What are the apparently extra files on a DVD, and what gets lost when the VOB file is converted to MP4?

Question

This is for curiosity only; I'm not trying to break copyright.
(I own the disc and there's a much better quality version available on YouTube).

I copy one file from the disc, convert it from .vob to .mp4 format and end up with a file that's less than 15% of the size of the original disc.

What exactly gets lost during this process?

$ find /tmp/Name -type f | wc
     13      13     405

$ du -s -m /tmp/Name/*
0   /tmp/Name/AUDIO_TS
1   /tmp/Name/AUTORUN.INF
7   /tmp/Name/IVI
192 /tmp/Name/VIDEO_TS

$ ls -l /tmp/Name/VIDEO_TS/
-r--r--r-- 1 ray ray     16384 Jan 10  2003 VIDEO_TS.BUP
-r--r--r-- 1 ray ray     16384 Jan 10  2003 VIDEO_TS.IFO
-r--r--r-- 1 ray ray    116736 Jan 10  2003 VIDEO_TS.VOB
-r--r--r-- 1 ray ray     20480 Jan 10  2003 VTS_01_0.BUP
-r--r--r-- 1 ray ray     20480 Jan 10  2003 VTS_01_0.IFO
-r--r--r-- 1 ray ray     75776 Jan 10  2003 VTS_01_0.VOB
-r--r--r-- 1 ray ray 200710144 Jan 10  2003 VTS_01_1.VOB

$ ffmpeg -i /tmp/Name/VTS_01_1.VOB  /tmp/Name/Name.mp4

$ ls -l /data/Music/Download/Hurt.mp4
-rw-rw-r-- 1 ray ray 28439944 Feb 15 15:38 /tmp/Name/Name.mp4

So:

• start with: 13 files, 198 Mb
• select: 1 file, 192 Mb
• convert: 1 file, 28 Mb

What was lost in the other 12 files?

What was lost in the 85% reduction during the conversion?

The video is 720p at 29.97 fps, and I don't see any significant difference between the two versions on a 4K monitor.

Answer 1

DVDs are usually encoded using an MPEG 2 codec. This is an older codec than h264, and encodes at a higher bitrate (and also encodes video differently) to a low bitrate mp4 encoder which uses an h264 codec.

The VOB files contain the video. The other files are metadata and other stuff you probably don't need if you're just looking to encode the video.

When you compress a video using any lossy compression algorithm, the codec looks for areas in each frame that the human vision system is less likely to notice. So it uses less data to encode areas with less detail in them.

So for example, someone's face might be clearly encoded, but when you look at the shadow on the wall behind them, you may see it looking blocky and those blocks jumping around, as the encoder struggles to render the original video to the best of its ability given the bandwidth constraints.

Watching on a well calibrated large display, it's usually possible to see the difference, especially on darker footage with lots of movement. A classic way to spot it is looking at things like confetti or water, where the whole screen is in motion. On low motion scenes, it can be hard to spot.