Density story is not good, people would do very “long” bass to have many slaves, and the problem would only be bigger, I’m also angry with this gameplay “limitation” story especially because I play at PVE.
But I don’t see this ‘density’ story as a way out, as this is an insent to fill the map even more with sandstones!
I don’t see the incentive to fill the map with more bases. I think you’re mixing up the cause and effect: people don’t have bases because they want to put thralls in them, they gather thralls so they can populate and/or defend their bases.
The two biggest non-malicious uses I’ve seen for follower thralls are:
Defending the base from purges and raids. If you’re using thralls for this, making your base larger makes no sense, because it only increases the attack surface without offering additional protection.
Aesthetics, such as having an inn with an innkeep, a few dancers and one or more guards; or having a zoo with pets; or a prison with guards and prisoners. If you’re using your thralls for this, making your base bigger also makes little sense, because the architecture and decoration of your build is more important than preserving thrall density.
Of course, if we’re talking about players who use their thralls maliciously – for trolling, griefing or exploits – then they might resort to claiming more territory, but that’s a different problem that won’t be solved by any technical solution. Griefers and trolls always find new ways to bother decent players.
I’m genuinely curious as to why not? Back when I worked in gamedev, we used a variant of CryEngine, so I don’t have experience with UE4. Is there any specific UE4 limitation that makes it unfeasible to do a simple radial gather operation on actors in a fixed radius to count them? Or is it that you doubt that reducing the density would reduce lag and improve server framerate?
That one. Local lag sure, but server calls? That one I’m more iffy on. AI is frozen when not in vicinity of a player, so it has that going for it (standard stuff of course). But what happens on the back end besides that? I have no idea, AI and server load is a bit beyond me (not afraid to admit that either). I’m thinking various back end things are still running though, probably can only limit it so much when not in action. Once Follower AI and features starts to get more complex and more fun tools are given to the player (and one could argue that the leveling system alone adds to this), that will require yet more performance demands.
Something’s gotta give eventually, has to be sacrificed to introduce more complex systems. Especially for consoles and players who run on lower end machines, who are already at the edge as it is. Luckily I’m not a coder so I don’t have to hurt my brain too hard to think on what needs sacrificing, I just use the tools that are available to make fancy stuffs
I’m aware, thanks. UE4 is pretty great, but it still relies heavily on it’s single game thread. Newer versions of the engine are better at this (guess we have one reason to thank Fortnite), but it’s still using the same core principles as the original UE. You can have the worlds greatest server with 15,000 cores, and if you are lucky, you might see a small performance difference with most UE4 based multiplayer games (though to be fair, it’s not just UE4 that has this problem).
That’s true, the server framerate part is iffy. Like you said, and like Funcom has explained on several occasions, the simulation for the actors on the server is not running when there’s no client in the rendering distance. They also explained that when someone enters the distance, they run the “catch-up logic” for those actors; so for example, they’re not ticking down the decay timers for the buildings or the thrall breaking timers in the wheels if there’s nobody there, but when someone gets there, they take the current timestamp and the last simulated timestamp and calculate the outcome based on the difference (the building decays into dust, the thrall goes from 10% done to 80% done, etc.)
My theory – and it’s just that, because I neither have UE4 experience nor insight into Conan Exiles implementation details – is that the density cap should help improve server framerate, too, because you go from a maximum of 40 players multiplied by an arbitrary number of followers to simulate, to a maximum 40 players multiplied by a fixed number of followers that can be within those players’ render distance and have to be simulated. So if the density cap is 50 followers per whatever is the radius of render distance, then you end up having to simulate a maximum of 2000 (40x50) followers on the server. This has the benefit of: 1) putting a hard, deterministic cap on the maximum of followers to simulate, 2) making sure that the server framerate degrades gracefully instead of exhibiting spikes.
Again, like you said, there could be important details we don’t know about that would make this impractical or unfeasible or completely worthless. But FWIW, that’s my rationale for my proposal.
if there are 400 thralls in 1 spot glurin that is usually in association with a base and chests, crafting stations… ect if the server can’t keep up with that you will lag regardless of whether your using a high end pc or a console.
I think you assume wrong. While I can’t speak for Funcom’s specific process in detail, a very significant amount of the work (code, art etc) is done once, for all platforms. Anything else would be much too expensive. The split into different builds (PC, Xbox, Playstation) probably comes very late in the process, and with the minimum amount of work required for each platform.
The reason we see PC patches drop quicker is not so much that they’re faster to create, it’s more to do with the deployment (console patches have to be verified by Microsoft / Sony before they’re available).
It would be nice and commercial that Funcom provide a tool, the time of the transition, so that players can remove thralls so they can keep their favorites. It would be more justice for those who have not abused, and who may be losing their favorites.
If this is the case it seems like it would be even easier to put set limits on console for the same features available on PC. As you wouldn’t be coding everything separately and could just put capped values on followers and such on console before converting it to the console version.
Given that these limits are set by the server owners, they wouldn’t even need to adjust them in the builds at all, it’s just a matter of setting it up the servers. Whether they want to differentiate between official console and official PC server settings is another matter, of course.
Private servers and single-players are already up to the server owner’s preference, so no change there.
I’ll watch the recorded video once I’m free from work to see if they properly addressed the Purge defense issues, but I really, really doubt it. The problem I have is not that I need strong thralls for defense, it’s that I need to place enough of them to cover all approaches. Since the AI is really buggy and dumb, and the target acquisition range for thralls is short, there’s no way I can cover the whole base with only a handful of thralls. Ideally, I could reduce the number of guards for my main base from 24 to 6, but that’s only if: 1) the AI can pick up targets from farther away than now, and 2) the Purge can’t spawn inside my base anymore. If #2 is still buggy, I have to use at least 12 thralls.
And that doesn’t even begin to cover the question of bases like this one (not mine, btw).
So yeah, I believe the most passionate PVE players who buy all the DLCs to build nice things are screwed. Then again, I know Funcom doesn’t care about that. DLCs are not really their source of income, they’re just a hook to reel new players in and maintain some interest in existing player base.