KSP 2 Multiplayer Discussion Thread

[Laxez]

The fact is that creating a stable console version is easier, since the console components are standard, while for PCs, it is practically never like this, to confirm the thing, on consoles there are never the graphical settings. it must also be considered that ksp uses so many commands that even if it is true that half are unused, they are still more than the buttons available on the controllers which makes the creation of video games on consoles a bit of a nightmare for the additional UI.

So it is normal that there are more bugs on pc than on console.

[Vl3d]

Just a FYI: I've been looking into how massive multiplayer is implemented in a game called Foxhole and I think there are technical gameplay elements that could be an inspiration for KSP (maybe in the future).

I'm taking mostly about the competitive aspects set in a persistent world and the specialized individual / team-based gameplay / the technology progression, research and base building elements. I think it makes for an interesting dynamic.

Edited October 13, 2022 by Vl3d

[Deadmeat24]

7 hours ago, Vl3d said:

Just a FYI: I've been looking into how massive multiplayer is implemented in a game called Foxhole and I think there are technical gameplay elements that could be an inspiration for KSP (maybe in the future).

I'm taking mostly about the competitive aspects set in a persistent world and the specialized individual / team-based gameplay / the technology progression, research and base building elements. I think it makes for an interesting dynamic.

Foxhole is not like ksp at all though. It is a logistics management and top down shooter game...

[JoeSchmuckatelli]

12 minutes ago, Deadmeat24 said:

Foxhole is not like ksp at all though. It is a logistics management and top down shooter game...

I think the point is about how persistence works in a multiplayer game where players can change the environment while online... and then come back later.

That aspect alone isn't sufficient to deal with one of the cornerstone features of KSP - specifically the ability to warp.

I've been consistently leery of KSP as an MMOg.  Like it really seems weird for a player to be able to join a game at the beginning of 'their' space program, and by the time they make their first Mun/Minmus landing... the moons are already populated by multiple bases by other players.

The 'small dedicated server' or rather peer-to-peer 'multiplayer'/co-op seems more feasible - but even then the problems of dealing with warp seem difficult to overcome.

[Deadmeat24]

2 minutes ago, JoeSchmuckatelli said:

I think the point is about how persistence works in a multiplayer game where players can change the environment while online... and then come back later.

That aspect alone isn't sufficient to deal with one of the cornerstone features of KSP - specifically the ability to warp.

I've been consistently leery of KSP as an MMOg.  Like it really seems weird for a player to be able to join a game at the beginning of 'their' space program, and by the time they make their first Mun/Minmus landing... the moons are already populated by multiple bases by other players.

The 'small dedicated server' or rather peer-to-peer 'multiplayer'/co-op seems more feasible - but even then the problems of dealing with warp seem difficult to overcome.

Yeah that's the part that I don't get. If you have 1000 people playing multiplayer together, how do you account for time warp? How do you account for the fact that there will be bases all over the moon, thus limiting the places to make one for new players? It just doesn't work.

[Vl3d]

1 hour ago, Deadmeat24 said:

If you have 1000 people playing multiplayer together, how do you account for time warp? How do you account for the fact that there will be bases all over the moon, thus limiting the places to make one for new players? It just doesn't work.

It works if you just don't use on-rails and physics time warp close to celestial bodies (like KSP1 doesn't allow on-rails inside the atmosphere). Will not start this discussion again, it's all theorized in another thread.

But yeah persistent world multiplayer is cool.

1 hour ago, JoeSchmuckatelli said:

Like it really seems weird for a player to be able to join a game at the beginning of 'their' space program, and by the time they make their first Mun/Minmus landing... the moons are already populated by multiple bases by other players.

But that's how it works when you create a new space company in real life. Besides, think about how cool it would be to progressively discover what other players have built or left behind. It's better than having in-game aliens.

[Deadmeat24]

4 minutes ago, Vl3d said:

It works if you just don't use on-rails and physics time warp close to celestial bodies (like KSP1 doesn't allow on-rails inside the atmosphere). Will not start this discussion again

That doesn't account for 2 people being around a celestial body at the same time in real world time, but the celestial body being in different places in the star system due to the in game time being different between the two players because of the time warp outside of the celestial body's no warp region.

[Vl3d]

2 minutes ago, Deadmeat24 said:

in game time being different between the two players

In game time would always be the same for players who can directly interact because it would be controlled by the server for each celestial body. Syncing would be required to join multiplayer and it would be done by looping in orbit or when landed.

[JoeSchmuckatelli]

Every 'solution' just seems more complicated than the last.

I'm still banking on:

• MP will be a 'feature to be developed later' upon release (much like KSP), or
• Peer-to-peer co-op with agreed upon warps, and a 'universal' clock that doesn't let P2P players warp past another's alarm.  (With the possibility of a DC'd or offline player's 'warp plan' being sidelined until they are next online; other players would then get to warp regardless of the flight plan in place for the offline player, and the next time they're online their ships come out of stasis and can execute the last saved plan)

Edited October 13, 2022 by JoeSchmuckatelli

[Deadmeat24]

6 minutes ago, Vl3d said:

In game time would always be the same for players who can directly interact because it would be controlled by the server for each celestial body. Syncing would be required to join multiplayer and it would be done by looping in orbit or when landed.

That's a lot of extra work just to play multiplayer. Not to mention the fact that syncing in order to go onto a celestial body could completely ruin any return trip plans, as the body's location in the system would move. If that's how it worked I think I probably wouldn't play it. There would certainly be a fair number of people (mainly the ones who want to play with their friends specifically) who would just not play multiplayer in that format. The changes to how the game functions in order to turn it into a mmo would hamper them actually playing the main portion of the game.

[Guest]

30 minutes ago, JoeSchmuckatelli said:

Every 'solution' just seems more complicated than the last.

Most solutions are harder to explain that they are to prototype and test on the current build of the game.

 

31 minutes ago, JoeSchmuckatelli said:

Peer-to-peer co-op with agreed upon warps, and a 'universal' clock that doesn't let P2P players warp past another's alarm. 

This doesn't solve the problem, it's simpler to implement, but only because it offloads all the complexity on the player's shoulders.

Voting for timewarp is more punishing people for wanting to play multiplayer than a true KSP2 multiplayer.

I can already do 100% of that in KSP1, it's called sharing the save on Dropbox while every player streams his monitor on Discord to the others. The turn-based gameplay is the same, it's only slightly more automated (but still requires the same level of communication and organization).

[JoeSchmuckatelli]

59 minutes ago, Master39 said:

solutions are harder to explain that they are to prototype and test on the current build of the game

Well - we are expecting a new Hype video to drop in a matter of weeks... One can hope it addresses MP and proves my trepidations unnecessary 

[Strawberry]

I seriously doubt that devs are going to release it latter because they know that unkept promises look bad, and they had several chances to go okay we had to cancel multiplayer heres why, and doing that wouldve looked far better in the long run. Likely theyll either use a hybrid of some of the solutions listed, or just stick to one and accept the flaws.

[Rutabaga22]

If there is anything that prevents freely warping multiplayer will be awful IMO

There is no way I am going to do the matt lowne no timewarp mission because a few people would rather have a kerbal run a circle around kerbin.

[Nazalassa]

Maybe a simple form of multiplayer, with only master-controlled TW?

[Guest]

29 minutes ago, Nazalassa said:

Maybe a simple form of multiplayer, with only master-controlled TW?

I was thinking the same, but I guess it will only be ideal for a small group of friends or players

Edited October 13, 2022 by Guest

[Deadmeat24]

4 minutes ago, Forked Camphor said:

I was thinking the same, but I guess it will only be ideal for a small group of friends or players

I suspect that is the type of multiplayer they are aiming for. Could be wrong but it seems to make the most sense to me, as the more people you add the more complex solving the time warp problem gets.

[Nazalassa]

Let's see -- in Minecraft there is time war (kinda).

Spoiler

It's called: "all the players are sleeping"

[Vl3d]

3 hours ago, Deadmeat24 said:

Not to mention the fact that syncing in order to go onto a celestial body could completely ruin any return trip plans, as the body's location in the system would move.

No it doesn't.. everyone sees his own correct configuration of the system when away from the body because they are in single-player gameplay during deep-space travel. Multiplayer would be available only after arriving and syncing to the server controlled area around the celestial body. Anyway I'm not falling down this rabbit hole again. The idea is simple: massive multiplayer persistent worlds KSP2 is possible without any of the game breaking issues people invent. A lot of players are so used to KSP1 and the solutions in mods that they cannot imagine other new way of doing things. Paradigm shift and accept that KSP2 WILL BE MULTIPLAYER AT ITS CORE.

1 hour ago, Rutabaga22 said:

If there is anything that prevents freely warping multiplayer will be awful IMO.
There is no way I am going to do the matt lowne no timewarp mission because a few people would rather have a kerbal run a circle around kerbin.

Use on-rails warp as much as you like. Multiplayer only requires not being able to physics warp when playing with other people. The very simple solution to this is: build faster vehicles!

Why do you guys think they are testing race cars?!

Edited October 13, 2022 by Vl3d

[Guest]

18 minutes ago, Vl3d said:

KSP2 WILL BE MULTIPLAYER AT ITS CORE

What? Confirmed by the devs?

[Pthigrivi]

I think persistence is absolutely critical so players can build in coop together, I just think all players need to be able to timewarp at-will any time they like or the game is basically unplayable. Even a single orbit around kerbin takes 30 mintues. If Im trying to dock with a station Im not waiting 30 minutes for it to come into position for launch, and then another 30 minutes while I approach. If I just got back from Minmus and happen to have entered orbit on the other side of Kerbin from the target station Im certainly not waiting 6 hours to either loop out or circle in a lower orbit so I can dock. Even if Im tinkering around in the VAB in the meantime the pacing would be excruciatingly slow. 

Edited October 13, 2022 by Pthigrivi

[Rutabaga22]

4 minutes ago, Pthigrivi said:

I think persistence is absolutely critical so players can build in coop together, I just think all players need to be able to timewarp at-will any time they like or the game is basically unplayable.

I think the best method would be what I suggested earlier here

 

[Guest]

1 hour ago, Nazalassa said:

Maybe a simple form of multiplayer, with only master-controlled TW?

1 is the player the others are just spectators. It's either that or the player controlling the timewarp just spend all their time micromanaging the timewarp schedule for the others (that are still playing in turns, but without voting and entirely relying on communicating with the King of Timewarp).

[MechBFP]

15 minutes ago, Master39 said:

1 is the player the others are just spectators. It's either that or the player controlling the timewarp just spend all their time micromanaging the timewarp schedule for the others (that are still playing in turns, but without voting and entirely relying on communicating with the King of Timewarp).

*about to land*
*time warp activated*
”lol sry!”

[Pthigrivi]

18 minutes ago, Rutabaga22 said:

I think the best method would be what I suggested earlier here

 

I think this is in the right direction. I also like t_v's stab at the problem, it just has a different set of pros and cons. In practice what you're describing creates basically KSP Git, which might be nuts, but at least from my experience managing CAD files across multiple collaborators isn't a terrible solution.