Man, I've got you guys trained well. Thanks for filling in!

And as a follow up:

I have yet to find an encoder that doesn't artificially insert a partial frame of silence at the beginning of a track...

I happen to use Audiocatalyst to do my encoding but I don't listen to any continuous mixes so I can't really tell you how well it is capable of doing.

...Including the Xing encoder that ships with AudioCatalyst, it has the same problem as all the others. I actually e-mailed Fraunhofer and asked about this once, and they basically said it's the nature of the data compression and there really isn't a way around it. I don't understand the underlying details of the data compression algorithm, so I'm not clear on exactly why this is necessary. If anyone finds an encoder that fixes this problem, let me know.

And the problem with the end of a track is that the frames are all a fixed timespan, and if the end of a track doesn't fall on the end of a frame, then you get a partial silent frame at the end, too.

There are a few theoretical ways around this problem, though:

1) Player software could be written to pre-decode the end and beginning of a song pair, then detect the silence and slide the wave data sections together so that they play back seamlessly. Not a trivial task, I haven't seen anything which does this. Even the gapless playback plug-in for WinAmp simply streams the songs together as-is, so if the source MP3s have partially silent frames, then the output of the plug-in will have the silence, too.

2) You could encode your entire album as one huge MP3 file, then use a utility like MP3split to slice it up into sections. A lot of work, but definitely do-able. And the results would be perfect.

3) My solution, as noted before, is to rip the album normally, then hand-trim frames and do trial-and-error previewing until the gaps sound OK. Far from perfect, but better than nothing.

When Hugo demoed the Mark2 prototype for me, he had a techno mix on the empeg that sounded like it was playing perfectly seamlessly to me. I don't know how he created the files, but I'm guessing it was with a method similar to (2) above, which made them sound so good. Also, the nature of the music (very rhythmic) was such that even if there were gaps, I might not have been able to hear the gaps.

___________
Tony Fabris