R. Sumi Matsumoto

MSLIS Portfolio, Class of 2021

Video File Formats & Normalization

Brief Project Summary

Although digital technology has been around for decades, archives are only recently having to manage digital content. There is no standard for archival digital formats, since many factors: from cost to equipment to expertise, impact institutions differently. Video files are particularly complex in their structure and compatibility. In addition to the wrapper (the file extension), a video file also contains at least one audio and video codec. All three formats (video codec, audio codec, and wrapper) must be compatible internally as well as with the playback software.

This project compared video container formats and codecs to better understand the options for video archiving. I was curious which transwrapping and transcoding mappings create the least loss in quality and how codec compatibilities impact normalization. I worked with a small collection of personal, born-digital videos that ranged in container formats, recording devices, and video codecs. I used the open-source, command-line tools MediaInfo and FFmpeg to evaluate, transcode, and transwrap each file into AVI, MOV, WMV, MP4, MKV, and MXF. I chose to use file size and bit rate as a quantitative metric of quality due to a lack of consumer-level tools for video quality assessment. Although transcoding and transwrapping allowed more wrapper options, the process often created a lower resolution file. Just transwrapping, when possible, made a smaller impact on the file’s quality. Transwrapping may be a useful tool in creating a normalized viewing format without too much quality loss or file manipulation however, codec compatibility is a major issue.

Read the full project summary & view a sample of the project data