Multiplayer oxygen not included

[CaRat]

I feel like multiplayer would be cool like controlling your own set of duplicates.It would make the game more challenging and fun(resource managment) but obviously I can only be 2 player or 3 player because it would be crazy with 5 people. Judge if you want but you cant crush my dream that it might happen someday ;-; I dont read forums I'm sorry;-;

[harmonium]

Yes that would be nice, we can create a portal to invade other world, maybe we could make co2 grenade, and when you kill all the duplicant you win a lootboxe, and we need a battle royal mod too 

[Glassyfo]

Not this again.

[Yunru]

I feel like having an entire subforum for things like this would be cool like a place you could go to post this named "suggestions" maybe with an "and feedback" so it's not too specific and then everyone could go there and- oh wait.

[Nightinggale]

7 hours ago, Yunru said:

It's the ONI Drinking Game! Take a drink every time you see a repeated suggestion!

We are running out of booze at this rate

[SharraShimada]

This forum needs a script. Everytime someone tries to make a topic with "multiplayer", the forum skips and displays a big fat "NO".

[Nightinggale]

9 minutes ago, SharraShimada said:

This forum needs a script. Everytime someone tries to make a topic with "multiplayer", the forum skips and displays a big fat "NO".

But what if Klei is planning on a multiplayer DLC?

[metallichydra]

maybe we should just ban all players who write the word "multiplayer" in the forums?

oh no!

*metallichydra was banned from the ONI forums

[SharraShimada]

4 hours ago, Nightinggale said:

But what if Klei is planning on a multiplayer DLC?

They already said, ONI is a single player concept. And they would have to redesign every single aspect of the game, to make it multiplayer-ish.

ONI was never intended to be a multiplayer game. So i dont think, we will ever see anything like that. Maybe some modder will crash the whole code and make something. But when released, the dev team shrinks to the minimum, and then they wont have nearly enough manpower to overwork the whole game.

[Nightinggale]

6 minutes ago, SharraShimada said:

ONI was never intended to be a multiplayer game. So i dont think, we will ever see anything like that. Maybe some modder will crash the whole code and make something.

The biggest problem with multiplayer is to make sure all computers play the same game, as in avoid desyncs. This isn't an addon, but rather a part of the foundation of your game engine, which you plan for when you start. Adding it later is... let's just say it will require a lot of work from very skilled people.

In a way I'm happy to learn that network sync issues will never be an issue. It makes modding a whole lot easier.

[Yunru]

21 minutes ago, Nightinggale said:

The biggest problem with multiplayer is to make sure all computers play the same game, as in avoid desyncs. This isn't an addon, but rather a part of the foundation of your game engine, which you plan for when you start. Adding it later is... let's just say it will require a lot of work from very skilled people.

Eurgh, don't remind me.

When minecraft switched from single player to multi...

Well, I guess it could of been worse.

[Neotuck]

24 minutes ago, Yunru said:

When minecraft switched from single player to multi...

Well, I guess it could of been worse.

That's a bit of a stretch

[KittenIsAGeek]

5 hours ago, Nightinggale said:

But what if Klei is planning on a multiplayer DLC?

We've discussed the problems with taking ONI multiplayer in a good number of threads on this exact board.  To summarize:

1. ONI is not server-based, which would be necessary for most multiplayer methods.  This means a complete redesign of the game.
2. For the methods where a server wouldn't be necessary (such as trading resources), you might as well just activate sandbox mode.
3. If you think ONI is slow now, imagine the wait while each tile's newly calculated information is synchronized across the network with however many players are joining you on your game.  The best-case scenario is that at the end of each tick, all new information gets bundled and compressed and sent.  This means your FPS will be, at most FOUR frames a second, unless your client runs with out-dated information.
4. Who will pay for the servers?  This isn't Quake where most of the work is just rendering graphics which is done by the clients.  Your average PC won't be able to work reliably as a personal server, which means you'll need some high-end equipment somewhere.
5. In the simplest case scenario, simply letting your buddy watch you play, Steam already has a mechanic in place with their broadcast system.  However, that isn't exactly what anyone asking for multiplayer wants.  Even so, this is still somewhat of a problem because you're broadcasting a video stream with the resolution of your monitor at no less than 22fps.  I don't know about everyone else, but my cable provider limits my upstream to 5Mbps.  The h264 codec needs 10Mbps of bandwidth for a 1080p stream at 30fps.  

[Yunru]

1 hour ago, SharraShimada said:

They already said, ONI is a single player concept. And they would have to redesign every single aspect of the game, to make it multiplayer-ish.

20 minutes ago, KittenIsAGeek said:

We've discussed the problems with taking ONI multiplayer in a good number of threads on this exact board.  To summarize:

I do believe you may be missing the joke?

That Klei would have to ban themselves be unable to post in such a case.

[KittenIsAGeek]

Just now, Yunru said:

I do believe you may be missing the joke?

I did not realize the original poster was making a joke.  It seemed like they were asking a serious question.  Which has been addressed multiple times, so I decided to point that out.  If you are correct and the original poster was making a joke, then they need to be more clear about it. 

[Nightinggale]

Strictly speaking, a server wouldn't be mandatory. There are games, which works with decentralized calculations, meaning all computers will have to do the same calculations. To implement this would require:

• Random values has to be the same on all computers (predictable randomness is major issue in itself)
• User input (such as clicking a button) would have to be transmitted and received before taking action, meaning the interface will be affected by network lag
• If the game can't keep up, it seems to make certain less important calculations less frequently. This isn't possible in multiplayer

That's most likely just the tip of the iceberg, but it is a path for multiplayer, which will work without a central server. The original battle.net server (at the Diablo 1 time) was just a single computer serving the entire world because all it had to do was telling IP addresses to people who wanted to play together. The CPU load of the server would then be placed on a computer of one of the players. This is the primary reason why Diablo 1 multiplayer went to cheat hell.

Another issue I have with multiplayer ONI is the different speed settings. For the game to stay in sync, all players have to use the same speed at all time. You can't have one who is waiting for stuff to finish at max speed an another who paused to plan a build.

[KittenIsAGeek]

2 minutes ago, Nightinggale said:

Strictly speaking, a server wouldn't be mandatory. There are games, which works with decentralized calculations, meaning all computers will have to do the same calculations.

<snip>

 The original battle.net server (at the Diablo 1 time) was just a single computer serving the entire world because all it had to do was telling IP addresses to people who wanted to play together. The CPU load of the server would then be placed on a computer of one of the players. This is the primary reason why Diablo 1 multiplayer went to cheat hell.

Uhm... Just to be clear.. the PC hosting the Diablo 1 multiplayer game IS the server for that particular method.  Steam already has a method for finding friends implemented.  However ONI is not designed to run as a server.  An additional framework would need to be developed to facilitate: 

1. Synchronizing the clients.
2. Disabling/enabling mods uniformly across the clients.
3. Verifying data integrity (to reduce one client exploiting the others) across the multiplayer network.

Beyond that, Diablo 1 is a very simple game by comparison.  Most of the actual computation work was simply rendering the graphics.  Even then, there were problems with multiplayer even before you consider the cheating that became prolific. 

[Nightinggale]

32 minutes ago, KittenIsAGeek said:

Synchronizing the clients.

That's done by transmitting user input. Joining players will get a savegame to load, hence getting the game stat. However that just made me realize a gigantic problem. A bunch of stuff resets on load and we get all those unconnected buildings and stuff. This minor issue is a killer for joining a multiplayer game.

32 minutes ago, KittenIsAGeek said:

Disabling/enabling mods uniformly across the clients.

That should be part of the join code, which is obviously not written.

32 minutes ago, KittenIsAGeek said:

Verifying data integrity (to reduce one client exploiting the others) across the multiplayer network.

Mod support and anti-cheat can't coexist. I can write a mod, which has some feature, but the player with my type of CPU/GPU combo will also gain free refined metal buttons and stuff like that. ONI can't prevent that from happening. This is actually another very good argument for not adding multiplayer.

32 minutes ago, KittenIsAGeek said:

Uhm... Just to be clear.. the PC hosting the Diablo 1 multiplayer game IS the server for that particular method. 

Ok, maybe it was a bad example. Marathon is a true non-server game. Whenever a player presses a key, that action is transmitted... not to a server, but to all clients and it's the same for all. There is one host, which starts a game, other computers can join on that address, but once the map has been picked and game has started, the host is no different from the other clients.

[KittenIsAGeek]

I still think you're comparing Tic-Tac-Toe to Chess. 

[bobucles]

How can a multiplayer scenario even work? The player has hardly any direct control over the world. Instead they delegate tasks to duplicants. A multiplayer scenario would mean that multiple players can place down build orders and fight over priorities. But if an ordinary save/load struggles to maintain persistence, then there's no way an internet game will ever stay synced.

[Nightinggale]

12 minutes ago, bobucles said:

How can a multiplayer scenario even work? The player has hardly any direct control over the world. Instead they delegate tasks to duplicants. A multiplayer scenario would mean that multiple players can place down build orders and fight over priorities.

I imagine it would be something like this: https://www.eurogamer.net/articles/2019-01-25-survive-space-together-with-this-rimworld-multiplayer-mod

[Heinermann]

6 hours ago, Nightinggale said:

I imagine it would be something like this: https://www.eurogamer.net/articles/2019-01-25-survive-space-together-with-this-rimworld-multiplayer-mod

Some people just want to play together with their friend or spouse, and it's unfortunate that the community here continues knocking that idea down for whatever reason when it isn't relevant to them. Taking player commands and networking would be trivial and is not worth talking about (besides correction). Almost all of the work would be in making the sim more deterministic while improving sim performance, followed up with a sim correction algorithm to catch the edge cases and careless mods. You can ignore most people in this thread because they don't know what they're talking about. Once 1.0 is released we can have a serious discussion about a multiplayer mod in a separate forum or discord.

[Yunru]

1 hour ago, Heinermann said:

when it isn't relevant to them.

Huh. And here I thought how developers invested there time in the game is always relevant to everyone, given that it's a finite resource and all :/

[KittenIsAGeek]

4 hours ago, Heinermann said:

Once 1.0 is released we can have a serious discussion about a multiplayer mod in a separate forum or discord.

We have had serious discussions.  You're over-simplifying the problems involved.  Besides, multiplayer is not a direction that the developers wish to go with this game, so it would have to be a third-party mod -- and I really don't think that's feasible with this particular game.

[Heinermann]

5 hours ago, KittenIsAGeek said:

We have had serious discussions.  You're over-simplifying the problems involved.  Besides, multiplayer is not a direction that the developers wish to go with this game, so it would have to be a third-party mod -- and I really don't think that's feasible with this particular game.

We haven't. I simplified it to outline the issues that need to be tackled without going into detail. I mentioned mod because I agree, the developers won't be doing it. Using my experience I believe such a thing is feasible.

8 hours ago, Yunru said:

Huh. And here I thought how developers invested there time in the game is always relevant to everyone, given that it's a finite resource and all :/

And? If enough people want a multiplayer it's not anyone's business but the devs to tell them they can't have it unless by modding, then people can continue the discussion about what they want or could be cool in the context of modding instead of getting a toxic community reaction by people who feel the need to continuously post their uninformed opinions and efforts to derail the thread. That's why I said it's better to have a serious discussion in a completely separate forum outside the Klei community.