You're doing it again. You are applying what they want to do, that isn't implemented yet, and using it to justify your argument. If we go that route then you are going to open the floodgates to a whole new level of argument cause then I can use any game that wants to do something and use it to justify my position. I know you don't like me but I think it's affecting how you reply to my posts since you use theoretical to justify your position to me but then say its too early to make a call when responding to phaserlight. You can't have your cake and eat it too.
And just for fun using your logic of say 24 players per instance and having multiple instances just puts COD:IW into the same category as a MMO and therefore beats SC out again.
Because Star Citizen is an MMO under development.
The game is already unique, once they get the network sorted, as I said, it will push the genre forward. I stand my claim as my view and opinion if it, something I am entitled to have as everyone else. So please just deal with that instead of continuing this childish baiting game.
Its called discussion, or arguing if you prefer. Everyone is entitled to their opinion but don't try to shut down points with the bait excuse because you've backed yourself into a corner.
That's quite a fancy plan. Yes, sure, you are entitled to your opinion; that doesn't make it valid.
It doesn't make it invalid either... It's opinions based on their ongoing and planned work for said part of the game, only time will tell that.
As for evidence, the core plan of what they are doing ties to Life is Feudal MMO is also doing, that core piece of tech with the borders and so forth lies within the same plan SC has, so I do can see how they are going to do it, ofc they need a far more complex integration but the base of it is the same.
Its called discussion, or arguing if you prefer. Everyone is entitled to their opinion but don't try to shut down points with the bait excuse because you've backed yourself into a corner.
You are the one who came after me for what I said my opinion is, and I do stand by what I said is my opinion of SC as one unique game that already is, and how I think it will push the genre forward once they have the network to support the rest of the game.
Because independent of all schematics used, this is one MMO actively under development.
That's quite a fancy plan. Yes, sure, you are entitled to your opinion; that doesn't make it valid.
It doesn't make it invalid either... It's opinions based on their ongoing and planned work for said part of the game, only time will tell that.
As for evidence, the core plan of what they are doing ties to Life is Feudal MMO is also doing, that core piece of tech with the borders and so forth lies within the same plan SC has, so I do can see how they are going to do it, ofc they need a far more complex integration but the base of it is the same.
The problem is, it looks like you are offering supported rebuttals to my pedantic questions, but unless you really dive deep into the links and history of this project you won't see that your opinion is based on Chris' word and a nebulous "core piece of tech" that is going to make all this happen.
That's OK; people have developed stranger beliefs for stranger reasons, but if I'm personally being asked to open my wallet to support this ongoing project, I'll need something a little more solid than that.
When this turns into a single player game with multiplayer elements, like Destiny, I won't be there writing "I told you so".
"The simple is the seal of the true and beauty is the splendor of truth" -Subrahmanyan Chandrasekhar Authored 139 missions in VendettaOnline and 6 tracks in Distance
Its called discussion, or arguing if you prefer. Everyone is entitled to their opinion but don't try to shut down points with the bait excuse because you've backed yourself into a corner.
You are the one who came after me for what I said my opinion is, and I do stand by what I said is my opinion of SC as one unique game that already is, and how I think it will push the genre forward once they have the network to support the rest of the game.
Because independent of all schematics used, this is one MMO actively under development.
I didn't come after you. I offered counter points and showed what you were saying is based on someone else's word and not what is currently implemented. That's generally how debate/discussion/arguments go. You have one opinion and I have another therefore we post something and put up evidence to support our claim and the other side posts counter evidence.
Im just poking holes in the evidence you are offering and letting nature take its course lol
The problem is, it looks like you are offering supported rebuttals to my pedantic questions, but unless you really dive deep into the links and history of this project you won't see that your opinion is based on Chris' word and a nebulous "core piece of tech" that is going to make all this happen.
That's OK; people have developed stranger beliefs for stranger reasons, but if I'm personally being asked to open my wallet to support this ongoing project, I'll need something a little more solid than that.
When this turns into a single player game with multiplayer elements, like Destiny, I won't be there writing "I told you so".
You don't have to if you don't want to, that's up to you.
This is not CR word, this is stuff talked about by the developers several times, more in-depth on that post but this is stuff ongoing, though we are talking about tasks are long and daunting, the nature of this type of work.
First, as we see the schedules they are working optimizing the netcode as much as they can, as this is mostly about making it run X players per X amount of CPU cores.
Then as this goes, we see they start the work on the first of the many tasks that englobe the object containers, the object containers are a core thing because compartmentalize the game world is the fundamental step to actually have localized servers processing a certain area of that game world, that was what Life is Feudal did, then it was about boundaries and that is where the whole streaming aspect of it comes to place, we see what's on the other server from our server as we move to it... SC designs a more complex approach, by making this dynamic so they can scale how much servers handle the game-world by the population per area. (something like this)
LIF does have its gameworld handled by 49 servers (like this), all within this core design that allows for one large amount of players on it, they are one small indie studio but they are already undergoing tests where this system is working.
If you need something solid then you'll have to wait until the large amount of network work, some ongoing, other still to start dev, startings being implemented.
You're right I don't play the game and while I don't really care about it I do find it hilarious how some people will blindly defend it using arguments like "this is what they want to do" or "its going to get better because". I've heard it far too many times from publishers about how this game(generalization, not talking SC lol) is going to blow your socks off or its going to be the next great thing and then when it releases its mediocre at best. SC for all it's posturing about how evil publishers are is pulling the same crap and hyping something and then delivering mediocre.
I know why I lags, I know how netcode works in general and how it's working on SC. It doesn't change the fact you made a claim you can't support. You saw DS saying on the internet the servers can't handle more than 8 players without falling apart (fun fact: we don't get to know how many peeps are in one instance), and you thought it was cute to use it as some factual piece of information on a forum internet discussion.
Does performance start to degrade as more people enter the game? Yes it does. This is posted about from multiple people. Is it 8 that you start to notice? depends on your computer and net connection I suppose, could be 2 or it could be at 24.
I didn't use the 8 before the server shits its pants as factual evidence. I used it to poke fun at the project. When you pushed back I offered up evidence from multiple sources of people playing the game and how the game degrades as more enter the instance. You can't argue it because even CIG knows it's an issue, hence the magical network patch that's been coming for however long that's supposed to fix everything.
I didn't come after you. I offered counter points and showed what you were saying is based on someone else's word and not what is currently implemented. That's generally how debate/discussion/arguments go. You have one opinion and I have another therefore we post something and put up evidence to support our claim and the other side posts counter evidence.
Im just poking holes in the evidence you are offering and letting nature take its course lol
And I refute your counter point because my point already considered the very important fact that this is one MMO actively under development. I do stand my claim that this game is unique as of what is currently implemented, what for me it is, nothing else around provides the type of gameplay this small fraction of SC that is the PU provides, something that will be even more obvious once 3.0 releases.
I'm about to close this thread as my question has been mostly answered, yet this thread seems to be spiraling into all types of topics.
Question for Erillion/MaxBacon - I wasn't aware that the current tech only allows 20-30 players in an instance... that's really low. Will 3.0 fix this? If not, what is the ultimate release goals around players in the same instance?
Not a big Eve player, but they had some epic PVP battles with hundreds of concurrent players... maybe because this is a space sim, I was expecting/assuming the same...
I didn't come after you. I offered counter points and showed what you were saying is based on someone else's word and not what is currently implemented. That's generally how debate/discussion/arguments go. You have one opinion and I have another therefore we post something and put up evidence to support our claim and the other side posts counter evidence.
Im just poking holes in the evidence you are offering and letting nature take its course lol
And I refute your counter point because my point already considered the very important fact that this is one MMO actively under development. I do stand my claim that this game is unique as of what is currently implemented, what for me it is, nothing else around provides the type of gameplay this small fraction of SC that is the PU provides, something that will be even more obvious once 3.0 releases.
That's fine that you stand by your claim, I'm cool with that. I take issue with you trying to classify SC as a MMO while saying other games can't be a MMO that fall under the exact same definition that you give.
This whole debate started with bab linking 4 videos that show flying in space and shooting and fps combat in response to me saying I have yet to see anything that pushes the genre forward or that hasn't been done before. You responded and we went from there.
You haven't really refuted my evidence, you've just offered up your opinion and what they hope to do which isn't evidence.
Since I need to goto work I'll also post this for Twisted to give you an overview of what they are up against for network issues. This was posted almost a year ago btw.
I'm about to close this thread as my question has been mostly answered, yet this thread seems to be spiraling into all types of topics.
Question for Erillion/MaxBacon - I wasn't aware that the current tech only allows 20-30 players in an instance... that's really low. Will 3.0 fix this? If not, what is the ultimate release goals around players in the same instance?
Not a big Eve player, but they had some epic PVP battles with hundreds of concurrent players... maybe because this is a space sim, I was expecting/assuming the same...
The current player count for 3.0 stands still at 24 players, there's a lot of network work ongoing but things set to release after 3.0, going up to 2018 of continued network work.
The most recent mention of player instance pops (you can read here) mentions that their goal on network is to keep all players on the same instance (single game world).
Also he replied a few posts bellow the same question as yours rewarding to the scale of EvE PvP Battles, EvE has a very cheap network model, a game like SC doesn't have that ability.
The problem is, it looks like you are offering supported rebuttals to my pedantic questions, but unless you really dive deep into the links and history of this project you won't see that your opinion is based on Chris' word and a nebulous "core piece of tech" that is going to make all this happen.
That's OK; people have developed stranger beliefs for stranger reasons, but if I'm personally being asked to open my wallet to support this ongoing project, I'll need something a little more solid than that.
When this turns into a single player game with multiplayer elements, like Destiny, I won't be there writing "I told you so".
You don't have to if you don't want to, that's up to you.
This is not CR word, this is stuff talked about by the developers several times, more in-depth on that post but this is stuff ongoing, though we are talking about tasks are long and daunting, the nature of this type of work.
First, as we see the schedules they are working optimizing the netcode as much as they can, as this is mostly about making it run X players per X amount of CPU cores.
Then as this goes, we see they start the work on the first of the many tasks that englobe the object containers, the object containers are a core thing because compartmentalize the game world is the fundamental step to actually have localized servers processing a certain area of that game world, that was what Life is Feudal did, then it was about boundaries and that is where the whole streaming aspect of it comes to place, we see what's on the other server from our server as we move to it... SC designs a more complex approach, by making this dynamic so they can scale how much servers handle the game-world by the population per area. (something like this)
LIF does have its gameworld handled by 49 servers (like this), all within this core design that allows for one large amount of players on it, they are one small indie studio but they are already undergoing tests where this system is working.
If you need something solid then you'll have to wait until the large amount of network work, some ongoing, other still to start dev, startings being implemented.
It looks like you are describing a form of On-Demand Computing, or maybe Grid Computing, both of which have been around for a while. These are not new solutions.
I can accept that microprocessors have come a long way, and computer servers, or server clouds, are capable have handling ever more complex computations and simulations. However, this still doesn't address the issue @Kefo mentioned, which is 'what happens when you try to stream that data out to all the different clients?' This is only something that comes up when you try to get lots of players together in the same area.
Again, layman's understanding here, but...
Let's say one player takes up 14kbits of data; that all has to be transmitted between the client and the server. The client has to tell the server 'I am here' and the server must say 'yes, I see you there, and by the way there are several dozen other players in your near vicinity each taking up 14kbits themselves'.
This scales to a certain degree across a network, and maybe you get several hundred players together in the same area.
Now let's say one player takes up a megabit of data. Regardless of how efficient the server is at juggling that data, it still must be streamed out to the client. The server can't tell the client 'there are a hundred other players in your near vicinity, here's what you need to keep track of them' because you don't have that much bandwidth.
Enter instancing... a common solution as games became more complex; 'you don't need to keep track of these other players in your area, I'll do that for you, just take my word that they're there. Of course, you're not going to actually render them because I can't give you that much information...'
You can easily see why this seems like BS to oldskool MMO players like me.
Fine, make a super-detailed game with unprecedented fidelity and immersion that allows you to get together with other players once in a while, but don't call it a MMO, and don't try and sell it as one.
"The simple is the seal of the true and beauty is the splendor of truth" -Subrahmanyan Chandrasekhar Authored 139 missions in VendettaOnline and 6 tracks in Distance
That's fine that you stand by your claim, I'm cool with that. I take issue with you trying to classify SC as a MMO while saying other games can't be a MMO that fall under the exact same definition that you give.
Again, Star Citizen is one MMO actively under development, again, under development, and one last time, under development. The fact the alpha right now does not have the MMO aspect to it does not make it not being one MMO under development (and why are you implying I said that?!). And again, I considered that within the post I made.
I can accept that microprocessors have come a long way, and computer servers, or server clouds, are capable have handling ever more complex computations and simulations. However, this still doesn't address the issue @Kefo mentioned, which is 'what happens when you try to stream that data out to all the different clients?' This is only something that comes up when you try to get lots of players together in the same area.
You can easily see why this seems like BS to oldskool MMO players like me.
Fine, make a super-detailed game with unprecedented fidelity and immersion that allows you to get together with other players once in a while, but don't call it a MMO, and don't try and sell it as one.
They already mentioned they are to figure a solution to the fact of many players trying to be one the same place at the same time, over the capability of the servers to handle it. And this is no boggie man either if anything they can even spawn overflow servers just like Guild Wars 2 originally did per area instance.
It does not seem like any BS, this designs exist around and have been proven possible, so they do not have to do any magical network model to sustain the complexity of the game.
Star Citizen is one MMO under development, not only on the mechanic that is created not only to support but has in mind this online persistent game-world, but also as that dev post shows the direction network is taking to tackle in the multiplayer scale of it.
I can accept that microprocessors have come a long way, and computer servers, or server clouds, are capable have handling ever more complex computations and simulations. However, this still doesn't address the issue @Kefo mentioned, which is 'what happens when you try to stream that data out to all the different clients?' This is only something that comes up when you try to get lots of players together in the same area.
You can easily see why this seems like BS to oldskool MMO players like me.
Fine, make a super-detailed game with unprecedented fidelity and immersion that allows you to get together with other players once in a while, but don't call it a MMO, and don't try and sell it as one.
They already mentioned they are to figure a solution to the fact of many players trying to be one the same place at the same time, over the capability of the servers to handle it. And this is no boggie man either if anything they can even spawn overflow servers just like Guild Wars 2 originally did per area instance.
It does not seem like any BS, this designs exist around and have been proven possible, so they do not have to do any magical network model to sustain the complexity of the game.
Star Citizen is one MMO under development, not only on the mechanic that is created not only to support but in mind of this online persistent game-world, but as that dev post shows the direction of network to support one higher scale.
Again, and last time I will say this; it's not the server, it's the network. Way to cut out the core of my argument when quoting me.
Unless they've found a way to drastically reduce the amount of data flowing between the client and the server, I'll have to assume that 20 per instance, everywhere, is about what we'll see in the final product.
I'm more than ready to step down from this opinion if some expert comes along and offers a compelling piece of evidence to the contrary.
"The simple is the seal of the true and beauty is the splendor of truth" -Subrahmanyan Chandrasekhar Authored 139 missions in VendettaOnline and 6 tracks in Distance
My understanding of the desire toward networking code is similar what many of the game engines already do as it pertains to Graphics. I am certain you have heard of LOD (Level of Detail) in graphics. Where the further you are from an object, the less detail you get.
This is how I would go about understanding what they are currently doing from a netcode perspective. Today, there is no filtering, anything that requires the client to be informed is sent. So if you increase the number of people in an instance, and even if you cant see them, you are getting that update.
They want to filter the data that each client gets based on what is relevant to their current position. Network LOD in essence.
I will preface, this is my understanding of their desire. There are far more complex responses, and many conversations on the actuals to make this happen.
Considering unless you have custom written an engine with this level of scale and fidelity, there has to be some retrofitting that needs to happen, that is the netcode challenge ahead of them. How do they proper filter the information to only send the client the relative information they require. If and when this addressed many of the restrictions as to number of players visible and able to interact will start to fade.
Again, and last time I will say this; it's not the server, it's the network. Way to cut out the core of my argument when quoting me.
Unless they've found a way to drastically reduce the amount of data flowing between the client and the server, I'll have to assume that 20 per instance, everywhere, is about what we'll see in the final product.
I'm more than ready to step down from this opinion if some expert comes along and offers a compelling piece of evidence to the contrary.
That is the nature of network, for some reason they are rewriting pretty much every piece of netcode they had, the objective is obviously to optimize it as much as they can and make it so they can have as much gameservers per VM as they possibly can, and this ties a lot to how much they can run per CPU core. (what @Izzin mentioned of network LOD's is indeed a good example of the type of work they're doing attm)
Star Citizen is obviously one complex game, so I don't think they'll ever be able to cheap their costs to the level of a normal MMO when it comes to resource cost per player.
If this is done decently enough, then scalability is not much of a big problem. I know this is more complex than just saying "throw the resources it would take to run 2 instances of 20 players, into one bigger instance of 40 players", but it's all work that needs to be done, either the instances are made of 20 or 100 players.
That's fine that you stand by your claim, I'm cool with that. I take issue with you trying to classify SC as a MMO while saying other games can't be a MMO that fall under the exact same definition that you give.
Again, Star Citizen is one MMO actively under development, again, under development, and one last time, under development. The fact the alpha right now does not have the MMO aspect to it does not make it not being one MMO under development (and why are you implying I said that?!). And again, I considered that within the post I made.
I'm not sure what you're getting at cause I think I'm reading your post wrong. What are you asking me that I implied you said something?
I'm not sure what you're getting at cause I think I'm reading your post wrong. What are you asking me that I implied you said something?
"I take issue with you trying to classify SC as a MMO while saying other games can't be a MMO that fall under the exact same definition that you give."
I never said SC is, right now, one MMO, what is released. I said and multiple times the game being developed here is one MMO instead, and that is what I implied from the very start.
If this is done decently enough, then scalability is not much of a big problem. I know this is more complex than just saying "throw the resources it would take to run 2 instances of 20 players, into one bigger instance of 40 players", but it's all work that needs to be done, either the instances are made of 20 or 100 players.
The problem is that the game, as currently designed with the that level of detail and realism they want, will not support instances of a much larger size than currently. Either realism or rather, level of detail in how you interact with the environment and how you can affect it, has to go down, or they will have to accept a low player count pr. instance or alternately accept greater room for possible desync issues. You cannot have all three. If you look at the high count games like EVE or Planetside 2, they are have made this tradeoff. SC will have to as well. No magic is gonna solve that specific issue.
While I am not saying it is the same, what you have here is a problem not unlike the CAP theorem. Go look it up.
I'm about to close this thread as my question has been mostly answered, yet this thread seems to be spiraling into all types of topics.
Question for Erillion/MaxBacon - I wasn't aware that the current tech only allows 20-30 players in an instance... that's really low. Will 3.0 fix this? If not, what is the ultimate release goals around players in the same instance?
Not a big Eve player, but they had some epic PVP battles with hundreds of concurrent players... maybe because this is a space sim, I was expecting/assuming the same...
Your best bet is to wait for a free flight weekend and try the game out. As to if 3.0 will fix the netcode, I don't think it will, my understand was that 3.0 was a foundation for fixing stuff. There's a schedule posted on their site that will give you a general idea of what they're working on. Best bet is to wait and see as schedules are changing a lot.
"We all do the best we can based on life experience, point of view, and our ability to believe in ourselves." - Naropa "We don't see things as they are, we see them as we are." SR Covey
I'm not sure what you're getting at cause I think I'm reading your post wrong. What are you asking me that I implied you said something?
"I take issue with you trying to classify SC as a MMO while saying other games can't be a MMO that fall under the exact same definition that you give."
I never said SC is, right now, one MMO, what is released. I said and multiple times the game being developed here is one MMO instead, and that is what I implied from the very start.
Ok still not quite clear but I'll run with it.
i don't care that it's a mmo under development. You keep using LiF as an example of how SC will eventually work as an MMO at some distant point in the future and I don't care about empty CR promises.
I asked for things that they are doing that are pushing the genre forward or have never been done before that exist in the current game. So far what has been posted has been done before or is still an idea on paper.
Its fine if you want to say they are a MMO still in development but don't come in here saying they are an MMO pushing the genre forward when they actually haven't done anything new or other games have beaten them to the punch
And this is why the game world is being compartmentalized by this very same thing the devs talked about this past times.
Games like Dual Universe are using pieces of tech like SpacialOS to allow one single-shard game universe on the cloud, things that were simply not possible/done before.
Seeing that most network rewrites and other related pieces of tech are still to come, it will come down to how to scalability, depending on that work, they can assign more resources per instance to make support higher player pops.
It would be feasible until the problem will be the player client instead, something we all know very well what happens when visiting one main player city HUB in many MMO's.
On Crusader Universe there is a limit of 24 players per server because servers also have to account for the ships which are as complex entity's as players. That's why in the ArcCorp city they can double the number of players and have as much as 48 players in the same instance.
The changes to item 2.0 , networking etc all go in the direction to alleviate that.
Its fine if you want to say they are a MMO still in development but don't come in here saying they are an MMO pushing the genre forward when they actually haven't done anything new or other games have beaten them to the punch
I say both, it is one MMO in development, and in my opinion, they will easily push the genre forward when they get the networking to support the gameplay mechanics and tech they already have.
As for the rest, it's not only ideas on paper, stuff is actively under development on the network front, they are long-term projects as Item 2.0 was (and today we are seeing the fruits and things they are now being able to do after such lengthy engine rewrite).
Time will tell what will their engineers be able to achieve on this.
Comments
As for evidence, the core plan of what they are doing ties to Life is Feudal MMO is also doing, that core piece of tech with the borders and so forth lies within the same plan SC has, so I do can see how they are going to do it, ofc they need a far more complex integration but the base of it is the same.
Because independent of all schematics used, this is one MMO actively under development.
That's OK; people have developed stranger beliefs for stranger reasons, but if I'm personally being asked to open my wallet to support this ongoing project, I'll need something a little more solid than that.
When this turns into a single player game with multiplayer elements, like Destiny, I won't be there writing "I told you so".
"The simple is the seal of the true and beauty is the splendor of truth" -Subrahmanyan Chandrasekhar
Authored 139 missions in Vendetta Online and 6 tracks in Distance
Im just poking holes in the evidence you are offering and letting nature take its course lol
This is not CR word, this is stuff talked about by the developers several times, more in-depth on that post but this is stuff ongoing, though we are talking about tasks are long and daunting, the nature of this type of work.
First, as we see the schedules they are working optimizing the netcode as much as they can, as this is mostly about making it run X players per X amount of CPU cores.
Then as this goes, we see they start the work on the first of the many tasks that englobe the object containers, the object containers are a core thing because compartmentalize the game world is the fundamental step to actually have localized servers processing a certain area of that game world, that was what Life is Feudal did, then it was about boundaries and that is where the whole streaming aspect of it comes to place, we see what's on the other server from our server as we move to it... SC designs a more complex approach, by making this dynamic so they can scale how much servers handle the game-world by the population per area. (something like this)
LIF does have its gameworld handled by 49 servers (like this), all within this core design that allows for one large amount of players on it, they are one small indie studio but they are already undergoing tests where this system is working.
If you need something solid then you'll have to wait until the large amount of network work, some ongoing, other still to start dev, startings being implemented.
I didn't use the 8 before the server shits its pants as factual evidence. I used it to poke fun at the project. When you pushed back I offered up evidence from multiple sources of people playing the game and how the game degrades as more enter the instance. You can't argue it because even CIG knows it's an issue, hence the magical network patch that's been coming for however long that's supposed to fix everything.
Question for Erillion/MaxBacon - I wasn't aware that the current tech only allows 20-30 players in an instance... that's really low. Will 3.0 fix this? If not, what is the ultimate release goals around players in the same instance?
Not a big Eve player, but they had some epic PVP battles with hundreds of concurrent players... maybe because this is a space sim, I was expecting/assuming the same...
This whole debate started with bab linking 4 videos that show flying in space and shooting and fps combat in response to me saying I have yet to see anything that pushes the genre forward or that hasn't been done before. You responded and we went from there.
You haven't really refuted my evidence, you've just offered up your opinion and what they hope to do which isn't evidence.
Since I need to goto work I'll also post this for Twisted to give you an overview of what they are up against for network issues. This was posted almost a year ago btw.
https://forums.robertsspaceindustries.com/discussion/335037/summary-of-server-and-network-issues
The most recent mention of player instance pops (you can read here) mentions that their goal on network is to keep all players on the same instance (single game world).
Also he replied a few posts bellow the same question as yours rewarding to the scale of EvE PvP Battles, EvE has a very cheap network model, a game like SC doesn't have that ability.
I can accept that microprocessors have come a long way, and computer servers, or server clouds, are capable have handling ever more complex computations and simulations. However, this still doesn't address the issue @Kefo mentioned, which is 'what happens when you try to stream that data out to all the different clients?' This is only something that comes up when you try to get lots of players together in the same area.
Again, layman's understanding here, but...
Let's say one player takes up 14kbits of data; that all has to be transmitted between the client and the server. The client has to tell the server 'I am here' and the server must say 'yes, I see you there, and by the way there are several dozen other players in your near vicinity each taking up 14kbits themselves'.
This scales to a certain degree across a network, and maybe you get several hundred players together in the same area.
Now let's say one player takes up a megabit of data. Regardless of how efficient the server is at juggling that data, it still must be streamed out to the client. The server can't tell the client 'there are a hundred other players in your near vicinity, here's what you need to keep track of them' because you don't have that much bandwidth.
Enter instancing... a common solution as games became more complex; 'you don't need to keep track of these other players in your area, I'll do that for you, just take my word that they're there. Of course, you're not going to actually render them because I can't give you that much information...'
You can easily see why this seems like BS to oldskool MMO players like me.
Fine, make a super-detailed game with unprecedented fidelity and immersion that allows you to get together with other players once in a while, but don't call it a MMO, and don't try and sell it as one.
"The simple is the seal of the true and beauty is the splendor of truth" -Subrahmanyan Chandrasekhar
Authored 139 missions in Vendetta Online and 6 tracks in Distance
It does not seem like any BS, this designs exist around and have been proven possible, so they do not have to do any magical network model to sustain the complexity of the game.
Star Citizen is one MMO under development, not only on the mechanic that is created not only to support but has in mind this online persistent game-world, but also as that dev post shows the direction network is taking to tackle in the multiplayer scale of it.
Unless they've found a way to drastically reduce the amount of data flowing between the client and the server, I'll have to assume that 20 per instance, everywhere, is about what we'll see in the final product.
I'm more than ready to step down from this opinion if some expert comes along and offers a compelling piece of evidence to the contrary.
"The simple is the seal of the true and beauty is the splendor of truth" -Subrahmanyan Chandrasekhar
Authored 139 missions in Vendetta Online and 6 tracks in Distance
My understanding of the desire toward networking code is similar what many of the game engines already do as it pertains to Graphics. I am certain you have heard of LOD (Level of Detail) in graphics. Where the further you are from an object, the less detail you get.
This is how I would go about understanding what they are currently doing from a netcode perspective. Today, there is no filtering, anything that requires the client to be informed is sent. So if you increase the number of people in an instance, and even if you cant see them, you are getting that update.
They want to filter the data that each client gets based on what is relevant to their current position. Network LOD in essence.
I will preface, this is my understanding of their desire. There are far more complex responses, and many conversations on the actuals to make this happen.
Considering unless you have custom written an engine with this level of scale and fidelity, there has to be some retrofitting that needs to happen, that is the netcode challenge ahead of them. How do they proper filter the information to only send the client the relative information they require. If and when this addressed many of the restrictions as to number of players visible and able to interact will start to fade.
-izz
Star Citizen is obviously one complex game, so I don't think they'll ever be able to cheap their costs to the level of a normal MMO when it comes to resource cost per player.
If this is done decently enough, then scalability is not much of a big problem. I know this is more complex than just saying "throw the resources it would take to run 2 instances of 20 players, into one bigger instance of 40 players", but it's all work that needs to be done, either the instances are made of 20 or 100 players.
I never said SC is, right now, one MMO, what is released. I said and multiple times the game being developed here is one MMO instead, and that is what I implied from the very start.
While I am not saying it is the same, what you have here is a problem not unlike the CAP theorem. Go look it up.
"We all do the best we can based on life experience, point of view, and our ability to believe in ourselves." - Naropa "We don't see things as they are, we see them as we are." SR Covey
i don't care that it's a mmo under development. You keep using LiF as an example of how SC will eventually work as an MMO at some distant point in the future and I don't care about empty CR promises.
I asked for things that they are doing that are pushing the genre forward or have never been done before that exist in the current game. So far what has been posted has been done before or is still an idea on paper.
Its fine if you want to say they are a MMO still in development but don't come in here saying they are an MMO pushing the genre forward when they actually haven't done anything new or other games have beaten them to the punch
And this is why the game world is being compartmentalized by this very same thing the devs talked about this past times.
Games like Dual Universe are using pieces of tech like SpacialOS to allow one single-shard game universe on the cloud, things that were simply not possible/done before.
Seeing that most network rewrites and other related pieces of tech are still to come, it will come down to how to scalability, depending on that work, they can assign more resources per instance to make support higher player pops.
It would be feasible until the problem will be the player client instead, something we all know very well what happens when visiting one main player city HUB in many MMO's.
The changes to item 2.0 , networking etc all go in the direction to alleviate that.
As for the rest, it's not only ideas on paper, stuff is actively under development on the network front, they are long-term projects as Item 2.0 was (and today we are seeing the fruits and things they are now being able to do after such lengthy engine rewrite).
Time will tell what will their engineers be able to achieve on this.