The PC is an Athlon 64 x2, 4400+, with 3GB of RAM, so it should certainly be able to handle three or four audio streams. As it happens, Multisequence can't even handle two. And when doing non-real-time rendering! If I use "Save as audio...", I get the same exact stutter.
This sure seems like a bug to me. So I created a test case, which you can download here:
The zip file contains a Multisequence project, two FLACs I used for input, and the output WAV file Multisequence produced (which I also converted to FLAC to save on space). Note the stutter on both the whistling and the speech (but only the first time).
Whaddya think?
larry