Bay 12 Games Forum

Please login or register.

Login with username, password and session length
Advanced search  
Pages: 1 [2]

Author Topic: First results from a borg-logic drop clock  (Read 4658 times)

Girlinhat

  • Bay Watcher
  • [PREFSTRING:large ears]
    • View Profile
Re: First results from a borg-logic drop clock
« Reply #15 on: March 16, 2012, 09:58:54 am »

Code: [Select]
#__ <-drop
#~#
#~#
#~#
#~#
#P+/# <- diagonal de-pressure, and ramp
#####
Put the pressure plate at the bottom of a column of water, like so.  In theory, any creature can fall from this and hit the ground stunned.  A good swimmer or a vampire will walk out unharmed.  Lesser creatures may drown, but should still trigger the plate.

Nil Eyeglazed

  • Bay Watcher
    • View Profile
Re: First results from a borg-logic drop clock
« Reply #16 on: March 16, 2012, 03:09:55 pm »

You can build a floating floodgate under a hatch and link both up to the same trigger.  The vamp will drop his stuff when the hatch opens, but neither he nor the item will fall until the floodgate opens, which I think has the 100 delay.

That's a cool idea, that I'm going to try to use for something, but I'd rather avoid it for this-- the idea is that an extra 100 ticks gives the vampire extra time to reach his drop destination, and if the hatch opens before then, then he won't make it onto the drop square in time.  (If you were to build a hydromechanical delay for the hatch, why not just use a fully hydromechanical system?)

Considering that vampires are pretty damn robust constitutionally, is there even a need to cushion them? I would think that a super-tough vampire would survive a brief fall.

Definitely a need.  The big delay for further testing has been that I've had an injured vamp at the bottom of a drop chute for several seasons.  I've been waiting for him to get better, not wanting to bring him to my hospital and risk his thirst.  I'm just now building a second drop chute for testing purposes, tired of waiting on his broken limbs.

How are you measuring ticks?

I pause the game and advance with the . key.  I use additional doors to make sure that I know exactly when signals are received.  When I say it takes 7 ticks for a dwarf to fall his first z-level, what I mean is that at 0, the hatch opens (as well as a nearby door); from 0-6, the dwarf hasn't fallen; and at 7, the dwarf is at the next z-level.

By the way, for anyone interested, I have a good hypothesis about why the first z-level takes an extra tick.  The game evaluates dwarves (maybe all creatures), then objects, in series.  What happens in tick 0 is that the dwarf is evaluated for falling BEFORE the hatch is opened, although both happen on that tick.  Because the dwarf is evaluated first, he doesn't fall, because the hatch doesn't open until later in that same tick.  Does that make sense?

Quote
Is the vampire taking injuries from his falls?  if so, is he healing?

The goal here is a consistent delay that can be used for a clock.  Any damage is unacceptable for my purposes.
Quote
Bridges.  Due to bridges seeming to fling objects in random directions, random distances, this is probably bad compared to retracting hatches/grates/bars.

They're fine, because the dwarf is constrained to a single square at time of drop-- there's no place to get flung to.

Put the pressure plate at the bottom of a column of water, like so.  In theory, any creature can fall from this and hit the ground stunned.  A good swimmer or a vampire will walk out unharmed.  Lesser creatures may drown, but should still trigger the plate.

I doubt that it's even necessary to depressurize it; in fact, submersion is a good method to get the vampire to return near reset position before it's even accessible, as they seem to walk in water to dry land, rather than spazzing out like regular, oxygen-dependent dwarves do.  Is that a surefire protection against damage?  I'll certainly add that to the testing queue (which, sorry, is going slow, kind of burning out on 34.x).  If it's surefire protection, and speed of falling through water remains constant, then it's probably better than the creature cushion for most purposes.
Logged
He he he.  Yeah, it almost looks done...  alas...  those who are in your teens, hold on until your twenties...  those in your twenties, your thirties...  others, cling to life as you are able...<P>It should be pretty fun though.

Girlinhat

  • Bay Watcher
  • [PREFSTRING:large ears]
    • View Profile
Re: First results from a borg-logic drop clock
« Reply #17 on: March 16, 2012, 03:31:23 pm »

It's surefire protection, but that's a side view.  If you don't depressurize it at the bottom then it will spill over that ramp.  The diagonal pressure gates are more for fortress protection and fluid dynamics rather than the actual falling.

Nil Eyeglazed

  • Bay Watcher
    • View Profile
Re: First results from a borg-logic drop clock
« Reply #18 on: March 16, 2012, 03:55:20 pm »

I think I have an application that people will actually find useful, that's not too complicated.

Two 100 on, 100 off hydromechanical repeaters.  The first is linked a floodgate, the second to a raising bridge.  The first repeater is started by lever, linked both to the activation gear of the repeater, and to a single z-level drop hatch.  The drop hatch pressure plate, a trigger once plate, is linked to the initialization gear of the second repeater.

The raising bridge and the floodgate give you a reliable live-fire bolt-recovery operation (possibly for cotton candy multiplication).  Because there are only 7 ticks during which there is a pathway through both the bridge and the floodgate, any bolts fired from further than 7 tiles away will hit the bridge and fall down.  7 ticks is sufficient time to acquire a target and shoot, at least in my experience.

The drop-based offset is the simple part of this design; the 100 on, 100 off repeater is the hard part.  2-step cyclic repeater gives you more like 199 on, 1 off.  Haven't tested it with a bridge, though; the refractory period of the bridge might work with it.  Using the drop to create a 7-tick offset is much simpler than using pumps and build order to create the offset.
Logged
He he he.  Yeah, it almost looks done...  alas...  those who are in your teens, hold on until your twenties...  those in your twenties, your thirties...  others, cling to life as you are able...<P>It should be pretty fun though.

Nil Eyeglazed

  • Bay Watcher
    • View Profile
Re: First results from a borg-logic drop clock
« Reply #19 on: March 17, 2012, 03:12:54 am »

Got a little bit more testing in.

The bolt recovery operation (set up with a drop-based delay) was a success.  A standard 2-cycle repeater works fine with refractory buildings like floodgates and bridges.  I'm left with a setup that should net me almost a bar of candy a day, once I mine some candy.  I consider this an improvement over the classic adamantine factory in terms of simplicity and automation.  Setting up a trigger-once plate was a pain (mechanics kept triggering on link jobs), and I ended up adjusting weight requirements and dropping a yak cow on it, instead of a dwarf.

Rebuilt my drop, rather than waiting for my vampire to recover (which he still hasn't).  Rebuilt as per original specification, sent a non-vampire to test.  No injury.  (EDIT: And drop times were consistent.)  Next test is with a bridge instead of a hatch; test after that is with forbidden doors adjacent to the drop (which I originally placed to let my mechanics set up a pressure plate without fear of crushing or dropping).  EDIT: Either modification could result in injury, which I find interesting, and unexpected.

Beyond that, still have to test submerged circuit and leashing zombies.  After that, it's time for a long-term longevity test, and after that, a synchronization test, which requires building a hydromechanical clock.

Doing a lot of work on build order at the same time.  Think I'm on to a grand unified theory of build order.  Few more tests in order, then maybe it's time for peer review (ie y'all), then a wiki page.

EDIT: Bridges were leading to injury.  Even with the exact same distance, drop a dwarf onto a cushion from a bridge, he gets hurt.  Drop him from a hatch, he's only stunned.  Who knows why.

Water consistently protects against injury.  Testing with 6 z-level fall, 5 z-levels of water.  The vampire doesn't even get stunned.  The problem is that the vampire is unwilling to path back to the surface, even using pressure tricks.  Trying to work around this right now.  Falling through water happens at the exact same rate, and pressure plates get set off on landing, same as ever.

Vampires will happily chain zombies, meaning that the critter cushion is viable and can last for forever.
« Last Edit: March 17, 2012, 11:37:34 pm by Nil Eyeglazed »
Logged
He he he.  Yeah, it almost looks done...  alas...  those who are in your teens, hold on until your twenties...  those in your twenties, your thirties...  others, cling to life as you are able...<P>It should be pretty fun though.
Pages: 1 [2]