We know the next big stage of server meshing is dynamic. What can you tell us, if anything, about that?

Well, so the dynamism in the mesh. So we're running out of words internally to define what the next step of server meshing is. Currently, we're working on quasi-dynamic server meshing, which is we're trying to iterate on the static mesh to bring it so that if an area is empty and nobody's there, we can spin down a server. And if people go to that area, we can spin up a server. To achieve all this, all of the mechanics required for a full dynamic mesh need to be there. So like mass authority transfer between servers, because when a server goes out and it has X amount of entities that it currently has authority over, it needs to give that to another one, and obviously you don't want that process to lock up the parent server. So that whole mechanism is what we're working on at the moment. In fact, over the break, the team has actually reached a milestone. We were able to run with our internal testers a test of quasi-dynamic server meshing internally. So that's good. I mean, it's the first test, so a little bit rough, but it basically requested, I think, 200 servers to run the mesh. Our ops teams, Ahmed specifically, I got a call from Ahmed immediately.

So you did your first test of this quasi-dynamic. Like, hey server, you try to determine how much you need, and the first thing it did was ask for 200 more servers?

Yeah, exactly. A little bug in the assignment system. So that's what we're working on right now. That's the next step towards the full dynamic mesh. Obviously, the goal with the dynamic mesh, just to recap it, is that instead of having a subdivision that we pre-make in advance, like currently, every patch we look at the territory assignment, is what we call it, of what's coming: oh, we have an event at this place and most players are going to go to this system, and so we adjust the mesh, right? And sometimes you get it wrong, and it's a very manually intensive process and we don't want to keep doing that. So the goal with the dynamic mesh is to be able to just adapt to where the players are, the cost of the location they're in, to be able to provision compute dynamically up and down. So there's also the unfortunate cost equation in this, right? The current static mesh will always provision the server for every territory that's configured. And so there's a little bit of savings we can go and get there. But more importantly, the dynamic aspect of it is to be able to always make sure that the proper compute is available for players when they're in an area doing something. So that's the goal.