Bay 12 Games Forum

Please login or register.

Login with username, password and session length
Advanced search  

Author Topic: Looking for a fail safe.  (Read 766 times)

jcochran

  • Bay Watcher
    • View Profile
Looking for a fail safe.
« on: August 27, 2013, 06:54:32 pm »

Doing a bit of experimentation with Forgotten Beast killing using minecarts. Found QuantumMenace's design in http://www.bay12forums.com/smf/index.php?topic=114394.0 as a nice start. However, that design causes the minecart to come to a full stop each corner and therefore the minecart really isn't going all that fast. It's fast enough to do a lot of damage to regular sized enemies, but it's a mere pinprick for a Forgotten Beast. I tried another design where I stretched into a long line and carved actual turns at each end (with either a wall or a bridge stationed to prevent derailment). Of course, in having actual track segments at each end, if the cart gets stopped there, it won't automatically resume motion. But that's easy enough to handle by having the FB enter the grinder past each end. But I've discovered a little problem. It seems that randomly, the cart will "jam" where ever there is a discontinuity in the wall. Can't predict when it will jam. Haven't really figured out a definite method of preventing jams, so I'm instead attempting to implement a 'fail safe' that detects if the minecart is actually moving. If the minecart stops moving, the failsafe should close a bridge sealing off the trap from the corridor leading into my fortress proper.

Unfortunate, I haven't figured out a way of detecting an extremely fast moving minecart. Seems that the pressure plates don't bother to trigger on something moving that fast. My tests have involved connecting three pressure switches in a row to hatch covers an seeing if any of the hatch covers open even momentarily. No joy.

So ... Does anyone have an idea on how to detect the passing of a minecart that is going at about 2.5 to 2.7 tiles per tick? And if the cart is rolling smoothly, it will pass that point again in less than 90 ticks or so later.

My initial attempts did manage to kill 3 FBs. But I'm a bit uncomfortable with the idea of the cart jamming and a passageway being left leading into my fortress with the hope of a drunken dwarf throwing a lever before the FB limps, slithers, or crawls into the fortress. Hence my desire for a fail safe.
Logged

Fayrik

  • Bay Watcher
    • View Profile
Re: Looking for a fail safe.
« Reply #1 on: August 27, 2013, 07:34:11 pm »

Even if you can't predict when the cart will jam, can you predict where it will jam?
It strikes me that the simplest possible solution is to reverse the state the cart is in apon detection.
Rather than try and detect it while it is moving, try detecting when it stops. With any luck, this situation will be made easier since the pressure plates will automatically ignore the cart when it's functioning properly.
Logged
So THIS is how migrations start.
"Hey, dude, there's this crazy bastard digging in the ground for stuff. Let's go watch."

Sutremaine

  • Bay Watcher
  • [ETHIC:ATROCITY: PERSONAL_MATTER]
    • View Profile
Re: Looking for a fail safe.
« Reply #2 on: August 27, 2013, 08:18:01 pm »

I have an idea.

Doors block water flow even from above, so you can make a small well with a door acting as a hatch at the bottom. Near the door is a water-sensitive pressure plate. When an FB comes along, it smashes the door and releases the water, allowing it to flow onto the pressure plate.

The pressure plate activates a minecart railgun, launching the heaviest minecart you can make into the FB. Hopefully it's enough to blow the FB apart, and the minecart carries on. It comes to rest on top of a pressure plate that closes off the access to the railgun, stopping something else from wandering in before the trap can be reset.

FB corpses have a variable weight, with 12,000L being about average. I'm not sure if I've even seen a 14kL corpse, but I'm pretty certain I've never seen a 15kL one.

Most of them don't need this treatment, and I just accidentally found a way of safely killing vapour-producing FBs using non-elite marksdwarves. Should work on deadly dusters too, but I haven't had a chance to test that yet. All you need is a diagonal fortification -- bolts can get through, but the vapour can't.
Logged
I am trying to make chickens lay bees as eggs. So far it only produces a single "Tame Small Creature" when a hen lays bees.
Honestly at the time, I didn't see what could go wrong with crowding 80 military Dwarves into a small room with a necromancer for the purpose of making bacon.

fricy

  • Bay Watcher
  • [DFHACK:ZEALOT]
    • View Profile
Re: Looking for a fail safe.
« Reply #3 on: August 28, 2013, 04:36:24 am »

I believe the pressure plate doesn't register the minecart because it's skipping over it. It's the same with track stops, if the cart is going fast enough it will ignore most of them. Predicting where a minecart will touch down at the speed is hard, if you build 3 plates in a row I think one of them will pick up a signal, or you can build a plate on the tile after a ramp, I know the minecart will touch that tile. (untested with p.plates, but track stops won't work there above 72k speed!)

jcochran

  • Bay Watcher
    • View Profile
Re: Looking for a fail safe.
« Reply #4 on: August 28, 2013, 08:35:49 am »

Well, my tests have had three pressure plates in a row. The test setup had each of the pressure plates connected to their own individual hatch cover and all of them also connected to a single common hatch cover to see if any of 'em would open. But no joy. Although the suggestion of having a plate at a known jam location is definitely a good one. I think that will have to be the route I take. Although in doing so, the probability of the cart being stopped increases (the jam points are where I have a drawbridge acting as a wall segment. I currently have three such points. One of them is at the extreme end for starting/stopping the loop (when the bridge is up, the loop is running. When the bridge opens, the cart derails, goes past the bridge, up a ramp, and loops back stopping at a dead end with a retracting bridge underneath it. Both drawbridge and retracting bridge connected to same lever. The other 2 jam points are the entrance for the FB and the exit leading into my fortress (the lure passage) so that the FB will attempt to path through the grinder.

On my earlier no failsafe version, a gold minecart travelling at approx 2 to 2.5 tiles per tick does a fair amount of damage to a FB, but it's definitely not a kill of said FB. But since it's on a loop, it just goes around and hits again (but slower. Not enough distance to build up to full speed). Eventually the FB is bludgeoned to death.
Logged

fricy

  • Bay Watcher
  • [DFHACK:ZEALOT]
    • View Profile
Re: Looking for a fail safe.
« Reply #5 on: August 28, 2013, 02:15:06 pm »

Yeah, sorry, I misread first that you already tried 3 plates in a row.Anyways, this made me curious, so I fired up my minecart testing course.

Results at max (265k) speed:


I was mistaken, the cart won't land on the first tile after a ramp, so no point building a plate there. But you can clearly see from this picture that high speed plate activation is possible, you can get a signal on the 3rd, 6th and 8th tile at this speed.