Well, I mean, I would have launched it first (as an AAA game), but I’m no game developer. 🤷 And neither are they, from the looks of it. Good at perpetually raking in money for himself and his family, though!

    • LordKitsuna@lemmy.world
      link
      fedilink
      English
      arrow-up
      4
      arrow-down
      1
      ·
      10 months ago

      400 on a single shard (server) that actually quite a lot. It is far far easier to just throw a bajillion different servers at the problem and only have a relatively small player count per server. Having 400 running smoothly on a single server is a very impressive optimization achievement

      • n3m37h@lemmy.dbzer0.com
        link
        fedilink
        English
        arrow-up
        1
        arrow-down
        1
        ·
        10 months ago

        Have you ever heard of CCP? This small game called EVE Online? They’ve been doing this for the better part of a decade

        • LordKitsuna@lemmy.world
          link
          fedilink
          English
          arrow-up
          3
          ·
          10 months ago

          Correct and eve has been very impressive with its per server player counts as well. It’s a completely different type of Beast mind you, trying to keep that many players synchronized over something like a first person perspective real-time movement game is a completely different ball game from keeping spreadsheet simulators synchronized.

          Still a very good achievement of optimization regardless but definitely a completely different ball game from synchronizing a first person type content where the players are free to just move in whatever fucking weird ways they want rather than linear vector paths

        • EchoCT@lemmy.ml
          link
          fedilink
          English
          arrow-up
          2
          ·
          edit-2
          10 months ago

          Yeah eve handles it by slowing down time in-game. So each player has less actions for the server to handle per cycle.

          Every game has their way of handling it. Cig is doing it via the replication layer and dynamic meshing. IE multiple servers talking to a “boss” server that scales based on needed load without Eve’s crutch of time scaling. Totally different technologies.

          Eve’s solution worked based on what they had and needed at the time, but it’s old hat now.