Recently I've become interested in textmode games. In my quest for a new project, I came across the roguelike genre and thought it would be a fun new horizon to explore. So...first, I should mention that I found an amazing resource for roguelike development. If you are interested in making a roguelike, and have no idea where to start, roguebasin.com is probably the best source of information you can find.

But getting back to this topic, I was wondering what you guys think of a cooperative roguelike. The player can choose between a single-player run or a co-op run, and the controls would be shared on a single keyboard.

• A S D W + Keys on the left-hand side
• LEFT DOWN RIGHT UP arrows + Keys on the right-hand side

Do you think I should spawn twice the number of enemies per room to maintain the overall difficulty of the game? What about items? Double the amount or share resources/inventory between the players? Also, should I make this game real-time or entirely turn-based like traditional roguelikes (where nothing happens unless the player presses a key).

I'm sorry if I'm being very vague. If you'd like me to clarify something, I'll definitely do so! I've always enjoyed co-op games, so I thought I'd give it a shot.

I suppose my main question is this. How should I alter the co-op gameplay to maintain the aspects players find fun in a single-player experience. This roguelike features the usual perma-death, randomized dungeons, and turn-based battles.

Technical details: I mentioned earlier that this is a textmode game, so it runs in a Windows console/command prompt window with double buffering (no flicker). I do have a variable timestep, so designing a real-time game won't cause any difficulties. Animation works great as well!

Please share your thoughts and suggestions with me.

You can add +50% amount of enemies and give them all +25% hp, this way your dungeon won't become too crowded. Balancing between 1 and 2 player mode should mainly be done during playtesting though. Btw could a single player control both characters/avatars?

You can add +50% amount of enemies and give them all +25% hp, this way your dungeon won't become too crowded. Balancing between 1 and 2 player mode should mainly be done during playtesting though.

Ok. I'll be sure to try it!

Btw could a single player control both characters/avatars?

Hmm. Well, I suppose it's possible to add AI to the second character so that the player can give commands similar to those of a FPS (hold position, follow, attack at will, defend, roam freely, etc.). That would be pretty awesome, actually. I believe this is how the Pokemon Mystery Dungeon series handles multiple characters controlled by a single player. The thing is, they also battle on their own as well (which means the player doesn't micro-manage each character during combat, although certain moves can be disabled for the AI). Whether this is a good or a bad thing depends on the player.

You can add +50% amount of enemies and give them all +25% hp, this way your dungeon won't become too crowded.

It's likely going to be more complex to balance things out then just tweaking enemy difficulty. One of the typically defining characteristics of roguelikes is the struggle for resources to remain alive. By splitting resources between two player characters, you've reduced the resources each player gets in half. In fact, if this includes experience as well, then anything less than a 100% increase in the number of enemies is going to result in slower character advancement in two player mode. If the game relies on any sort of consumables for survival, than that needs to be factored in as well.

I'm not sure which are still alive, but there have been online versions of a few angband variants. They used slow realtime so you maintained the rougelike sense of pacing without having to wait for everybody on the server. With local co-op, I'm not sure how turn-based woul work out. You could always do a combination, where the turn increases every two seconds or when both players have entered their commands.

I wouldn't go with 2x enemies, large swarms tend to either be not that much harder (you funnel them through a choke point) or devestating (you're hit with 50 fire breath attacks and die). And even given that, a pair of players is going to be more than twice as strong as one. A wizard can hang back and fire powerful spells while the warrior holds the monsters back. Or even with two warriors, you can stand in a hallway swapping spaces so one of you is safe to heal. You'll have to play test to balance, but I'd probably increase all the monsters' stats, reduce the players', and do maybe 25-50% more monsters.

Personally, I'd probably go with a shared inventory. It makes resource management a little too easy (if you have one potion of *healing* you don't need to guess who will need it), but micromanaging getting the shield to the other player, and dividing up potions and scrolls evenly, doesn't seem terribly fun. There's so much inventory management in most roguelikes, you want to keep it as streamlined as possible. The only caveat would be if you have fewer, more powerful items (e.g., rods of healing instead of potions), where it becomes a lot of binary choices (do you get the staff of fire or do I?) instead of "I'm low on teleportation scrolls, give me another 10"

In fact, if this includes experience as well, then anything less than a 100% increase in the number of enemies is going to result in slower character advancement in two player mode.

I haven't played many roguelikes, but I do play alot of coop games, and IMO shared experience is the way to go. Anytime either player kills a monster, both get an equal copy of the experience (same with money - any money picked up gets duplicated between both players into separate wallets). It's not fun in coop games when one player starts outleveling another because the other character is a support class. Shared experience and shared money drops and shared inventory (but equipping an item makes it unavailable to the other player), but independant stats, skills, and upgrades. i.e. both characters level up at the exact same time, but each gets their own stats and their own skill points. Both get the same amount of money, but each has a separate wallet so they can freely spend it on themselves.

I'm not sure which are still alive, but there have been online versions of a few angband variants. They used slow realtime so you maintained the rougelike sense of pacing without having to wait for everybody on the server. With local co-op, I'm not sure how turn-based woul work out. You could always do a combination, where the turn increases every two seconds or when both players have entered their commands.

Slow real-time seems reasonable. On another forum, a number of people mentioned that using turn-based co-op gameplay could potentially cause an issue because the turns would go twice as fast if the enemies moved each time either player moves. Yeah... So, at least in co-op mode, slow real-time might be the way to go. Though I'll probably keep the turn-based gameplay in single-player mode.

I wouldn't go with 2x enemies, large swarms tend to either be not that much harder (you funnel them through a choke point) or devestating (you're hit with 50 fire breath attacks and die). And even given that, a pair of players is going to be more than twice as strong as one. A wizard can hang back and fire powerful spells while the warrior holds the monsters back. Or even with two warriors, you can stand in a hallway swapping spaces so one of you is safe to heal. You'll have to play test to balance, but I'd probably increase all the monsters' stats, reduce the players', and do maybe 25-50% more monsters.

You're right. Play-testing is the best way to figure out what works the best in this situation. If I can't find anyone to test it locally, I'll likely post a beta version somewhere for feedback.

Personally, I'd probably go with a shared inventory. It makes resource management a little too easy (if you have one potion of *healing* you don't need to guess who will need it), but micromanaging getting the shield to the other player, and dividing up potions and scrolls evenly, doesn't seem terribly fun. There's so much inventory management in most roguelikes, you want to keep it as streamlined as possible. The only caveat would be if you have fewer, more powerful items (e.g., rods of healing instead of potions), where it becomes a lot of binary choices (do you get the staff of fire or do I?) instead of "I'm low on teleportation scrolls, give me another 10"

I guess I wasn't using the KISS method when I approached the inventory issue. I'll implement a shared inventory system.

In fact, if this includes experience as well, then anything less than a 100% increase in the number of enemies is going to result in slower character advancement in two player mode.

I haven't played many roguelikes, but I do play alot of coop games, and IMO shared experience is the way to go. Anytime either player kills a monster, both get an equal copy of the experience (same with money - any money picked up gets duplicated between both players into separate wallets). It's not fun in coop games when one player starts outleveling another because the other character is a support class. Shared experience and shared money drops and shared inventory (but equipping an item makes it unavailable to the other player), but independant stats, skills, and upgrades. i.e. both characters level up at the exact same time, but each gets their own stats and their own skill points. Both get the same amount of money, but each has a separate wallet so they can freely spend it on themselves.

Often, I wish every game did this (even single-player) so I wouldn't have to worry about constant swapping of characters mid-battle to gain equal experience across the board. Thanks for the suggestions!

The reality is if you have 2 players you need to MORE THAN DOUBLE the number of enemies - most likely. This is because with players playing in co-op they can work together more efficiently than most AIs can. Basically two players can and will find ways to 'game the system.' and you'll have to compensate for that.

This includes things like foodclock mechanics, where you will LESS than double them. Why? Lets use Stone Soup as an example if you did co-op:

One player plays a Troll (or something that focuses only eats meat) the other plays a Spriggan (An herbavore) - If the food is increased, even a little bit, they're actually gaining a statistic advantage over the foodclock by playing co-op.

HOWEVER - all that said, the reality is you want your players to have fun. There IS fun to be had in thinking of creative ways to game the system together to make it easier! You may not want to go too far into chasing the difficulty in co-op, leaving that a more casual experience while single player will be the "normal" difficulty of a roguelike.