Script is always the fastest way, but manually you can do it by zooming in to the timeline enough (you can zoom in so much that your whole visible timeline is showing just one frame) to get the required accuracy.

It can be done, I just did and it was not even too difficult. The biggest problem was that the main camera was mounted to tripod and shot continuously from balcony and the other was manually held and shot at floor level from several positions around the ring. Thus I was forced to synch each manually shot event to the continuous timeline/event. I could use the soundtrack only for rough synching and then needed to use visual cues for the final touch. I used just the main cameras soundtrack for final product to prevent any shifts in voice quality.

I even mixed HDV with SD and the results are satisfactory. Color correction took most of the time, though...

The event was a floorball-game.

I do it manually al the time.. aside from sync issues, which are dead easy with V7 upwards(each frame is shown on teh yellow arrow with frame number counts as you move) these spanky multicam tools just dont come close to a good ol manual Pan and crop PiP

MRe: One way to simplify is to let both cameras run continuously. Then you only have to sync once.

Jack

I do this on almost every project I produce. 1) While taping, I run both cameras continuously as JackW states, 2) load in Vegas and place on tracks above each other, 3) find a spot in the audio where there is a clear peak (you may have to zoom in closely; if one of the audio tracks isn't loud enough, increase the volume in another program such as Sound Forge to raise the peaks visibily), 4) group the events together, 5) then you can split or use one of the Multi-Cam programs to edit.

Heck, you don't need SoundForge or anything to make the peaks visible, just split the audio track on either side of your area of interest, then right-click into Properties and Normalize.

Once you've gotten that slice of track synced, mate both side pieces to it, then delete the slice. You can then drag the edge of one piece over to cover up the hole, or you can delete one side piece completely and just drag the remaining piece over to create a complete (uncut) audio track.

Sounds harder than it is. I takes only a few seconds to boost levels, sync, and return to normal levels this way.