Matchmaking with 500 to 1000 players?

How I:B will handle this ? Did you find a good solution ?

Easy. With those sort of numbers per server there won’t be many servers, so you let players pick for themselves.


I think it is not that easy. When does the game start (min player required ?), When you stop accepting new players, how you balance multiple instances of 500 to 1000 players if you have between 3000 and 4000 permanent players a day etc…

After your clarification I see what you are asking.

Here’s something related, some of your concerns are discussed:

Currently it’s not clear how I-Novae will do this. A lot more has to be considered compared to a MP game with <20 Players per match or even <100 players per Match. Even with low playernumbers per match there are certain problems that can arrise.

It was hinted by Flavien that individual Battlescapes inside a “Full Match” may have quite different sizes and play times, required players. Battlescapes could be seen as “Sub Matches” inside a bigger “Full Match”. They could be used to compensate for fluctuating player numbers.
Now, bigger Battlescapes may require hundreds of players … and what happens if suddenly, half of them go watch a football match or something … many things that have to be considered.

Important things to keep an eye out in my oppinion are:

  • Critical Mass of Match and Battlescapes
  • Scaling
  • Incentive for people to join “empty” Server … kickstarting Matches / Battlescapes.
  • Team balance

the link is maybe for dev-tier support only :smiley: (no access with planetary support (if related))



Reposted from the dev channel (I originally posted it there but there is nothing private within it).

Its a sad fact that most multiplayer games that are released struggle to retain the player numbers required to keep their game servers populated. When numbers of players begin to drop it creates a vicious circle where people leave the game because they have no opponents to play against.

There is a great article by Dan Marshall that describes this problem better than I can:

There’s also a slightly more optimistic article by Daniel Cook that goes into much more detail about the things you should be considering when designing a multiplayer game:

What can be done to cross this hurdle?

There are a few options and none of them are guaranteed.

Option 1. Carry on and hope that the game is good enough and gets good enough reviews to sustain a healthy community of players.

The problem is maintaining concurrent active users. As a rough guide your daily active users (DAU) will be in the region of 25% to 50% of your downloads in the first few weeks after launch and (usually) tailing off from there based upon how good the game is. Your concurrent number of users (CCU) is usually your DAU divided between from 50 to 100. As the sessions on I:B will be normally be long we could estimate the lower end of that range.

That means to maintain 50 concurrent users in the first few weeks will require ten thousand downloads of the game (50 * 50 * 4). There are just under six thousand backers of the game on Kickstarter and even with additional downloads after release that puts the game in the danger zone for active users. Also bear in mind that is at the peak time of the first few weeks after release and the alpha and beta releases will spread out the concurrent usage even further.

All that taken into consideration, ignoring the problem and hoping it will be okay will only be successful if the game is so good that it bucks the trend of usage tail-off. That isn’t impossible but it is a long bet to make.

Option 2. Provide enough solo content that other players are not needed in order for users to have an enjoyable play session.

There is an aspect of this already in Battlescape as simply flying around and enjoying the scenery is a satisfying experience. Unfortunately the commitment to an offline exploration mode will only make the problem worse. If players are offline, they are not populating your servers and aren’t helping to form a healthy community of active players.

For this option to be successful there would need to be enough solo content in the online game to make it worthwhile being online when the servers are underpopulated. This would mean additional content in the game and the small development team is already stretched with their current game design.

Solo content could be of the form of bots to fight against or activities that advance your own side and give an advantage for when more players come online. Another option would be the ability to find things within the playing arena that give your craft a boost. For these activities to feel meaningful there would probably have to be some persistence between game sessions. I don’t think that kind of persistence is within the scope of the current game design.

Option 3. Design the game to concentrate players during certain time slots.

Another option if you are expecting low server population could be to concentrate your daily active users into specific times of the day. This could be done by scheduling battles at certain times and telling users when they are allowing them to come back at that time.

This often happen in an informal manner when much loved games are in their dying days. Members of the player community will organise events to experience the multiplayer gameplay as it was intended. I’ve not heard of this saving a game in the long run but it could help to extend the lifetime of a game.

1 Like

Repost from the Dev thread, posted on July 22 2016:

1 Like

Guys rather than reposting stuff let’s just move the thread. I’ve posted in that thread and I’ll just wait a couple of hours in case anyone has a reason not to move it.

Edit: I’ve moved the other thread. It’s quite long but please try to read the discussion that has already taken place before contributing to it. :+1:


You don’t. You keep the server open for people to join at any point during the match. With high numbers like this, a drop-in system’s the only real way to go. It’s not a good idea to have people waiting about in a lobby for the other few hundred players to show up and click a ready box, or have the player count steadily drop during a battle as people leave.


It starts when the server starts. If a single player joins they shouldn’t be prevented from spawning and doing stuff.
Maybe the NPCs don’t start skirmishing until a minimum number of players are on, eg 10.

Are you thinking they should be staggered so they don’t all start and end at the same time?
I don’t think that’s necessary or practical because matches will last different times depending on loads of factors.
To start with there’ll probably just be an NA server, then one in Europe… they’ll be added with demand.

1 Like