ehm.. oni multiplayer

[NeoDeusMachina]

1 hour ago, darknotezero said:

Throughout this thread, I think that more of what has been talked about that is "pro multiplayer" has been about conceptual implementation with some incorrect base assumptions about the ease of the technical/programmatic implementation.

I think it is important that we can have a discussion about our interests for multiplayer or no multiplayer, how we see it could work from a gameplay perspective etc. but it seems like a lot of the argumentation revolves around the technical feasibility in terms of coding and so on. If I am being honest here, I think it belongs to Klei to figure that out. They are developing the game and if they decide one day that they want to go ahead and develop a multiplayer avenue for ONI, whether it is a DLC or a completely separate game, I trust that they will think about how to implement it in a way that makes sense and that they will figure things out. I also trust that they will not jump into it if they don't think it is doable, worth the hassle, or that it won't be profitable for the company. Again, that is their responsibility to figure out - if they even want to.

I don't know how helpful it is to simply say "no" or to try to dismiss the idea because we as players, no matter our technical backgrounds, believe it is doable or not on a technical level. I am not sure how that really contributes anything constructive to the idea of how multiplayer could or could not work in game..

 

 

[Zarquan]

On 10/5/2021 at 1:32 PM, lee1026 said:

Honestly, I think multiplayer can be implemented easily. This is 2021, the world of pretty good internet options.

 

The host runs the sim. The game is literally unchanged from what we have now. The players can all issue orders, and the dupes will carry out orders without caring about who issued the orders. The non-host players just send requests to the host as to what is in the view window of the game, and the non-host players just render the game.

 

It would enrich the game a lot as ideas would mix in multiplayer game servers as people see each other's designs and can interact with them. Being able to work together on a base also seems like fun, especially as dupe count grows to be able to support more than a single project at a time.

I would use this in single player.  I would put the host on my main monitor and 4 clients watching jobs on my second monitor.  I spend way too much time staring at a sequential job waiting for my dupes to finish a step when I could be doing other things or I go off to do other things and forgetting about the important sequential jobs.

8 hours ago, pether said:

https://steamcommunity.com/sharedfiles/filedetails/?id=2612332708

Here, catch this. This is as far as ONI multiplayer can go. Anything that would allow for shared control of a base would end up being a disaster (confrimed by experiments already done on this forum) and Klei is doing right thing not even thinking about it. ONI is a singleplayer game and if you spent 10 minutes about thinking of all the things that could go wrong during multiplayer game you wouldn't try to force it into this game.

Honestly, that experiment was extremely limited.  Of course a colony managed by a group with no joint goals or communication would be a mess.  ONI is a game of careful planning.  I don't think anyone wants ONI to be an MMO.

[darknotezero]

1 minute ago, NeoDeusMachina said:

I don't know how helpful it is to simply say "no" or to try to dismiss the idea because we as players, no matter our technical backgrounds, believe it is doable or not on a technical level. I am not sure how that really contributes anything constructive to the idea of how multiplayer could or could not work in game.

fair point.

[babba]

I would rent or host a dedicated server for ONi 3 multiplayer in the year 2029

64 player slots, one big 1280x1280 map, anybody can build whatever they like, every 24hours there is a map reset. No server password to join

Team kill cuddles on, every 1 hour the knife toaster bread round happens, where everyone has to bake loads of buns in the game to avoid getting auto kicked - The baked kitchen buns needs to be sent off in rockets to planet "Duh Dah".

Also NPC Baal will run around at the bottom of the map, in the lava channels...

P.S. It will rain gold, regolith and oil from the sky + teddy bears !

5 hours ago, Smife said:

Cuisinart CPT-180, gets the job done and it's a 4 slicer btw.

[gabberworld]

32 minutes ago, babba said:

1280x1280 map

 

that must be some big wet dream. im not kind off sure if current state oni are able handle that

but then again off thinking. player only needs that data what he should see at screen other should handle server self. soo if think like that its even possible

[Astrologic]

9 hours ago, darknotezero said:

Throughout this thread, I think that more of what has been talked about that is "pro multiplayer" has been about conceptual implementation with some incorrect base assumptions about the ease of the technical/programmatic implementation.

I'll grant that I don't know anything about the actual ONI code, but I'd say it's a high probability that @SharraShimada is correct that implementing multiplayer would require an entire base code rewrite if you want it to run efficiently and effectively. Sure, they could probably 'tack on' code that would make it work, but that's a short-term solution that would cause more headaches in the long-run.

Consider the approach to Don't Starve/Don't Starve Together. Klei clearly doesn't conceive of DST as an expansion/DLC to Don't Starve - it's a completely different product. They have their own purchase pages on Steam and they both have independent sets of DLC. They both have their own discussion forums and bug reporting sections on this site rather than a large discussion about both.

Thinking about those games as separate is key for me. It means that if there's a bug in DST, the devs are working on a patch release that just exists for DST without pollenating the code in DS and vice versa. If DST were a component of the original DS, the "if you don't like it, just don't play it" mentality doesn't work because any patches/DLC code will likely affect my DS experience whether I play the multiplayer or not.

For ONI, we've already seen some of that bleed - ONI base game players now have a much different product now than pre-DLC even if they've never played the DLC. They've have had to deal with some of the bugs and issues as a result of the DLC once the codes were merged. If I wasn't participating in the DLC, i would find that frustrating; i wouldn't want my game to get bloated from things that I'm not interested in/going to be a part of.

Also, as a Klei programmer/coder/developer, at some point I would want to put ONI/SO to rest.

That's not to say that ONI Together doesn't have potential and maybe could be a future Klei project, but for me, if that were to be greenlit, I would want it to work the same way that DS does - as a completely separate product with a big "Stay Off My Lawn" sign and fence around the ONI base game/DLC.

 

The difference between DS and ONI is that in DS, you control a single character.  The actions need to be synchronized, you need to implement little tricks like predicted movement in order to get things to match up properly.  There's PVP as well.
Each character has their own inventory and things they can/can't do.  That's why it had to be built separate from the base game.
With ONI, you don't have separate inventories unless you're on a different planetoid (which is already implemented).
You're literally sending commands for units to follow.  
Aside from synchronizing the graphics, all you'd need to do is allow a second source of input for those commands, the input being a socket on the network.
If they decided to create a brand new product for this feature, that's a design choice, and not a necessity.  
I know it's not THAT easy, but it's also not THAT difficult either.
As for graphic synchronization, that's probably the real challenge here.
If Nisbet is catching her breath on tile 45,46, it needs to appear as tile 45,46 for both players, not tile 45,47.
The host could of course send those coordinates to the client, but it really depends how it'd be implemented.
Is the host basically streaming their screen to the client?  Is the host sending tons of information so the client can render it's own game in the same way?  How much data can be spared from the network if it's assumed that the same actions at the same time produce the same results on a given seed? 

I think another challenge would just be planning it out.  Would there be separate task lists for dupes belonging to another player?  That might be more complicated to implement, especially with shared resources on the planet.  Have to make sure dupes don't take resources that are no longer there.  However, that's already managed between multiple dupes, so I don't know how problematic that'd be.  As I've said, I haven't looked at the source code.

I bet even a skilled modder could pull these things off.


 

[cpy]

Multiplayer you say?

 

Spoiler

 

 

[Yunru]

18 hours ago, NeoDeusMachina said:

I think it is important that we can have a discussion about our interests for multiplayer or no multiplayer,

No, it's not. Because we've already had it, endlessly.

This isn't a novel concept, this is just another poster not bothering to search to see if their idea's already been suggested, or even go to the right subforum for it.

[NeoDeusMachina]

Just now, Yunru said:

No, it's not. Because we've already had it, endlessly.

This isn't a novel concept, this is just a poster not bothering to search to see if their idea's already been suggested, or even go to the right subforum for it.

Am I understanding correctly that everything that could ever be said about the topic has already been said? That the forum users have never changed over time, no new user has ever joined the forum and no one could ever bring some new perspective to a topic that has been discussed in the past by other people?

The game is not finished, the DLC is still being developed, which means new gameplay mechanics recently implemented could open up new opportunities for multiplayer or any other topic for that matter.

I wonder how many new ideas are not based on any older concepts or ideas

[Yunru]

1 hour ago, NeoDeusMachina said:

Am I understanding correctly that everything that could ever be said about the topic has already been said? That the forum users have never changed over time, no new user has ever joined the forum and no one could ever bring some new perspective to a topic that has been discussed in the past by other people?

Yes. Literally everything that could be said on the issue has. Unless the game fundamentally changes its core gameplay, every variation of every iteration of the concept has been discussed to death, Ground up, and sold in Tescos as beef. 

[Astrologic]

10 minutes ago, Yunru said:

Yes. Literally everything that could be said on the issue has. Unless the game fundamentally changes its core gameplay, every variation of every iteration of the concept has been discussed to death, Ground up, and sold in Tescos as beef. 

With all due respect, if every idea had been discussed, there'd be a more viable way to implement multiplayer that most people would agree is a good idea.  We haven't come to that conclusion yet, because there's more to discuss.
If you've been exhausted by the conversations about it, that's understandable, I feel there's always more to be discussed, but you don't need to be part of those conversations if you're exhausted by them, or if you can't come up with unique ideas.

 

[Yunru]

2 hours ago, Astrologic said:

With all due respect, if every idea had been discussed, there'd be a more viable way to implement multiplayer that most people would agree is a good idea.  We haven't come to that conclusion yet, because there's more to discuss.

We haven't come to that conclusion yet, because that conclusion doesn't exist.

[Gurgel]

29 minutes ago, Yunru said:

We haven't come to that conclusion yet, because that conclusion doesn't exist.

I agree. At this point, it is pretty plausible that there is no good way to do ONI multiplayer. 

This is not a word-association thing. I have no idea why people continue to ask for it. Just because multiplayer works in some game does not mean it works in others. Might as well ask for a 3D ONI RTS that has the look&feel or the original game. Cannot be done.

 

[lee1026]

14 hours ago, Astrologic said:

Is the host basically streaming their screen to the client?  Is the host sending tons of information so the client can render it's own game in the same way?  How much data can be spared from the network if it's assumed that the same actions at the same time produce the same results on a given seed? 

Streaming entire screens is pretty feasible in this day and age, and that would be lazy way to do it.

[silverbluep]

Since it's established beyond doubt that ONI multiplayer cannot be done; any dupes suggesting multiplayer from this moment onwards shall be sentenced to hug Experiment 52B.

[pether]

2 minutes ago, lee1026 said:

Streaming entire screens is pretty feasible in this day and age, and that would be lazy way to do it.

not screens. entire game. simulation is not predictable and every time a sim updates your game, you would need to send entire map to the other client. it's like you made save-load game 5 times per second. it is possible to do this, but nobody would like to play a game with less than 1 fps

[lee1026]

No, render the screen that they need and send them the graphics. Or just the tiles that they are looking at, depending on how lazy we want to be.

 

Thin clients vs fat clients. The other clients does no calculations, and only requests things from the host.

[pether]

11 minutes ago, lee1026 said:

No, render the screen that they need and send them the graphics. Or just the tiles that they are looking at, depending on how lazy we want to be.

and how will you determine what off-screen gas pump is sucking in? off-screen automation connected to the power sources? what about random movement of off-screen animals? germs? just after one minute of such game you would have 2 totally different bases

[gabberworld]

11 minutes ago, pether said:

and how will you determine what off-screen gas pump is sucking in? off-screen automation connected to the power sources? what about random movement of off-screen animals? germs? just after one minute of such game you would have 2 totally different bases

that is run by backround in server

[pether]

1 minute ago, gabberworld said:

that is run by backround in server

...and must be synchronized with both bases every game tick unless you want to create two different realities

[gabberworld]

Just now, pether said:

...and must be synchronized with both bases every game tick unless you want to create two different realities

of-course but we talking year 2021 atm. this kind stuff done at year 2010 already in games who had separate servers app with them

[lee1026]

27 minutes ago, pether said:

...and must be synchronized with both bases every game tick unless you want to create two different realities

This is very much correct. Think of GeForce now: the client only gets the pixels and runs no game logic, and this all works fine.

[darknotezero]

potential practical implementations aside, if we're going to loop back around to conceptual, I echo the thoughts of others in this thread that trying to add multiplayer to ONI is like trying to wedge a square peg into a circle hole. If i have a deck of 52 cards with intent to play solitaire and someone else says "hey do you want to play a game together?", i'm not going to suggest we find a way to collaboratively play solitaire, i'm going to choose a game that's designed for 2+ people.

I consider ONI to be solitaire.

3 hours ago, Astrologic said:

if every idea had been discussed, there'd be a more viable way to implement multiplayer that most people would agree is a good idea.

i can concede that thought experiments can sometimes bring out fresh perspectives and an unusual random thought can bring out unknown innovations.

I challenge, however, that "most people would agree this is a good idea" is a pretty baseless assumption. It seems to me that if you were to ask the player base what level of enthusiasm they have from 1 to 10 about multiplayer implementation, you'd get a very wide spectrum. Some people think it's a great idea and would love to see it happen, but just as many "some" people feel neutral about it or vehemently oppose it, as is evidenced not only in this thread but in other archived threads to this point.

That's another reason why I feel it would be better off as a standalone project. I would want ONI multiplayer to be truly multiplayer, not the current ONI experience with some multiplayer tacked on. That way, those that advocated for it get a product that carries the ONI experience but has a true multiplayer intent, and those that feel neutral or oppose it will not have their single player ONI experience distracted by it even if they would never partake of the option.

 

[Astrologic]

The calculations for the game already happen off camera, you'd just need a remote viewport.

[pether]

16 minutes ago, lee1026 said:

and this all works fine.

except it doesnt and reports of poor ONI performance are more frequent than poor multiplayer proposals.

but it all doesn't matter, because even if we had superb hardware and ONI could fly with 120 fps even during late game - multiplayer with shared control over the base would still be a bad idea and the final gameplay would be terrible. sure, some folks could play with their friends, but those game would be very short with little chance of playing it again. even if it could be done (still, it can't), ROI for such project would be not enough to justify it. ONI is just not this kind of game.

There were some valid ideas how to allow player interaction in ONI, I even modded one of them, but nothing happening in real time would be a good idea. Instead of trying to convince everybody how great and fun it would be (it wouldn't), try to thing for a few moments about everything that would not be fun in such game. To make it somehow possible, core ONI mechanics and designs would need to be changed for multiplayer approach, making it totally different game. And I'm still not sure if that would be a good idea