A configuration file is a case-sensitive list of <parameter> = <value> separated by [<section>] with <section> being one of INPUT, DYNAMIC_INPUT, OUTPUT, GOP, RATE_CONTROL, SETTINGS, RUN. Each value is defined to be of specific type(s):
| <string> | Sequence of characters without spaces |
| <path> | Valid file path |
| <enable> |
DISABLE (or FALSE),
ENABLE (or TRUE) |
| <ae> | Arithmetic expression: <decimal or hexadecimal> (<operator> <decimal or hexadecimal>)* |
| <array> | <decimal or hexadecimal> [<decimal or hexadecimal>] |
The definition of syntax elements used in <ae>:
| <unit> | One of K Ki M Mi G Gi with i meaning that it is an 1024 is used as factor instead of 1000. |
| <unary> | One of - + |
| <operator> | One of - + * / |
| <decimal> | <unary>? ([0-0]*[.])?[0-9]+<unit> |
| <hexadecimal> | 0 [xX] [0-9a-fA-F]+ |
Files that start with # are comments. For example:
#simple example configuration file
[INPUT]
YUVFile = input.yuv
Width = 1280
Height = 720
Format = NV12
[OUTPUT]
BitstreamFile = output.265
| Parameter | Section(s) | Format | ||||
|---|---|---|---|---|---|---|
| Basic IN/OUT | ||||||
|
YUVFile --input |
INPUT DYNAMIC_INPUT |
<path> | ||||
| YUV input file path. | ||||||
|
BitstreamFile --output -o |
OUTPUT |
<path> Default: Stream.bin |
||||
| Elementary stream output file path. | ||||||
|
RecFile --output-rec -r |
OUTPUT | <path> | ||||
| Optional output file path for reconstructed YUV pictures. | ||||||
|
Width --input-width |
INPUT DYNAMIC_INPUT |
<ae> 80 – 8192 (AVC), 256-8192 (HEVC) |
||||
| Input frame/field width in pixels. | ||||||
|
Height --input-height |
INPUT DYNAMIC_INPUT |
<ae> 96 – 4096 (AVC), 128-4096 (HEVC) |
||||
| Input frame/field height in pixels. | ||||||
|
Framerate --framerate |
INPUT RATE_CONTROL |
<ae> Default: 30. Range: 1-240 (or 0 for INPUT section) |
||||
| Number of frames per
second of the YUV input file. When this parameter is not present in the
INPUT section (or the value is 0), its value is set to what is specified
in section RATE_CONTROL. When its value differs from what is specified
in section RATE_CONTROL, the encoder will drop or repeat some frames. --framerate specifies output frame rate. |
||||||
|
MaxPicture --max-picture |
RUN |
<string>/<ae> Default: ALL. Values: ALL or 0-MAX. |
||||
| Number of frames to encode. | ||||||
|
FirstPicture --first-picture |
RUN |
<ae> Default: 0. Values: 0 - MAX |
||||
| The first frame to encode. | ||||||
|
Loop --loop |
RUN |
<enable> Default: DISABLE |
||||
| Loop at the end of the input stream or when MaxPicture is reached. | ||||||
| Pixel Format | ||||||
|
Format --input-format --rec-format |
INPUT OUTPUT |
<string> | ||||
|
The output format specifies the reconstructed output The FourCC input/output format: (Note a C typically stands for 12 bpp, an A for10 bpp) I4CL I4AL I444 YUV 444 planar, 16 or 8 bits. I4AM (Y416) YUV 444 planar, 16 bits with 12 bits in most significant bits P212 P210 NV16 YUV 422 semi-planar, 16 or 8 bits. P012 NV12 P010 YUV 420 semi-planar, 16 or 8 bits. Y012 Y010 Y800 YUV 400 single plane (monochrome), 16 or 8 bits. T64C T64A T648 Planar YUV444, 64(32)x4 tiles, 8, 10 or 12 bits. T54C T54A T548 Planar YUV444, 64(32)x4 tiles, 8, 10 or 12 bits. T62C T62A T628 Semi-planar YUV 422, 64x4 tiles, 8, 10 or 12 bits. T52C T52A T528 Semi-planar YUV 422, 32x4 tiles, 8, 10 or 12 bits. T60C T60A T608 Semi-planar YUV 420 64x4 tiles, 8, 10 or 12 bits. T50C T50A T508 Semi-planar YUV 420 32x4 tiles, 8, 10 or 12 bits. T6MC T6MA T6M8 Single plane Y 400, 64x4 tiles, 8, 10 or 12 bits. T5MC T5MA T5M8 Single plane Y 400, 32x4 tiles, 8, 10 or 12 bits. Non-hardware supported formats (conversion in software) P410 NV24 Semi-planar YUV444, 16 or 8 bits. P216 Semi-planar YUV422, 8,10,12 bpp in 16 bits, in most significant. P016 Semi-planar YUV420, 8,10,12 bpp in 16 bits, in most significant. YUVP Packed YUYV 422, 16 bits (10 bpp) YUY2 Packed YUYV 422, 8 bits Y800 YUYV 422, 8 bits I2CL I2AL I422(YV16) Planar YUV422, 16 or 8 bits. I0CL I0AL I420(IYUV) Planar YUV420, 8 bits. Also, YV12 = I420 with inverted U and V |
||||||
|
Bitdepth --out-bitdepth |
SETTINGS |
<string> Default: 8. Values: 8, 10, 12 |
||||
| Number of bits used for each pixel component (Y, U and V) in the encoded stream. | ||||||
|
ChromaMode --chroma-mode |
SETTINGS | <string> | ||||
|
Set the expected chroma mode (4:4:4, 4:2:2, 4:2:0 or 4:0:0) of the encoded stream. Depending on the input FourCC, this might lead to a conversion. Together with BitDepth, these options determine the final FourCC the encoder is expecting. CHROMA_4_0_0 (=CHROMA_MONO), CHROMA_4_2_2, CHROMA_4_2_2, CHROMA_4_4_4 |
||||||
| SrcFormat | SETTINGS | <string> Default: RASTER | ||||
|
RASTER Source pixels are in raster format, 8 or 16 bits. RASTER_MSB Source pixels are in raster format, 8 or 16 bits, with relevant bits in highest bit (for 12-bit for example: <12 bit pixel><4 bit 0s>) TILE_32x4 4x4 tiles with packed pixels, stride multiple of 32. TILE_64x4 4x4 tiles with packed pixels, stride multiple of 64. |
||||||
| VideoMode | SETTINGS | <string> Default | ||||
|
Video is progressive or interlaced. In interlaced mode, the corresponding flags in header and SEI message will be added. PROGRESSIVE Progressive video mode INTERLACED_BOTTOM Interlaced video mode bottom field before top field. INTERLACED_TOP Interlaced video mode top field before bottom field. |
||||||
| ColourDescription | SETTINGS | <string> Default: COLOUR_DESC_UNSPECIFIED | ||||
| The colorimetry information that is written in the SPS/VUI header. | ||||||
| ColourMatrix | SETTINGS | <string> Default: COLOUR_MAT_UNSPECIFIED | ||||
| The matrix coefficients used in deriving luma and chroma signals from RGB (HDR setting). | ||||||
| TransferCharac | SETTINGS | <string> Default: TRANSFER_UNSPECIFIED | ||||
|
The reference opto-electronic transfer characteristic function (HDR setting). TRANSFER_ followed by one of: BT_1361 Rec. ITU-R BT.1361-0 extended colour gamut system BT_2020_10B Rec. ITU-R BT.2020-2 (10 bit system) BT_2020_12B Rec. ITU-R BT.2020-2 (12 bit system) BT_2100_HLG Rec. ITU-R BT.2100-2 hybrid log-gamma (HLG) system BT_2100_PQ Rec. ITU-R BT.2100-2 perceptual quantization (PQ) system BT_470_SYSTEM_B Rec. ITU-R BT.470-6 System B, G BT_470_SYSTEM_M Rec. ITU-R BT.470-6 System M BT_601 Rec. ITU-R BT.601-7 525 or 625 BT_709 Rec. ITU-R BT.709-6 IEC_61966_2_1 IEC 61966-2-1 sRGB or sYCC IEC_61966_2_4 IEC 61966-2-4 LINEAR Linear transfer characteristics LOG Logarithmic transfer characteristic (100:1 range) LOG_EXTENDED Logarithmic transfer characteristic (100 * Sqrt( 10 ): 1 range) SMPTE_240M Society of Motion Picture and Television Engineers 240M SMPTE_428 Society of Motion Picture and Television Engineers ST 428-1 UNSPECIFIED Characteristics unknown or determined by the application |
||||||
| VideoFullRange | SETTINGS | <enable> Default: DISABLE | ||||
| Use full range YUV instead reduced range (CCIR601) legacy range. | ||||||
| AspectRatio | SETTINGS | <string> Default: ASPECT_RATIO_AUTO | ||||
| The display aspect ratio of the video sequence to be written in SPS/VUI, after applying any changes due to cropping. | ||||||
|
CropPosX CropPosY CropWidth CropHeight |
INPUT OUTPUT |
<ae> Default: All 0 (no cropping) |
||||
|
Crop window with position X, Y, width, and height. Position must be a multiple of 32 for raster 8-bit, 16 for raster 10-bit/12-bit, 32 for tile 32x4 and 64 for tile 64x4. Width and height must be a multiple of 8. |
||||||
| HEVC | AVC | |||||
| MIN | MAX | MIN | MAX | |||
| X | 0 | 7936 | 0 | 8112 | ||
| Y | 128 | 3968 | 96 | 4000 | ||
| Width | 256 | 3840 | 80 | 4016 | ||
| Height | 128 | 3968 | 96 | 4000 | ||
| Rate Control | ||||||
|
RateCtrlMode --ratectrl-mode |
RATE_CONTROL |
<string> Default: CONST_QP |
||||
|
Selects the way bitrate is controlled: CONST_QP No bitrate control is used; all pictures use the same QP (SliceQP). CBR Constant Bitrate Control encodes a stream with a fixed average bitrate, regardless of complexity of the scene. VBR Variable Bitrate Control attempts the specified bitrate but is allowed to use less/more bits when the scene is easy/complex (up to MaxBitrate). CAPPED_VBR Similar to VBR but a PSNR value is used to limit it to a maximum quality. LOW_LATENCY Low-latency VBR mode. PLUGIN Use a customized rate control plugin. Requires RISC-V code to be generated and linked to the firmware binary. |
||||||
|
Bitrate --bitrate |
RATE_CONTROL |
<ae> Default: 4000 (Kbits/s) |
||||
| Target bit rate in Kbits/s. Unused if RateCtrlMode=CONST_QP | ||||||
|
MaxBitrate --max-bitate |
RATE_CONTROL |
<ae> Default: 4000 (Kbits/s) |
||||
| Maximum bit rate in Kbits/s. This is used in VBR modes. This should be the maximum transmission bandwidth available (the encoder should not exceed this value on average on a moving window of CPBSize seconds). For CBR, Bitrate is used instead of this. | ||||||
| CPBSize | RATE_CONTROL | <ae> or <string>AUTO (default) | ||||
| Size of the Coded Picture Buffer as specified in the HRD model, in seconds. | ||||||
| EnableFillerData | SETTINGS | <enable> Default: DISABLE | ||||
| Specifies whether Filler Data is to be inserted when target bitrate cannot be achieved in CBR mode (RateCtrlMode). | ||||||
| EntropyMode | SETTINGS | <string> Default: MODE_CABAC | ||||
| AVC only. Selects the entropy coding mode: MODE_CABAC or MODE_CAVLC. | ||||||
| InitialDelay | RATE_CONTROL | <ae> Default: 1.5 (seconds) | ||||
| The initial removal delay as specified in the HRD model, in seconds. Not used for CBR. The value cannot be greater than the CPBSize value. | ||||||
| MaxPSNR | RATE_CONTROL | <ae> Default: 42. Values: 20.00 – 50.00 | ||||
| Maximum/minimum Peak Signal to Noise Ratio for the CAPPED_VBR rate control mode. | ||||||
|
MaxPictureSize [.(IPB)] MaxPictureSize InBits[.(IPB)] |
RATE_CONTROL | <string> DISABLE (default) or <ae> | ||||
| Coarse picture size in Kbits (or bits) that should not be exceeded. Optionally specify .I.P or .B for the specific picture types. | ||||||
| EnableSkip | RATE_CONTROL | <string> DISABLE (default) or <ae> | ||||
| Allows rate control to replace a frame by a skipped frame when the target bitrate cannot be reached. | ||||||
| MaxConsecutiveSkip | RATE_CONTROL | <string> DISABLE (default) or <ae> | ||||
| Maximum consecutive skip pictures allowed. | ||||||
| ScnChgResilience | RATE_CONTROL | <enable> Default: DISABLE | ||||
| Enables a rate control feature allowing a more conservative behavior to improve the video quality on scene change. However, the overall quality decreases. | ||||||
| SCPrevention | RATE_CONTROL | <enable> Default: DISABLE | ||||
| Enable Scene Change quality loss prevention. Limits the QP change to maintain encode quality when a scene change occurs. | ||||||
| CostMode | SETTINGS | <enable> Default: ENABLE | ||||
| Reinforce the influence of the chrominance in the RDO choice. Improves visual quality of video sequences with low saturation. | ||||||
| Profile/Level/Tier | ||||||
|
Profile --profile |
SETTINGS |
<string> Default: AVC_MAIN or HEVC_MAIN |
||||
|
Profile that applies to the bitstream. AVC_BASELINE, AVC_CAVLC_444, AVC_CAVLC_444_INTRA, AVC_C_BASELINE AVC_C_HIGH, AVC_HIGH, AVC_HIGH10, AVC_HIGH10_INTRA, AVC_HIGH_422, AVC_HIGH_422_INTRA, AVC_HIGH_444_INTRA, AVC_HIGH_444_PRED, AVC_MAIN, AVC_PROG_HIGH, XAVC_HIGH10_INTRA_CBG, XAVC_HIGH10_INTRA_VBR, XAVC_HIGH_422_INTRA_CBG, XAVC_HIGH_422_INTRA_VBR, XAVC_LONG_GOP_HIGH_422_MXF, XAVC_LONG_GOP_HIGH_MP4, XAVC_LONG_GOP_HIGH_MXF, XAVC_LONG_GOP_MAIN_MP4 HEVC_MAIN, HEVC_MAIN10, HEVC_MAIN10_INTRA, HEVC_MAIN12, HEVC_MAIN_422, HEVC_MAIN_422_10, HEVC_MAIN_422_10_INTRA, HEVC_MAIN_422_12, HEVC_MAIN_444, HEVC_MAIN_444_10, HEVC_MAIN_444_10_INTRA, HEVC_MAIN_444_12, HEVC_MAIN_444_INTRA, HEVC_MAIN_444_STILL, HEVC_MAIN_INTRA, HEVC_MAIN_STILL, HEVC_MONO, HEVC_MONO10 |
||||||
|
Level --level |
SETTINGS |
<string> Default: 5.1 |
||||
|
Level that applies to the bitstream. AVC: 0.9, 1, 1.1, 1.2, 1.3, 2, 2.1, 2.2, 3, 3.1, 3.2, 4, 4.1, 4.2, 5, 5.1, 5.2, 6, 6.1, 6.2 HEVC: 1, 2, 2.1, 3, 3.1, 4, 4.1, 5, 5.1, 5.2, 6, 6.1, 6.2 |
||||||
|
Tier --tier |
SETTINGS |
<ae> Default: MAIN_TIER |
||||
| HEVC tiers. HIGH_TIER or MAIN_TIER. | ||||||
| Quantization Parameters | ||||||
| QPCtrlMode | SETTINGS | <string> Default: UNIFORM_QP | ||||
|
How to generate the QP per CU. ADAPTIVE_AUTO_QP Same as AUTO_QP, but the lookup table parameters can be updated at each frame depending on the content AUTO_QP The QP is chosen according to the CU content using a pre-programmed lookup table LOAD_QP The QPs of each CTB come from an external buffer loaded from a file. The file shall be named QPs.hex and located in the directory specified by QPTablesFolder RELATIVE_QP Can be use with LOAD_QP using the ‘|’ operator to fill the external buffer with relative values instead of absolute QP values ROI_QP Region Of Interest values. UNIFORM_QP All CUs of the slice use the same QP |
||||||
|
MaxQP[.<IPB>] MinQP[.<IPB>] |
RATE_CONTROL | <string>/<ae> Default:51(0 for MinQP), Values: AUTO or 0-51 | ||||
| Maximum/Minimum QP allowed (optionally specified per picture type). | ||||||
| SliceQP | RATE_CONTROL | <string>/<ae>. Default: AUTO. Values: AUTO or 0-51 | ||||
| Quantization parameter. When RateCtrlMode == CONST_QP, the specified QP is applied to all slices. In other cases, the specified QP is used as initial QP. When RateCtrlMode = CBR the specified QP is used as initial QP. SliceQP = AUTO is the recommended mode when RateCtrlMode is CBR or VBR. | ||||||
|
IPDelta PBDelta |
RATE_CONTROL | <string>/<ae> Default: AUTO. Values: AUTO or 0-51 | ||||
| Corresponds to the value added to the QP of the I-picture to get the QP of the P-picture (IPDelta), or the value for P- to B-picture (PBDelta). | ||||||
| Gop.TempDQP | GOP | <array> Default 1 2 3 4. Range -51-51 | ||||
| Specifies a delta QP for pictures with temporal id 1 to 4 | ||||||
| QpTablesFolder | INPUT | <path> | ||||
| The location of the files containing the QP tables to use for each frame. This is used when QPCtrlMode is set to LOAD_QP. | ||||||
| ROIFile | INPUT | <path> | ||||
| File containing the Regions of Interest used to encode. | ||||||
|
PicCbQpOffset PicCrQpOffset SliceCbQpOffset SliceCrQpOffset |
SETTINGS |
<ae> Default: 0 Values: -12 <= PicCbQpOffset + SliceCbQpOffset <= 12 -12 <= PicCrQpOffset + SliceCrQpOffset <= 12 |
||||
|
The QP offset for the chroma channel (Cb or Cr) at picture(slice) level. Slice level offsets only applies to HEVC. Picture level offsets apply to AVC and HEVC. |
||||||
| CuQpDeltaDepth | SETTINGS | <ae> Default: 0 Values: 0-2 | ||||
| HEVC only. The QP per CU granularity, used only when QPCtrlMode is set to one of: LOAD_QP, AUTO_QP, ADAPTIVE_AUTO_QP. Delta QP available down to CU = MaxCUSize / 2N with N in [0..2]. | ||||||
| ScalingList | SETTINGS | <string> Default: DEFAULT | ||||
|
The scaling list mode. CUSTOM A file defining the coefficient of the quantization matrices shall be provided with FileScalingList. DEFAULT Default value. FLAT Flat scaling list; all matrix coefficients are set to 16 |
||||||
| FileScalingList | SETTINGS | <path> | ||||
| If ScalingList is CUSTOM, the file containing the scaling matrices | ||||||
| Group of Pictures | ||||||
|
GopCtrlMode --gop-mode |
GOP |
<string> Default: DEFAULT_GOP. |
||||
|
The Group of Pictures (GOP) configuration mode. ADAPTIVE_GOP Advanced GOP pattern with an adaptive number of consecutive B-pictures DEFAULT_GOP Basic GOP setting DEFAULT_GOP_B Basic GOP setting with only B-pictures. LOW_DELAY_B GOP pattern with a single I-picture at the beginning followed with B-pictures only. Each B-picture uses the previous picture as first reference; the second reference depends on Gop.Length. LOW_DELAY_P GOP Pattern with a single I-picture at the beginning followed by P-pictures only. Each P-picture uses the previous picture as reference. PYRAMIDAL_GOP Advanced GOP pattern with hierarchical B-pictures. The size of the hierarchy depends on Gop.NumB. PYRAMIDAL_GOP_B Advanced GOP pattern width hierarchical B-pictures, includes only B-pictures |
||||||
|
Gop.Length --gop-length |
GOP |
<ae> Default: 30. Values: 0-1000 |
||||
| GOP length in frames including the I-picture. Used only when GopCtrlMode is set to DEFAULT_GOP or PYRAMIDAL_GOP or LOW_DELAY_B. Should be set to 0 for Intra-only. | ||||||
|
Gop.NumB --gop-numB |
GOP |
<ae> Default: 0. Values: 0-4. |
||||
| Maximum number of consecutive B frames in a GOP. Used only when GopCtrlMode is set to DEFAULT_GOP, ADAPTIVE_GOP or PYRAMIDAL_GOP. | ||||||
| Gop.FreqIDR | GOP | <string>/<ae> Default: DISABLE. Values: DISABLE, SC_ONLY, >= 0 | ||||
|
The minimum number of frames between two IDR pictures. Actual IDR insertion location depends on the position of the GOP boundary. SC_ONLY Intra Decoding Refresh on scene change only. |
||||||
| Gop.EnableLT | GOP | <enable> Default: DISABLE | ||||
| Enables the Long-Term Reference Picture (LTRP) feature | ||||||
| Gop.FreqLT | GOP | <ae> Default: 0 | ||||
| The Long-Term reference picture refresh frequency in number of frames | ||||||
| Gop.GdrMode | GOP | <enable> Default: DISABLE | ||||
|
When GopCtrlMode is LOW_DELAY_(PB), this parameter specifies that Gradual Decoder Refresh (GDR) is used. DISABLE/GDR_OFF Disable GDR GDR_HORIZONTAL GDR using a horizontal bar moving from top to bottom. GDR_VERTICAL GDR using a vertical bar moving from left to right. |
||||||
| Gop.FreqRP | GOP | <ae> Default: MAX. Values: 0-MAX | ||||
| The number of frames over which a picture is refreshed in case of GDR. | ||||||
|
Gop. WriteAVCHeaderSVC |
GOP |
<enable> Default: ENABLE |
||||
| For AVC only. AVC Scalable Video Coding extension (Annex G) defines the standard of AVC scalability and the syntax of the NAL unit header for this extension: nal_unit_header_svc_extension. When temporal scalability is used to encode a stream, this specifies if this header is written. | ||||||
| Gop.DoubleRef | GOP | <enable> Default: DISABLE | ||||
| Allows a P-picture to reference two pictures. | ||||||
| Low Latency | ||||||
|
NumSlices --num-slices |
SETTINGS |
<ae> Default: 1. Values 0-128(HEVC) or 256(AVC) |
||||
|
Number of slices used for each frame. Each slice contains one or more full LCU rows and are spread over the frame as regularly as possible. The maximum value is determined according to the maximum picture height and the minimum LCU size. The maximum value may not be reachable as the number of slices are also dependent to level conformance and multicore encoding for instance. |
||||||
| DependentSlice | SETTINGS | <enable> Default: DISABLE | ||||
| Additional slices are dependent slice segments (Applies only when NumSlices > 1 or SliceSize > 0). | ||||||
|
SliceLat SubframeLatency --slicelat --framelat |
SETTINGS |
<enable> Default: DISABLE Note: --framelat disables subframe latency mode. |
||||
| Enable the subframe latency mode. This option requires multiple slices per frame. In this mode, each slice is forwarded as soon as encoded. In frame latency mode this happens after a complete frame (all slices of a frame) is encoded. | ||||||
| NumCore | SETTINGS | <string>/<ae> Default: AUTO (=0). Values: AUTO, 0-2 | ||||
| The number of cores to be used for encoding. | ||||||
| ConstrainedIntraPred | SETTINGS | <enable> Default: FALSE | ||||
| AVC Only. The value of constrained_intra_pred_flag syntax element. This limits the prediction modes that can be used by the encoder. It is a trade-off between complexity and quality. | ||||||
| Deblocking | ||||||
| LoopFilter | SETTINGS | <enable> Default: ENABLED | ||||
| Use deblocking loop filter. | ||||||
|
LoopFilter. CrossSlice |
SETTINGS | <enable> Default: ENABLED | ||||
| Deblock across the left and upper boundaries of each slice of the frame. | ||||||
| LoopFilter.CrossTile | SETTINGS | <enable> Default: ENABLED | ||||
| For HEVC only. In-loop filtering across the left and upper boundaries of each tile of the frame. | ||||||
|
LoopFilter. BetaOffset |
SETTINGS | <ae> | ||||
| Deblocking algorithm beta offset. | ||||||
| LoopFilter.TcOffset | SETTINGS | |||||
| Deblocking algorithm alpha_c0 offset for AVC or Tc offset for HEVC | ||||||
| Miscellaneous | ||||||
| CabacInit | SETTINGS |
<enable>/<ae> Default: DISABLE Values: 0-1 (or 2 for AVC). |
||||
| As defined by the standard, it specifies the CABAC initialization table | ||||||
| DirectMode | SETTINGS |
<string> Default: SPACIAL Values: SPACIAL, TEMPORAL |
||||
| DirectMode exploits the spatial or temporal correlation on B-frames by deriving its motion vector from previously encoded information. | ||||||
| EnableSEI | SETTINGS | Determines which Supplemental Enhancement Information are sent with the stream | ||||
|
SEI_NONE No SEI message is inserted SEI_ALL means SEI_PT | SEI_BP | SEI_RP SEI_ATC Alternative Transfer Characteristics SEI SEI_BP Buffering Period SEI SEI_CLL Content Light Level SEI SEI_MDCV Mastering Display Colour Volume SEI SEI_PT Picture Timing SEI SEI_RP Recovery Point SEI SEI_ST2094_10 Recovery Point SEI SEI_ST2094_40 ST2094-40 SEI |
||||||
| ForcePpsIdToZero | SETTINGS | <enable> Default: DISABLE | ||||
| Every PPS ID is set to 0. Not compatible with reordering. By default, the PPS id increases by 1 each time a PPS with different settings must be sent. | ||||||
| EnableAUD | SETTINGS | <enable> Default: DISABLE | ||||
| Determines if Access Unit Delimiters (AUD) are added to the stream or not | ||||||
| StartCode | SETTINGS | <string> Default: SC_AUTO | ||||
|
Start code size alignment. SC_3_BYTES Force start-code 3-bytes alignment: 0x00 0x00 0x01 SC_4_BYTES Force start code 4-bytes alignment: 0x00 0x00 0x00 0x01 SC_AUTO Software select the start code bytes alignment |
||||||
| PCM | SETTINGS | <enable> Default: DISABLE | ||||
| The encoder should consider the PCM tool. | ||||||
| UniformSliceType | SETTINGS | <enable> Default: DISABLE | ||||
| Enable uniform slice types encoding. This means that all the slices will have the same type. The encoder will use a slice_type between 5 and 9. | ||||||
| SAO | SETTINGS | <enable> Default: DISABLE | ||||
| Enable the Sample Adaptive Offset filter. | ||||||
| WeightedPred | SETTINGS | <string> Default: DEFAULT Values: DEFAULT, IMPLICIT | ||||
| Specify the weighted prediction mode applied on bi-predictor blocks. | ||||||
| LambdaCtrlMode | SETTINGS | <string> Default: AUTO_LDA | ||||
|
The lambda values used for rate-distortion optimization AUTO_LDA Automatically select between DEFAULT_LDA and DYNAMIC_LDA depending on RateCtrlMode and GopCtrlMode. DYNAMIC_LDA Uses lambda factors depending on GOP pattern. LOAD_LDA Uses lambda factors from a provided file called ‘Lambdas.hex’ in same folder as the application. |
||||||
| LambdaFactors | SETTINGS | <array> Values: 0.00 – 1.00 | ||||
| Specifies a lambda factor for each picture. The factors are ordered as: I, P, B(temporalId = 1), B(temporalId = 2), B(temporalId = 3), and B(temporalId = 4). A factor on LOAD_LDA lambdas is available for each picture type and temporal ID with the LambdaFactors table. Example: LambdaFactors = 0.20 0.35 0.59 0.60 1 1 | ||||||
| Enables the look-ahead encoding mode specifying the number of frames in advance for the first analysis pass. This option increases the encoding latency and the number of required memory buffers. | ||||||
| Enables the two-pass encoding mode and specifies which pass is considered for the current encoding. In two-pass encoding, during the first pass information is stored in a file (see TwoPassFile). During the second pass this information is read back. | ||||||
| File containing the first pass statistics. | ||||||
| During first pass, to encode faster, enable only the scene change detection | ||||||
| UseL2C | SETTINGS | <enable> Default: FALSE | ||||
| The encoder must use the L2 cache | ||||||
| ForceStreamBufSize | SETTINGS | <ae> | ||||
| The size of the bitstream buffer at the output of the encoder. Will be automatically increased in subframe latency mode. | ||||||
| CmdFile | INPUT | <path> | ||||
| File containing the dynamic commands to send to the encoder. | ||||||
| ScnChgLookAhead | RUN | |||||
| When CmdFile is used with defined scene change position, this parameter specifies how many frames in advance the notification should be sent to the encoder. | ||||||
|
MapFile --map |
INPUT | <path> | ||||
| The map file name. | ||||||
| HDRFile | INPUT | <path> | ||||
| Name of the file specifying HDR SEI contents. | ||||||
|
GMVFile --gmv |
INPUT | <path> | ||||
| Optional input file for Global Motion Vector feature. When this parameter is not present, the GMV feature is disabled. The GMV file format is specified in Global Motion Vector File Format. | ||||||
| BitrateFile | RUN | <path> | ||||
| The generated stream size for each picture and bitrate information will be written to this file. | ||||||
|
InputSleep --input-sleep |
RUN |
<string>/<ae> Default: DISABLE Values: DISABLE, >0 |
||||
| Adds the specified interval (in milliseconds) between frame processing. | ||||||