I'm sorry, I still don't get the supposed benefits of this round-robin patch over the legacy GIL.   Given that using interpreter ticks as a basis for thread scheduling is problematic to begin with (mostly due to the fact that ticks have totally unpredictable execution times), I'd much rather see further GIL work continue to build upon the time-based scheduler that's been implemented in Python 3.2.  For instance, I think being able to specify a thread-switching interval in seconds (sys.setswitchinternal) makes much more sense than continuing to fool around with check intervals and all of this tick business.

The new GIL implementation is by no means perfect, but people are working on it.   I'd much rather know if anything that you've worked out with this patch can be applied to that version of the GIL.