pours himself a cuppa
Alrrriiiiggghhhtty then.
Our Plans for Multiplayer
At release, Last Epoch will support both offline-only, locally stored characters and online-only, server-side characters played in a server-authorative environment. Competitive aspects of the game (such as ladders) will cater to the latter. While we don’t plan to ever allow previously offline characters to be brought online, we are open to the prospect of allowing players to download a copy of an online character for offline play. This was an idea I initially had when considering how best to drive online players to try offline play (and also the reverse - online or offline, we want people playing Last Epoch as much as is healthy and responsible!).
We do not presently have any plans for hideouts, player housing, or guild halls. While some or all of these may be included in post-launch updates, they’re very unlikely to feature in 1.0. You will see other players who aren’t in your group while in town hubs such as End of Time.
Loot will be instanced. Group leaders may be able to change this - I’m not 100% sure. I can say that it isn’t presently possible, however I’d advise not trying to read too much into that.
Discussion & Implementation
We recognize that it may take some time to get multiplayer ‘right’ - and I might remind folks that there isn’t presently a release window offered for when Last Epoch might leave Early Access.
While multiplayer will impact balance in ways that only become apparent when the community begins playing multiplayer en masse, at the risk of saying The Wrong Thing™, I do believe the concerns raised by some in this thread to that effect are exaggerated. The majority of skills, specialization tree nodes, and class passives will not need to be changed. A Mage casting Fireball doesn’t really care whether they are alone or part of a group.
One of the most significant factors in whether an online game is considered a success is how well the game servers, databases, and other infrastructure weather the storm of the community first using them. We thus want to be sure that the online version of the game is ready - and we are ready to adaquately support it - before making it available to the community. Fixing these problems after the fact is seldom enough to salvage a tarnished reputation.
Multiplayer Status
Currently, I’d say the most common group size in internal testing is three. This isn’t mandated by any technical limitations - it’s just a product of our team size and how many people we have currently participating in such testing. We typically perform testing with latency values ranging from approximately ~25 ms to approximately ~125 ms. The game feels very playable across this range, although not all of the niceties have been implemented yet.
One example of such a missing feature is the server skipping parts of a skill depending on your latency. We’re likely to cover this system in our Multiplayer-focused post later this year.
There is no difference in stability between our internal multiplayer-enabled builds and Patch 0.7.10c. (Technically the former may actually be better, due to being more recent.)
Our internal testing has involved servers from multiple providers. I believe we’ve decided which company we wish to partner with but haven’t yet signed anything to that effect (it’s possible that my knowledge here is outdated, but I doubt it - I expect it would be big news internally!).
We have not yet chosen a specific server model or manufacturer. To-date the servers we’ve used have been powered by a variety of Intel Xeon CPUs. I can’t promise we’ll announce which hardware we end up using, however I imagine we’d be willing to confirm that the decision has been made. We will be using identical hardware across all data centers in all regions.
P.S. On a semi-related note, we’re totally replacing in-game chat in Patch 0.8 later this year.