I've lost track, and I'm too drunk to look back. But "Star Engine" was a set of bolt-ons for CryEngine, and there's no reason they cant't be bolted onto Lumberyard with no fuss. the whole "OMG they changed engines" is a storm in a teacup.
The Star Citizen Thread v5
- Thread starter Brett C
- Start date
- Status
I've lost track, and I'm too drunk to look back. But "Star Engine" was a set of bolt-ons for CryEngine, and there's no reason they cant't be bolted onto Lumberyard with no fuss. the whole "OMG they changed engines" is a storm in a teacup.
Correct SE is modified CE 3.7....LY is based on 3.8...but what do you expect from some usual shilizen who think that switch on LY is magically going to solve ALL NETWORK PROBLEMS IN THE WORLD...
Last edited:
1. draw diagram explaining how StarEngine and Lumberyard branched from different CryEngine versions
2. quote CR claiming they're based on the exact same version
3. act as if that confirms what you said
His confidence that all the problems have been solved is justified by the thousands of players in an instance in 2.6.
Once again a typical example of SC dreamweaving. This guy does not provide any proof that LY will make it happen but instead simply tries to dumb down Chris Roberts Statement how its supposed to work and just regurgitates non-information promising something that isnt yet a thing on the horizon. Zero information in that video.
Last edited:
What are you talking about what 1000 players in the same instance in 2.6????When did that happens???Are you for real or you are been sarcastic???
Now If you logg in PU 2.6 the MAX number of players in same instance is 24!!!Same goes for the SM only in Arcorp was maybe around 40 players wich is totally expected as you can´t do almost nothing in there........
Last edited:
No doubt about it, I m just saying he surely has taken a bit of the cake already, I know thats the first thing I would do to secure myself.
You paint a sad picture there but I agree with you on those points. Still it would mean hes richer by millions, thats more then honest working people can say and have to make do with in order to keep living.
I m surprised and happy to see you delving into this (theoretical) scenario so willingly. It shows you are aware of possebilities but made a conscious choice without being ignorant to other realities. I respect that
Said this before.
To me this is an Investment
It might or might not Succeed.
Risk vs Possible Gain.
Just like Poker or any other type Investment from Stocks to anything else. You should never Invest more into this than you can afford to lose.
This not only makes it much less frustrating when it Fails after all. It also allows you to keep a Cool Head on it and not go delving into Dreams or Nightmares about it.
Its something I really dislike about this.
SC is a Monstrous Project. And on both sides of the Coin you see an Complete Overreaction.
One Side is Painting an Super Dread Worst Case Scenario of how this will be the End of the Work and can never Possibly Work.
While the other Side is Painting an Super Dream Fantasy Scenario of how everything is fine and will be coming without Problems
If People would just Sit down calmly and rationally. And stop investing so much Emotion into this. They would Realize. That in terms of Development. Star Citizen is making pretty much the Progress thats Normal for such an Project.
Anyone looking at this Calmly knew right away that the Ridiculous Schedules they had could never Work.
Their Schedule was still that of the 500k Dollar Game that they Originally had on Kickstarter. Up to this Day I wonder how anyone could possibly even consider this to work out.
WoW back then Cost roughly 60-70 Million for the Base Game to Develop. And took 5 Years to be Finished by an Well established Big Company with Experience on Large Scale Projects and all the Systems for this in Place.
Star Citizen does not have that. They did not have such an Network and Preset much less the Experience for this. Meaning they had to Grind up all of this out of the Ground and Build it First. So you can assume it took just 1 Year to even get Development Started Properly because they had to somehow get up an Company of the Scale to even Start on a Project like this.
And by now their Stretch Goals include up 70 Million as well. This is likely an Complete False Measure tough. Just getting all the Stuff needed must have Cost over 10 Million. So the actual Project Cost is likely about 100 Million at least.
Leaving them 40 Million of Buffers and Overtime Expenses.
An Realistic Estimation for Development of a Game on this Scale would have been 7-8 Years. Possible Delays and Failures along the Way adding Delays of 1-2 Years thus getting it up to 8-9 Years.
Said this before. But anyone Expecting an Properly Playable Game with an majority of the Promised Features before 2019 is Dreaming at best
We might see some Open Beta by the end of 2018 which will be Playable but likely lack a good part of the Promised Features and also a good part of the World. But that depends on their Priorities.
This is one Mistake they Made to be Honest.
They apparently wanted to get this out Finished.
But thats Ludicrous
They would have been far better off. Rushing for an Open Beta. Doing the Physics, Flight Models etc Properly. Adding NPCs and Mission Content and Working an Part of the World so they could have gone into an Open Beta by 2016-2017. This would have solved alot of the Worries of People because even if alot of Features are missing. The Base Game with its Mechanics would have been there already.
But I assume that this Shows their Inexperience with Large Scale Projects. They kept handling this like the 500k Dollar Game they Originally wanted to make. Trying to Finish all the Parts before Building the Ground Frame.
Thanks to that they now got a Giant Puzzle Box of real high Quality Parts. But have not really been working on the Base Frame. Meaning that they now have to get the Frame Working First to get something to Present.
The Positive Tradeoff on this is. That once the Frame is working. The Beta itself will be starting with much more Content than People are used to from Games. And the Progress in Adding more Content will be Feeling Fast. Because they have alot of Preworked Parts that will be Added throughout the First Year
Ok.
Long Text small Message.
Keep a Calm Head.
Leave Emotions out of it.
And given they dont Fail Badly. Expect them to Put out something that can be called a Game in 2019
Last edited:
The thing is, CR hasn't actually promised anything that's impossible - on it's own.
1000 player instances? possible at a low data rate. look at reddit - Millions of "players" in one instance.
Realistic, real-time animations? possible with fast data turn-around and low fidelity. Guildwars/WoW
Fedelitiousness like reflections in eyeballs? possible with very low numbers of players & lots of processing time. Anything from Quantic Dream
But 1000 players in photo-realism and real time? Basic laws of information theory and physics say no. Somehow, nobody is questioning that.
1000 player instances? possible at a low data rate. look at reddit - Millions of "players" in one instance.
Realistic, real-time animations? possible with fast data turn-around and low fidelity. Guildwars/WoW
Fedelitiousness like reflections in eyeballs? possible with very low numbers of players & lots of processing time. Anything from Quantic Dream
But 1000 players in photo-realism and real time? Basic laws of information theory and physics say no. Somehow, nobody is questioning that.
Last edited:
I understand that actual development time of SC now probably is 10 years or more, but problem here is that CR is lying every day about release dates (like v3.0 in 2016) to get that sweet backers money
But that would mean everything the guy in the video was saying was nonsense!!!???
Because >95 % of backers have zero idea how networking in games works...those who dare to question "The vision" gets downvoted and/or banned...
To be honest is even painful to watch such dreamcrafting from incompetent people...
Correct....I mean even with lame Crytek netcode you can have hundreds of players in same instance as an example I played for years Entropia Universe wich is based on CE2 atm but this type of games are not Fast FPS action type it´s more like traditional mmo icon based where some lags are tolerated and netcode calc are on much lower demands.....I mean that is what we can see in Arcorp in SC where atm we have more players in same instance but they can only chat/dance/walk/run there is no shooting or anything that demand higher net-calculations........
- - - Updated - - -
But that would mean everything the guy in the video was saying was nonsense!!!???
Last edited:
Of course not, because digging for any such proof would unearth the statements from Amazon that the engine and backend are not suitable for any kind of persistent MMO-style game that requires constant server connections and continuous database writes to store and manipulate world state — especially not in real-time. Reality has a strong anti-dream bias, so it's not really something you'd want to mix in with any discussion of what someone as inexperienced as Chris might dream up.
If there's any hype that is stronger than that of SC, it's the hype of the cloud. Obviously, such silly laws do not apply to someone else's computer.
Last edited:
"Blue Mars" is a good one - basically a CE-based Second Life. Many people, low datarate...
Yep...exactly....and many other games that are based on much better and more suitable game-engines for those type of games,EVE is the pioneer and probably still the game with higher numbers of players per instance....but when it comes to the Action/based games only Planetside2 reach some crazy numbers and still that game doing some kind net-"trickery"and thats not actual number of players in same instance as I recall......
Last edited:
Planetside 2's virtual instancing is pretty impressive. It's very server-intensive and it doesn't quite work properly but it's basically witchcraft.
Planetside 2's virtual instancing is pretty impressive. It's very server-intensive and it doesn't quite work properly but it's basically witchcraft.
I'll take your word for it as I never really look deep into PS2 and I barely played few times......still PS2 is not even close to what´s been promised in SC when it comes to the data transfer between server&players........
Last edited:
Planetside 2's virtual instancing is pretty impressive. It's very server-intensive and it doesn't quite work properly but it's basically witchcraft.
Yeah, PS2 instancing is not really instancing, as just visual reference system combined with clever LoD trickery. As many have pointed out you can't really interact with everyone in PS2 unless they are in clear vicinity.
It is not that SC can't have such instancing - it can - but what's really goal there. Impression of huge battlefield? If scale is correct no one will see that far anyway.
Again, fact that even CIG really haven't figured out how to proceed there raises many warning bells.
I understand that actual development time of SC now probably is 10 years or more, but problem here is that CR is lying every day about release dates (like v3.0 in 2016) to get that sweet backers money
Possible.
Its also Possible that he (known as an Dreamer and someone making impossible demands of his Employees anyways) due to his Inexperience did simply not realize that meeting such Deadlines was never Possible.
I doubt that he ever dreamed about the scale this has become.
Back when I was an Apprentice. We were given a Large Scale Project to Manage.
This also Included giving out Schedules and Plans.
The Result was something that any of our Teachers back there knew Perfectly well. It was the same Result they always get when setting the Apprentices onto an Project far outranging their usual small Groups Scale.
Working on a Project which Involved over 10 different Teams. Coordinating and getting the Work done Building the Parts and Connecting them etc etc. Showed very fast that all our Predictions were completely off the Mark.
We estimated to be Finished in 2-3 Hours. Because thats how much time we Expected to be doing for the Work that needs to be Handled by each of us.
But we did not expect this to be this hard at all. We were Seperated from each other as Teams. And the Leadership Team ended up doing no Work at all but pretty much only walked around between the Different Rooms they Put the teams into. To relay messages and changes as well as Problems. And thats the thing. Nothing can be Preplanned 100% so any Change to Plans. Any Problem arising. Any earlier decision overturned. Resulted in the other Groups doing double or triple work and scrapping it again.
We needed nearly 2 Days (16 Hours as we had 8 Hours per Day) to Finish the Base Job. And it was Riddled with Problems which would have need another few Hours to Fix.
Roberts can be considered similar to this.
He might have had the Training to do actual Work himself.
But if you check the Projects where he had to take of Coordination and Leadership. You notice Fast that this was usually stuff where he had one Group working on things.
He has Zero Experience on handling 10-20 different Groups working on the same Project.
So while the Possibility cannot be Denied of course.
The Chances are that he has no Ill Intention or something there.
But that he himself simply did simply not know just how much the Timetable is thrown up by increasing the Scale of an project.
Its good that he Got 140 Million for his Goals which Reached 70 Million.
He would never have managed this Project with 70 Million.
Much less with such an Inexperienced Team and Leadership.
1000 Players in the same World. And 1000 Players in the same Instance is not the Different thing
The Trick is to have Seamless Transitions between the Instances. And to make sure that each Place only has one Instance so Players cannot walk into the same Spot in the Game without meeting each other.
This Generally done by doing an Grid Field of Instances. (Or in a Space Game it likely will be an Cube of Cubes
) Each Instance is an Cube of 10x10x10km Size for example.
The Game in the Backround will Load the Cube your in. As well as any Cube Adjacent to yours.
And anything within your Cube or within the Neighboring Cube can be Spotted and Interacted with.
Moreover you will Preload all Cubes Adjacent to those.
Each Cube is its own Instance.
But each Instance even the ones Neighboring you will already Calculate you.
So you got
1 Cube where your in and which Defines your own Position.
26 Further Cubes Surrounding that Cube which are what you can Fire upon and what you can See.
and another 98 Cubes around that which are Preloaded but dont actually Affect anything you do right now.
This will be done in Initial Loading. So it wont Affect your Performance.
Later you will then Move around. So when your at the End of your Cube. The Game will Connect the Preloaded Cubes to your actual Interactive World. And also start Preloading the next line of Cubes in that Direction. Thats 25 Cubes then Loading in the Backround while the Row behind you will be Disconnected.
This way Players can Play all around. All in the same World. But still being Calculated by Several Different Instances
Of course. If the Players then Decide to do like they did in EVE Online.
And have an Insane Battle with over 7000 Players in the Same Place. xD
This whole System will (just like in Eve Online) Break apart and go Super Slow and at some point just get Stuck xD
But for the General Playing. This System will work Fine
- - - Updated - - -
The System works fine tough.
Because thats the thing. Beyond your Weapon Range you wont be Interacting with anyone anyways.
So them being Possible for Interaction or not doesnt affect you.
The moment you come closer to each other. You will also be able to Interact.
So for you this is Seamless and entirely Fine.
As an Player an Instancing being there doesnt matter as long as you dont notice it.
Planetside 2 made an Awesome Job on this actually ^^
They're entirely different.1000 Players in the same World. And 1000 Players in the same Instance is not the Different thing
The former is pathetically low; the latter is almost impossibly high.
The first means you're not having that many people in an instance; the latter means you've just lost one of the main benefits of having instances: the ability to have a ton of people in the same place without running into population locks that keep everyone out. The rest of your description is just a complete confusion between what an “instance” is, and the kind of visibility/communication culling that a game such as PS2 employs. Oh, and loading new data as you certainly affects your performance — increasingly so as the world around you gets populated with more and more people, and definitely so if you use fixed sizes rather than a proper dynamic setup.
A couple of problems with that comparison. For one, EVE does not use instances. For another, the whole point of CCP's setup is that it didn't break and you didn't get stuck, in spite of not instancing up the world the way a less ambitious design would do. TiDi certainly made things slow, but it made them controllably and reliably slow, rather than just unresponsive due to computational overload. So almost nothing about EVE is applicable to how SC is supposed to work (doubly so since SC won't have the custom server architecture required to do so, but rather on a generic setup that very specifically can't handle that kind of work).
In other (Archered) news, a trip down memory lane to a more innocent time.
Note the post date: December '13…
Last edited:
Good luck programming netcode for SC if there is fast FPS action on one ship (Idris) and several ships are flying nearby and shooting a lot of bullets and lasers at same time.
And do it in LY engine that currently is in Beta stage.
So Beta engine, Pre-Alfa game what here could possibly go wrong
And do it in LY engine that currently is in Beta stage.
So Beta engine, Pre-Alfa game what here could possibly go wrong
- Status