I found a “shortcut” that can save a lot of time with “difficult” conversions. I redid my second test movie, the one with the ac3 audio stream. I loaded the source directly into WinFF. I created a new preset with exactly the same line suggested by themichael, but I added “-async 1” at the end. As per the documentation, this option tells ffmpeg to “stretch/squeeze” the audio stream to match the timestamps. The parameter is the maximum samples per second by which the audio is changed. -async 1 is a special case where only the start of the audio stream is corrected without any later correction.
As expected, when I ran my avi through WinFF with this preset, ffmpeg gave a lot of errors but it completed the job. I ran the output through SMC. The result is that the video and audio streams are in sync.The only minor problem is that the audio has minor glitches: once or twice per minute, you hear a very short “missing sound”. This imperfection is very tolerable in my opinion (much more so than the out of sync audio I’ve been trying to resolve since I acquired my e260v2.) In my first post above, I gave a procedure to get perfect audio, but it is a lot more work than the shortcut I just described.
A final word on the “-async 1” option. If you find that your audio is a bit out of sync, you might want to create another preset with “-async 2” instead and rerun your conversion in WinFF. This will spread the corrections more evenly throughout the avi and it might be just what you need to get more consistent tracking.