A bit of both.

A player building a normal sized base (one large enough to hold crafting tables and storage for normal play) is probably never on their own going to affect server performance. There is ways for a player to cause issues on their own, but these are very extreme builds and usually involve someone using knowledge of such to cause server hits. In these cases its very rare and next to impossible for a player to accidently do this.

Now where you see servers that have hard limits there is a pretty decent reason for this. On PC for example, it isn’t uncommon for popular servers to have very strict limits. Thinks such as 200-300 building piece limits per player in a clan and so many placeables, etc. The reason for this is these servers will routinely see 80/80(10)+ players on at any given time during the day. When you have communities large enough to be full like that every hour in a 24 hour day. You don’t merely have 90 players on the server. You have closer to around 1500-2500 or more.

Unrestricted building in that case you will see the entire map could potentially have over 10,000,000 building pieces and placeables to go with. Those servers struggle badly and in many cases don’t even load up properly. So when you have large communities you need to cut that down drastically.

On Funcom servers its hard to say how many people actually play. Many players in PVE for example do serial refreshing, where they will go weeks or months without actively playing but continue to log in once a week to refresh their decay timers. So a server that barely has a dozen players during peak could easily have a community of several hundred players.

Where this really causes lag on the server side is when multiple people login. You may have heard that server FPS tanks around 20 players. But not many really understand why. If those 20 players were in a single area and there is only 5000 building pieces to load, the server FPS should be around max.

But if there is 5000 pieces on one side, another 5000 on another, and another and those 20 players are in each one. Not only is the server loading around 100,000 building pieces, but also around a dozen different locations. And those players’ actions have to be relayed to the others logged in or at least to adjacent areas. That’s where the computation starts taking a dive.

When you play in single player, you could probably build 100 different camps in different areas with each being around 10,000 pieces. And you know what? You wouldn’t see any lag. SP loads only the host’s (player) area and not the others. So it is very efficient compared to multiplayer (this has other issues such as meteors and such not spawning). You likely won’t see lag, even on console.

There is something else that contributes to lag. Players leaving bodies in game when they log out. Anecdotal evidence on this suggests it inflates the database by 2-3x is size as if it didn’t have this setting checked. Again this is something singleplayer doesn’t deal with. FC should really consider turning this setting OFF on PVE servers.

Now everything said is for server side.

Client side is a bit different. Client lag is only attributed by the local area and the hardware’s ability under the settings selected to show everything in the immediate area. I’ve seen buildings with 50-100k building pieces not phase even weaker PC hardware. I’d assume console is likewise unaffected here.

What is affected is placeables. These have loads and loads of geometry to load compared to buildings, and they are frequently loading dozens and dozens of different types. Whereas a foundation used 1000 times doesn’t have too much effect outside of geometry count (and what the GPU has to show), it should simply be instanced in ram for every other feature attributed to it.

The real taxing one is those placeables with animations and those with both animations and lighting causing shadows. Hence why torches are recommended to be used sparingly, and if one has to light an area, to prefer using radium (as they have no animation attached). While they won’t use much ram or resources in themselves, their effects are unique per torch and it can bog down a system in short order if overused.

Unfortunately testing both things in singleplayer will get very skewed results. Need a live server to really get some decent testing in. And even then, doing it by yourself is not easy. Having built structures of 50k+ and more… even with creative that can take a long long time. And you really need to hit the million mark to really see affects on servers.

Its better to try and get a hold of a used database that had high traffic to really see what effects things have. I’ve personally not been able to lag a server by simply building using normal buildings.

But building a line of foundations from one end of the map to the other, and have enough people logged in to keep the whole line loaded in at once… and you’ll see something. Don’t do this on any live server that isn’t your own. Its a fast way to get banned otherwise.

1 Like