Jump to content

View more

Image of the Day

Adding some finishing touches...
Follow us for more
#screenshotsaturday #indiedev... by #MakeGoodGames https://t.co/Otbwywbm3a
IOTD | Top Screenshots

The latest, straight to your Inbox.

Subscribe to GameDev.net Direct to receive the latest updates and exclusive content.

Sign up now

Zombie MMO - Spreading Players Out

4: Adsense

Old topic!

Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.

  • You cannot reply to this topic
10 replies to this topic

#1 Legendre   Members   


Posted 24 November 2012 - 02:04 AM

Hypothetical zombie survival horror mmorpg: how we do spread the players out so they don't over crowd an area and trivialize the survival aspects?

I got interested in this topic while reading the other zombie MMO thread where this was brought up. Suppose everyone has to be in one world so we don't have the luxury of isolating players in separate games like DayZ.

Although I don't intend to make such an MMO, this is actually quite relevant to what I want to make: PBBGs - persistent browser based games. Usually, in PBBGs, players are not given the option to move around. They control some kind of base or kingdom with a fixed permanent coordinate. If you allow base relocation or make it so they control movable characters, you suddenly run into overcrowding problems.

One solution would be to set a quota for each area and artificially stop players from moving in once its full. Any other ideas?

#2 Waterlimon   Members   


Posted 24 November 2012 - 02:37 AM

make food come from farms and water from some other limited source etc. So if you live in a single place you need to bring food from far away defenseless farms. It should make more sense to have many smaller places centered around stuff like this.

Maybe there are ammo storages but its not easy to move the ammo so bases will form around then (too many players means ammo runs out soon)

Maybe theres some easier to defend areas but they cant fit enough things for many players (places to sleep, food storage etc.)


#3 Hodgman   Moderators   


Posted 24 November 2012 - 03:58 AM

In the context of a zombie game, you could penalize overcrowding by making crowded areas more dangerous, e.g.
* spawning more zombies in areas that have more players in them to mitigate the protection benefits of large groups.
* reducing item spawns in crowded areas so that large groups are more likely to starve to death by using up all available resources quickly.
* increasing the likelyhood that hordes of zombies will be attracted to such areas, causing players to become surrounded/trapped there (and teaching them not to put themselves in such a situation next time).
* allowing diseases to spread via human-to-human contact, making it more likely to become sick if you're in a large group of people.
* allow players to exploit each other -- e.g. 90% of the people in an area form a "guild", and kill/loot the other 10%. Repeat the procedure (kicking the weakest 10% out of the guild and looting them) until the group is sufficiently small.

Edited by Hodgman, 24 November 2012 - 04:01 AM.

#4 Mratthew   Members   


Posted 25 November 2012 - 12:09 PM

-multiple possible objectives that split up the group
-overrun players with a huge horde when a survivor cluster reaches certain sizes (demanding players run, to survive)
-limit the capacity of safe heavens (safe house cap. 12 players, mobile safe van cap. 6 players, ect)
-morally split people with right and wrong choices assigning them to certain objectives given the "moral" choice they choose

#5 Waterlimon   Members   


Posted 25 November 2012 - 12:21 PM

I think it largely depends on the game mechanics. Is it scavenging based? Do you want the players to set up bases, but only small ones? Do you just give the players weapons and a ton of ammo all around the map and let them run around shooting stuff?


#6 Telcontar   Members   


Posted 25 November 2012 - 11:09 PM

Lots of good suggestions above, and I'll sum them up with a major thought of my own: in the majority of apocalyptic stories, what prevents the survivors from living in harmony with one another? (hell, in the real world, what prevents us ALL from living in harmony?)

Answer: Resource scarcity.

Without modern technology, we could only sustain about one twentieth of the current world population - and that is without the undead thirsting for our flesh. There won't be enough food, water, fuel, weaponry, etc etc etc, for everyone. Someone is going to have to go without. People will fight for these resources, ESPECIALLY when "going without" basically means "starvation."

Place a lot of pressure on a group and you will get a lot of fractures. People will go off on their own, hoping that they can find more resources away from the giant group of resource-hogs they are currently with. Often, they will be wrong (at least, according to the chances of most zombie survival games).

I Create Games to Help Tell Stories

#7 Ashaman73   Members   


Posted 26 November 2012 - 07:59 AM

The simple recipe: people must be more terrifying than the zombies and you need a way to detect people around.

The first is easy, people will kill each others really soon. The way to detect others people can incoperated in the game. Whenever a player chat you hear a radio noise or see a visual,fading icon on the minimap, getting clearer when getting closer to the source. Larger populations would need to utilize the chat more often, leaving enough warnings to approaching peoples.



Gnoblins: Website - Facebook - Twitter - Youtube - Steam Greenlit - IndieDB - Gamedev Log

#8 Iron Chef Carnage   Members   


Posted 01 December 2012 - 12:07 AM

Item scarcity, PvP and scaling difficulty are all good ways to influence player distribution, but as long as you have finite space you can't really lock down player density without capping the server.

Are you committed to solving this through gameplay considerations, or are you open to some technological cheating? Here's something I've been noodling with, using the Chernarus map from Arma2 (default DayZ map) as an example:

Step 1: Divide the world into zones. Do this in an organic way, so the boundaries are not obvious to the players. If I walk east from Kamenka (far SW of the map) to Chernogorsk, I'll cover about 5km of ground. Imagining that each square kilometer is a "zone", I'd start in the Kamenka zone, go through a wilderness zone, then hit the Komarovo zone, the Balota zone and finally approach Cherno through another wilderness zone.

Step 2: Give each zone an approximate "target population". Kamenka might generally have 5 or 6 players in it, the wilderness only 2 or 3, and Chernogorsk 50 or more. Playtesting and prototyping will refine this idea and polish it up.

Step 3: Aim for the target population using a blend of matchmaking and social network algorithms. When I spawn in Kamenka, I'll be sharing it with a few random dudes. If I have a "buddy" online in that area, he'll spawn before a random does, so I can link up with my buddies without having to relog a million times. This will work like drop in/drop out multiplayer matches in your favorite FPS. If a guy heads north to Zelenogorsk, then the server will try to keep Kamenka's visible population in the target zone either by spawning in a new login or having the next guy who walks into town from outside get linked up with our instance.

Step 4: Think ahead. Track all adjacent zones to maintain fluidity and prevent glaring shifts in the immediate vicinity.

Some examples of how it can work and not work:

Example 1: Managing population. As a given instance of a zone approaches its target population, new players who approach that zone will become less and less likely to be put into that instance. Like a casino's poker room, the goal is to have just the right number of "tables" open that every player gets a seat and every table is filled. If you have 5 instances with 2/10 players, then they'll start to get smashed together, with players bumping into each other. If you have 3 at 10/10 and one at 1/10, some shifting is called for. Tastefully timing the transfers and picking who moves becomes important to avoid glaring "pop ins" and "pop outs".

Example 2: Picking who moves. In addition to tracking the population of each zone, tracking each player's interactions becomes important, so they don't feel like people are ghosting around. Giving interplayer relationships a weight will solve this. All interactions, from a sighting to a trade to a gunfight, will build up a connection between the players in question. Obviously, banding together or becoming in-game friends will also be a factor, as will player reviews. Ideally, people should be able to use the same kind of "prefer/avoid", personality profile and party system that influences their Halo matchmaking to guide the roster of people that they get saddled with when they enter a new zone.

Example 3: Travelling together. Since a player is simultaneously present in more than one zone, for purposes of pre-loading and priming matchmaking, it makes sense that players in close proximity will remain together as they move through the zones. Whether it's cooperative or adversarial, it's important to be able to keep the same people together when they're in the same neighborhood.

Example 4: Overpopulation. This is what the thread's about after all, eh? This is where I stop rambling and start talking about the solutions to the problem, albeit in the context of this idea rather than in a more general sense. There are two main problems to be addressed: Stagnant zones and armies. The three ways of handling them have already been addressed in this thread: Resource scarcity, PvP and difficulty scaling.

A stagnant zone is a zone instance that's at its target population, so that nobody new can get in, but all the players are affiliated with one another so strongly that the matchmaking heuristics won't break them up, i.e. in a party. Item scarcity's the obvious fix here. When an instance gets stale, stop spawning loot/food/ammo/whatever it is that people are looking for, and the herd will move on to greener pastures. If that doesn't work, difficulty scaling's a great solution, just start spawning more and tougher baddies until they're literally run out of town. Finally, the PvP option is closely tied into the army problem, so I'll just go talk about that.

An army is a large group of socially connected players that's steamrolling like a boss at all times. They're big enough that they can take control of high-population instances by forcing out the unorganized survivor rabble, and in smaller zones they get an instance all to themselves because their sheer size confounds the matchmaking algorithms and the server winds up just spawning a new empty countryside and plopping them in it by themselves. Item scarcity can do wonders here, since armies march on their stomachs, and when a pack like that gets flagged by the system, you can just reduce the loot they get so that it's tough for them to resist splitting up into smaller squads and spreading out over the map. Scaling difficulty won't be as effective here, since forty guys are a match for most anything zombies can throw at them, so unless you're willing to ramp up zombie HP and damage to Concordokken levels, that's not a good fix. The best fix, and my favorite, is the PvP Armageddon Maneuver.

That's where you identify these armies on the fly and use the power of the matchmaking and zones to smash them into one another. Nothing more beautiful than two half-starved bands of apocalypse survivors going bananas on each other. It's epic, it's fun, it's going to be on YouTube the next day with hilarious TeamSpeak throughout. Best of all, the bulk of your players will never know it's happening, because the server took it upon itself to build these mooks a private Thunderdome to have it out in.

It accommodates a vast array of play styles, it allows population density to be managed on the fly, and it lets every player have a chance of meeting every other player, without having to build a world big enough for them all to share at once.

#9 aattss   Members   


Posted 01 December 2012 - 01:15 PM

Scarcity. There can only be so much resources in a certain group. If you have better/higher level/better geared players then you could have players go on scavenging trips to feed the other players, but even then there's still a limit of how many people you can feed based on efficiency.

#10 ManuelMarino   Members   


Posted 03 December 2012 - 11:08 AM

resource scarcity is the more obvious, realistic and also fun. This can lead to interesting trading aspects between the players (I give food, you give ammo), also alliances and fights.

I remember a game of the past where scarcity was very important, but don't remember the name, anyway, a post apocalyptic game.
Electronic, Hard House, Film Music

88 preview tracks to listen to online + artist forums

And my projects Vanethian, and X-tivity Factor

#11 ShiftyCake   Members   


Posted 10 December 2012 - 03:35 AM

There are some nice ideas here, but in zombie games we are aiming for the humans to "group together" or "help each other out" rather then punish them for it. Anyone watch resident evil? If you payed attention to those movies, you would realize a mechanic in them is that zombies have a sort of "human radar" as such. That is, the longer one stays in a certain place, the more zombies swarm over their establishment. This is because they sense the humans, and move to their location. therefore moving constantly is the best option, which does not discourage grouping together (just discourages staying still).

Nothing else needs to be changed, just use that mechanic for the problem.

If, at any point, what I post is hard to understand, tell me. I am bad at projecting my thoughts into real words, so I appreciate the knowledge that I need to edit my post.


I am not a professional writer, nor a professional game designer. Please, understand that everything you read is simply an opinion of mind and should not, at any point in time, be taken as a credible answer unless validated by others.

Old topic!

Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.