|
2 | 2 |
|
3 | 3 | **AV1Studio** is a GUI for AV1 encoding via Av1an and SVT-AV1-PSY, written in Rust using egui. If you're allergic to well-written programs and good-looking UIs, this one is for *you*! |
4 | 4 |
|
| 5 | +## Usage |
| 6 | + |
| 7 | +| | **Default Value** | **Description** | |
| 8 | +|:-------------------------: |:---------------------------: |:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: | |
| 9 | +| **Av1an-verbosity Path** | (command) `av1an-verbosity` | Full path to the Av1an-verbosity binary. | |
| 10 | +| **Input File** | None | Full path to the input MKV file. | |
| 11 | +| **Output File** | None | Full path to the output MKV file. | |
| 12 | +| **Scenes File** | None | Full path to a scenes file. (Check out [Trix's Auto Boost Script](https://github.com/trixoniisama/auto-boost-algorithm).) | |
| 13 | +| **Zones File** | None | Full path to a file specifying zones within the video with differing encoder settings. (Check out [Trix's Auto Boost Script](https://github.com/trixoniisama/auto-boost-algorithm).) | |
| 14 | +| **Source Library** | BestSource | Method to use for piping exact ranges of frames to the encoder. BestSource is now, supposedly, the best best and most accurate option, but slightly slower than L-SMASH and ffms2. L-SMASH can sometimes fuck up the frame orders completely. ffms2 might corrupt frames on problematic sources. | |
| 15 | +| **File Concatenation** | mkvmerge | Method to use for concatenating encoded chunks and audio into output file. If you don't know what you're doing, just go with the default option. | |
| 16 | +| **(Output) Resolution** | None | Resolution to resize the output video to. | |
| 17 | +| **(Output) Pixel Format** | yuv420p10le | FFmpeg pixel format to use. It's best to go with yuv420p10le (10-bit color format), even if the input video has 8-bit colors. | |
| 18 | +| **Preset** | 4 | Encoding preset to use. A very simple explanation is that you trade quality for encoding speed, the lower you go. Can be set from a range of 0-13. Generally, the sweet spot will be between 2-4-6, of course, depending on how powerful your CPU is, you might want to go higher. | |
| 19 | +| **CRF** | 27.00 | Sets CRF value. A simple explanation is that you trade file size for quality, the lower you go. Can be set from a range of 0-70, can be set in quarter steps (0.25). Generally, the sweet spot will be between 27-23. | |
| 20 | +| **Synthetic Grain** | 0 | Sets the strength of the synthetic grain applied to the video. | |
| 21 | +| **Custom Encoder Params** | None | Provides SVT-AV1-PSY custom encoder parameters on top of the already included parameters. | |
| 22 | +| **Thread Affinity** | 0 | Pin each worker to a specific set of threads of this size. Leaving this option unspecified allows the OS to schedule all processes spawned. | |
| 23 | +| **Workers** | 0 | Number of workers to spawn. It's generally recommended, if you have enough RAM, to set this to the total amount of CPU cores you have for better encoding speeds. Leaving this at the default value will allow Av1an to figure out the amount of workers to spawn automatically. | |
| 24 | + |
5 | 25 | ## Development |
6 | 26 |
|
7 | 27 | As of now, **AV1Studio** is barely functional and has a lot of bugs. |
|
0 commit comments