Monsters occupying multiple squares

Packbat · 2015-01-26T03:39:21.000Z

I feel as if the vehicle system could be repurposed to make multiple-square monsters. Has this been suggested? Is it planned? How hard would it be?

kevin.granade · 2015-01-26T04:57:10.000Z

The AI would be a pain to deal with, but not impossible. I’m not sure leveraging the vehicle system would be worthwhile unless you wanted the monster to literally be a vehicle, like a wheeled or tracked robot.

EkarusRyndren · 2015-01-26T10:58:29.000Z

For whatever reason I viewed this as multiple monsters in one square, which could make things interesting, but not much could take up one square…

Muaddib · 2015-01-26T11:51:11.000Z

EkarusRyndren:

For whatever reason I viewed this as multiple monsters in one square, which could make things interesting, but not much could take up one square…

Rats. Insects. Imagine how much harder rat or spider filled basements would be if they were too small to bottleneck?

ramesesniblet · 2015-01-26T15:20:54.000Z

Kevin Granade:

unless you wanted the monster to literally be a vehicle, like a wheeled or tracked robot.

That would be pretty cool actually. Actual tank drones, with all the weapons mounted on vehicle sections that drive around. It’d be more interesting to fight imo (taking out the weapons, dodging it trying to squish you, hoping it runs out of juice, maybe even trying to board it to rip out the AI circuits), and slightly more possible to escape from via chokepoints. Vehicle drones with AI would make the robot side of things very interesting indeed, with the potential to build your own Z seeking tank drones. Mechanics skill gain would need a nerf though… strip down a schoolbus and suddenly you have the skill to build a remote controlled Deathstar. Also I imagine pathing for AI vehicles would be nightmarish.

EkarusRyndren · 2015-01-26T17:23:48.000Z

ramesesniblet:

[quote=“Kevin Granade, post:2, topic:8756”]unless you wanted the monster to literally be a vehicle, like a wheeled or tracked robot.

That would be pretty cool actually. Actual tank drones, with all the weapons mounted on vehicle sections that drive around. It’d be more interesting to fight imo (taking out the weapons, dodging it trying to squish you, hoping it runs out of juice, maybe even trying to board it to rip out the AI circuits), and slightly more possible to escape from via chokepoints. Vehicle drones with AI would make the robot side of things very interesting indeed, with the potential to build your own Z seeking tank drones. Mechanics skill gain would need a nerf though… strip down a schoolbus and suddenly you have the skill to build a remote controlled Deathstar. Also I imagine pathing for AI vehicles would be nightmarish.[/quote]

Tank drones aren’t that big, they’re not full-fledged tanks. That said it’d be more of an electrical feat than a mechanical one anyway, probably both. Build the hull, but it’s useless without a computer, build the computer, but it’s useless without a hull.

Packbat · 2015-01-27T01:03:07.000Z

I agree that it takes both Mechanics and Computers, but I think Computers is the real bottleneck because of the machine-vision problem. Based on what I know about SICP, and given that in Cataclysm:DDA it can take your computer skill from 5 to 8, I think it would take a minimum of Computers 7 to install a rudimentary AI on your would-be Bolo, and at least Computers 9 to write all the AI modules.

Also, agreed on the swarms-in-one-space, but I think you’d want to add two new key-controls if so: “move through instead of attacking” and “attack instead of moving through”.

Edit:

Kevin Granade:

The AI would be a pain to deal with, but not impossible. I’m not sure leveraging the vehicle system would be worthwhile unless you wanted the monster to literally be a vehicle, like a wheeled or tracked robot.

Good point - I was thinking about the way the game supports rotation of multiple-square objects more than I was about how hard it would be to use the code.

Coolthulhu · 2015-01-27T03:06:53.000Z

ramesesniblet:

Also I imagine pathing for AI vehicles would be nightmarish.

Yeah, DDA vehicles have almost every possible thing that makes pathfinding hard:

[ul][li]Multiple tiles[/li]
[li]Orientation-dependent multiple tiles[/li]
[li]Limited turning capability[/li]
[li]Limited acceleration and deceleration[/li]
[li]Inability to turn in place[/li][/ul]

And that’s even before we consider performance.

Would be doable, just that it’s mountains of hard work.

StopSignal · 2015-01-27T04:43:16.000Z

Muaddib:

[quote=“EkarusRyndren, post:3, topic:8756”]For whatever reason I viewed this as multiple monsters in one square, which could make things interesting, but not much could take up one square…

Rats. Insects. Imagine how much harder rat or spider filled basements would be if they were too small to bottleneck?[/quote]

Paper and lighter, please. Hurry.

Llamageddon · 2015-01-27T04:52:55.000Z

Kevin Granade:

The AI would be a pain to deal with, but not impossible. I’m not sure leveraging the vehicle system would be worthwhile unless you wanted the monster to literally be a vehicle, like a wheeled or tracked robot.

If rotation wasn’t possible for said monsters, it’d probably be enough to generate a secondary movement graph, and whether that would be heavyweight is questionable. Probably not worth the effort, either way.

And personally I think that it would be worthwhile to try and leverage the vehicle system for more stuff. Perhaps not monsters, but furniture, maybe items. That’d make it easier to do electricity stuff in the (probably very distant) future, and I wouldn’t mind being able to disassemble/reassemble/modify furniture. Then again, not like there is a lot that would benefit from it currently. Recycler, garbage containers, maybe fridges/ovens.

kevin.granade · 2015-01-27T05:12:58.000Z

Yea, I’m familiar with enlarging obstacles to simplify pathing, but as you point out, rotation kind of hoses that. If we do NPCs driving vehicles (nightmarish) we’d probably leverage the same code to make self-driving tanks or similar, but both of those are stupid hard to do, and I’m not sure if the payoff is worthwhile.

More likely is large non-rotating (or technically infinitely rotating but symmetric) monsters, or snake-like monsters, or amorphous colony monsters (that’s what I was originally planning for the brain blob, but making all the blobs stay adjacent proved difficult, so I went for a different implementation)

Llamageddon:

And personally I think that it would be worthwhile to try and leverage the vehicle system for more stuff. Perhaps not monsters, but furniture, maybe items. That’d make it easier to do electricity stuff in the (probably very distant) future, and I wouldn’t mind being able to disassemble/reassemble/modify furniture.

That’s the plan for making electrically active furniture, making them vehicles under the hood, but disallowing interactions with the vehicle menus and substituting a custom interaction menu.