[ANIM]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
[SAV1]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
The number of threads used by decoder.
Note: Default is 1.
Type: UInt32
The number of frames that can be processed in parallel.
Note: Default is 1.
Type: UInt32
[DAV1]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
Configures the number of CPU threads to use.
Note: Defaults to 16.
Type: Int32
Maxmimum number of frames to compute in parallel.
Note: Set to 1 for low-latency decoding.
Internally clipped to Threads.
Defaults to 8.
Type: Int32
Selects an operating point for scalable AV1 bitstreams (0 - 31).
Note: Defaults to 0.
Type: Int32
Apply grain.
Note: Defaults to true.
Type: Boolean
Output all spatial layers of a scalable AV1 biststream.
Note: Defaults to true.
Type: Boolean
[AVUI]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
[CANO]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
Configures the maximum number of threads to use for encoding.
Note: Defaults to 0 (automatic).
Type: Int32
[FFV1]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
[F266]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
Size of the threadpool allocated for decoding.
Set to 0 for single-threaded execution and -1 to allocate one thread per core available.
Note: Defaults to -1.
Type: Int32
Maximal number of parsed frames waiting for reconstruction.
Increases the decoding latency but improves MT scaling.
Note: Set to -1 to allow one parse frame delay per allocated thread.
Defaults to -1.
Type: Int32
Verify picture, if digest is available.
If true, checks hash in SEI messages if available, otherwise ignore SEI message.
Note: Defaults to false.
Type: Bool
[FJXS]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
Enables proxy image extraction and selects the size of the proxy image.
A proxy image is a scaled down version of the image encoded in the codestream. This property defines the scale parameter as an exponent to the basis of 2. If the parameter is set to 0, proxy extraction is disabled and the original image size is used, if the parameter is 1, an image of half the width and half the height is reconstructed, a parameter of 2 extracts an image of a quarter of the width and height. Whether proxy extraction is possible, or how many proxies are available depends on the encoding parameters, in particular the number of vertical decomposition levels. For the default parameter of 1 vertical decomposition level, a single proxy of half the dimensions is extractable, for 2 vertical decomposition levels, 2 proxies are included in the codestream. The case of 0 vertical decomposition levels does not include any proxies.
Note: Defaults to 0.
Type: Int
If the processors array is supplied and AllocCores is true, then the filter pins its threads to the CPU ids provided in this array. For that, the number of elements in this array shall be at least as large as the number of threads started.
If the processors array is not supplied or nullptr, and AllocCores is true, the filter pins its threads to logical processors #1 and up, in sequential order. Logical processor #0 remains unallocated and available for the system or the calling thread. Note that this may not be ideal, depending on the topology of the host system.
Note: Defaults to nullptr.
Type: VariantArray of Ints
The number of parallel threads to be launched for the decoding process.
Decoding speed scales almost linearly with the number of available threads.
Note: Defaults to 8.
Type: Int
If this property is true, the filter sets the CPU affinity of its threads to CPU cores, i.e. it pins its threads to particular logical processors.
This may help to improve overall system performance as the operating system is disallowed to migrate worker threads between logical processors.
Note: Defaults to false.
Type: Bool
Enables or disables a decoding mode that uses a more compact internal representation of the image data and thus lowers the required memory bandwidth.
If fast mode is enabled, the decoded images will deviate slightly from the normative reference output of the standard-defined reference decoder model, and hence, the decoder is — strictly speaking — no longer fully conforming to the standard. However, in many applications the quality loss due to the fast mode is negligible, but improves the performance of the decoder by up to 10%.
Note: Defaults to false.
Type: Bool
[VPX ]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
Configures the number of CPU threads to use.
Note: Defaults to 16.
Type: Int32
[JPEG]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
[MFV ]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
Represents the device index to use.
If the compute mode indicates either GPU, OpenCL, Cuda, or Metal, this determines the device to use.
Note: type: Int32
Determines the mode of computation.
This property is used to select which mode of computation to use.
If any GPU compute mode is specified, use the DeviceIndex property to determine which device to use when more than one is pressent.
Note: type: ComputeModes
Valid values for the ComputeMode property.
| Value | Name | Description |
|---|---|---|
0 |
CPU | CPU computation mode |
5 |
D3D9 | D3D9 computation mode (not supported) |
6 |
D3D11 | D3D11 computation mode |
[PNG ]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
[SOSt]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
The number of threads the filter should use
Type: Int
Decompress at 1/4 resolution
Type: Bool
Decompress at 1/2 resolution (i.e one field, full width)
Type: Bool
[SORa]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
The number of threads the filter should use
Type: Int
Decompress at 1/4 resolution
Type: Bool
Decompress at 1/2 resolution (i.e one field, full width)
Type: Bool
[APPR]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
Flag indicating if the alpha channel should be preserved. This property is only meaningful for input formats with an alpha channel.
Type: Bool
Set true if the filter should run as a low priority task
Type: Bool
The number of threads the filter should use
Type: Int
Decompress at 1/4 resolution
Type: Bool
Decompress at 1/2 resolution (i.e one field, full width)
Type: Bool
[APRR]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
The number of simultaneous processing threads to spawn.
Set this to 0 to have the decoder spawn according to the number of processors detected in the system.
Type: Int32
The down-sampling method.
Type: DownSampleModes.
The desired output primaries of the color-space.
Note: Can be either Primaries::BT_709 or Primaries::BT_2020.
Type: Byte
Valid values for the DownSampleMode property.
| Value | Name | Description |
|---|---|---|
3 |
SixtyFourthSize | 1/8 width x 1/8 height |
0 |
FullSize | Full size |
2 |
SixteenthSize | 1/4 width x 1/4 height |
1 |
QuarterSize | 1/2 width x 1/2 height |
[CFHD]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
[MC64]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
[MC65]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
[MCM2]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
Perform native decompressor deinterlace if available
Type: Bool
Flushes every frame that is entered in the decompressor (e.g. for Main Concept I frame rendering)
Type: Bool
Perform native decompressor deinterlace if available
Type: Bool
[MCM4]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
[MCDV]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
[WMV ]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
[CPJP]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
Number of quality layers to be discarded.
Use 0 to not discard any quality layers.
Note: Defaults to 0.
Type: Int32
Configures the level of logging.
Valid values are enumerated by the LOG_LEVEL DIB_PROPERTY_ENUM.
Note: Defaults to the current Dib Console log level.
Type: LOG_LEVEL
Number of the highest resolutions to be discarded.
Each discarded resolution reduces the image size to half (e.g., 1 discarded resolutions ends up in half-size image, 2 discarded resolutions ends up in quarter-size image). Use 0 to decode the whole image.
Note: Defaults to 0.
Type: Int32
Configures the number of CPU threads to use.
Note: Defaults to 0 (use all available cores).
Type: Int32
Number of the least significant bits to be discarded from all image samples.
Use 0 to not discard any least significant bits.
Note: Defaults to 0.
Type: Int32
Maximum number of the most signification bits to keep in all image samples.
Use 0 to not limit the maximum number of the most significant bits. When both DiscardBitDepth and KeepBitDepth parameters are nonzero, first the given number of the least significant bits is discarded and then the resulting bit depth is limited by the maximum allowed number of the most significant bits which can be kept.
Note: Defaults to 0.
Type: Int32
Maximum number of quality layers to keep.
Use 0 to not limit the maximum number of quality layers. When both DiscardLayers and KeepLayers parameters are nonzero, first the given number of quality layers is discarded and then the resulting number of quality layers is further limited by the maximum number of allowed quality layers which can be kept.
Note: Defaults to 0.
Type: Int32
Limits overall memory usage of specified device (in mega-bytes).
Use 0 for no limit.
Note: Defaults to 0.
Type: Int32
Limits number of tiles to be decoded at one moment.
Use 0 for no limit.
Note: Defaults to 0.
Type: Int32
Configures the decoder to limit the number concurrent decoders.
Limits number of images to be decoded at one moment.
Note: Defaults to 30.
Type: Int32
Determines the mode of computation.
This property is used to select which mode of computation to use.
If either CM_CUDA or CM_OPEN_CL compute mode is specified, use the DeviceIndex/PlatformIndex properties to determine which device to use when more than one is pressent. If CM_CPU mode is selected use the FilterThreads property to determine the number of threads to use.
If either CM_CUDA or CM_OPEN_CL are specified and unavailable, then the compressor will fallback to CM_CPU. After Begin has been called, this property may be updated to indicate the which compute mode was actually selected.
Note: Defaults to CM_CPU.
Type: COMPUTE_MODE
Represents the platform index to use.
If the compute mode indicates OpenCL, this determines the platform to use.
Note: Defaults to the first valid platform.
Type: Int32
Represents the device index to use.
If the compute mode indicates either OpenCL or Cuda, this determines the device to use.
Note: Defaults to the first valid device.
Type: Int32
Represents the availablilty of the device.
Determines if the devices in the Devices array is supported.
Note: Read only.
Type: Bool
Determines the number of devices and their capabilities.
This property is used to determine the available devices.
Each item in the array contains two or three properties used to indicate the device index, if it is supported, and optionally the open cl platform index. Use the DeviceIndex, Supported, and PlatformIndex properties to obtain this information.
Note: Defaults to null.
Type: VariantArray
Valid values for the \refLogLevel property.
| Value | Name | Description |
|---|---|---|
0 |
LL_SILENT | Silent: Disable logging |
10 |
LL_ERROR | Error: Log only error messages |
20 |
LL_WARNING | Warning: Log only warning and error messages |
40 |
LL_INFO | Info: Log only informative, warning and error messages |
100 |
LL_DEBUG | Debug: Log all possible messages |
Valid values for the ComputeMode property.
| Value | Name | Description |
|---|---|---|
1 |
CM_CPU | Cpu computation mode |
2 |
CM_CUDA | Cuda computation mode |
3 |
CM_OPEN_CL | OpenCL computation mode |
[ADNH]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
The number of threads the filter should use
Type: Int
[QSNC]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
[NVDA]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool
Specifies the preferred Nvidia device ID.
If the GPU with the specified device ID supports all the required features, it will be used; otherwise, the first GPU that meets the requirements will be used by default.
Type: UInt32
[XPVC]Four character code that specifies the output video stream format
Type: Four
True if the filter is using hardware acceleration.
Note: This property is read only.
Type: Bool