Development continues

[owen]

Posting a few more of my thoughts and questions about campaign mode and the game in general.

1. How big is the campaign going to be, and how is permadeath going to work? I really wonder if it is feasible to design a campaign that is intended to last 30+ hours, all on a single life, without the game becoming too boring and forgiving. The Rogue-likes and -lites that I’ve played (DOS Rogue, Faster than Light, Convoy, and recently, Risk of Rain) can all be completed in a few hours, but it is expected that the player is supposed to die. In many cases, it’s expected that the player dies more often or as much as he succeeds, especially if he is not a master of the game. This is tolerable if each run through only takes a couple of hours, but not if the game takes days or even weeks.

   Secondly, do you think that this game will be increasingly enjoyable as the campaign length increases? I think certain genres (RPGs, for example) obviously benefit as the playthrough length increases; however, three two-hour rogue-like sessions might be more fun than one six-hour session, because a lot of the fun in a rogue-like is dealing with variable conditions early on, and how the choices that result from these early decisions affect the late game. In other words, the more the player starts a new game, the more fun he has (to some extent).

   In some ways, this game is becoming more and more like fantasy XCOM, heh.

2. I like the idea of being limited to one base. Managing dozens of dungeons everywhere would be a mess. Since we’re limited to one base, though, I think that the size and/or complexity of the base has to increase proportionally to the player’s progression through the overworld. I think it would be VERY BORING if the player were able to carve out a dungeon and mine out the home sector of the map to build his facilities all on the first day, and then just wait for reinforcements to arrive for the rest of the game. It would be more interesting if you fought a battle, carved out some dungeon, fought another battle, carved out more, and so forth.

   Given that the dungeon carving is a such a big part of the game, I wonder if designing dungeons could involve more strategy than it currently does. For example, some regions of the mountain could provide certain production boosts for different types of minions, the mountain could become more difficult to mine the deeper you go in, and so on.

   I’m drawing on inspiration from Faster-than-Light and Convoy, games in which <b>every transaction involving resource allocation is important.</b> I believe that every mechanism that a strategy game introduces should be developed on as extensively as possible. In Faster-than-Light, for example, rooms can catch on fire. Fire also interacts with the oxygen and crew mechanisms, because fires can be extinguished by draining oxygen from their immediate vicinity or by sending a crew member to deal with it. This results in somewhat interesting decisions that involve extinguishing fire. In the heat of battle, he has to determine how the fire should be extinguished, either by sending crew or by purging oxygen, and who should be sent. If the doors are damaged while there is a fire, the response becomes more complicated if large sections of the ship are evacuated.

   Secondly, there are certain types of crew members (rockmen) who are immune to fire (as well as those who are not), and there are certain types of weapons that can start fires. There are also augments that can suppress fire or mitigate low oxygen. The point is, though, that <b>all of these systems are intertwined.</b> And this is just one little mechanism of the game.

   I don’t really see such a network of concerns present in KeeperRL. Such a system exists to some degree, but it’s not very well balanced. For example, the minerals are just sitting there in a cache within the mountain. You mine them and then you spend them. Does mining have any costs to it? Well, it occupies my imps, but that’s not really a big deal because they generally don’t have anything better to do. It also exposes my imps to danger, but they’re usually enough that they can run away, and there’s generally an alternate way to grab the minerals if I really want to get them quickly. Gold attracts bandits, and tree stumps spawn tree spirits, but neither of those are big threats.

3. Is the “campaign” going to be as simple as plunking the keeper in the middle of a world and letting him go anywhere he wants, or is there going to be some sort of directional progression? I think that too much freedom can be a bad thing, because players will almost always select the best option. Secondly, forcing the player into a limited set of moves makes him think harder, and it adds replay value to the game.
4. How will the hunger clock work in a campaign mode?
5. What happens if you’re attacked on a map that you’re not playing on at the moment.)

[owen]

Keeperman wrote:

Hi,

I need to say that if you decide your campaign mode is achievable, I’d still like to help.

In fact, I am offering to make you a selection of 1/4 size dungeons that you can install with alpha 18. You would find this especially helpful if you are unable to convert any of the alpha 17 dungeons cropped to the right size.

This is a good idea if there are going to have prefabricated dungeons. The developers of Xenonauts, an XCOM-like, had the controversial idea to use prefab maps instead of randomly generated ones, but I think it’s a good idea. Handmade dungeons are always going to look more interesting than automatically generated ones (although there’s no reason why you can’t have a mix of both), and it’s easy for the community to step up to make more maps.

I suggested a rating system for dungeons, but why not have someone who is willing to make the maps do it himself? It’s one less thing to program, and you’ll probably get better dungeons in the end.

[owen]

On tactics:

Currently, battles don’t involve much more than dancing your keeper back and forth while minions bash each other over the head (or shoot arrows, if they’re clever). I mean, it’s fun for a while, but it really doesn’t offer the sort of tactical game play that XCOM or even Final Fantasy might. Basically, you just use your spells and items to keep the keeper alive and hope that you have a big enough army. If SHTF, then run your keeper back to the dungeon where your traps can kill inordinately large numbers of enemies.

Well, that begs the question, how can this game be more tactically interesting? I guarantee you that simply marching around an ever-increasing death ball of orcs and ogres while it auto-attacks the enemy will not make for an exciting game.

• In a game that has potentially dozens of units per side and battles that proceed for hundreds of turns, I don’t think it’s reasonable to expect the player to micromanage every single unit like he would in XCOM or Final Fantasy. A corollary of this statement is that it shouldn’t be possible for the player to gain a (significant) advantage through micromanagement of this sort. Single-player games that can be exploited through tedious micromanagement are the worst; it feels bad to win (because it’s boring) and it feels bad to lose (because you know you could have won if you’d just done more repetitive actions). They’re entirely antithetical to a game like chess, where you get one move, and the skill is in finding the best move.
• Many units have item slots, and this mechanism could be exploited to increase the strategy in the game. Units could also be given preset behaviours, or skills. Essentially, the tactical decision would come down to pre-equipping your band with a certain item combination, and then dancing your keeper (or other team leader) around while your team fights it out.

• Perhaps giving orders as a team leader, and the reliability with which orders are heeded could be a game mechanic in itself. For example, the team leader might have a general “give order” ability, which might have a cooldown of 10-20 turns (just so that you can’t spam-micromanage orders) and costs one turn to execute. By casting this ability on to a unit within visual range of the team leader, and then selecting the particular details of the order (which are dependent upon the combination of team-leader and team-member), there is some method of minion control given to the player.

[Keeperman]

Campaign size. Dimensions of the oveworld map could be adjusted after developing it.

[Keeperman]

Permadeath. dont send your keeper on invasions!

[Keeperman]

Attacked on a map you aren’t currently playing.

Complex and tricky. For alpha18 AIs and players might have to take it in turns to attack squares on the overworld.

This means AI cant interupt until an invasion has ended. Then it is the other forces who attack while you must defend. It is their turn. etc.

[Keeperman]

Alpha 18 map making volunteers. There would be plenty. You may find it hard to unit test Alpha 18 without some keeper maps that you can invade. In Alpha 17 you can only invade them with adventurers. It will be a new thing.

[owen]

On strategy:

• – Currently, minions in production facilities spit out random weapons, charms, and spells. I wonder if it would be more strategically interesting if this were further refined. As there are going to be many more villains to conquer, perhaps individual items could only be created after conquering a quadrant of the map. For example, the recipe for a sleeping potion could be won after a dwarf colony is conquered. The same system could be extended to weaponry, charms, traps, and even to skills that are taught to beasts. The idea is that certain major villains would either be more susceptible or resilient to certain technologies, and your current production abilities would influence your plan of attack.
• – Similarly, the same system could work for hiring minions. You can only hire minions from tribes whom you discover on your quests. This limits your tactical options on each play-through, forcing you to think on your feet.

[Keeperman]

Unlocking potions, weapons and traps.

Would need a technology for each recipe and design.

Also would need to block research in the library until you actually find the books that exist on the map/campaign.

[Keeperman]

Unlocking minions.

Maybe in Alpha19 campaigns?

[Keeperman]

Combat tactics.

Yes, the AI of my team could be a problem. They may have a tendency to go running off into traps. Traps need to be taken out using magic (or recruiting bandits;)

[owen]

Keeperman wrote:

Attacked on a map you aren’t currently playing.

Complex and tricky. For alpha18 AIs and players might have to take it in turns to attack squares on the overworld.

This means AI cant interupt until an invasion has ended. Then it is the other forces who attack while you must defend. It is their turn. etc.

Yeah, I thought that something like this might work. There are a couple of scenarios that I imagine:

1. AI (or player) wants to reinforce an active battle with units from a neighbouring square. In this case, the units can just stream in.
2. The AI wants to start a new battle while there is already an active battle. In this case, the time at which the new battle is supposed to start would be recorded. After the first battle finishes, then time-space “rewinds” back to the start and location of the second battle, and all of the activity in the square of the first battle is frozen until the time of the second battle reaches the time at which the first one ended.

The problem with all of this is that you need some sort of indication of the start and the end of a battle. In the current single-map system, you can control and release units at will. I would suggest that, in the campaign mission, any time there is a hostile unit in a square that is owned by the player, or vice versa, that the player should be forced into unit control mode

[Keeperman]

Games like civilisation, chess and risk use turned based strategy.

You have 1 battle and choose the troops you wish to attack with. The battle must then be resolved win/lose and casualties. Then your turn ends and the next player has a turn at attacking.

While the battle takes place all other players and units are idle.

A more complex model for later releases?

[Keeperman]

Ah. I see your point. Each invasion would have to wind back the game clock to the start of the round. A round starts when someone invades someone else.

[Keeperman]

Maybe the game clock should just keep running?

[Author]

Posts