Further discussion on IRC raises the question of why did this never
happen before, and whether the countdown timers used to be reset before
this code was reworked.

The question boils down to "what changed in the maintenance of the
variable te_surrender?"

Either nothing was changed, or something was changed.  The case for
nothing being changed is that we've very rarely been able to trigger the
situation; a near collapse of one race, everyone swaps to another?

On the assumption that something changed, relevant patches for review
may be:

Sat Dec 15 04:21:05 EST 2007  Karthik Arumugham <karthik at karthik.com>
  * Fix "20 minutes left when T mode resumes" on geno countdown in T mode
      daemonII.c: te_surrender_pause was not getting updated to
          TE_SURRENDER_PAUSE_OFF until the first minute fuse occured
          after the countdown started.
          Use the TE_SURRENDER defines rather than integers.

Above reviewed, no change to maintenance of te_surrender.

Tue Aug  7 23:16:51 EST 2007  quozl at us.netrek.org
  * surrender detail, and struct teams prefix change
        Feature requested by Niclas Fredriks on 2007-07-19, change to
        mention surrender more frequently in last five minutes, and
        provide seconds as well as minutes on TIME command.

Above reviewed, no change to maintenance of te_surrender.
Prior to the above patch, te_surrender was s_surrender.

Finished review back to Mon Mar 21 16:23:48 EST 2005.

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