StarCitizen Server Meshing Architecture
sc-server-meshing.info
sc-server-meshing.info
I really hope server meshing pans out, but I'm skeptical that RSI has the chops to pull it off. The number of objects in SC is huge and the world is seamless. This not something that's entirely new though. EVE Online has meshed servers for persistence of tens of thousands of concurrent players in a single universe successfully for decades now.
All have their faults, but there are obvious alternatives to StarCitizen out there that's not vaporware.
None of those offer all the things as SC even at it's current state. Elite Dangerous is closest and even then they're just different games inherently.
> there are obvious alternatives to StarCitizen out there that's not vaporware.
You can download and play Star Citizen right now. It is, by definition, not vaporware.
Starcitizen is for better or worse unlike any other. Until another game comes where me and my friends can hop onto a capital ship and travel to planets etc seemlessly - ill switch. But nothing on the horizon yet.
Also love the ground vehicle gameplay too and its quite underrated.
I suppose Elite's one saving grace is that it doesn't claim to be in alpha.
A lot of great gameplay mechanics have been added over the years, but it hasn’t all been forward progress.
No Man’s Sky is the total opposite and I really want to come back to that at some point and get lost in it the way Elite was in VR.
Truly transported you elsewhere and gave you presence in an empty uncaring galaxy that made the tedium and grind still feel like something.
On the other hand, over the years they've left so many things broken/unfinished or just in general avoided certain things, makes me wonder if maybe the foundation itself was unstable. Eg powerplay being ignored even when it wasn't working right, many years between the addition of new ships and the lack of creativity in terms of the ship's capabilities, similarly with SRVs, weapons bugs, completely failing to capitalize on the excellent PvP mechanics they have, the relative standstill of the Thargoid plot until people largely lost interest, the stagnation/disconnect of Colonia from everything else etc.
If you want to be seriously competitive in PvP, it's required to do Engineering.
It's funny, we've seen games released by the previous generations leading lights in game design and they're all disappointing so far. Relying on boring grinds, poor difficulty curves, etc.. Julian Gollop, David Braben, Chris Roberts, Peter Molyneux, etc.
It's like they can't adopt the new ideas from the next generation to make their games fun. Quite eye opening on how old age or perhaps success, not sure which, can make your thinking rigid.
Elite Dangerous is one of the worst. It's simply not fun. So pointlessly grindy for what is predominantly a single player game. I was so excited for the first 5-10 hours, and then so disappointed as it was obvious that every mechanic was just another massive progress bar that barely moved after hours of play.
No Man's Sky was much better, 40-50 hours fun play before the procedural nature of it became too obvious for me to enjoy it anymore. I played it 2-3 years after launch though, after all the updates. I'd probably play it again over a holiday weekend if I ever get the MS game pass again.
Speak for yourself, I had a lot of fun for hundreds of hours and revisited the game multiple times over the years. I did space trucking and mining and combat at various times, and enjoyed never paying a monthly fee for an MMO.
Elite Dangerous was announced at the same time as Star Citizen, for comparison, and it's so old now that the main reason I don't play it anymore is because I did everything I could do in single player and the concept has finally lost its allure.
I have a friend who played thousands of hours because he was more social than I and wound up in a large player guild.
I love(d) Elite so much that I bought a VR headset and a joystick solely for it, and I don't regret these purchases.
Lots of E:D's own player base acknowledge this:
https://www.reddit.com/r/EliteDangerous/comments/qk6kqs/why_...
I bought the game based on what I see people doing in it already, and it was fun enough for me. That definitely won't be the case for everyone and I recognize that. So to me it's not vaporware and there are lots of other people still enjoying the game today.
It's a cool project though, and I do enjoy the game, I just find that their plans and announcements should always be taken with a liberal grain of salt.
That toggle adds so much depth and skill to combat in Elite, it's a shame it isn't a more common thing in space games (also a shame that Elite itself doesn't really take advantage of it).
It was fun in the old first Elite games :)
And I follow SC just from the outside, but at least partially, I get that impression. So sure, shiny new ships are coming out. But the game itself does not seem to get much more stable and .. playable.
https://www.youtube.com/watch?v=-tB3cark5lA
But I'm in a similar boat now, backed on Kickstarter almost eleven years ago now, and playing the game is just too much of a job.
In its current state, you spend 20 minutes riding elevators and trains around to get off a planet, and then crash into an invisible asteroid and explode, respawn and start over. And add a bunch of errands flying around the system to replace all your gear that blew up.
When it works it's very cool. But unless you have a lot of free time, not worth all the times when it doesn't work.
I hope they finish it, once it stops wiping progress and is possible to slowly accumulate longer term progress I'd be more willing to put in the effort. As-is, I'm not putting in any more money. They need some pressure to actually finish the damn thing (especially Squadron 42), and the bottomless fountain of crowdfunding with no publisher calling the shots clearly isn't working.
As do we all, but CIG doesn't make it easy to hold on to that hope when their transparency has all but disappeared. Squadron 42's store page has been removed from their website for months now, where any links to "pledge" lead to a 404 error. The only acknowledgement of it from them was buried as a reply to a days-old, user-posted forum thread where they claim it was removed for a price change. That was months ago. Meanwhile, multiple ships' prices have been updated.
EVE uses a different, simpler model where each system is a process, and players connect to proxy servers that keep track of which system they're on (and therefore which server they need to connect to) [0]. I'm sure there are layers of caching and other optimizations built on top of that basic structure, but that's just Gall's Law in action: a working complex system invariably evolved from a working simple system.
I don't think it's a coincidence that the architecture that I can describe in a single sentence is the one powering the legendary 20-year-old game, while no one seems to be quite able to explain what "server meshing" will look like except that it seems to have a lot of moving parts and still hasn't been finished.
[0] https://www.engadget.com/2008-09-28-eve-evolved-eve-onlines-...
Note that I don't play or follow either game, but I think you're making an unreasonable comparison.
SC was from the start supposed to be a giant multiplayer world, yet they picked cryengine, which can barely handle simple shooter multiplayer gameplay. Half of the updates Chris has given on development are just over-explaining simple multiplayer game engine functionality that they've had to implement from the ground up, while you can download unity or unreal and replicate it in the next hour.
I've always found him to be full of bullshit. The kickstarter video had him spend like ten minutes waffling on about how the ships have thrusters on them and all the physics are calculated off those actual thrusters, as if 1) that's not just a useless implementation detail that changes nothing about the actual game, 2) as if that's impressive in 2012 instead of the exact thing the rest of the industry was doing, 3) as if that was hard work instead of just a normal thing a video game developer should be comfortable building, 4) AS IF IT MAKES THE GAME MORE FUN
I don't understand how he sold anyone off that video. The sales pitch wasn't even aspirational, it was clearly and obviously a grift, though maybe unintentionally, from the very beginning.
It's a great example of someone thinking "more simulation detail === more better". Every game developer should really attempt to take a class on the human psychology of games.
Star Citizen seems to have 100k+ entities per system[2], some with extremely complex collision (eg player models, ship interiors), many of them interacting with each other on each timestep, probably very few of them sleeping.
You can't fix that with faster processors. All those updates trying to go out to players, all subject to checks, all needing updates. It's not in the same class of difficulty.
[1]: https://en.wikipedia.org/wiki/Battle_of_B-R5RB [2]: https://www.reddit.com/r/starcitizen/comments/qbqv7v/heaps_m...
Improbable seem to have gone a bit... off the rails(?) in recent years, diving into Web3, metaverse, and defence contracting, but I assume the core tech is still there.
I’m not exactly an expert in them or anything, but the last time I looked at them at all (admittedly many years ago) the whole concept of what they do seemed both very expensive and mostly useless for real games.
More so than Star Citizen?
I don’t know if their Unreal offering is comparable to what they were doing with unreal.
You should do tech improvment to enhance the gameplay, and currently I don't see many gameplay improvements that have real value.
Starfield maybe the first space opera fantasty that will deliver.
If this game ever actually comes out and doesn't cure at least one type of cancer the weirdest online people are going to be furious.
And, part of the game at this stage is watching it develop.
The amount raised/sold so far is massive, but people keep buying because they’re getting something out of it.
I know I left a few comments like this on this thread already, but I'm a big fan of the game. As a game dev, I have a lot of respect for their ambition, tech, and recent transparency. I'm not sure why people always fixate on the money - people invest and donate to worse things. RSI isn't predatory about it, they're very honest about the progress and expectations (especially recently)
PVE - Cargo delivery missions (from small to large size) - Bounty hunting (in space and on planet) - Mining - Salvaging - Cave exploration/rescue - Escaping from prison
PVP - Bounty/pirate hunting
Just search for Star Citizen content on YouTube. You fill find a lot. My favorites are https://www.youtube.com/@CaptainBerks and https://www.youtube.com/@Morphologis
Squadron 42, the cinematic, single player, space campaign, was supposed to come out in 2014!
Perhaps you don't know how evocative that is, and how cynical it is for you to compare that to procedural MMO quests.
They're very different things. It's like paying for a gourmet meal and then being given a month's supply of tasteless protein bars.
Isn't StarCitizen funded by their Kickstarter though?
So it would be more like someone asks you for money so they might make you a meal, of unknown quality with unknown timeline, and won't promise you anything even after you pay them. Which is standard for a Kickstarter.
The game has major ups and downs, but nothing really comes close to the same feeling of being on a capital ship with your friends pulling out from a beautiful space port. I wish there was a better alternative but there isn't, nothihng really comes close.
Performance has vastly improved from 50 players per server to 110 now i beleive? And i think theres talks of increasing it further.
Also free fly weekends are when the game is at its worst, the servers are bought to its knees.
Again, no fault to the player and its completely on CIG but its a shame that a lot of players never get to see the game in its best light because free fly is a nightmare.
It fails to load geomoetry on slower disks sometimes, i.e. insides of an elevtator which causes the player to clip through and the game tries to compenstate and cause funky reactions.
It fails to load geomoetry on slower disks sometimes, i.e. insides of an elevtator which causes the player to clip through and the game tries to compenstate and cause funky reactions.
There had/have been recurring problems with elevators and inventory. I would get ejected into space; no leg breaking.
I think just the way they mesh the world together makes that something that has been hard for them to fix.
The inventory issues are what got me to stop playing. Long inventory load times, and then 75% of the time, inventory would be lost or permanently broken (items lost/can't add new things to inventory).
Still fun for a few days. Pretty graphics. Space flight was fun. FPS missions needed a lot of work, but functional.
Another thing that made me quit, was the tedium of setup. You respawn quickly, from death into the hospital (unless you end up in prison, for shooting an at FPS guard that looks 90% exactly like an FPS enemy (then you get the prison play loop, which is fun once or twice)). But, then you need to take the elevator (chance of instant death) down to the ground floor. find the train. wait for the train. take the elevator (chance of death; or stairs if you are lucky [ i feel like there's a 'I'm gonna git you sucka' joke in here ]) to the shopping district. Wander around the shopping district (which is different on each planet) to hopefully find new Armor, Rifle, Mining tool, Gravity-gun tool (not all planets have the things you need, even)... then get back to the train and repeat for the space dock. then elevator to your ship bay. ... that's like 10-30 minutes. it just gets tedious. - probably more fun playing with groups of people, but it got too tedious playing solo.
Was an early Kickstarter backer, so I'll try it again at some point when it is more polished.
Agree that what I paid for was mainly a modern Wing Commander/Freelancer type game, and they are overreaching by a lot. I think the First-person-shooter stretch goal was added later in the funding campaign.
I also feel like a lot of the kick-starter ships have been power-creeped (and they also seem to be lagging behind in quality updates; because CIG has to design and sell new ships to make more money to fund development, and then most of that money goes to make new ships).
They should not be making new ships at all at this point, unless those ships are required for the single player campaign. Focus should be on game systems and game play.
It's not an MMO as such, with something like 50 players per server. The Server Meshing God-tech that they've been banging on about for years now is practically impossible as described. To paraphrase, "you'll be able to shoot a gun from one ship and the bullet will pass through several cloud servers as it hits a player in another ship". Latency and n^2 are not your friends here. Honestly, based on this alone I'm surprised by the amount of Gell-mann Amnesia in the thread.
In my experience it's a "make your own fun" sort of game. Try it on a free play weekend, but IMO don't get sucked in and donate tens of thousands of dollars to a broken CryEngine demo running on promises.
World of Warcraft ("WoW") is the elephant in the room. It killed the previous elephant (Everquest aka EQ) but has reigned supreme in a shrinking space for 18 years. It partially shrank because it used to have a social component that has mostly been replaced by social media but there are other factors too.
But what I scratch my head at is all the bad competitors that came afterwards and predictably failed because they made product decisions. The most common one is focusing on PvP. MMO is an interesting genre because only something like 10% of the population at best cares about PvP. So right off the bat you've reduced your potential market by 90% without doing a thing. But no, you get assured that "our PvP is so good, casuals will want to PvP".
There's another factor here too: in a persistent world where your power increases, PvP players fundamentally want an unfair fight. They want to be rewarded for various grinds. Contrast this with deathmatch and battle royale games where everyone is basically equal other than player skill.
Anyway, Star Citizen has now raised (and spent) over half a billion dollars on the same bad premise of PvP.
Server meshing aims to create a single persistent universe but PvP, which most will want to avoid, is fundamentally unfair because the people with the best ships, paid for with real money, will win. This is such a fundamentally bad and limiting design, it's hard to comprehend.
Imagine instead SC had defined a small core game loop and gone for a Sea of Thieves like world, both PvP and non-PvP. SoT has like 8-12 players on a server and they can PvP. Interestingly, a lot of players don't want this so they monopolize a server and agree not to PvP. Even in WoW, which has faction PvP, the players have basically voted out world PvP by concentrating a single faction on each server with very few exceptions.
But SC is doing things like server meshing when after a decade they still don't have a core game loop. I occasionally check in on the SC roadmap and progress. Not because I'm interested in playing but because I like watching a car wreck in super slow motion. Constant missed deadlines, new (incomplete) features being promised (and a few added), more promised features to fix the bad previous features and so on. It's a disaster.
I would like to see an MMO that really focused on a wide range of gameplay systems rather than content. It's been attempted before, one example I think is Archeage, but unfortunately ruined by monetization.
For those unfamiliar, WoW came out in late 2004 and has had some ~8 expansions since. Each expansions adds new lands, new cosmetics, new dungeons, new raids and new systems. Additionally, these expansions changed existing systems. So much so that what is now refered to as "retail WoW" is considered too complex with too big of a barrier to entry by many. It's partially why the player base has shrunk as the game has increasingly catered to the top 1% to 5% ta the expense of the casual player base.
There has been a rich history of so-called private servers. These are privately run WoW servers that stick to older versions of the game, particularly the vanilla game and each of the first two expansions. These are illegal and never last long.
At Blizzard's annual conference, famously this came up when an audience member asked about it and the head of the game said "You think you do but you don't" [1]. This one guy probably did more to move this cause forward than anyone as it galvanized the community behind this idea.
Then 4-5 years ago Blizzard announced they were releasing "classic WoW". This is basically (but not entirely) the original game with some changes. Some were understandable, some less so. But it was basically the original game.
This has been massively popular and continues to be, so much so that Blizzard just released "hardcore classic WoW", servers where your character is over when you die. Lots of games have hardcore modes but it's interesting in WoW that it may takes ~6 days played to get to max level and a lot of people still want hardcore mode.
But my point is that this 18 year old game with about the same graphics continues to be massively in demand. You don't need to constantly churn out new content to a large scale like you suggest.
Also, SC has over a half a billion dollars and continues to somehow raise a lot through ship sales (which, to me, is bizarre). They very much could afford to continually pump out new content. But there's still only one system with the second years delayed.
Studios like PvP because it falls into user-generated content. The players entertain themselves and that costs almost nothing. But they fail to acknowledge that they're, at best, catering to 5-10% of the potential market.
Besides, in the last 6 months performance has drastically improved, as anyone that actually plays will tell you. I play regularly, do a variety of delivery/bounty/mining/recovery missions, explore planets/stations, have funny interactions surrounding all of that with other players, etc. Rome wasn't built in a day.
Do I have to accept all those excuses, which like those pictures of ships never run low in this company, to comment on the performance? I don't think so.
They promise a lot. They don't deliver. They break things which were already working and I'm still waiting. Those are clear signs of terrible Project Management and I think, I don't have to know all those cool insiders-excuses to be able to comment on this failure.
So please. Spare me your epigrams. They won't change anything about the facts.
This is a ridiculous statement. They've collected half a billion dollars from customers and have been in alpha for over 10 years.
For anyone curious what PES does, when a player places an object somewhere or for example their ship explodes, that object will remain on the server in the same place even if you disconnect. Seems fairly simple from a user perspective but required a lot of changes to make work apparently.
IF you have the hardware to run it and you're a scifi nerd its mindblowing the first time around, even after 3000 hours i still have moments when i'm mindblown. (Recently had a tank that shot my friends ship from the sky during an ingame event)
Except this isn't actually impressive if you've been playing games. Supreme Commander lets you block artillery shells with your planes if you are lucky, with zero scripted event, and that game released in 2007, is single threaded, is an RTS, and allows you to play 8 vs 8 games where EACH PLAYER can control 1000 units and play on a 50km by 50km map. Now that is impressive IMO.
Of course, if you attempted those things with a 2007 computer, you get about 2fps, but it works great nowadays. Also, that company was never handed half a billion on a silver platter before releasing anything.
More to your point, here are a bunch of other games that let you shoot a plane out of the sky with a tank, from before star citizen, that show just how mundane of a situation that is: Arma 2, Battlefield (all of them), Planetside 2, Damn Halo!, Various airplane and helicopter wargames from before the millennium, various tank games from before the millennium etc etc etc.
Hell, Atari's Battlezone from 1980s had helicopters as a target for your tank!
Starcitizen is probably one of the most technologicaly advanced games out there currently barring something rockstar produces in the near future.
You're just ignorant.
But this MMO is the most expensive video game of all time (although there's others that have had more invested in them over time, I'm thinking of World of Warcraft or GTA 5, possibly upcoming titles like GTA 6), therefore the server technology also looks like some of the fanciest of all time. They have 1100 employees working on a game that's been in development for over 10 years now, funded to the tune of $600 million so far, with about 50-60 million still coming in every year just from pre-orders / donations / in-game purchases / ships for a possible game that may come out at some point.
Except it shouldn't take almost 1000 people more than ten years to make a single player space game.
~Im assuming this has to do with a service mesh and kubernetes of some sort. But I'm a bit lost.~
Edit: Ah! Seems like this is a game. But I still have no idea what any of this means :)
Ambitious and such; the only other game I know of that tried it was Eve Online, which has tens of thousands of players in the same world, but which is limited to hundreds of players that can actually see each other.
At 10 Hz that is 40 million updates a second.
There are caches and prediction that can help but I ignore all gameplay in that number, it is just flying around.
That's how.
Combat is all about range, and if you're close enough to actually see the enemy ship models, you're probably about to explode.
I'll be amazed if Star Citizen ever comes close to the scale of EVE battles, though. There's so many technical challenges in making an MMO scale, especially when a game is so focused on graphical fidelity.
(Then again, EVE is a 20-year-old game at this point, and was designed for the limited PCs and bandwidth of the time. But I'm assuming that they're using modern top-end hardware for the servers, and they still struggle with the biggest battles)
I would be surprised if they reached "hundreds" of players in a single instance. I don't think thousands is ever going to be attainable for them. Unless they are somehow able to make some sudden breakthrough that's alluded them in the last decade plus of development.
They have several youtube videos on their channel where their game director has come out to talk about these projects in a technical manner. It's pretty neat.
[I say these things as a backer patiently waiting for more game]
It's a pretty cool vision, but it's just unlikely to succeed in my eyes, and that's why i didn't buy in.
The words they choose must serve the goal of their performance first, conveying technical meaning is a optional secondary requirement
Quantum is a system for simulating activity with economic impacts and determines how much of raw material or assembled products are available and where, how much things cost, how busy and how dangerous or safe space travel lanes are, and other such aspects of economic scaling.
Subsumption AI is used to drive NPC behavior up close. The game is supposed to have a population that is 90% AI NPCs that drive economic variation and action in general. This makes the game world seem alive even if there are no players currently with you in game. The NPCs have the ability to decide what to do, how to move around, make jokes and idle banter, and can use various equipment in the game as well as fight either in person or as spaceship pilots.
Not totally sure about pCache but it appears to be used on the server to capture changes to state in the universe that should be stored persistently. This is part of the server container streaming feature which allows servers to stream in areas and sets of objects as players approach and then stream them out when they are left empty for a time.
I loved the Wing Commander games, but they had people in charge who could actually direct a project. Chris Roberts can't get out of his own head. The video was a clear indication that he was going to ruin it.