Non-ergodic (timed and first-exit) runs now parallelise arbitrarily. This has been tested on a 64-core (128 worker threads) Debian GNU/Linux machine, along with many two- or four-core machines. The default setting remains as a single worker thread. To accomplish this, most of the allocation and worker thread data have been shifted around and reconstituted. There's still work to be done to remove overhead, but single-worker simulations should run at the same speed.