mirror of
https://github.com/master-of-zen/Av1an.git
synced 2024-11-24 18:18:06 +00:00
51 lines
No EOL
1.6 KiB
Markdown
51 lines
No EOL
1.6 KiB
Markdown
### 7
|
|
- Starting including rust code
|
|
- FFMPEG info, hash path, adapt probing rate to rust
|
|
|
|
### 6
|
|
- ~1.7x faster probes for svt-av1
|
|
- ~2x faster probes for aomenc
|
|
- Changed rav1e settings, disable rav1e scene detection
|
|
- Temporally removed VVC support until it's 100% ready and working
|
|
- Added overwrite promt
|
|
- Speed up for SVT-AV1 target quality
|
|
|
|
|
|
### 5
|
|
- Fixed fatal errors with ffms2/lsmash
|
|
- Added vmaf validation on each time when VMAF initialized
|
|
- Fix running not required frame probe
|
|
- Chunk restarting
|
|
- Fixed ffmpeg segmenting
|
|
- `color_range 0` by default for pipes
|
|
- aomenc target quality probes to be 8 bit
|
|
|
|
### 4
|
|
- Refactored Args to Project class
|
|
- Removed dead Rust code
|
|
- Default encoder settings changed
|
|
- Better Vapousynth error handling
|
|
- Target Quality settings balanced
|
|
- Target Quality score calculation fixed and improved
|
|
- Default extra_splits set to 240 frames.
|
|
- Extraction and Concatenation to copy all streams.
|
|
- Scenes file to save total frame count, faster restart/start up with scenes file.
|
|
- Fix concat on windows with ffmpeg.
|
|
- Revorked args to classes.
|
|
- Skip files in queue if they already encoded.
|
|
- Default chunk method to ffms2.
|
|
- Per frame target quality for SVT-AV1
|
|
- Added none split method option
|
|
- Added webm output
|
|
- Refactored VMAF and Target Quality
|
|
- VMAF is now separate, ready for import module
|
|
- Changed Target Quality probing rate
|
|
- Refactored module structure/names
|
|
|
|
### 3
|
|
- Default pix format to be yuv420p10le
|
|
- Default scene change interval to be 120 frames
|
|
|
|
### 2
|
|
- Target VMAF renamed to Target Quality
|
|
- Changed Algo for Target Quality score calculation |