss .IPs vme=<0\-5> motion estimation method: .RSss -0: none (very lq) +0: none (very low quality) .br 1: full (slow) .br -2: log (lq) +2: log (low quality) .br -3: phods (lq) +3: phods (low quality) .br 4: EPZS (default) .br 5: X1 (experimental) +Note: 0\-3 currently ignores the amount of Bits spent, so quality may be low. .REss .IPs vhq\ \ -high quality mode, encode each macro block in all modes an choose the smallest -(slow). -(default: HQ disabled) +high quality mode, encode each macro block in all modes and choose the best. +This is slow but results in better quality and file size. +(default: disabled) .IPs v4mv -4 motion vectors per macroblock (slightly better quality). +Allow 4 motion vectors per macroblock (slightly better quality). (default: disabled) .IPs keyint=<0\-300> -interval between keyframes in frames. -Larger numbers mean slightly smaller files, but less precise seeking, 0 means -no key frames and values >300 aren't recommended. -For a strict mpeg1/\:2/\:4 compliance this would have to be <=132. +maximum interval between keyframes in frames. +Keyframes are needed for seeking as seeking is only possible to a keyframe but +keyframes need more space than other frames so larger numbers here mean +slightly smaller files, but less precise seeking, 0 means no key frames. +Values >300 are not recommended as the quality might be bad depending upon +decoder, encoder and luck. +For a strict MPEG1/\:2/\:4 compliance this would have to be <=132. (default: 250 or one key frame every ten seconds in a 25fps movie) .IPs vb_strategy=<0\-1> strategy to choose between I/\:P/\:B frames (pass\ 2): .RSss 0: always use the maximum number of B frames (default) .br -1: avoid B frames in high motion scenes (bitrate mispredictions) +1: avoid B frames in high motion scenes (will cause bitrate misprediction) .REss .IPs vpass=<1\-2> -Activates internal 2pass mode (default: disabled): +Activates internal two pass mode, only specify if you wish to use two pass +encoding (default: disabled). .RSss 1: first pass .br 2: second pass .REss +.IPs +Tip: Try to use constant quantizer mode for pass one (vqscale=<quantizer>). +.br +huffyuv: +.RSss +pass 1 saves statistics +.br +pass 2 encodes with an optimal Huffman table based upon pass 1 statistics. +.REss .IPs aspect=<x/y> Store movie aspect internally, just like MPEG files. Much nicer solution than rescaling, because quality isn't decreased. @@ -2719,7 +2747,7 @@ (default: 800) .REss .IPs vratetol=<value> -approximated filesize tolerance in kbit. +approximated filesize tolerance in kBit. 1000\-100000 is a sane range. (warning: 1kBit = 1000 Bits) (default: 8000) .IPs vrc_maxrate=<value> @@ -2745,21 +2773,25 @@ .br else .br -do normal ratecontrol (dont lock to next P frame quantizer) and set -q= -q * v{b|i}_qfactor + v{b|i}_qoffset +do normal ratecontrol (dont lock to next P frame quantizer) and +set q= -q * v{b|i}_qfactor + v{b|i}_qoffset +.IPs +Tip: To do constant quantizer encoding with different quantizers for +I/P and B frames you can use: +vqmin=<ip_quant>:vqmax=<ip_quant>:vb_qfactor=<b_quant/ip_quant> .IPs vqblur=<0.0\-1.0> -quantizer blur (pass1): +quantizer blur (pass1) Larger values will average the quantizer more over +time (slower change). .RSss 0.0: qblur disabled .br 0.5 (default) .br -1.0: average the quantizer over all previous frames, larger values will - average the quantizer more over time (slower change) +1.0: average the quantizer over all previous frames .REss .IPs vqblur=<0.0\-99.0> -quantizer gaussian blur, larger values will average the quantizer more over -time (slower change) (pass2) (default: 0.5) +quantizer gaussian blur (pass2) Larger values will average the quantizer more +over time (slower change) (default: 0.5) .IPs vqcomp=<value> quantizer compression, depends upon vrc_eq (pass\ 1/\:2) (default: 0.5) .IPs vrc_eq=<equation> @@ -2784,19 +2816,19 @@ .br avgTex: average texture complexity .br -avgIITexaverage: intra texture complexity in I frames +avgIITex: average intra texture complexity in I frames .br -avgPITexaverage: intra texture complexity in P frames +avgPITex: average intra texture complexity in P frames .br -avgPPTexaverage: non intra texture complexity in P frames +avgPPTex: average non intra texture complexity in P frames .br -avgBPTexaverage: non intra texture complexity in B frames +avgBPTex: average non intra texture complexity in B frames .br -mv: bits used for MVs +mv: Bits used for motion vectors .br -fCode: maximum length of MV in log2 scale +fCode: maximum length of motion vector in log2 scale .br -iCount: number of intra MBs / number of MBs +iCount: number of intra macro blocks / number of macro blocks .br var: spatial complexity .br @@ -2821,28 +2853,27 @@ sin, cos, tan, sinh, cosh, tanh, exp, log, abs . .IPs vrc_override=<options> -User specified quality for specific parts (pass\ 1/\:2). -The options are <start-frame, end-frame, quality[/\:start-frame, end-frame, -quality[/...]]>: +User specified quality for specific parts (ending, credits, ..) (pass\ 1/\:2). +The options are <start-frame>, <end-frame>, <quality>[/\:<start-frame>, +<end-frame>, <quality>[/...]]: .RSss quality 2\-31: quantizer .br -quality \-500\-0: quality correcture in % +quality \-500\-0: quality correction in % .REss .IPs vrc_init_cplx=<0\-1000> initial complexity (pass\ 1) -.IPs vqsquish=<0\-1> +.IPs vqsquish=<0,1> specify how to keep the quantizer between qmin and qmax (pass\ 1/\:2): .RSss -0: use cliping +0: use clipping .br 1: use a nice differentiable function (default) .REss .IPs vlelim=<-1000\-1000> single coefficient elimination threshold for luminance. Negative values will also consider the dc coefficient (should be at least -4 -or lower for encoding -at quant=1): +or lower for encoding at quant=1): .RSss 0: disabled (default) .br @@ -2857,8 +2888,8 @@ .br 7 (JVT recommendation) .REss -.IPs vstrict=<-1\-1> -(strict) standard compliance. +.IPs vstrict=<-1,0,1> +strict standard compliance .RSss 0: disabled (default) .br @@ -2866,12 +2897,26 @@ reference decoder .br -1: allows non-standard YV12 huffyuv encoding (20% smaller files, but -can't be played back by the official huffyuv codec) +cannot be played by the official huffyuv codec) .REss .IPs vdpart data partitioning. -Adds 2 byte per video packet, improves error-resistance when transfering over +Adds 2 Bytes per video packet, improves error-resistance when transferring over unreliable channels (eg.\& streaming over the internet) +Each video packet will be encoded in 3 separate partitions: +.RSss +1. MVs (=movement) +.br +2. DC coefficients (=low res picture) +.br +3. AC coefficients (=details) +.REss +.IPs +MV & DC are most important, loosing them looks far worse than loosing the +AC and the 1. & 2. partition. (MV&DC) are far smaller than the 3. partition +(AC) meaning that errors will hit the AC partition much more often than the +MV&DC partitions. Thus, the picture will look better with partitioning than +without, as without partitioning an error will trash AC/DC/MV equally. .IPs vpsize=<0\-10000> video packet size, improves error-resistance (see \-vdpart option too): .RSss @@ -2896,7 +2941,8 @@ .REss .IPs idct=<0\-99> idct algorithm. -Note: all these IDCTs do pass the IEEE1180 tests afaik: +Note: To the best of our knowledge all these IDCTs do pass the IEEE1180 +tests. .RSss 0: automatically select a good one (default) .br @@ -2906,7 +2952,7 @@ .br 3: simplemmx .br -4: libmpeg2mmx (inaccurate, DONT USE for encoding with keyint >100) +4: libmpeg2mmx (inaccurate, DON'T USE for encoding with keyint >100) .br 5: ps2 .br @@ -2916,8 +2962,8 @@ .REss .IPs lumi_mask=<0.0\-1.0> luminance masking. -Warning: be careful, too large values can cause disasterous things. -Warning2: large values might look good on some monitors but may look horrible +Warning: Be careful, too large values can cause disastrous things. +Warning2: Large values might look good on some monitors but may look horrible on other monitors: .RSss 0.0: disabled (default) @@ -2940,7 +2986,8 @@ spatial complexity masking. Larger values help against blockiness, if no deblocking filter is used for decoding. -Crop any black borders to get better quality: +Tip: Crop any black borders completely as they will reduce the quality +of the macro blocks (also applies without scplx_mask). .RSss 0.0: disabled (default) .br @@ -2971,6 +3018,7 @@ .REss .IPs qpel use quarter pel motion compensation +Tip: This seems only useful for high bitrate encodings. .IPs precmp=<0\-2000> comparison function for motion estimation pre pass .IPs cmp=<0\-2000> @@ -2986,15 +3034,15 @@ .br 3 (DCT): sum of absolute dct transformed differences .br -4 (PSNR): sum of the squared quantization errors +4 (PSNR): sum of the squared quantization errors (don't use, low quality) .br 5 (BIT): number of bits needed for the block .br -6 (RD): rate distoration optimal, slow +6 (RD): rate distortion optimal, slow .br 7 (ZERO): 0 .br -+256: use chroma too, doesnt work with b frames currently ++256: use chroma too, doesn't work (correctly) with B frames currently .REss .IPs predia=<\-99\-6> Diamond type and size for motion estimation pre pass @@ -3045,7 +3093,8 @@ .RSss 0: (default) .br -a: will use 2a+1 x 2a+1 MB square of MV predictors from the previous frame +a: will use 2a+1 x 2a+1 macro block square of motion vector predictors from +the previous frame .REss .IPs preme=<0\-2> motion estimation pre-pass @@ -3058,14 +3107,20 @@ .REss .IPs subq=<1\-8> subpel refinement quality (for qpel) (default: 8). -Note: this has a significant effect on the speed +Note: This has a significant effect on the speed. .IPs psnr print the psnr (peak signal to noise ratio) for the whole video after encoding -and store the per frame psnr in a file with name like 'psnr_012345.log'. +and store the per frame psnr in a file with a name like 'psnr_012345.log'. Return values are in dB (decibel), the higher the better. .IPs mpeg_quant use MPEG quantizers instead of H.263. (default: disabled) (i.e.\& use H.263 quantizers) +.IPs aic +advanced intra prediction (H.263+ only) +Note: vqmin should be 8 or larger. +.IPs umv +unlimited MVs (H.263+ only) +Allow encoding of abritarily long MVs. .RE . .TP