New players, maps

jere · 2013-05-31 08:37:29

I'm trying to put myself into the shoes of a new player and the thought is terrifying: any newcomers are living on the same block as the most hardened criminal masterminds. When defending, one has to have an in depth knowledge of the most complex wiring in the game or they stand no chance. Any of the old strategies that a new player might try are immediately defeated with blueprints. When robbing, they have to have that same knowledge. Most simple houses are going to be robbed clean by experienced players. The remaining ones are naturally going to be really well defended (combo locks have now been beaten out by obfuscated combo locks as the optimal strategy).

I was an EE/CPE major and loved playing around with redstone in Minecraft. I find wiring in TCD to be fairly straightforward, but I'm wondering if it might be a brutal learning curve for most players.

Maps are really neat, but I feel like we lost something. Let me quote Jason:

--A simple maze, made up of walls only, is more interesting than a combo lock.
--A room with only dogs and walls, where you have to figure out where to walk, in what order (DROD-style), is more interesting than a combo lock.

I'm not sure if Jason is only referring to undiscoverable locks, but either way I guess I'm saying I miss mazes a lot. The tension of walking down the wrong hallway is completely gone. The fog of war means very little now, whereas before it was truly terrifying. The promise of decoys, trickery, and diversions is no longer, not including wire obfuscation. I feel like the game we have now is really interesting, but mostly for highly experienced players who are trying to play optimally. But the game we had was really interesting too, was more intense, and was probably better for new players (i.e. mazes are easier to construct).

Thus, it'd be nice to try to maintain both kinds of games. Doing so would require a change to maps and I certainly realize that involves massive rework. Perhaps the game is OK as is. After all, it's a game about violation and it's meant to be brutal. On the other hand, I'd like to see more people playing and a shallower learning curve may help.

Anyway, the basic idea is to segregate, either sharply or fuzzily, players into map or no-map buckets. There are a few ways to go about it.

The most kludgy: players could start in a separate queue where no maps are available. Once they get enough playtime, money, or self-elect they move into the normal queue where all maps are available.
Newly spawned houses have no maps until they cross a certain value threshold.
Alternatively, newly spawned houses have expensive buyable maps, but the cost of the map comes down every time it is purchased or simply every hour until it becomes free.

This essentially recreates both kinds of gameplay: the old style where mazes are feasible and the new optimal play style. Experienced players might actually stick to the latter because they have too much invested and mazes are too risky. New players should have less to lose and they can stick to both building and robbing mazes until they learn the ropes.

Thoughts?

dalleck · 2013-05-31 10:26:41

Yes this IS an issue. As you say, a steep learning curve has been introduced with blueprints and also we have lost a great deal of variety (not to mention tension).

I am not sure of a definitive solution, but this issue needs balancing out before release, so players can experience a wide range of play styles in the game.

jere wrote:

The most kludgy: players could start in a separate queue where no maps are available. Once they get enough playtime, money, or self-elect they move into the normal queue where all maps are available.

I think this idea of a kind of 'beginners lounge' just wouldn't fit into the brutal theme of the game. As you said, it's kludgey.

jere wrote:

Newly spawned houses have no maps until they cross a certain value threshold.

Yes, some form of this is the likely solution. Jason would need to decide on what qualifies a house for a blueprint.
I also feel quite strongly that if the thieves guild only 'release' blueprints for popular houses, then there should be some delay before that blueprint becomes available.

As I see it, blueprints should always be there in the end, but there should also be an opportunity for the unknown, mystery and suspense.

jere wrote:

Alternatively, newly spawned houses have expensive buyable maps, but the cost of the map comes down every time it is purchased or simply every hour until it becomes free.

I was against free blueprints initially, but now having played it I wouldn't like to see them having a cost.

ukuko · 2013-05-31 10:35:05

jere wrote:

Newly spawned houses have no maps until they cross a certain value threshold.

This would be my preferred route. I think this would temper the difficulty curve for new players and also reintroduce some of the fog-of-war tension when robbing new houses. I can see it now — the $6000 empty-looking honey trap...

I guess the blueprint threshold would be based on either ranking, time unbeaten, attempts, deaths or a combination of the above. I don't think you should have to pay.

Last edited by ukuko (2013-05-31 10:36:43)

jere · 2013-05-31 11:03:37

I was against free blueprints initially, but now having played it I wouldn't like to see them having a cost.

I can see it both ways. The idea of buying a blueprint and gaining secret knowledge though is rather compelling (e.g. Many Bothans died to bring us this information.).

Anyway, to me it's just another tool in your backpack. Except in the case of extremely complex circuitry (what optimal play has become), it shouldn't be required. Also, if you stand to gain $50k, then buying a blueprint for say $1k seems like a bargain. And either way, it'd end up free eventually.

jasonrohrer · 2013-05-31 15:05:26

Great points here.

Obviously, this space needs a lot of exploration before we can find the right combination of game mechanics that makes the game work on all levels.

The simplest thing is just a value threshold for blueprints. Hard houses accumulate money over time very quickly, both through dropped loot and salaries that are paid as long as the house remains unbroken. At the current rate of $140 + $280 (if wife is alive), you earn $10K in a day if your house is not broken.

So, houses with less money simply wouldn't have blueprints, ever. Of course, if they were actually hard, they would gain more money and get blueprints.

This would also let us bring starting money back down to $2000, and the creepy maze house would return as a viable low-level house strategy. Then it would rise in value, eventually cross the blueprint threshold, and then get broken instantly.

Also, this would bring back the joy of watching tapes to see people fall for tricks. That's kindof gone from the current experience.

There's still the lurking problem of cheaters peeking at the map anyway for these low-level houses. How do we stop expert players from doing this and "ruining" the game for beginning players? Maybe ignore this edge case? Ugg...

Certainly, disabling the protocol print-out would make it a bit harder for most people (would require a real mod to see the map).

I've also thought a LOT about segregating the community into two servers, etc. The problem is that the community isn't large enough yet to make this viable.

In the last 24 hours, 125 people have visited their house. In the last hour, 8 people have visited their house.

zed · 2013-05-31 16:12:48

Bringing back unmapped houses could be fun. But I'd be wary of having the criterion for mappedness being under the owner's control. If it's a matter of having a certain amount in the vault, players would be motivated to (i) build the guessing-game-based houses we thought we'd put behind us (ii) log in regularly to sell tools and spend surplus cash. This would be fun for no-one.

I do quite like the suggestion of having maps cost money at first, but becoming common free knowledge once enough people have bought them. You'd need to not have it reset when some small changes are made to the house; maybe if it was free before the edit, a map for an edited house could have an initial cost corresponding to the money spent making the modifications.

jere · 2013-05-31 17:16:02

How do we stop expert players from doing this and "ruining" the game for beginning players? Maybe ignore this edge case? Ugg...

Yes, ugg! When I said massive rework, I was referring to switching to a snapshot being sent to the client for each turn. I don't know how feasible this is (server constraints?) and if it is feasible, I'm sure it's a bunch of work. But we're talking about a pretty huge payoff IMHO. The unmapped, maze portions of the game were really fun and tense. I suppose this would be a long term strategy and I don't know what your plans are for supporting the game over time.

Maybe you could make it so if a house does have a blueprint, it uses the normal client simulation mode and if it doesn't it relies on the server. More complex, but that might reduce the overhead.

But I'd be wary of having the criterion for mappedness being under the owner's control.

I think you're right. That might be a real problem. Also, I like a buyable map that eventually goes to zero because it matches other systems that drive at a house inevitably being beat (e.g. salary system).

In the last 24 hours, 125 people have visited their house. In the last hour, 8 people have visited their house.

Not sure how to solve this. I guess the game is somewhat niche? I've tried on 3 different channels to get someone to take my code and found no takers, which is a huge bummer because I think this game is amazing and deserves more exposure.

Last edited by jere (2013-05-31 17:19:27)

DrNoid · 2013-06-01 06:49:02

Maybe a way to improve the situation is adding npc houses so that there are always 25 to 50 unbroken houses that are earning income. If one of those houses gets robbed it is replaced with a fresh new one. There should be plenty of designs in the database that can be used.

dalleck · 2013-06-01 08:29:37

DrNoid wrote:

Maybe a way to improve the situation is adding npc houses so that there are always 25 to 50 unbroken houses that are earning income. If one of those houses gets robbed it is replaced with a fresh new one. There should be plenty of designs in the database that can be used.

But the problem isn't actually an economic one, it is an issue of privacy.

In v7 each house begins exposed, thus eliminating many simple and fun designs, and suffocating beginners.

jasonrohrer · 2013-06-01 10:40:34

Running the whole game server-side, step-by-step, is not feasible for loads of reasons. Yeah, it's a ton of work, but it's also really protocol heavy. The server currently runs as a PHP/MySQL on a web server. That means each message is an HTTP request/response. This is an amazingly simple and easy way to handle this. I'm not sure if this is the world's first MMO written from the ground up by one person, but if not, it's one of the first. I was able to get this done in one year by myself because of this approach to client/server communication.

But if you need a message sent for every move that every player makes in the game, I don't think that HTTP request/responses would be feasible for that. I think you'd need to have an open socket connection for every player in the game. And that means moving away from PHP/MySQL, most likely.

The win would be... stopping a few people from cheating, assuming that some houses do not have blueprints. Much more sane to just have blueprints for every house. Something is lost there (sneaking around a maze), yes.

I'm not convinced that this game is as hard on new players as it seems to be. I mean, yeah, it is a BRUTAL game, in terms of what it expects from you, but it's brutal even if you're playing by yourself, on a server of one person.

So, in that way, it scares away new players.

But, in terms of interacting with other players, the interaction is so limited. I mean, in the beginning, for your first few hours, you could fiddle around in your own house, testing it and dying, without ever being influenced by expert players. (I've seen game recordings of people doing this. They play FOREVER without even realizing that there are other houses out there. Then they finally get to the menu, and think, "Whoa, what's this?")

Then you finally leave your house, try robbing someone, and die. Then you come back and fiddle around some more. Then you leave again, try robbing, back out, try to come home, and find yourself robbed. You watch your tape. One guy robbed you. Was it an expert player? It doesn't really matter. You get to watch how it happened, and you can learn from that at your own pace. With each "failure," it happens once, and then you have as much time as you want to react to it.

FPS deathmatch is much more brutal on new players... you're constantly beat down by experts. You hardly even get to play, and spend most of the game waiting to respawn.

I'm thinking that we need a quick video to give new players basic tips. Like the fact that you can always back out the front door... or that you can test your house with non-lethal stand-ins before you install the lethal stuff...

bey bey · 2013-06-01 13:48:01

The value threshold shouldn't be safe content but building cost. Say it costs more than 10000 --> there's a blueprint. (The building process has alerted the thieves guild.)

jere · 2013-06-01 14:06:49

Ok, fair enough. Clearly, running the whole thing server side is too costly.

Let me make one more comment and then I'll leave it alone. Ideally, we'd have zero cheating. But I think it's still worth having mazes/tension/trickery/information asymmetry even if there's the possibility of cheating using the current client simulation system. Here's why.

Cheaters would have it no easier than it already is. Honest players get the upside of mazes/tension/trickery/information asymmetry and new players get, unless cheaters are really rampant, some solace in being able to get by for a while with simpler houses. That first part is there regardless of cheaters. Even in the worst case, new players would simply be up against cheaters that would have as much knowledge as they do with the current build.

How do we stop expert players from doing this and "ruining" the game for beginning players?

See, that can't be considering "ruining" unless ruining also describes the current state of the game.

The only real downside I can see is if the false sense of security drives some players to build mazes that cheaters could instantly beat. But I'm not convinced they're not already building houses that can instantly be beat with blueprints (that's my concern here anyway ... we're not all blips, zeds, or ukukos).

There's also the nagging thought that cheaters have an unfair advantage. Maybe we shouldn't be bothered by this though. For one, you'd only be able to use this tactic on low value houses anyway. Also, cheating doesn't help you defend that money any better. So what if you spend all day stealing a bunch of money? You're going to be departed from that money rather quickly unless you play the defensive game legitimately. All you've done is make the game less interesting for yourself; you won't have anything to show for it. So all that makes me question the motivation behind cheating. Getting rid of the low hanging fruit (i.e. protocol print out) might be enough to dissuade most.

It's not necessary, but there's an interesting question about how to deal with cheaters if and when they do arise. I'm going to ramble about this idea I had even though I'm fairly sure it's doomed (feel free to ignore): honeypots. If a player thinks they see a cheater on a security tape, they could flag the robber. If flagged X times, a special honeypot house is shown only on the robber's queue. The honeypots would all be extremely complicated mazes or guessing games. A successful robbery on the first attempt, especially without using items would be a strong indicator of cheating. The fun part would be generating them randomly, Rogue style.

I'm thinking that we need a quick video to give new players basic tips.

That would be great. The most important part for me was realizing suicide was viable and getting comfortable with dying. Before that I kept thinking "how am I going to figure all this out without testing it?"

FPS deathmatch is much more brutal on new players... you're constantly beat down by experts. You hardly even get to play, and spend most of the game waiting to respawn.

Yes, this drives me crazy when I'm playing multiplayer games with experienced friends. To be fair though, a lot of modern online games have fairly sophisticated matchmaking systems. So you are usually paired up with players close to your skill level.

I'm not convinced that this game is as hard on new players as it seems to be

What I mean is that defending your house for any considerable length of time is hopeless without expert level play (e.g. an understanding of of electronic memory).

I mean, in the beginning, for your first few hours, you could fiddle around in your own house, testing it and dying, without ever being influenced by expert players.

Good point! I hadn't considered this because it's been so long since I was at that stage.

The value threshold shouldn't be safe content but building cost. Say it costs more than 10000 --> there's a blueprint.

But you could make a house that's impossible (without blueprints) for fairly cheap. And you could make a costly maze. The safe value is driven by how hard the house is to beat over time, which I think is the proper way to do it.

Last edited by jere (2013-06-01 14:18:04)

jasonrohrer · 2013-06-01 20:23:48

I'm going to wait a while on this one.

The idea that new players have no hope of defending for any length of time.... well, that's kind of the point of the game. It's about a false sense of security. Everyone should be violated eventually. Yeah, not everyone is Zed, but even he got his in the end. There's nothing wrong with a new player getting their house broken into. I mean, if they are NEVER broken into, that would be much worse, at least in terms of my aesthetic goals.

In v5, the home owners had a huge advantage (hidden information). New players had no chance of breaking into those 50+ houses. No one had a chance, unless they were cheating. Robbers are risking permadeath trying to break in. Their one advantage was the weakness of the family, but it was still way out of balance. I mean, a homeowner doesn't risk permadeath at the hands of a robber.

In v6, the robbers now have a bigger advantage, which matches their risk better. Homeowners have an added advantage in the family behavior, though. I think it's much more balanced, both for new and expert players.

I remember seeing a bunch of "Let's Play" videos from v5. Some new player, the reviewer, was going out to try robbing houses for the first time. Try 1, of course, is the top house on the list. Step in door, get instantly fried. Try 2, second house on the list, step in the door, get instantly fried. "How are they doing that?" It was baffling for them, because they could see nothing about how the trap was working. Now at least they could check the blueprints, see the complexity behind the walls, and move on.

Also, even a new player making a dog maze is likely to make something that ends up being one-time robbable. Protecting the family is pretty natural, too. So, they get robbed, but not mowed to the ground. They fix up their house and try again.

I haven't made up my mind one way or the other about whether blueprints for all houses should remain. I'm just kicking it down the field for a while.

dalleck · 2013-06-05 07:54:07

Just read a brilliant suggestion on the Youtube comments for Aavak's videos:

Destron - Youtube wrote:

I feel like they should only show blueprints the second someone has placed a pressure plate or a piece of wiring (to make houses that arent combo locks and other unbreakable houses slightly fairer)

What a great idea! (although not the wiring part, just switches) You cannot make any tricks without switches, just a maze with dogs and doors, electric floors etc.

So if those designs are left alone without blueprints, that leaves essentially two types of house, the maze house and the (blueprinted) switch house.

Once a switch exists in the house design, it has a blueprint available. Simple as that!

DrNoid · 2013-06-05 07:59:49

Sounds good!
I'm not sure what use electric floors are without switches though...

dalleck · 2013-06-05 08:15:25

DrNoid wrote:

Sounds good!
I'm not sure what use electric floors are without switches though...

They could be used to lure dogs in to their death in a particular maze puzzle.

I think there are perhaps some interesting puzzles you can create without switches... especially if the robber is in the dark.

Anyway it would just be nice to have blind maze houses exist again!

jere · 2013-06-05 08:23:19

I'm not sure what use electric floors are without switches though...

At the risk of revealing some of my techniques: you can use electric floors to dispose of pets. Cheaper than pits at scale! For example, you could have 20+ pitbulls guarding a safe and you have to get all of them to suicide onto electric floors before passing. I actually used this trick once to get to a pitbull guarded family.

I think the switch->blueprint idea is pretty clever, though I still think each house would receive a blueprint sooner or later. You could still make a really hard house with no switches.

jasonrohrer · 2013-06-07 21:27:29

Also... trial and error in a blind maze.... Especially if the maze is well-designed and not full of easily-exploitable flaws.... it's just not interesting to try to figure out how to get through such a maze.

And yes, it's possible to design a very hard house with no switches or wiring. You can essentially build a combo-lock out of a maze, where you have to walk in exactly the right way to get the dogs placed in the maze to move and get caught on things in exactly the right way. One wrong step and the dogs get out of whack and get you. The owner knows the right stepping to get through unscathed. With blueprints, solving such a thing is still really hard. Without blueprints, solving it is impossible, in practice.

nashable · 2013-06-12 13:26:43

jasonrohrer wrote:

Also... trial and error in a blind maze.... Especially if the maze is well-designed and not full of easily-exploitable flaws.... it's just not interesting to try to figure out how to get through such a maze.
And yes, it's possible to design a very hard house with no switches or wiring. You can essentially build a combo-lock out of a maze, where you have to walk in exactly the right way to get the dogs placed in the maze to move and get caught on things in exactly the right way. One wrong step and the dogs get out of whack and get you. The owner knows the right stepping to get through unscathed. With blueprints, solving such a thing is still really hard. Without blueprints, solving it is impossible, in practice.

I'm also a new player (started with v7) and found blueprints, while helping me evaluate a house to rob, also intimidated me from trying to rob anything. I played a lot trying to build my own house but always felt that it was going to get cracked when I saw way more epic houses being cracked.

What I would suggest in terms of design would be to remove the blueprints but add a sliding scale to the house building. The more you have of a specific component (except walls) would increase it's cost exponentially. You tune the factor at which they increase based on their base cost and what kind of behavior you're trying to encourage/discourage (e.g. wire and dogs increases faster). This way a house that specializes into only a few components would be very expensive to build vs a house that offers a maze with different types of traps (which I would argue is more interesting).

Another solution is to have non-reachable squares from the start show up on the blueprint initially. Then only areas discovered/walked by a robber is added to the blueprint (for all to see). The blueprint is reset each time the vault is reached.

Last edited by nashable (2013-06-12 13:27:05)

jasonrohrer · 2013-06-12 15:23:39

Good idea here about house object costs changing based on how many of a given item you're using in your current house design.

In v9, blueprints will be removed, and the way the backpack functions will change dramatically. I think that should address some of the issues that you're describing, but we'll see how it goes.

The Castle Doctrine Forums

#1 2013-05-31 08:37:29

New players, maps

#2 2013-05-31 10:26:41

Re: New players, maps

#3 2013-05-31 10:35:05

Re: New players, maps

#4 2013-05-31 11:03:37

Re: New players, maps

#5 2013-05-31 15:05:26

Re: New players, maps

#6 2013-05-31 16:12:48

Re: New players, maps

#7 2013-05-31 17:16:02

Re: New players, maps

#8 2013-06-01 06:49:02

Re: New players, maps

#9 2013-06-01 08:29:37

Re: New players, maps

#10 2013-06-01 10:40:34

Re: New players, maps

#11 2013-06-01 13:48:01

Re: New players, maps

#12 2013-06-01 14:06:49

Re: New players, maps

#13 2013-06-01 20:23:48

Re: New players, maps

#14 2013-06-05 07:54:07

Re: New players, maps

#15 2013-06-05 07:59:49

Re: New players, maps

#16 2013-06-05 08:15:25

Re: New players, maps

#17 2013-06-05 08:23:19

Re: New players, maps

#18 2013-06-07 21:27:29

Re: New players, maps

#19 2013-06-12 13:26:43

Re: New players, maps

#20 2013-06-12 15:23:39

Re: New players, maps

Board footer