Loading...
  OR  Zero-K Name:    Password:   
Title: COOP vs AI (chickens or CAI): All welcome!
Host: CZrankIodine
Game version: Zero-K v1.1.11.9
Engine version: 91.0
Battle ID: 210546
Started: 11 years ago
Duration: 30 minutes
Players: 4
Bots: True
Mission: False
Rating: None
Watch Replay Now
Manual download

Team 1 Lost
XP gained: 25
USrankyanom died in 31 minutes
NZrankKaira died in 26 minutes
CArankAltus died in 31 minutes
USrankQuantum_Menace died in 31 minutes




Preview
Filter:    Player:  
sort
I've seen this bug before.

About 27:59 my Inferno missile explodes on the south silo while launching, and there are no air units present. With Alt-B it looks like the silo starts building the next missile on that spot before the launched one is clear, so the launched one collides with the nanoframe.
+0 / -0


11 years ago
Interesting bug.
+0 / -0
It's known, and supposedly here's what happens:

1) You queue launch for missile before it's ready.
2) You have another missile queued for construction.
3) It fires right on completion while silo's buildpieces are still perfectly aligned to build the next missile.
4) Upon firing, missile stops being unit and becomes projectile.
5) Projectiles don't block build space. Silo is ok to go build next missile.
6) Since all pieces are still perfectly aligned, silo creates nanoframe for new missile during almost the same frame.
7) But projectile is still there....
8) BOOM.

Possible pathways to fix:
- disable colliding with friendlies for missile
- forced delay on building transitions in silo animation
+0 / -0
FIrankFFC
11 years ago
or make it not start building next missile at same spot?
+0 / -0


11 years ago
Won't work if there's already 3 missiles and this was fourth.
+2 / -0
FIrankFFC
11 years ago
umm... well not then
+0 / -0
An evening of descent into ever deeper, layered, levels of the silo's workings have yielded me knowledge that this has happened before - and a safeguard was put in place by allowing the missile silo workers some sleep so as no to lose their vigil.

However, one of approx. 60 launches still resulted in an accident.

So i demanded the manager to let them sleep twice more - now, for a full second. The incidents ceased shortly hereafter, and i'm now observing 11239'th launch since last accident.
+5 / -0


11 years ago
A great victory for workplace safety!
+2 / -0

11 years ago
I'm not sure if a second delay between building missiles is welcome when you need to get those missiles up asap, but it's better than nuking your own silo i guess...
+0 / -0


11 years ago
There was already a 500ms delay. I only pushed the delay into 100% safety margin, because between 500ms of timeout and 1/60 chance of misfire, i think the missiles deserve their sleep.
+0 / -0
I know, what i meant was that a proper workaround would've been cooler because it would be win on two fronts :)

But again, it's now better than it was before.
+0 / -0
The tech porn adept in me actually disagrees. I mean, code-wise yeah it could be nicer without the wait. But actually having the silo unable to build during launch makes physical sense.

(this is not a rationalization for the fix, i actually love units that work plausibly, don't clip their pieces, have little bits rotating for their actions, etc)
+0 / -0

11 years ago
Hm alright, i'll have to agree with that.
+0 / -0


11 years ago
OSHA wants to have a word with you about your workplace. Apparently loading heavy equipment in an area prone to attack by giant robots violates regulations.
+2 / -0