[http://sourceforge.net/p/ufoai/bugs/3163 Item 3163] imported from sourceforge.net tracker on 2013-01-28 20:02:17

First, the game tries to use autostand when moving an actor to or form a position that can only be walked while crouching (like under the wing of a harvester), resulting in the actor not moving at all and still losing 6 TU, as seen in autostand0.jpg and autostand1.jpg

Second, I know you are aware that the calculated move length of an autostanding is often short by some TUs, resulting in the actor ending its move with more TUs than expected (the longer the move the more extra TUs), but I'm not sure if you know that this miscalculation can also result in the actor being short of TUs at the end of its move, resulting in it being unable to crouch again once reached its destination (unpleasant surprise if it results in your soldier being headshoot because of the inability to crouch back again) as seen in autostand2.jpg -- I thought I'd mention this in case you are interested.

Using latest master It is reproducible on demand easily.
===== Comments Ported from Sourceforge =====

====== darkrainx (2012-03-06 19:11:44) ======

Try to autostand to a low ceiling position
====== darkrainx (2012-03-06 19:12:47) ======

Try to autostand from a low ceiling position
====== darkrainx (2012-03-06 19:13:48) ======

Autostanded and failed to crouch again at the end of move
====== aduke1 (2012-03-12 21:45:53) ======

Confirmed. Happens when walking more than squares diagonally.
I agree, that's a nasty little trap.
But there is a workaround (don't use autostand), thus only prio 4.

And I have an idea what happens here...we'll see ;)
====== aduke1 (2012-03-12 21:48:16) ======

* more than four squares...
====== aduke1 (2012-03-13 22:55:31) ======

darkrainx, you opened a can of worms...
The following is only partially targeted at you, I'm just taking notes.

The culprit is SHOULD_USE_AUTOSTAND.
1. the formula is BS imho.
2. even if corrected, if calculates for the whole move and doesn't take into account that pathfinding rounds the TUs down for each single step. => rounding difference
3. it's called in *two* places and I don't yet understand why. Should be just *one* imho.
4. The decision to autostand does NOT consider any obstacles on the path that force you to crouch again midway.

==> no clue how to fix it atm :(
====== aduke1 (2013-01-09 01:58:27.687000) ======

The TU calculation issues should be fixed with commit f7e8908.
====== aduke1 (2013-01-10 02:19:08.174000) ======

The "don't autostand if you can't stand" issue should be fixed in 987f520.

Darkrainx, plz check if you can find any remaining issues.