x264 preset class parameters detailed explanation
Reprinted address:http://www.cnblogs.com/zyl910/archive/2011/12/04/x264_presets.html
x264 version-
x264 core:116 r2074 2641b9e
Add the “FullHelp” parameter to view the full version of X264
x264 --fullhelp
The explanation of the preset parameters is–
Presets:
--profile <string> Force the limits of an H.264 profile
Overrides all settings.
- baseline:
--no-8x8dct --bframes 0 --no-cabac
--cqm flat --weightp 0
No interlaced.
No lossless.
- main:
--no-8x8dct --cqm flat
No lossless.
- high:
No lossless.
- high10:
No lossless.
Support for bit depth 8-10.
--preset <string> Use a preset to select encoding settings [medium]
Overridden by user settings.
- ultrafast:
--no-8x8dct --aq-mode 0 --b-adapt 0
--bframes 0 --no-cabac --no-deblock
--no-mbtree --me dia --no-mixed-refs
--partitions none --rc-lookahead 0 --ref 1
--scenecut 0 --subme 0 --trellis 0
--no-weightb --weightp 0
- superfast:
--no-mbtree --me dia --no-mixed-refs
--partitions i8x8,i4x4 --rc-lookahead 0
--ref 1 --subme 1 --trellis 0 --weightp 1
- veryfast:
--no-mixed-refs --rc-lookahead 10
--ref 1 --subme 2 --trellis 0 --weightp 1
- faster:
--no-mixed-refs --rc-lookahead 20
--ref 2 --subme 4 --weightp 1
- fast:
--rc-lookahead 30 --ref 2 --subme 6
--weightp 1
- medium:
Default settings apply.
- slow:
--b-adapt 2 --direct auto --me umh
--rc-lookahead 50 --ref 5 --subme 8
- slower:
--b-adapt 2 --direct auto --me umh
--partitions all --rc-lookahead 60
--ref 8 --subme 9 --trellis 2
- veryslow:
--b-adapt 2 --bframes 8 --direct auto
--me umh --merange 24 --partitions all
--ref 16 --subme 10 --trellis 2
--rc-lookahead 60
- placebo:
--bframes 16 --b-adapt 2 --direct auto
--slow-firstpass --no-fast-pskip
--me tesa --merange 24 --partitions all
--rc-lookahead 60 --ref 16 --subme 11
--trellis 2
--tune <string> Tune the settings for a particular type of source
or situation
Overridden by user settings.
Multiple tunings are separated by commas.
Only one psy tuning can be used at a time.
- film (psy tuning):
--deblock -1:-1 --psy-rd <unset>:0.15
- animation (psy tuning):
--bframes {+2} --deblock 1:1
--psy-rd 0.4:<unset> --aq-strength 0.6
--ref {Double if >1 else 1}
- grain (psy tuning):
--aq-strength 0.5 --no-dct-decimate
--deadzone-inter 6 --deadzone-intra 6
--deblock -2:-2 --ipratio 1.1
--pbratio 1.1 --psy-rd <unset>:0.25
--qcomp 0.8
- stillimage (psy tuning):
--aq-strength 1.2 --deblock -3:-3
--psy-rd 2.0:0.7
- psnr (psy tuning):
--aq-mode 0 --no-psy
- ssim (psy tuning):
--aq-mode 2 --no-psy
- fastdecode:
--no-cabac --no-deblock --no-weightb
--weightp 0
- zerolatency:
--bframes 0 --force-cfr --no-mbtree
--sync-lookahead 0 --sliced-threads
--rc-lookahead 0
The text looks a little dizzy, so I sorted it into a form.
1 Profile (grade)
| preface | name | Name | 8x8dct | cqm | bframes | cabac | weightp | interlaced |
| 1 | baseline | baseline | no | flat | 0 | no | 0 | no |
| 2 | main | no | flat | Unlimited | Unlimited | Unlimited | Unlimited | |
| 3 | high | high | Unlimited | Unlimited | Unlimited | Unlimited | Unlimited | Unlimited |
| 4 | high10 | high 10 bits | Unlimited | Unlimited | Unlimited | Unlimited | Unlimited | Unlimited |
BFRAMES: The maximum value of the number of B-Frames can be inserted between the I frame and the P frame, with a range of 0-16.
CQM: Custom Quantization Matrices. There are FLAT and JVT default.
WeightP: Make X264 can improve the compression of the P frame using clearly weighted weighted forecasting. It also improves the quality of fading / out. The higher the mode, the slower.
8X8DCT: Elastic 8X8 discrete string conversion (Adaptive 8×8 DCT).
CABAC: The binary coding of elastic content (CABAC: Context Adaptive Binary Arithmetic Coder).
inter -.
2 Preset (preset)
B-Adapt: Set the elastic B frame position decision algorithm. This setting controls the X264 how to decide to place the P frame or B frame.
BFRAMES: The maximum value of the number of B-Frames can be inserted between the I frame and the P frame, with a range of 0-16.
Direct: “Direct” dynamic vector prediction mode. There are two modes available: Spatial and Temporal. You can specify the None to stop the Direct dynamic vector and specify the auto to allow X264 to switch to a suitable mode between the two.
ME: Full-Pixel Motion Estimation algorithm.
MERANGE: The maximum range of controlling motion estimates (unit is pixel). For Hex and DIA, the range is limited from 4 ~ 16. For UMH and ESA, it can increase to a wider range of dynamic search that exceeds silence 16, which is very useful for HD video and high dynamic lenses. Note that for UMH, ESA, and Tesa, increasing Merrange will slow down the coding speed.
Partitions: H.264 Video is divided into a macro block of 16X16 during the compression process. These blocks can be further divided into smaller divisions, which is part of this option to control.
RC-LOOOKAHEAD: Set the number of frames used in MB-Tree bit rate control and VBV-LOOKAHEAD. The maximum allowable value is 250. For the MB-Tree section, the number of frames increases better results but will be slower. The maximum buffer used by MB-Tree is min (RC-LOOKAHEAD,-Keyint).
Ref: The size of the control decoding picture buffer (DPB: Decoded Picture Buffer). The range is from 0 to 16. In short, this value is that each P frame can use the number of previous frames as the number of reference frames (the number of B frames can be used is less or two, depending on whether they are used as a reference frame). The minimum reF number that can be referenced is 1.
Subme: Set Subpixel estimated complexity. The higher the value, the better. Level 1 ~ 5 is just a refinement intensity. Rado 6 is the mode decision -making RDO, while the level 8 is the dynamic vector and internal prediction mode to enable RDO. The RDO level is significantly slower than the previous level.
TRLLIS: Perform TRLLIS Quantization to improve efficiency.
WeightP: Make X264 can improve the compression of the P frame using clearly weighted weighted prediction. It also improves the quality of fading / out. The higher the mode, the slower.
no-8x8dct: Disable elastic 8X8 discrete string conversion (Adaptive 8×8 DCT).
AQ-Mode: Elastic quantitative mode. When there is no AQ, the X264 is easy to allocate the number of bites with less details to the less details. AQ is used to better allocate the number of usable places between all macro blocks in the video.
No-Cabac: Dental: CABAC: Context Adaptive Binary Arithmetic Coder) Data flow compression, the variable length encoding of the elastic content with lower elastic content of the switching efficiency (CAVLC: Context Adaptive Varial LENGTH) system Essence Reduce compression efficiency (usually 10 ~ 20%) and hardware demand for decoding.
no-deBlock: Disable loop screening (Loop Filter. It is also known as the continuous circulation (Inloop Deblocker).
No-MBTREE: Disable Macroblock Tree bit rate control. The use of macro -area tree bit rate control will improve the overall compression rate, and weighted it accordingly by tracking the time of the time of tracking the time.
no-mixed-reFS: Mixed reference will be selected for reference based on each 8X8 division, not based on each macro block. When using multiple reference frames, this will improve the quality, although it is necessary to lose some speed. Setting this option will be discontinued.
Scenecut: Set the threshold of the I/IDR frame position (scene change detection).
no-Weightb: Reference to the “weighted” B frame is discontinued.
Slow-Firsstpass: Slow Pass.
No-Fast-PSKip: Early Skip Detection of the P frame is discontinued. It is very slightly improved, but it is necessary to lose a lot of speed.
3 Tune (adjustment)
Ref: The size of the decoding picture buffer (DPB: Decoded Picture Buffer). The range is from 0 to 16. In short, this value is that each P frame can use the number of previous frames as the number of reference frames (the number of B frames can be used is less or two, depending on whether they are used as a reference frame). The minimum reF number that can be referenced is 1.
BFRAMES: The maximum value of the number of B-Frames can be inserted between the I frame and the P frame, with a range of 0-16.
Deblock: Control loop screening (also known as continuous circulation (Inloop Deblocker)).
PSY-RD: The first number is the strength of PSY-RDO (required subme> = 6). The second number is the strength of PSY-TRLLIS (requires trellis> = 1).
AQ-Stringth: Elastic quantitative intensity. Set the strength of the macro block of AQ bias to low details (plane). It is not allowed to be negative. Values other than 0.0 ~ 2.0 are not recommended.
no-DCImate: DCT Decimation. DCT DeCimation will abandon it that it considers “unnecessary” DCT blocks. This will improve coding efficiency, and the reduction of quality is usually insignificant.
DEADZONE-Inter/Intra: Set the size of the Inter/intra brightness quantitative reaction area (Deadzone). The range of the reaction area should be 0 ~ 32. This value settings X264 will be arbitrarily abandoned without trying to keep the level of subtle details. Very subtle details are difficult to see and consume the number of bites. Discarding these details can not waste the number of bites on the low -income picture of the video. The reaction area is incompatible with –sRLLIS.
ipratio: Modify the average increase in the target of the quantification value of the I frame compared to the quantization value of the P frame. The greater the value will improve the quality of the I frame.
PBRATIO: Modify the average reduction in the target of the quantification value of the B frame compared to the quantization value of the P frame. The larger the value will reduce the quality of the B frame. When MBTREE is enabled (activated by default), this setting has no effect, Mbtree will automatically calculate the best value.
QCOMP: Quantitative value curve compression coefficient. 0.0 is a fixed level rate, and 1.0 is a fixed quantization value. When MBTREE is enabled, it affects the strength of MBTREE (the larger the QCOMP, the weaker the MBTREE).
AQ-Mode: Elastic quantitative mode. When there is no AQ, the X264 is easy to allocate the number of bites with less details to the less details. AQ is used to better allocate the number of usable places between all macro blocks in the video.
No-Cabac: Dental: CABAC: Context Adaptive Binary Arithmetic Coder) Data flow compression, the variable length encoding of the elastic content with lower-efficiency of switching efficiency (CAVLC: Context Adaptive Varial LENGTH) system) Essence Reduce compression efficiency (usually 10 ~ 20%) and hardware demand for decoding.
WeightP: Make X264 can improve the compression of the P frame using clearly weighted weighted predicate premictIn. It also improves the quality of fading / out. The higher the mode, the slower.
Force-CFR: If you use FFMS2 or LAVF separator and output files are not raw formats, copy the time code from the input file. This option shuts down this function and forced X264 to produce itself. When you use this option, it is estimated that you will also set -FPS.
No-MBTREE: Disable Macroblock Tree Bit Bittage Rate Control. The use of macro -area tree bit rate control will improve the overall compression rate, and weighted it accordingly by tracking the time of the time of tracking the time.
Sync-Lookahead: Set the frame cache size used for thread prediction. The maximum value is 250. It is automatically closed in the second time more coded or based on a piece -based thread. Set to 0 will close the thread prediction, which will reduce delays, but at the cost of reducing performance.
Sliced-Threads: Open a piece-based thread. Low -quality and low -default, but not delayed coding.
RC-LOOOKAHEAD: Set the number of frames used in MB-Tree bit rate control and VBV-LOOKAHEAD. The maximum allowable value is 250. For the MB-Tree section, the number of frames increases better results but will be slower. The maximum buffer used by MB-Tree is min (RC-LOOKAHEAD,-Keyint).