A database focused media conversion utility that converts video files to the HEVC video codec with a focus on reducing disk usage in media libraries. This script attempts to be as safe as possible, however encoding to HEVC is a lossy operation. though it should be unnoticable it is recommended to test first. Backups are encouraged.
x265-videoconverter needs python3 and ffmpeg The latest version of ffmpeg is 4.1.4 when testing, download from ffmpeg.
main.py -t /path/to/media -s
main.py -n 10
usage: main.py [-h] [--crf int] [--errors] [--database DATABASE] [--focus PATH] [--list-paths] [--low-profile] [--number NUMBER] [--nvenc] [--preset PRESET] [--track PATH] [--saved-space] [--scan] [--quiet] [--verbose]
A database focused media conversion utility that converts video files to the HEVC video codec with a focus on reducing disk usage in media libraries. This script attempts to be as safe as possible, however encoding to HEVC is a lossy operation. though it should be
unnoticable it is recommended to test first. Backups are encouraged.
optional arguments:
-h, --help show this help message and exit
--crf int CRF paramater to be passed through to ffmpeg, determines quality and speed with lower values being slower but higher quality
--errors, -e list errors
--database DATABASE name of database to be used
--focus PATH, -f PATH
immediately begin conversion on target directory
--list-paths, -lp list tracked paths
--low-profile for weaker devices, convert to 4-bit HEVC including downgrading 10-bit hevc
--number NUMBER, -n NUMBER
transcode from tracked paths limit number of files to be converted
--nvenc transcode using NVENC compatable GPU
--preset PRESET string for ffmpeg paramater, accepts ultrafast, superfast, veryfast, faster, fast, medium, slow, slower, veryslow and placebo, slower speeds have a higher filesize and better quality
--track PATH, -t PATH
add a new path to be tracked
--saved-space display HDD space saved by transcoding into x265
--scan, -s scan tracked directories for new files
--subtitles copy external subtitles
--quiet, -q only produce minimal output
--verbose, -v produce as much output as possible