No, Zach, this problem has been reproduced by me and others several
times over the past few years, but it hasn't been important enough to be
fixed until now.  For a short time during development last week or the
week before, the SP_PING packets were sent five times faster, and so the
chances of the problem happening were much greater, and so finding the
problem much easier.

It was what we call a race condition problem.  Without this opportunity
created by the temporary bugs introduced into the server, it would have
been very hard to find.  Strict timing would be necessary, over a large
range of timing possibilities.  We would have had to craft a client
modification that would delay UDP initialisation by different amounts of
time, from zero to two seconds, in hundredth second increments.

The change that returned SP_PING to normal on the server, was the change
to efticks() macro.

Please review all the changes.  They aren't all related to 50 fps.
Don't focus on just that feature, to the exlusion of others.

-- 
James Cameron    mailto:quozl at us.netrek.org     http://quozl.netrek.org/