

Grain can be artistically desirable, but is generally speaking expensive from a bitrate standpoint to represent in the samples. By capturing it as a separate field, stripping it from the samples and then reapplying it at the point of decoding, you get the same image for less bitrate, which is basically what codecs are all about










Sure, you’d need a second exploit to escalate from there.
ffmpeg is expected to run for extended periods of time, given its use in transcoding.