View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0008441 | Dwarf Fortress | Dwarf Mode -- Environment | public | 2014-10-18 00:40 | 2014-11-30 08:02 |
| Reporter | PatrikLundell | Assigned To | lethosor | ||
| Priority | normal | Severity | minor | Reproducibility | have not tried |
| Status | assigned | Resolution | open | ||
| Platform | PC | OS | Windows | OS Version | 7 |
| Product Version | 0.40.13 | ||||
| Summary | 0008441: Quicksand water slope tiles | ||||
| Description | For some reason a caravan guard entered a watery slope 6/7-7/7 depth, and never managed to get up again. While his caravan was trading I was attacked by a titan that found the guard to be a suitable target, killed him, and in turn is stuck in the adjacent water slope tile. Both tiles are part of an original murky pool on the map. I have built a bridge two tiles off to one side, and I have a fortress wall two tiles off to the other side. | ||||
| Additional Information | Using DF Starter Pack with Dwarf Therapist, DF Announcement Filter 1.1, and DFHack Pure Bugfixes+Performance Tweaks. I've got a save that I can upload, if it is of interest. | ||||
| Tags | No tags attached. | ||||
|
|
Are the stuck creatures drowning? Is the path out of the pool blocked? If not, a save uploaded to http://dffd.wimbli.com/ would be helpful. |
|
|
The guard was not drowning, and the titan isn't drowning.There are several ways out of the water; one directly up onto land, and others via adjacent slopes. It turned out the titan wasn't completely stuck: if a victim came close enough, it could attack with its fireball attack and get out of the water. However, its fireball attack seems weird, as it sort of seems to ride its own fireballs when attacking, as the speed when attacking is greatly larger than when moving normally, so I wouldn't count getting unstuck as a result of attacking as a proof that it isn't stuck for normal movement. Leaving the gates to the fortress open is not sufficient to get the monster to move, nor was a caravan leaving or immigrants arriving (again, provided they stayed out of a fairly short attack detection distance). Save uploaded as http://dffd.wimbli.com/file.php?id=9941. |
|
|
Sounds exactly like the same type of problem as 0003700, especially the FB/titan. The one with the guard is probably closer to 0008348, but it's still the same pathing problem. |
|
|
Well, the titan has now been replaced by 4 goblin invaders... Why not use my perfectly good bridge one tile to the side? The remaining invaders took the bridge only when the water was full... I'd say they take the shortest route. It might be possible to improve the behavior if water tiles are given a higher pathing cost than land, unless you're a perfect swimmer. I'd say there are at least two issues: one is getting into the water in the first place, and the other is the reluctance to leave it. Further info: When I approached the bathing goblins they leapt out of the water when I came quite close (5 tiles, 3?). Thus, I would guess it's at least partially a perception issue: When down on the watery slope they may no longer perceive their target, and they probably have a goldfish memory, so they don't remember the juicy dwarves on the other side of the open drawbridge. They have no problem detecting when I lower the drawbridge on the remote side of the fortress (when out of the water), though, and forgotten beasts do not seem to have any problem with the caverns twisting in three dimensions, so it's not purely a z-level difference issue. |
|
|
I haven't observed the behavior of FBs in the caverns yet in 40.xx, but in 34.11, while the FBs would beeline for the fort entrance (or any available structure that it wants to knock over), the instant (more like after a few tiles, but pretty much instantly) they went into water, they would forget where the entrance was and enter wander mode. I've had cases then where the FB would refuse to move from a particular spot or something and they would even return to said spot if I did manage to make them move. |
|
|
I haven't seen any water in the caverns, and when FBs have showed up I have immediately fled and sealed the door, as I don't think I'm ready to take them on, but they have went directly for my entrance (or possibly the dwarves running towards the entrance). When the titan HAS moved out of the water (triggered by goblin invaders), it has charged my fortress after eliminating the nuisances. Yet another observation, although I cannot swear it's correct: I believe I've seen goblin invaders pass through drained murky pools (1/7-2/7 depth due to rain) without any problems, so I'd guess swimming depth water is required to induce amnesia. |
|
|
Another case of a similar but inverted FB behavior uploaded as http://dffd.wimbli.com/file.php?id=10153. In this case a flying FB appears in the upper Caverns, leaves them via surface Caves (but no spores were released until I breached the caverns), kills off one or two wild geese, and then hovers in the air 3 z levels above the ground approximately straight above the center of my fortress. When I saw it in the Caves I immediately turtled, leaving a tunnel with two doors at the ends and repeating spike traps in between as the only entrance. I even unlocked the outer door in the vain hope it would enter without destroying the door. A naked mole dog keeps moving towards the FB, panicks, runs away, and then starts moving in the FB direction again, repeating the cycle. Opening the drawbridges fully does not get it to move. Moving a 4 dwarf squad to an area quite near to the FB causes it to attack (only 2 shows up, the other two are busy). By restationing the squad to inside the fortress I can lure the FB away (they can't attack it as it is flying). After it kills the two dwarves it settles beside the door to the tunnel. I presume it it trying to destroy the door, but 10 minutes or so later I can see no damage to the door. I'm using PyLNP with DF 0.40.16, Annonucement Filter, Dwarf Therapist and DFHack with Workflow - production targets, Performance Tweaks, and Pure Bugfixes. |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2014-10-18 00:40 | PatrikLundell | New Issue | |
| 2014-10-18 07:13 | lethosor | Note Added: 0030658 | |
| 2014-10-18 07:13 | lethosor | Assigned To | => lethosor |
| 2014-10-18 07:13 | lethosor | Status | new => feedback |
| 2014-10-18 14:11 | PatrikLundell | Note Added: 0030664 | |
| 2014-10-18 14:11 | PatrikLundell | Status | feedback => assigned |
| 2014-10-18 22:09 | smjjames | Note Added: 0030668 | |
| 2014-10-19 09:51 | lethosor | Additional Information Updated | |
| 2014-10-19 12:27 | PatrikLundell | Note Added: 0030674 | |
| 2014-10-19 13:23 | PatrikLundell | Note Edited: 0030674 | |
| 2014-10-19 15:55 | smjjames | Note Added: 0030675 | |
| 2014-10-19 18:31 | smjjames | Note Edited: 0030675 | |
| 2014-10-19 23:43 | PatrikLundell | Note Added: 0030688 | |
| 2014-11-30 08:02 | PatrikLundell | Note Added: 0031232 |
