The Star Citizen Thread v5

[Rilzi]

Think it through though - is that actually possible in any way at all? What kind of back end do you think you'd need to handle 10 million NPC lives complete with careers, desires, long term goals and daily choices? Then how do you hold it all in a single shard so a player in another instance doesn't cause their galaxy to diverge from yours?

^ this.
I mean, it would be genuinely awesome if they could get that kind of system working, it would put SC right back into contention of BDSSE. But lets be real, does anyone think CIG could build a system that complex, big, one that has never been done before by anyone? With the time and money they've spent, they haven't even managed to get a basic Freelancer clone working. Basic functions in the game are either broken or MIA. Everything great has been 'just around the corner' for years now and at some point, no matter what they're promising, you gotta take anything they say with a few grains of salt. Or, you know, just outright disbelieve them until they can show proof. Personally? Well, in the words of Rowan Atkinson, I wouldn't trust them to sit the right way on a toilet seat.

 

[Skizomeuh]

Is it possible? Probable.
From CIG? Unlikely.
Why? <<USE>>
Period.

- - - Updated - - -

perhaps its time to stop building fantasizing systems and building an actual game.

Fixed.

Also, we all know how implementation would end:

Replace leaf with player's behaviour. Multiply the situation by players.

 

[Cotic]

I don't know. I'm no programmer. My guess is that most of the people on this thread arent btw.
I have no doubts that SC as it is envisioned by a lot of people will never see the light of day (I'm with you guys, don't shoot me please).

There's a lot of negativity surrounding SC on this thread and IMO not all of it is deserved.
A couple of pages back there's somebody ridiculing SC backers wanting to have unique ID numbers for all their parts and how impossible that would be.
Then somebody else posts that E : D allready does this and that.

My point is. Just because an idea is floated by CIG or its supporters doesn't nessecarily mean that it's infected with the plague.

That's not an unfair assessment tbh. There's certainly a good dosing of ridicule for ridicule's sake (of which I participate in as well).

 

[frostypaw]

My point is. Just because an idea is floated by CIG or its supporters doesn't nessecarily mean that it's infected with the plague.

I don't think people do assume that - but you should still stop and think these things through and see if it's at all possible before breaking out the champagne

 

[duffbeeeer]

*insert witty comment here*

sorry guys, I tried to think of something funny, or profound, but honestly at this point posts like this just make me tired.

You know back in 1998, i have read in an german magazine "Der Spiegel" an article about a game called Ultima Online. The world was described as alive with food chains and consequences. Predators would eat herbivors and if their food source was gone, theyd be starting to attack the nearest village. It all sounded just great in my mind and got me into the whole MMO sandbox thing. Have been playing EVE Online for nearly a decade.

The thing is it never made it into UO. The developers realized pretty quickly that this would not be feasible in an Multiplayer Enviroment. The challenge was not technical or something to be solved with more computing power. It was purely the presence of everyone trying to be the hero. IT JUST DOESNT WORK. Chris Robert trying to resell those ideas even 20 years later makes him look a like snake oil sales man to me.

 

[Zetta]

You know back in 1998, i have read in an german magazine "Der Spiegel" an article about a game called Ultims Online. The world was described as alive with food chains and consequences. Predators would eat herbivors and if their food source was gone, theyd be starting to attack the nearest village. It all sounded just great in my mind and got me into the whole MMO sandbox thing. Have been playing EVE Online for nearly decade. The thing is it never made it into UO. The developers realized pretty quickly that this would not be feasible in an Multiplayer Enviroment. The challenge was not technical or something to be solved with more computing power. It was purely the presence of everyone trying to be the hero. ITS JUST DOESNT WORK. Chris Robert trying to resell those ideas even 20 years later makes him look a like snake oil sales man to me.

What I find ironic about that is the guy who was making UO was Roberts' mentor. Both are now ex-EA guys who have gone on to crowdfund their individual games in a very divisive way.

You can take the guy out of EA but you can't take the EA out of the guy...

 

[duffbeeeer]

What I find ironic about that is the guy who was making UO was Roberts' mentor. Both are now ex-EA guys who have gone on to crowdfund their individual games in a very divisive way.

You can take the guy out of EA but you can't take the EA out of the guy...

Atleast one of them made it into space IRL. [big grin]

 

[Ian Skippy]

I don't know. I'm no programmer. My guess is that most of the people on this thread arent btw.
I have no doubts that SC as it is envisioned by a lot of people will never see the light of day (I'm with you guys, don't shoot me please).

There's a lot of negativity surrounding SC on this thread and IMO not all of it is deserved.
A couple of pages back there's somebody ridiculing SC backers wanting to have unique ID numbers for all their parts and how impossible that would be.
Then somebody else posts that E : D allready does this and that.

My point is. Just because an idea is floated by CIG or its supporters doesn't nessecarily mean that it's infected with the plague.

You misunderstand. You need to be a programmer to know if a specific idea can be implemented in your language. You dont need to be a programmer to know that some ideas are fundamentally impossible. Some ideas about SC are fundamentally illogical: no amount of hardware, magic germans or super languages can make x==x+1 work. If you have an idea that is logically internally consistent, only then do you need programmers to tell you if it is feasible/practical in this or that situation/context.

 

[Viajero]

*insert witty comment here*

sorry guys, I tried to think of something funny, or profound, but honestly at this point posts like this just make me tired.

Try with this:

It’s not difficult to create a large, randomly generated world. The problem is in making a vast universe that’s actually interesting – that’s able to hold your attention for an extended period of time. On the mission side, for example, we’re aiming to algorithmically construct a lot of our missions, but we’re going to leverage our designers’ capabilities, instead of trying to entirely replace them with mathematical equations. We’ll have the traditional hand-crafted missions whereupon the player gradually progresses through a detailed storyline, but as with all MMOs the problem with that approach is that you’re always struggling to create enough new content to keep the player base entertained, and if that’s your only method of content generation the quality tends to suffer as a result. In your quest to deliver a large quantity of content, you tend to lower your standards and create a lot of really simplistic missions that aren’t very interesting, that are identical whether you’ve played it once or a dozen times, that aren’t even remotely fun to repeat, and that often have little to no direct linkage with the previous or subsequent missions you’re offered. For Star Citizen, we’re going to try and do something a lot different. Designers are going to hand-craft individual mission components, and then specify how those pieces can be customized at run-time and linked to others to form coherent chains that effectively represent small, unique stories consisting of multiple sequential mission objectives. Responding to a simple distress call, then, becomes a lot more interesting because – just like in real life – you never know what might unfold as a result. Prevent a freighter from being destroyed by brigands, and you might collect your reward and leave. Search the computer core of the attackers’ ship, though, and you might ascertain the location of one of their remote outposts. Infiltrate that base and you might learn where they stash their stolen booty and make off with a fortune…if you can figure out how to defeat or draw away the heavily armed ship defending it. I think that the end result is going to be considerably greater than the sum of the individual pieces, and provide players with a much more diverse and interesting world to explore, where there are a practically infinite supply of threads that, when pulled, can dynamically instantiate long strands of a story that the player can choose to pursue or ignore as they see fit.

Or this:

Every NPC will have their own individual 24-hour schedule periodically constructed for them based upon things like how the economy is doing and what’s happening and within the solar system. A prosperous, growing economy might be reflected by the streets being filled with businessmen, tourists, and shoppers by day, and bar patrons, restaurant diners, and revelers at night. Criminal activity – muggings, vandalism, pickpocketing, and even assault – might rise significantly once the sun sets, especially in those areas off the beaten path or on planets facing economic hardship. Some stores will close at night, leaving travelers arriving at odd hours and in need of their services and/or products to either burn some time at one of the establishments open around the clock, or trying to find an alternative source for their needs, which might mean trying to strike a late-night deal in a dark alley with a character of questionable integrity.

Or this:

The composition of characters that you see at the various landing zones – criminals, vagrants, factory workers, businessmen and women, tourists – won’t be a constant but will instead vary according to the results of a simulation tracking the individual desires and actions of millions of NPCs. That simulation – which will be impacted by the actions of the players – will also dictate the prices of various goods and services, the type of jobs available and what those jobs pay, the amount of criminal activity in an area, and many other such things. As a result, the visible fortunes of a given area will ebb and flow with the overall economy, which will make planets seem more like a living, breathing entity than a static area where you occasionally stop to refuel your ship. If an asteroid belt is discovered that’s laden with valuable elements, the economy of a nearby planet might explode as more and more NPCs are attracted to mining those assets, and fewer are willing to work in the factories. More cargo ships would be needed, which would in turn spur demand for component parts, raising prices on a whole slew of other things. Hijackers would be drawn to the lightly armed transport ships, which would result in more conflict in the shipping lanes and, eventually, significant increases in things like the cost of weaponry, ammunition, and protective escorts for that area. As the resources of the mine were gradually depleted, or more NPCs or players showed up to exploit the situation in one way or another, the planetside economy and activity within the system would reflect those changes. Over time, then, systems will face considerable change across a range of different areas.

 

[duffbeeeer]

Try with this:



Or this:



Or this:

The last paragraph reads exactly like the UO article back in the 90s. Almost literally, just exchange ships and npcs with dragons and heroes. I even found it:

http://www.spiegel.de/spiegel/print/d-7861892.html

Sorry its in german but this is the original copy & paste from 16/1998.

 

[ZeeWolf]

Ok i don't really see why you would write a background simulation to track the "hopes and desires" of NPC's over a 24 hour period. The end result is always the same.

Player enters instance and sees some NPC's.

You can get all the results based on the economy and state of the universe at the point you generate the NPC's in the instance. It's like the virtual pet example I gave earlier in the thread, there is no point tracking what something is doing when all you need is to know a few parameters at the point you ask the sever for the data, when it was last seen and what's going on now. All you need to do is convince players that it isn't in another instance.

I don't know if it makes that much difference to players whether someone else is seeing the same NPC's anyway, but you still don't need some elaborate background system. So I'm not sure what the goal is there.

 

[Ian Skippy]

Ok i don't really see why you would write a background simulation to track the "hopes and desires" of NPC's over a 24 hour period. The end result is always the same.

Player enters instance and sees some NPC's.

You can get all the results based on the economy and state of the universe at the point you generate the NPC's in the instance. It's like the virtual pet example I gave earlier in the thread, there is no point tracking what something is doing when all you need is to know a few parameters at the point you ask the sever for the data, when it was last seen and what's going on now. All you need to do is convince players that it isn't in another instance.

I don't know if it makes that much difference to players whether someone else is seeing the same NPC's anyway, but you still don't need some elaborate background system. So I'm not sure what the goal is there.

If it doesnt matter you just blackbox it yet tell people you model it anyway. Thats what CR does a lot: windowdress (sensible) shortcuts. If the player dont notice either way, there is little harm in making it sound cool as long as the devs can just create a proper approach.

I like how the first quote is basically ED's multi-chain missions plus wrinkles. And how they claim 'designing parts of PG missions by hand' is new, rather than the standard.

Next up: rather than making pictures of giraffe vomit and layering that on our meshes, we've employed real-world artists to handcraft what we call 'textures'.

 

[duffbeeeer]

Ok i don't really see why you would write a background simulation to track the "hopes and desires" of NPC's over a 24 hour period. The end result is always the same.

Player enters instance and sees some NPC's.

You can get all the results based on the economy and state of the universe at the point you generate the NPC's in the instance. It's like the virtual pet example I gave earlier in the thread, there is no point tracking what something is doing when all you need is to know a few parameters at the point you ask the sever for the data, when it was last seen and what's going on now. All you need to do is convince players that it isn't in another instance.

I don't know if it makes that much difference to players whether someone else is seeing the same NPC's anyway, but you still don't need some elaborate background system. So I'm not sure what the goal is there.

Its the stuff childhood dreams are made of. Unknown worlds with their own rules. Which teen male doesnt get excited by stories like that? It also implies that all the parameters can be manipulated somehow and you as a avatar have a real impact on the world. Imagine I shoot down your favourite quest giver and he wont be there anymore. Thats the stuff power dreams are made of too. But yeah that stuff is pretty funny anyway because a computer game is the definition of the illusion of the mind. CR is just selling his illusion as more real than others.

This whole premise can work if you replace the npcs with players. EVE Online did that and was pretty succseful while doing so. But to enjoy that game on that level you have throw your whole life away and live a virtual one. At some point one must accept that these are still virtual worlds on no real grounds or just buy the VR headset and be gone with it. I prefer the first option.

 

[n4p0l3on]

what many people still fail to understand is that, even if CIG or anyone else, or literally all programmers in the world successfully create such complex and complete systems for sc, how are they really going to make it run for from hundreds to millions of players from all over the world?

we have already seen lags on games like wow and eve, and those games only have like 1/20-1/50 the complexity of sc, performance nightmare is going to be a real problem...

then there are exploits, bugs and glitches that needed to be dealt with, we have seen this on elite, like bounty systems that already evolved with different iterations... with the complexity of sc, changing or fixing bugs will be a mess, maybe beyond nightmare... I already gave example here with crates trolling... it's extreme, but so was station camping on elite... the thing is, the more complex your entire program is, the more issues you will face when changing something...

and they have been working on this for 4-5 years, with no apparent substantial result (not even many of the fundamentals are in strong foundation), not to mention the whole squadron 42 thing, their money won't last forever, even the white knights and fanboys will tire from waiting eventually and thus no more income for them, how will they continue to work then?

 

[Asp Explorer]

Black boxing Star Citizen unfortunately only leads to three results.

T pose
Explosion
Crash to desktop

 

[DoctorDanny]

You misunderstand. You need to be a programmer to know if a specific idea can be implemented in your language. You dont need to be a programmer to know that some ideas are fundamentally impossible. Some ideas about SC are fundamentally illogical: no amount of hardware, magic germans or super languages can make x==x+1 work. If you have an idea that is logically internally consistent, only then do you need programmers to tell you if it is feasible/practical in this or that situation/context.

Thank you for explaining this to me. I had no idea. I allways though that shards and instances have to do with networking and coding that I know zero about. Turns out that I can have an opinion without knowledge

 

[ZeeWolf]

Its the stuff childhood dreams are made of. Unknown worlds with their own rules. Which teen male doesnt get excited by stories like that? It also implies that all the parameters can be manipulated somehow and you as a avatar have a real impact on the world. Imagine I shoot down your favourite quest giver and he wont be there anymore. Thats the stuff power dreams are made of too. But yeah that stuff is pretty funny anyway because a computer game is the definition of the illusion of the mind. CR is just selling his illusion as more real than others.

This whole premise can work if you replace the npcs with players. EVE Online did that and was pretty succseful while doing so. But to enjoy that game on that level you have throw your whole life away and live a virtual one. At some point one must accept that these are still virtual worlds on no real grounds or just buy the VR headset and be gone with it. I prefer the first option.

Yeah, but you'd shoot your NPC in an instance right? What happens to your NPC when nobody is in an instance with him is completely inconsequential.

I don't want to upset anyone, but when your off fighting Dragons in Skyrim your PC isn't updating all of the NPCs in your local village in real time.

 

[duffbeeeer]

Yeah, but you'd shoot your NPC in an instance right? What happens to your NPC when nobody is in an instance with him is completely inconsequential.

I don't want to upset anyone, but when your off fighting Dragons in Skyrim your PC isn't updating all of the NPCs in your local village in real time.

To me the whole 24h cycle idea was based on persistence. With instancing that doesnt really matter, you are absolutely right. But I thought CR promised to make a real world D) and not an instanced one? Isnt that what all those people critize about Elite Dangerous and hope SC will fix all of it? The instancing?

Also if you are not being consistent on all instances, everyone plays his own version of SC and not a real multiplayer world. Again its the biggest point of critique about Elite Dangerous.

The longer SC takes time to develop the more I get the feeling that FD had the right idea of what is possible and what is not. I know many people were upset about the latest Q&A with Sandy about space legs being far off but to me this was brutaly honest. I like that much more than all these pipe dreams CIG is selling.

I will be really suprised if SC can pull all these things of with out getting repetitive or grindy.

 

[Zyll Goliat]

Ok i don't really see why you would write a background simulation to track the "hopes and desires" of NPC's over a 24 hour period. The end result is always the same.

Player enters instance and sees some NPC's.

You can get all the results based on the economy and state of the universe at the point you generate the NPC's in the instance. It's like the virtual pet example I gave earlier in the thread, there is no point tracking what something is doing when all you need is to know a few parameters at the point you ask the sever for the data, when it was last seen and what's going on now. All you need to do is convince players that it isn't in another instance.

I don't know if it makes that much difference to players whether someone else is seeing the same NPC's anyway, but you still don't need some elaborate background system. So I'm not sure what the goal is there.

The goal is to pump as much adrenaline as possible into the whale-citizens bloodstream so that those people again can admire how complex&real SC is going to be........but the reality is that this is just one of the latest story in Roberts books of tales and he love to do this kind of magic tricks again&again.....let´s just remind ourself with previous few:Intelligent Flight Control System (IFCS)-and all complexity behind that but reality is opposite and FM feels just like crap,Oxygen&preassure realism ahhh yeah we all know how that goes when you walking around totally naked in space???,DEATH OF SPACE MAN-this one is still pending I guess but never the less it was the one of the tales that rise the Hype levels out of the charts....etc.....there is so many more tales and none of them so far become REALITY that we can feel it in current RL demo of the SC after the 6 years of development.....

 

[ZeeWolf]

To me the whole 24h cycle idea was based on persistence. With instancing that doesnt really matter, you are absolutely right. But I thought CR promised to make a real world D) and not an instanced one? Isnt that what all those people critize about Elite Dangerous and hope SC will fix all of it? The instancing?

Also if you are not being consistent on all instances, everyone plays his own version of SC and not a real multiplayer world. Again its the biggest point of critique about Elite Dangerous.

The longer SC takes time to develop the more I get the feeling that FD had the right idea of what os possible and what is not. I will be really suprised if SC can pull all these things of with out getting repetitive or grindy.

An instance is a group of connected clients with or without a server. SC must have instances of some kind by definition. I don't know exactly what you mean by "real world" but when it comes to simulation type games there are obvious limits to how much current networks can keep in sync