KSP 2 Multiplayer Discussion Thread

[Pthigrivi]

One worry I have is that the demands of multiplayer might make single player a less interesting game, mainly as it effects habitation and consumables, be they LS or even just power requirements. 

Say Player A has built a station orbiting Laythe and allowed Player B access.  The station is nuclear powered with 2 years of fuel. Player B leaves Kerbin and time-warps 3 years to reach Laythe. Do they find the station bricked? If so, are there negative consequences for leaving a station or colony without power for a year? And if not, does that mean there’s no point in providing power or snacks or anything else for habitation? Maybe those requirements and other progression mechanics are just turned off for multiplayer, but Id hate for the single player game to be constrained by that. 

Edited April 9, 2022 by Pthigrivi

[SkyFall2489]

It's all very complicated. However, my hope will be that player A has remembered to set up some supply chain, either from KSC or a base, sending nuclear fuels to the station. This supply chain feature has been confirmed available, so it's just a matter of player A remembering. And if player A doesn't, they'll know for next time

[Vl3d]

24 minutes ago, Bej Kerman said:

Wasn't the discussion about not being able to timewarp in orbit?

You play real-time on / in-proximity-of the celestial body when not on a inter-SOI journey.

When planning and starting a journey you exit the multiplayer real-time bubble and you time-warp.

There could also be a hybrid system in which you would be allowed to auto-warp from orbit to other orbiting space stations, similar to travelling to moons, but the server would have to deal with local system configuration desync, maybe by limiting your map view and hiding time or a hard-resync.

Most important things would be that other players don't see you teleport, you don't break causality, you don't have cheating-like advantages over other players like pseudo-time travel and you let the server control time, system positional configuration.

 

[Lettuce]

4 minutes ago, Vl3d said:

You play real-time on / in-proximity-of the celestial body when not on a inter-SOI journey.

Therefore you would have to wait for the right spot in your orbit around kerbin to start your burn, you can’t start a burn to the mun from just any point in that orbit you have to wait for your craft and the mun to form the right angle which could take anywhere from 5 to 40 minutes

[Vl3d]

59 minutes ago, Pthigrivi said:

The station is nuclear powered with 2 years of fuel. Player B leaves Kerbin and time-warps 3 years to reach Laythe. Do they find the station bricked?

Logically yes, if it's not self sustainable or there's no supply chain. You have to first create a supply chain, like mentioned by @SkyFall2489

But in the system I suggested time / solar system configuration varies for each celestial body. There would objectively be no centralized solar system time that you as a player control. But you could have your own game time that is used to calculate resources usage.

There would be multiple parallel present(s) representing each celestial body's real-time bubble current time, each controlled by the server. When you switch craft or colony from one SOI to another you travel back and forth in time as decided by the server. So how do you fix this desync in gameplay? I think easily: the server just hides the local time from the player and show "remaining resources estimated time".

The thing is that it doesn't really impact multiplayer. Other players can't see inside your craft or colony to check if kerbals are alive or dead. They just see the buildings / crafts on rails. So you're actually playing single player KSP in an MMO context and you're not breaking causality.

Because the server controls time-regions, it can simply calculate resources based on your personal game time. The Kerbals without supplies perished for you. When you arrive you just sync you to the real-time bubble and find them dead. Causality preserved.

 

Edited April 9, 2022 by Vl3d

[Bej Kerman]

47 minutes ago, Vl3d said:

1 hour ago, Bej Kerman said:

Wasn't the discussion about not being able to timewarp in orbit?

You play real-time on / in-proximity-of the celestial body

Vl3d, not allowing timewarp just isn't something I want to do. I haven't seen anyone in support of this idea yet.

47 minutes ago, Vl3d said:

There could also be a hybrid system in which you would be allowed to auto-warp from orbit to other orbiting space stations, similar to travelling to moons, but the server would have to deal with local system configuration desync, maybe by limiting your map view and hiding time or a hard-resync.

If I understand this right, you're suggesting we just allow timewarp anyway?

Edited April 9, 2022 by Bej Kerman

[Vl3d]

1 minute ago, Bej Kerman said:

If I understand this right, you're suggesting we just allow timewarp anyway?

Yes but only when on a journey, not in the multiplayer real-time bubble.

[Guest]

56 minutes ago, Pthigrivi said:

One worry I have is that the demands of multiplayer might make single player a less interesting game, mainly as it effects habitation and consumables, be they LS or even just power requirements. 

Say Player A has built a station orbiting Laythe and allowed Player B access.  The station is nuclear powered with 2 years of fuel. Player B leaves Kerbin and time-warps 3 years to reach Laythe. Do they find the station bricked? If so, are there negative consequences for leaving a station or colony without power for a year? And if not, does that mean there’s no point in providing power or snacks or anything else for habitation? Maybe those requirements and other progression mechanics are just turned off for multiplayer, but Id hate for the single player game to be constrained by that. 

I would say that whatever system they use for time warp unloaded crafts, stations and bases would work on a "last seen" basis.

Let's say you build a Duna base in year 1, I use it in year 2 to launch my 10 year jool mission, I do my mission and now I'm on Laythe, on year 12 but the Duna base is still "frozen" at year 2, you would have to timewarp up to year 2 after I left to use it, but not all the way to year 12.

Obviously if you timewarp to year 12 to interact with me with your previously launched Laythe plane you wouldn't be able to go back on the Duna base at year 2, but you can't warp back in time in KSP 1 either.

[Bej Kerman]

2 minutes ago, Vl3d said:

4 minutes ago, Bej Kerman said:

If I understand this right, you're suggesting we just allow timewarp anyway?

Yes but only when on a journey, not in the multiplayer real-time bubble.

The multiplayer real-time bubble only has to surround a vessel. There's no need for all this tedious designing a system built to hinder players and make mission times 100s of times longer in real time.

[Vl3d]

43 minutes ago, Lettuce said:

to start your burn, you can’t start a burn to the mun from just any point in that orbit you have to wait for your craft and the mun to form the right angle

You don't manually start your burn. In KSP2 you would just start your journey. You set a destination and maneuver nodes in advance. Then when you press Start Journey the game would automatically warp you to your maneuver nodes and do the burn for you, outside of the real-time multiplayer bubble.

[Bej Kerman]

1 minute ago, Vl3d said:

47 minutes ago, Lettuce said:

to start your burn, you can’t start a burn to the mun from just any point in that orbit you have to wait for your craft and the mun to form the right angle

You don't manually start your burn. In KSP2 you would just start your journey. You set a destination and maneuver nodes in advance. Then when you press Start Journey the game would automatically warp you to your maneuver nodes and do the burn for you, outside of the real-time multiplayer bubble.

What's the difference between this and manual timewarp?

[Guest]

5 minutes ago, Bej Kerman said:

The multiplayer real-time bubble only has to surround a vessel. There's no need for all this tedious designing a system built to hinder players and make mission times 100s of times longer in real time.

This, the physics rage already extends beyond visual range, the only thing past that that you could see is shiny dots if they implement some sort of distant objects enhancement equivalent, which would be nice to have but sincerely not worth losing all control over timewarp and transfers.

Edit: And, it wouldn't even have to be completely absent from the game, it could be just disabled for MP or limited to on-rail crafts in your timeframe.

Edited April 9, 2022 by Master39

[t_v]

 

56 minutes ago, Vl3d said:

Most important things would be that other players don't see you teleport

Umm, how does your solution accomplish this? When people go on journeys, they will vanish as they leave normal time. This doesn’t fix the immersion error that you were talking about. In fact, the subspace bubbles solution is the one that allows you to see ships that have already gone ahead still orbiting. If you want full immersion, subspace bubbles is the way to go. And if you are fine with seeing people warp (which doesn’t have to be people vanishing by the way, it can be them speeding up in warp or it can have a very slow fade) then there is no reason to keep real time in SOIs. The majority of missions will be between SOIs, so people popping in and out of journeys is inevitable.

[Vl3d]

1 minute ago, Bej Kerman said:

What's the difference between this and manual timewarp?

Other players don't see you warping / teleporting / hyper-driving on / in proximity of the celestial body. You don't get to manually travel back in time, you can only cancel a journey and return where you were. You don't break causality. Only the server controls celestial body configuration and local time, you do not. All players are in sync.

[Bej Kerman]

Immersion is barely worth anything, let alone doing away with timewarp or other such critical features.

2 minutes ago, Vl3d said:

10 minutes ago, Bej Kerman said:

What's the difference between this and manual timewarp?

Other players don't see you warping / teleporting / hyper-driving on / in proximity of the celestial body. You don't get to manually travel back in time, you can only cancel a journey and return where you were. You don't break causality. Only the server controls celestial body configuration and local time, you do not. All players are in sync.

Why does it matter? As I said, removing critical features for the sake of immersion is just futile. This being implemented would only destroy multiplayer. Can you imagine waiting hours just to get a space station rendezvous?

The golden rule of multiplayer is that it has to work with time warp, not around.

Edited April 9, 2022 by Bej Kerman

[Vl3d]

11 minutes ago, t_v said:

they will vanish as they leave normal time

No they don't. You see their projection starting a burn and leaving the SOI. You just can't interact with them anymore.

As for arrival, there's a huge difference between seeing people arrive from outer SOI and seeing them warp around on the planet or in lower orbit. You could just arrive outside the visual range of other players.

Edited April 9, 2022 by Vl3d

[SkyFall2489]

Well, what we can agree on is that It should be similar enough to KSP1 that KSP1 players will want to play KSP2.

I'm thinking we should keep timewarp about the same for this reason.

I'm also thinking @Lisias is being a little pessimistic about the whole DRM thing - plus, we could get the main KSP2 devs to get our other timewarp mods in as well. 

[Pthigrivi]

33 minutes ago, Master39 said:

I would say that whatever system they use for time warp unloaded crafts, stations and bases would work on a "last seen" basis.

Let's say you build a Duna base in year 1, I use it in year 2 to launch my 10 year jool mission, I do my mission and now I'm on Laythe, on year 12 but the Duna base is still "frozen" at year 2, you would have to timewarp up to year 2 after I left to use it, but not all the way to year 12.

Obviously if you timewarp to year 12 to interact with me with your previously launched Laythe plane you wouldn't be able to go back on the Duna base at year 2, but you can't warp back in time in KSP 1 either.

This is really the only good solution Ive seen. Because all your transfers depend on where bodies are in relation to each other you need to have a consistent causal timeline and players just leapfrog each other as they advance. 

[Vl3d]

It's somewhat understandable that players with synapses defined by years of playing thousands of hours of KSP1 would be resistant to and suspicious of change.

I feel like some people just refuse to let go of the God-like superpower of manually controlling time and solar system configuration in exchange of a cool multiplayer game. And this is why they can't change paradigm. It's normal.

22 minutes ago, Pthigrivi said:

58 minutes ago, Master39 said:

I would say that whatever system they use for time warp unloaded crafts, stations and bases would work on a "last seen" basis.

Let's say you build a Duna base in year 1, I use it in year 2 to launch my 10 year jool mission, I do my mission and now I'm on Laythe, on year 12 but the Duna base is still "frozen" at year 2, you would have to timewarp up to year 2 after I left to use it, but not all the way to year 12.

Obviously if you timewarp to year 12 to interact with me with your previously launched Laythe plane you wouldn't be able to go back on the Duna base at year 2, but you can't warp back in time in KSP 1 either.

This is really the only good solution Ive seen. Because all your transfers depend on where bodies are in relation to each other you need to have a consistent causal timeline and players just leapfrog each other as they advance. 

This is describing a single-kerbal game, right? Your game would be limited to where your character is physically, right? No managing colonies, no switching vessels, no KSP... All this because you refuse to let the server control the simulation.

Why can't you just switch colonies / crafts and go back and forth in server controlled state / configuration, not objective time?

I'll always repeat that in orbit time is relative, except for life support. And that can be fixed by setting an individual per-player inner game time which is used to calculate resource usage and stocks. It does not impact multiplayer because other players can't look at your colony/craft remaining resources.

Edited April 9, 2022 by Vl3d

[Bej Kerman]

1 minute ago, Vl3d said:

It's somewhat understandable that players with synapses defined by years of playing thousands of hours of KSP1 would be resistant to and suspicious of change.

I feel like some people just refuse to let go of the God-like superpower of manually controlling time and solar system configuration in order to have a cool multiplayer game. And this is why they can't change paradigm. It's normal.

Are you blaming us for not wanting mission times to become much, much larger for a bit of pointless immersion? I feel like we should be beyond saying things like "Vl3d refuses to let go of their idea" by this point. Didn't the mods say blame the opinion, not the person behind that opinion?

[Vl3d]

I'm not blaming anyone. I'm saying I understand why we are stuck in limbo talking about this time-warp superpower. I'm trying to explain a paradigm-shift that allows us to have huge multiplayer. Just let the server control the state of the in-game simulation.

[Bej Kerman]

6 minutes ago, Vl3d said:

I'm not blaming anyone.

Just last message you were saying we're not complaining because it's a bad idea, but because we're "resistant and suspicious to change". I can only translate this to "stubborn".

6 minutes ago, Vl3d said:

I'm saying I understand why we are stuck in limbo talking about this time-warp superpower.

Because timewarp is literally needed for KSP 1. It isn't a superpower. It's a critical feature; every space simulator has timewarp. Even before the Mun was a thing, KSP had timewarp simply because it is so important for gameplay. If you can find anyone else in the whole wide forum who doesn't use timewarp in Low Kerbin Orbit, I'd enjoy being able to converse with them too. 

6 minutes ago, Vl3d said:

I'm trying to explain a paradigm-shift that allows us to have huge multiplayer.

You don't need to remove time warp for huge multiplayer to work.

 

Edited April 9, 2022 by Bej Kerman

[Vl3d]

8 minutes ago, Bej Kerman said:

You don't need to remove time warp for huge multiplayer to work.

Yes you do, on / in-proximity-of the celestial body. It's my opinion.

[BowlerHatGuy3]

Hi I’m new to this thread, could someone explain how this theoretical multiplayer works or send me to a document?

[Bej Kerman]

2 minutes ago, Vl3d said:

12 minutes ago, Bej Kerman said:

You don't need to remove time warp for huge multiplayer to work.

Yes you do, on / in-proximity-of the celestial body. It's my opinion.

An opinion:

"I think there are ways multiplayer could work without time warp"

A fact:

"You need to get rid of timewarp near a celestial body for multiplayer to work"

Regardless of your opinion, you don't need to remove time warp for huge multiplayer to work. It's not an opinion. It's a fact you could probably affirm with a beefy server, 30 people and a Dark Multiplayer session.

Edited April 9, 2022 by Bej Kerman