The above media is the worst H264 example anyone can give. (sorry)
- The attached video was created with the oldest AVC codec profile (Main@L3.1), it was part of the original release of the standard and has been present since 2003!
Most up-to-date software uses at least 4.0, if not a higher profile. - The video bitrate is so tiny that it is compressed to a level that extreme processing is needed to play it back, which requires time, which requires audio time shift so it will playback ‘in-sync’ with the visual. A good practice is to export H264 with at least 3 Mb/s
- It looks like the video is a rerender of another (original) H264 video, so the audio shift was done twice, which makes it impossible to determine the correct total audio offset just by analyzing the container’s metadata.
In short, if I receive such material from a client, I always say, “Sorry, it is unusable.”, Please send me a higher-quality video.
Nevertheless, for all of the above (and many more reasons), never (ever) trust the audio sync for a video that comes embedded with an H264 video. That is exactly why professionals invented 2-pop!
If you want something you can 100 percent trust, ask for DNxHD or ProRes videos. Those are the only professional codecs that were designed for pro production. H264 is nice to have for its small size, but it was created for end-users as a final product, not for studio collaboration.
Just my two-cent opinion.
Edit:
That said. If you convert the sample video above with my ProResER Pro app using the settings below. You will get separate video and audio files. Importing both video and audio files into Cubase/Nuendo or any other DAW will give you perfectly synced audio if the audio is aligned exactly with the video start.
