Well, that's a big problem when you're taking video sources into account - in video sources, there are usually additional or missing black frames at the beginning, which would be accounted for by a simple audio delay, so it's hard to say what is actually frame #0001. That is why I synced all three at a cut from shot to shot and then you can start comparing frame differences, because a frame difference, which is there from the very first black frames and continues to be the same throughout is practically inconsequential.