🔒 Locked

Starting as an Extraction Shooter, then adding 5v5 and Survival modes later in Unreal? Is this a player-base/technical trap?

Started by YourFxV Jun 9 at 2:59 AM 1 replies 350+ views

Original Post

Jun 9 at 2:59 AM

Hi everyone,

My ultimate dream has always been to create a massive, realistic, survival game (like DayZ), but I also have loved the idea of things like extraction shooters (Escape from Tarkov), and 5v5 kind of games (like Rainbow Six Siege). I was wondering: what if I started a long-term project where I created a realistic extraction shooter (like Tarkov), then, once I get a larger team and more funding, I add a 5v5 mode (Siege) and a persistent survival open-world mode? (DayZ)

Is this realistic or possible? Or will changing the core game loops across the modes create an impossible amount of technical debt?

Would housing all three of these modes make the game too large and laggy?

Would offering three vastly different modes end up dividing the player-base too much and kill matchmaking/server populations? (for example there could be 500 people playing the actual game, which looks like alot, but split between part of that 500 on 5v5s, another part playing extraction, and another part playing in their own servers within the survival, that means there may be 500 total but you loading/matchmaking times would be quite long because of how spread out everyone is, (right?))

Anyone who has any advice or experience, please let me know. I'm kinda new to this all, so I'm not quite sure which direction to even go in! Thank you!

PS: I was going to make it a first-person, using blueprints within Unreal Engine!

frob

Jun 9 at 5:07 PM

YourFxV wrote:
what if I started a long-term project where I created a ...

You must design and implement with all the phases in mind. You can't implement features as though they're single player even if you're currently single player, they must be built as though it's multiplayer, happens to potentially be 5x5, but at the moment only one player is active.

It's easy for developers to accidentally paint themselves into a corner, that the game is always a certain way instead of happens to currently only have a special case, but must handle the general case.

To get working as you described, you must always consider that the game is a multiplayer game, there may be any number of clients in the world, and any game client can have any number of players. The easiest assumption both from early local multiplayer games where it was hard to sit around a screen, and from the design of game consoles, is that there will be up to 4 players per game client, and an unbounded number of game clients. Nintendo bumped that up to 8 local players. You can't just get player 0 and assume it will work because you know in the future it won't. You also can't assume that the game client has all the information, because you know in the future it's going to be distributed and clients only know a subset of the world, and clients always lag because you can't break physical world causality.

YourFxV wrote:
Would housing all three of these modes make the game too large and laggy?

That's not where those problems come from. To large is easily solved with sublevels and streaming. Laggy means you screwed up in other ways.

YourFxV wrote:
Would offering three vastly different modes end up dividing the player-base too much and kill matchmaking/server populations? (for example there could be 500 people playing the actual game ...

That is extremely ambitious for a hobby project game. 500 concurrent users is going to need on the order of 10K-20K daily users, which is going to need a half million or so total buyers.

Either be prepared for empty games, or you'll want to include server's ability to generate bots of varied skill level. The good news is that those bots give a good service during development, you can spawn bot-on-bot automated tests to catch an awful lot of bugs.

YourFxV wrote:
PS: I was going to make it a first-person, using blueprints within Unreal Engine!

The engine gives you enough to get going quickly and relatively easily. I've been in projects where we had 40 player matches less than a month after starting the project. It's certainly possible, although still a tremendous amount of work.

Topic Locked

This topic has been locked by a moderator. New replies are not allowed.