The Castle Doctrine Forums

Discuss the massively-multiplayer home defense game.

You are not logged in.

#1 2013-06-15 07:29:54

FunnyMan
Member
Registered: 2013-06-15
Posts: 8

Encouraging desirable player behaviour

While I certainly applaud the courage and effort it takes to drop back and reexamine the core of Castle Doctrine's game design, I feel that there's still a fundamental mistake that needs to be corrected.  Namely, the root of CD's major problems is that the reward structure encourages building houses that are as close to unsolvable as possible.

Look at it from the homeowner's point of view.  A successful house is one that nobody can rob, because it keeps your money safe.  The ideal result of a robbery isn't so much that the robber dies to your house, but that the robber leaves and never returns.  Dying is a nice perk, but dead robbers have a nasty tendency to reincarnate and try again, while a robber who is bored or frustrated enough to ignore your house will leave it alone for good.  Needless to say, this style of house is not fun for robbers, as evidenced by the various styles of uninteresting top houses.

As long as the homeowner's goal runs against the game design's goal, the homeowners are going to win in the long run.  No matter what changes Jason makes to the setup, people will tend towards optimal house designs, and that means they're going to make houses that are as frustrating as possible to rob.  v6 had combination locks.  v7/8 had ridiculous puzzles.  v9 is still too new to have settled to an optimum, but I guarantee it's going to be something that people don't like.  None of the changes in v9 altered the core problem: homeowners are optimizing for houses that aren't fun to rob.

I won't claim to have found a perfect solution, or even a great one, but I have an idea for somewhere to start:

  1. Thematically, establish a Bureau of Population Reduction, who enacted the Castle Doctrine in response to food shortages.
    --  This gives us a convenient excuse for mechanics that otherwise seem a bit odd: the ultimate goal of CD is to get people to kill themselves, because that's what the BPR needs to do.  That is a good thing for the game design, because it lets us embrace a classic roguelike premise, best formulated (by the Dwarf Fortress community) as Losing is Fun.  Far more so than being bored away from a house, at least.

  2. When a player dies to your house, half of the loot they dropped (rounded down) is automatically collected by the BPR, who store it securely until you have a chance to collect it from them.  (i.e. until you return home)
    --  This gives players a "nest egg" that survives being robbed, reducing the penalty for "letting the robber win" while still increasing the house value with each failure.

  3. When a player ignores your house, they file a complaint with the BPR, who fine you 1/10th of the amount they have stored (rounded up).
    --  By penalizing the nest egg, we discourage houses that robbers ignore.  That kind of house is, nearly by definition, exactly the kind of house we don't want to have.

  4. When a player successfully robs your house without using any tools, an additional 1/10th of the amount stolen is deposited into the BPR account of both players, as a bonus for taking/encouraging extra risk.
    --  Since we reward the homeowner, they want the thieves who are determined enough to solve the house to be able to solve it "properly".  And since we reward the thief, they want to solve it that way, too.

The end result of all of that should be that homeowners optimize for an ideal roguelike experience: Very dangerous, fun to attempt, and difficult but not impossible/unreasonable.  It probably needs some tweaking, but as I said, it's a starting point.  The most important thing is that it's a major step towards reversing the current situation, making the homeoners work towards Jason's goal for the game, not against it.

(This post was heavily based on one I made over on reddit that I felt needed to be mirrored over here for greater attention.  Incidentally, the "anti-spammer" question here doesn't reflect the v9 prices yet.)

Offline

#2 2013-06-15 10:07:09

bey bey
Member
Registered: 2013-04-20
Posts: 386

Re: Encouraging desirable player behaviour

Weeeell, I think the main point is that Jason also wants a feeling of violation and danger to occur in the game. I wouldn't wonder if he weren't happier with a situation were some people protect their houses with all they have and somebody else just cuts through all that is within it since that's deeper in the emotional sense. I think there is no objective of the robber and the robbed "both being happy" in the plan underlying TCD. I do agree that a lot of the frustration with the game comes from preserving this while still keeping it playable...


In fact you can be batman.
(if he robbed houses and murdered families.)
- Dalleck

Offline

#3 2013-06-15 10:24:09

Matrix
Member
Registered: 2013-04-06
Posts: 137

Re: Encouraging desirable player behaviour

bey bey wrote:

I do agree that a lot of the frustration with the game comes from preserving this while still keeping it playable...

Yes smile

If we could just drop all restrictions then it would be too easy... you would just take the mechanics from another game that is already released and working and build the burglary theme around it.

The idea is to come up with suggestions and solutions that could potentially work by keeping some restrictions in the game. Of course some restrictions might be changed or dropped eventually, but not before a few iterations are attempted to actually solve the problem at hand.

Last edited by Matrix (2013-06-15 10:25:24)

Offline

#4 2013-06-15 10:27:08

FunnyMan
Member
Registered: 2013-06-15
Posts: 8

Re: Encouraging desirable player behaviour

bey bey wrote:

Weeeell, I think the main point is that Jason also wants a feeling of violation and danger to occur in the game.

I don't disagree with that, but any time you pit your users against the game design, the users are going to win.  He can shuffle parameters around all he likes, but either there won't be any viable houses (as currently seems to be the case), or the most successful houses will be the ones nearest to optimal.  As long as that optimum is a house that nobody can (or wants to) rob, people are going to keep finding new ways to build houses that nobody can (or wants to) rob.

I don't claim to have a monopoly on ways to change what the optimum house is.  Hell, I don't even claim that my method would work the way I intend it to.  All I did was give an example of the kind of change we need.

It's unreasonable to expect the players to make houses that create "a tactical, player-generated, Roguelike, every-cycling arms race" when you reward them for making houses that are effectively unsolvable.

Offline

#5 2013-06-15 10:40:28

bey bey
Member
Registered: 2013-04-20
Posts: 386

Re: Encouraging desirable player behaviour

I do keep thinking along similar lines as you do, and I somehow think that the user-base is not unified on that. I think that secretly, quite a few of us (who got it) were in love with v7/8 that made houses robbable, but only to those who got it. I think this experience of a close circle of people admiring and solving each others stuff is not coming back (and it's alright since few people "got" the whole thing and it was a love-hate-thing to me). Most people like to play instead of think about a map, and it's moving into that direction again.

I'd just say that the essential feeling I get from the game, especially when having a very smart house, is a mixture of the enjoyment of having people play it and the ultimate fear of people breaking / destroying it (or even my family). This mixture would be lost to a more harmonic gameplay where everybody is rewarded.


In fact you can be batman.
(if he robbed houses and murdered families.)
- Dalleck

Offline

#6 2013-06-15 11:01:47

FunnyMan
Member
Registered: 2013-06-15
Posts: 8

Re: Encouraging desirable player behaviour

bey bey wrote:

I'd just say that the essential feeling I get from the game, especially when having a very smart house, is a mixture of the enjoyment of having people play it and the ultimate fear of people breaking / destroying it (or even my family). This mixture would be lost to a more harmonic gameplay where everybody is rewarded.

Well, look at it this way.  There are three final outcomes to any series of robbery attempts:

1. Robber dies.
2. Robber stops visiting the house.
3. Robber reaches vault.

The balance of death vs success is the central game dynamic.  Ideally, we want lots of death, a little robbery, and no giving up at all.  How houses are designed plays a major role in the balance of the three results.  Too little difficulty, and the robber always wins.  Design the house well, and you get a mix of deaths and robberies.  Too much obvious difficulty, and most people just give up.

There's nothing in that that requires the homeowner to want people to rob him.  Even with the system I described, he always prefers a dead robber to a successful one.  There's still plenty of room to make thefts scarier once we've got a basic setup that encourages the kind of house design we want.  Of all the changes I proposed, #3 is the one that most directly attacks the problem, but I didn't propose it alone because it's also the most prone to abuse if we isolate it form the others.

Offline

#7 2013-06-15 11:07:54

Matrix
Member
Registered: 2013-04-06
Posts: 137

Re: Encouraging desirable player behaviour

FunnyMan wrote:

It's unreasonable to expect the players to make houses that create "a tactical, player-generated, Roguelike, every-cycling arms race" when you reward them for making houses that are effectively unsolvable.

But technically unlimited backpack doesn't reward houses that are effectively unsolvable. Because with enough tools any design can fall.

So the optimal house is a house that is an elaborate trap that is too hard to figure out (hidden logic) and at the same time too expensive to break through (to get to the logic or vault). But if you place enough money into the vault then it's not too expensive anymore and instead it becomes "worth the risk".

In an ideal world this game would be about house owners changing and expanding their house over time to raise that "worth it" threshold while still keeping enough money in the vault to remain a good target (after all you get more money if people fail and leave some tools behind). At the same time robbers would be able to pick targets they think they can handle, sometimes failing and sometimes succeeding. An owner who lost his money would then just go rob others to fill his vault again.

But in the real world (well the virtual TCD world tongue) this might be impossible to achieve without certain limitations and "forced" mechanics. It might be also impossible to achieve without a change to the reward system, I agree, that's a possibility.

Offline

#8 2013-06-15 11:25:19

FunnyMan
Member
Registered: 2013-06-15
Posts: 8

Re: Encouraging desirable player behaviour

Matrix wrote:

But technically unlimited backpack doesn't reward houses that are effectively unsolvable. Because with enough tools any design can fall.

Right now, we seem to be in a degenerate "no viable houses" situation.  Talking about "the optimum house" is pretty meaningless when all houses have the durability of a mandala on a windy day.

That said, I agree with your analysis, I just count the kind of house you describe as "effectively unsolvable".  Instead of a combination lock that can be bruteforced in a couple dozen years (like in v6), we've got a combination lock that can be bruteforced in a couple dozen years or dismantled with an unreasonable number of tools.  Personally, I don't consider that a substantial improvement.

All that an unlimited backpack does is add a second type of brute force.  That doesn't significantly change what's optimal, it just adds a maximum storage amount to optimal houses.  It treats the symptom, not the cause.

Offline

#9 2013-06-15 13:29:52

Matrix
Member
Registered: 2013-04-06
Posts: 137

Re: Encouraging desirable player behaviour

FunnyMan wrote:
Matrix wrote:

But technically unlimited backpack doesn't reward houses that are effectively unsolvable. Because with enough tools any design can fall.

Right now, we seem to be in a degenerate "no viable houses" situation.  Talking about "the optimum house" is pretty meaningless when all houses have the durability of a mandala on a windy day.

Yeah but let's talk about the core problem and not the v9 state. The v9 state is obviously unplayable at the moment.

FunnyMan wrote:

That said, I agree with your analysis, I just count the kind of house you describe as "effectively unsolvable".  Instead of a combination lock that can be bruteforced in a couple dozen years (like in v6), we've got a combination lock that can be bruteforced in a couple dozen years or dismantled with an unreasonable number of tools.  Personally, I don't consider that a substantial improvement.

...or dismantled is the important part here. And that's what we want -- tools that can be used to dismantle things so that we can break into a house/vault. In v5 tools in optimal houses were just obsolete, while in v8 they had practical use only after you understood the idea behind the trap, and even this could be argued upon, because a real optimal house just makes any tool useless, i.e. it allows no real shortcuts by using tools.

FunnyMan wrote:

All that an unlimited backpack does is add a second type of brute force.  That doesn't significantly change what's optimal, it just adds a maximum storage amount to optimal houses.  It treats the symptom, not the cause.

I agree with most those statements but at the same time I think that the second type of bruteforce that the unlimited backpack provides is the "playable" kind of bruteforce. Technically it's also a different kind of bruteforce, since you are not searching the problem space (i.e. the puzzle solution space), but instead you are just searching the real house space (which is much more limited), with the ability to mix both, if you think that you can save time or money by doing so. Also searching the house space requires only basic game knowledge (what tools do to different house tiles), while searching the solution space requires advanced (sometimes even complete) game knowledge. And, since logic gates are part of the game, it means that knowledge from other fields is practically required in order to understand the whole puzzle.

FunnyMan wrote:

That doesn't significantly change what's optimal...

This is the part I don't agree with. So how exactly do we define the "optimal" house? When talking about that we would have to agree on what are we trying to optimize.
- Minimizing house cost while maximizing total tool cost to reach the vault while not knowing anything about the house?
- Minimizing house cost while maximizing tool cost to reach the vault when knowing the vault location in advance?
- Minimizing house cost while maximizing tool cost to secure the house escape route?
- Minimizing house cost while maximizing tool cost to access all relevant house defense logic areas?
etc.

And even if we agree on one of those (it's already non trivial to pick the best one, but it should be possible since game mechanics and costs are known), it's not a trivial thing to build an optimal house, because you would need to do it in stages. You would have to make sub-optimal houses until you accumulate enough money to change the house into an optimal one. And by definition those sub-optimal houses are the ones that are worth being robbed. And if you manage to reach that optimal house state you could still be robbed by someone who made a net loss out of it, but it doesn't matter because he probably robbed some other houses to get that extra money.

All these optimizations were trivial in v5-v8; you just ruled tools out with Electric grids and 9+ walls. Of course in v8 you still had to build an obfuscated system behind that, but no one had to even think about the optimal house problem, since it was right there in front of us right from the start. I am pretty sure that it won't be that easy with unlimited tools and probably that's part of the reason why people think it won't work. We (me included) don't know what's the best house design to aim for. While that's exciting for some, it makes others think they don't have control and maybe even scares them.

Offline

#10 2013-06-15 20:27:55

dalleck
Member
Registered: 2013-04-13
Posts: 250

Re: Encouraging desirable player behaviour

In my first minutes of v9 I saw the future and it was good.  By this, I mean there was a brief moment where the game was functioning as I believe Jason intended.

I spent my money building a simple house design, which left me with $0.

I scouted some houses.  Hmmm.. one had a mechanism that released a few dogs.. if only I had some meat.  I needed a hundred dollars or so.  I found a house with one dog in it protecting the entrance.  Ok, so I just need 1 dog meat.  Then I found an easy prey which had $100 in it.  I stole enough to buy a couple of meats, robbed the house with one dog, made a few bux, bought some more meat.  Robbed the house with a pack of dogs, now I had some more money.

I then expanded my house.  Had a few robbers leave tools, sold them, and then expanded my house more.

So slowly I was making a more elaborate deathtrap, enticing more people in, made it more difficult for robbers to use fewer tools to get through my maze and I enjoyed myself thoroughly.

Shortly after this I died testing my house and then we had this vandalism problem.

But I saw the future.  And it was good.

=======

I don't believe we need any elaborate solutions to this puppy.  Houses will be houses.  People can make what they want.  If you have a tough house with $10,000 in your vault then someone with $5,000 is going to try to break in.  Maybe they have to spend all their money to buy tools to break you, but no matter how difficult you have made your house, you are going to go down.

If you have made something uninteresting to people, then your house will just be ignored.  Sure you could accrue lots of money and stay at the top of the ladder.  Does this make you 'win'?  No.  Winning this game is just playing it.  Everyone is actually a loser here.

The only problem we have at this moment is vandalism through starting cash.


The rich aren't safe. Nobody is safe. -jere                   ...but the smell wafts out from the pit, obviously. - Jason Rohrer

And the more dickish they are, the more I feel like beating a house to destruction after finally figuring it out. -bey bey

Offline

#11 2013-06-15 23:51:08

Matrix
Member
Registered: 2013-04-06
Posts: 137

Re: Encouraging desirable player behaviour

dalleck wrote:

But I saw the future.  And it was good.

=======

I don't believe we need any elaborate solutions to this puppy.  Houses will be houses.  People can make what they want.  If you have a tough house with $10,000 in your vault then someone with $5,000 is going to try to break in.  Maybe they have to spend all their money to buy tools to break you, but no matter how difficult you have made your house, you are going to go down.

If you have made something uninteresting to people, then your house will just be ignored.  Sure you could accrue lots of money and stay at the top of the ladder.  Does this make you 'win'?  No.  Winning this game is just playing it.  Everyone is actually a loser here.

Exactly!

dalleck wrote:

The only problem we have at this moment is vandalism through starting cash.

Are you sure that vandalism through robbed money is not a problem? I am playing the game non stop now and yeah there is no vandalism at the moment, but I wouldn't just leave that loophole in there.

Offline

#12 2013-06-16 22:45:27

jasonrohrer
Administrator
Registered: 2013-04-01
Posts: 1,231

Re: Encouraging desirable player behaviour

What counts as vandalism?  Trashing someone's house when there's nothing left to take from the vault?

I don't think there's a way to automatically detect it otherwise (like if you trash a bunch of stuff AND reach the vault and take something).

Vandalism is certainly very expensive right now.... so... getting enough money to do it would require DAYS of work on the part of the perpetrator.

There was a lot of it right after the v9 transition, when the top houses fell and suddenly there were huge money and tool balls passing around (I just got 300 saws?  What am I going to do with these?).

Offline

#13 2013-06-17 05:28:48

dalleck
Member
Registered: 2013-04-13
Posts: 250

Re: Encouraging desirable player behaviour

jasonrohrer wrote:

What counts as vandalism?  Trashing someone's house when there's nothing left to take from the vault?

I don't think there's a way to automatically detect it otherwise (like if you trash a bunch of stuff AND reach the vault and take something).

Vandalism is certainly very expensive right now.... so... getting enough money to do it would require DAYS of work on the part of the perpetrator.

There was a lot of it right after the v9 transition, when the top houses fell and suddenly there were huge money and tool balls passing around (I just got 300 saws?  What am I going to do with these?).

I wouldn't want to punish people for behaving in this way, because of course it is a valid way to play the game.  However, the model of $2000 beginning cash & unlimited tool-set is ripe for abuse.  Take for example, double accounts (or friends) and you have an instant bump up to $10,000+.  Now you have a player who can take down any house at all, and then burn it to the ground and piss on the ashes in the process.

Prior to v9, starting cash was not unbalanced because players would actually struggle to spend the entire $2000 (or more) on tools.  Now it is all too easily abused.

Just saying.  I think something needs to be put in place to curb it or players will continue to abuse it.

Edit: Oh, and another exploit, increasing the value of a house by buying tools and dying in it over and over.

Last edited by dalleck (2013-06-17 06:20:04)


The rich aren't safe. Nobody is safe. -jere                   ...but the smell wafts out from the pit, obviously. - Jason Rohrer

And the more dickish they are, the more I feel like beating a house to destruction after finally figuring it out. -bey bey

Offline

#14 2013-06-17 06:16:35

bey bey
Member
Registered: 2013-04-20
Posts: 386

Re: Encouraging desirable player behaviour

Yup. The current build has people running all over the map burning down everything in sight just for kicks. The ruins of most houses together with the ugliness and practical unplayability of the top house testify to the issue that so far, nothing much has been achieved...

Probably the biggest step backwards is that the family has now become the biggest liability. People exploit them to cause immense damage AND their mere existence makes you become a target even quicker. I cherished my wife and kids in v8, in v9 the first service somebody could do me is to kill them off so nobody can use them to raze my house...


In fact you can be batman.
(if he robbed houses and murdered families.)
- Dalleck

Offline

#15 2013-06-17 12:17:25

jasonrohrer
Administrator
Registered: 2013-04-01
Posts: 1,231

Re: Encouraging desirable player behaviour

Ah... I just added a feature where as admin, I can search for tapes where YOU were the victim.

I got to watch the trials and tribulations of Kevin Troy Trujillo (aka, bey bey) and a house under siege.

What I've seen so far is pretty good.... people (or one person) struggling to figure out a hard house over the course of many scouting trips, and figuring out the best places to "save damage" on those trips through the killing of family members.

Clearly, though, coming in with 20 crowbars to kill 20 pitbulls seems over-powered.  Especially since that's $4000 worth of pitbulls killed!  In tight corridors, crowbars work as well as guns, because the dogs have limited mobility.  That's how they reached your family...

After that, though, there was a lot of exploring and precision-placed permanent damage.  One guy came in with 3 ladders and laid them, then killed family and died.  But the next guy, Miller, the successful one, came in with 6 ladders to finish the job for $12715.

But the interesting thing?  The guy who dropped 3 ladders was a different player from the one who succeeded with 6 ladders later!    Where did he get the 6 ladders?  Maybe from die-and-drop?  NO, actually, he's been alive forever, he's currently still alive as the top house!  So, he got those 6 ladders from hard, rung-by-rung work.

After that, Miller came back for the tools that he dropped, destroying nothing, but he didn't return after that.

One person later came back and wandered around with tools, but died in the house.  Another guy came in with 2 guns and died.

After that, you robbed a bunch of easy houses, and fixed up your house slowly.  One guy came into the fixed-up house, but left.

Finally, a bunch of people came in with full tool sets to try breaking your house.  None succeeded.

And that's the end of the story.  Did you die testing your own house, or suicide?

Anyway, except for all those dogs killed so trivially (which meant that you couldn't really protect your family), it seemed pretty close to what I'm envisioning for what it will be like to break a hard house.

Though.... why have the family so close to the front door?  Why not reserver part of your house for a maze, with dogs placed here and there, where it's hard to even FIND the family?  Or spread them out, so that finding one doesn't mean you've found the others. 

Also, there's this mechanism to prevent crowbars from being used on dogs:

9070350518_f4f79251f5_o.png

It ensures that the dog is always one tile away from you, from the moment you come around the corner, all the way out of the house.  Gun, yes.  Drugged meat, yes.  Crowbar, no. 

Though I realize that with meat so cheap, you could carry both meat and crowbars to bypass this device----I should change it in v10 so that sleeping dogs can't be clubbed or woken up, so drugged meat will be much more tactical.  Maybe THAT would be the fix for crowbars, other than simply jacking their price way up?  I mean, I don't want to encourage a hallway with 20 pitbulls anyway.... I want to encourage clever placement of pitbulls (like in the image above).

Oh, yeah, in your house, on the first big dog-killing trip, Grayson used bricks to kill the cats, which gave him a safe place to stand to club the dogs.  After that, it was an easy club-walk through the rest.

Offline

#16 2013-06-17 12:48:10

zed
Member
Registered: 2013-04-16
Posts: 161

Re: Encouraging desirable player behaviour

jasonrohrer wrote:

Oh, yeah, in your house, on the first big dog-killing trip, Grayson used bricks to kill the cats, which gave him a safe place to stand to club the dogs.  After that, it was an easy club-walk through the rest.

Yes, corpses providing safe ground, like some grisly scroll of Elbereth, is
what renders futile any attempt to do anything cleverer with dogs than just
lining them up along a corridor. The knock-on effect of that is that
protecting the family is really dull - other than just trying to hide them,
which isn't going to keep them alive very long, your best bet really does seem
to be to make such a corridor and protect it as well as you can afford, hoping
that you can make it expensive enough that no-one will decide to spend the
tools to get through. This isn't interesting at all, and no cost-tweaks are
going to make it more interesting.

Getting a dog at distance 2 is rarely helpful for multiple reasons - even if
the drug+club approach were ruled out and there's no corpse to fall back to,
you only have to find somewhere to use a single tool to switch the parity and
get the dog in clubbing-range.

So how about letting pets walk past corpses? I suspect this would be enough to
make protecting the family interesting, since it would really open the
possibility for pitbull-based deathtraps to be tripped by anyone going for
the family.

I know there are sound game-architecture reasons for not wanting to just let a
pet occupy the same tile as a corpse, so how about having them swap positions?
Not particularly realistic (particularly in the case of pit-crippled pets),
but I don't think that matters.

Offline

#17 2013-06-17 13:43:56

jasonrohrer
Administrator
Registered: 2013-04-01
Posts: 1,231

Re: Encouraging desirable player behaviour

Damn, I forgot about using a tool to "stand still" while time passes.... hmm...  like cut a wall.

I wonder if THAT part changing would help anything.  Would have the bad side-effect of making you SAFER in most cases (you could stand there and club four surrounding dogs without being hurt).

But it's strange that otherwise, you have to "step" to pass time, but you can use a tool to pass time while stationary.  Maybe I will change it...

Still, it doubles the cost of killing a dog to set them up this way.  You can't just come in with 20 crowbars and kill 20 dogs.  You'd need 20 crowbars plus 20 saws, which you can't afford.  The "line of dogs" is just waiting to be plowed through.

Also, it's not just dogs stepping on corpses that's a problem, but the fact that they might DIE on a corpse as well (and pile up there, or what?  Visually messy, complicated for tool tips, etc.)  Also, it's tactically interesting for them to block each other (instead of just advancing and advancing no matter what you do

Offline

#18 2013-06-17 14:08:31

largestherb
Member
From: england
Registered: 2013-05-27
Posts: 381

Re: Encouraging desirable player behaviour

i think tunnel 'o dogs that can just be mowed down is a flaw of the home designer, not the game engine. you can take up a bit more space and make dogs that are a bit harder to wrangle.. i've seen a few hard to wrangle dog herds in the past!

Offline

#19 2013-06-17 14:38:40

zed
Member
Registered: 2013-04-16
Posts: 161

Re: Encouraging desirable player behaviour

The dogs-at-two-paces technique doesn't really scale up, though, because once
the robber has dealt with one he has a corpse to fall back to for any more.

largestherb: have you really found a way to set things up protecting the
family which is more cost-effective than the dog-filled corridor? I have tried
extensively, but I don't see that it can be possible. The robber can basically
just follow in reverse the exit path of the family, filling it with corpses
wherever dogs threaten, falling back to the last corpse where necessary.
Occasionally with a lot of effort it can be possible, using powered doors
and/or carefully arranged cats, to force the robber to deviate from that path.
But you're not going to kill a robber who's paying attention that way, so you
can get the same effect for less by just adding another pitbull to a
corridor...


Having dogs switch positions with corpses rather than double up with them
would deal with UI issues. I agree it's tactically interesting to use corpses
as blockades, but as it is currently it makes dealing with pitbulls almost
entirely just a matter of packing enough crowbars.

To preserve some tactical element, here are a couple of suggestions:
(i) rather than switch places with corpses, pitbulls (and I'd say only
pitbulls) are able to push a corpse off to one side if there's nothing in the
way. So e.g. if a pitbull wants to move right but there's a corpse in the way,
it will push the corpse down if the tile is free, else up if that's free, else
the dog will be blocked in that direction (and will consider going a different
way instead). Here 'free' means 'passable under current rules' (so in
particular containing no corpse). So you'd still be able to use corpses as a
blockade, you'd just have to be more clever about the use of space.

(ii) use the 'swap positions' version for moving past corpses, but introduce a
new tool: caltrops. These can be placed on an adjacent passable position; they
then block any pets from moving past them (could be implemented by counting
them as mobiles which happen not to move). The player can move past - but only
by clearing away the caltrops, removing them from play. I think this would be
an interesting tool anyway - I can imagine strategic use, blocking off
unexplored corridors to protect against traps which release hounds behind you.

Offline

#20 2013-06-17 14:46:29

largestherb
Member
From: england
Registered: 2013-05-27
Posts: 381

Re: Encouraging desirable player behaviour

certainly not cost-effective, certainly takes up a lot more space and certainly hopes that a player will make a mistake estimating where a dog is going to step next!

caltrops sound interesting smile

Offline

#21 2013-06-17 14:47:51

nathan
Member
From: A ditch somewhere
Registered: 2013-06-15
Posts: 61

Re: Encouraging desirable player behaviour

jasonrohrer wrote:

Ah... I just added a feature where as admin, I can search for tapes where YOU were the victim.

Do you need the email for that? Because my PayPal email is different than the email I signed up for this forum with. If you want, I can send you the other email address.


"I just robbed Mr. Rogers." -Ludicrosity "The wood is my desk, and I'm knocking it with my head." -Blip
"I'd rather pack 25 meats than 1 crowbar if you know what I mean..." -Jabloko
"This is one of the most disturbed things I have seen in quite a while. I blame global warming." -bey bey
"that seems like more resources than I'm willing to put into having my kids killed." -cbenny

Offline

#22 2013-06-18 00:08:28

jasonrohrer
Administrator
Registered: 2013-04-01
Posts: 1,231

Re: Encouraging desirable player behaviour

Yeah, I was doing it by email address.  Bey Bey used the same email for both, so I could spy on his or her game.

In that case, I was just trying to see what made Bey Bey feel a certain way about being vandalized or overpowered by robbers.

I'll ask for your other email if I ever need it!

My goodness, Nathan.... your profile picture is amazing!

Zed, not sure about caltrops.  12 tools is a nice number.  And I'd like to avoid "corpse pushing" if there's any way to.

In general, though, if there's SOME way to set up tactical situations with dogs that require a gun to pass, the crowbar stops being overpowered against them.  I'm not worried if the "line of dogs" is easy to pass, or if the "room packed with dogs" is easy to pass.  In fact, if the design DISCOURAGES those layouts, that's a good thing.  Anything that would make those layouts more powerful (like dogs pushing corpses out of the way to get you) would make the game less interesting, because those layouts aren't particularly interesting.  Little weird mechanical inventions, like the one pictured above (if it actually worked) are what makes the game interesting.

(I mean, I remember people all the way back to v5 complaining about the "room full of dogs" being lame).

Clearly, pitbulls are central to protecting the family.  I like that, because it ads mechanical variety (they don't seem to be central to protecting the vault these days), and because it's thematically appropriate (your family can't hang out around trapdoors, but their pets are safe around them).

Offline

#23 2013-06-18 03:03:35

Matrix
Member
Registered: 2013-04-06
Posts: 137

Re: Encouraging desirable player behaviour

jasonrohrer wrote:

Though I realize that with meat so cheap, you could carry both meat and crowbars to bypass this device----I should change it in v10 so that sleeping dogs can't be clubbed or woken up, so drugged meat will be much more tactical.  Maybe THAT would be the fix for crowbars, other than simply jacking their price way up?  I mean, I don't want to encourage a hallway with 20 pitbulls anyway.... I want to encourage clever placement of pitbulls (like in the image above).

This could be an interesting change, but let's try to think of any reasons why this might not be a good idea (it seems ok at a first glance).

jasonrohrer wrote:

Damn, I forgot about using a tool to "stand still" while time passes.... hmm...  like cut a wall.

I wonder if THAT part changing would help anything.  Would have the bad side-effect of making you SAFER in most cases (you could stand there and club four surrounding dogs without being hurt).

Don't change this, please. Like you said, this adds more danger when dealing with animals, but also makes some advanced mind tricks possible, for example making a trap where using a tool on a seemingly safe tile next to you causes your death. Granted, most players don't use such tricks, but I've seen some traps that used them to some degree. And things like this just open up more possibilities to trick people, while still not making it overpowered in any way since anyone can use tools beforehand to secure/scout the trap.

This change could also introduce some engine inconsistencies, since you would still have to update the electricity state after a tool is used, but not update animal movement? And if you went for consistency then the game would allow the player to cut off electricity while standing on a powered (safe) trapdoor, which would only update (lose power) once the players makes the next step (which could be onto a safe tile)?

Last edited by Matrix (2013-06-18 03:10:18)

Offline

#24 2013-06-18 06:26:32

dalleck
Member
Registered: 2013-04-13
Posts: 250

Re: Encouraging desirable player behaviour

Matrix wrote:

This change could also introduce some engine inconsistencies, since you would still have to update the electricity state after a tool is used, but not update animal movement? And if you went for consistency then the game would allow the player to cut off electricity while standing on a powered (safe) trapdoor, which would only update (lose power) once the players makes the next step (which could be onto a safe tile)?

Electricity could be kept as an instant effect while still keeping animal movement fixed to your actual steps.

To me, the interesting part of separating actions from movement is it allows you to run away from dogs while cutting through walls, throwing ladders, crowbarring doors, etc..  Hmm you would also be a complete ninja with meat.

If we are trying to balance pit bulls why not take away the ability to use a crowbar or a gun on a dog, and sleep is the best you can do?
Robbers will have to completely rethink how they deal with dogs when they become an impassable square...  Maybe a moving pit bull wakes a sleeping one?  Then you could slow a group of dogs, but not stop them.

Would this be too dangerous?

----

Ahhh...

jasonrohrer wrote:

instead of just advancing and advancing no matter what you do

Ahem.. this would indeed be 'advancing and advancing', but maybe with a hint of 'I can still get away from them'...

Last edited by dalleck (2013-06-18 06:52:14)


The rich aren't safe. Nobody is safe. -jere                   ...but the smell wafts out from the pit, obviously. - Jason Rohrer

And the more dickish they are, the more I feel like beating a house to destruction after finally figuring it out. -bey bey

Offline

#25 2013-06-18 06:48:57

zed
Member
Registered: 2013-04-16
Posts: 161

Re: Encouraging desirable player behaviour

jasonrohrer wrote:

In general, though, if there's SOME way to set up tactical situations with dogs that require a gun to pass, the crowbar stops being overpowered against them.  I'm not worried if the "line of dogs" is easy to pass, or if the "room packed with dogs" is easy to pass.  In fact, if the design DISCOURAGES those layouts, that's a good thing.

Agreed. I think it's fine that lines of dogs do nothing more than force the
robber to use that many crowbars, I'd just like that not to be the optimal
strategy for protecting the family.

My thinking was to make viable traps which involve releasing dogs on the
robber, which currently are pointless because the robber can just club them
all (or fewer), using corpses to stay safe.

Maybe it's worth comparing with DROD, actually - imagine how boring a game
that would be if you could use corpses the way you can in CD!

Offline

Board footer

Powered by FluxBB 1.5.8