I must retract my previous statements regarding wind effect on on the
"cruise" STF, i.e the true airspeed to fly to achieve the fastest
speed over the ground between two ground points. Further thought, and
working through a specific example to convince myself, tells me that
airspeed is _independent_ of a head/tailwind. The way I now think
of it is that the "optimal" interthermal-glide airspeed produces the
fastest obtainable "cruise" airspeed, i.e. with no altitude loss over
a thermalling/gliding cycle (for the given thermal strength
and glider polar) - a wind affects the resultant groundspeed but not
that optimal value through the air. This means it is possible to have
a wind for which no forward progress can be made without a net loss of
altitude.

I now believe I've made an incorrect assumption somewhere in script I
wrote to predict "optimal" speeds over a track (most likely neglecting
the thermal drift with the wind) and will need to correct that.

Jack