Yeah XRM, stationary tyr with M6 floating around and laser selection algorithm changed to favour slow lasers.
Was highly ineffective shooting PSPs at around the 1-2km mark where the M6 are flying around. I've experimented a little this evening and plan some changes, although most likely I won't be switching in PSP's unless the target is within 1km.
However the balancing will take some time because it makes a huge difference when flying an M6 or M7, because distance to target doesn't take into account the size of the player ship and where the turrets are located relative to the origin, so fixing one case can screw over others.
your bailing settings work just fine even in TC

I can't believe you got that far
Actually the bailing stuff works based off shield and hull damage per shot, so much of the tracking would still be accurate. But things like this will fall over:
Code: Select all
411 |$Laser.Flags = = get bullet flags of laser $Laser
412 |* Can't strip shield with mass drivers
413 |skip if not ( $Laser.Flags & [Bullet.IgnoreShield] )
414 ||endsub
Because the bullet flags command isn't available in TC.
I'd love to create a unified single version. They'd need to backport the new script commands to TC though. They haven't backported any commands, just checked. The command they reused for getting laser refire delay (get default launch time difference for missile type), is returning strange numbers in TC.
I suspect what probably happened in your khaak case, is the refire delay was reported by that API as something very low, so Smart is firing them extremely quickly. The TC version of Smart uses the laser training feature to estimate laser refire delays.
I've added version detection to the 4.3 setup script. If it's not AP it'll send a message to the player and exit.