This tutorial, given by Hugo Cornelis at GUM*02, discusses the use of
the hsolve method for faster and more accurate implicit solution of the
model equations. He discusses a number of internal secrets of hsolve
that explain how it gets its extra performance: examples of channel and
table rearrangements, updated simulation schedule, internal data
streams, interoperability with synchans via discrete events. It also
treats the use of findsolvefield and interactions with external objects.
