Chris Roberts Back in your wallet - STAR CITIZEN

[calapine]

I anticipate this game having terrible performance and major technical problems. I saw the quote where Roberts said that this would be a old school "pc" game -- well, I remember the era of PC games before console ports took over.  A big part of what made companies like id and blizzard popular was that their games actually worked, while EA and Origin would consistently shoot for the moon performance-wise and end up putting out buggy messes.  I'll make a prediction now:  no matter how good the design is, the story about this game in the first six months after its release is going to be about game-breaking hardware requirements and bugs.

Quite likely, in a good and a bad way. Good in the sense that it might replace Crysis as the benchmarking game du jour, something to drive up hardware sales and push some boundaries. PC gaming has been held back quite a bit by the need to be console portable, especially the last years with the current consoles being ever more apart from a modern high end gaming setup. Even the "next gen" ones are graphic performance wise only on a rough level with Radeon 7750. Meh!

In the "its buggy" sense (which I think was what you meant) I agree. Besides the "OMG Ponzi scheme" hysterics here (yes, I said it) there was another argument insofar that they are promising quite a lot, biting of more than they can chew. That's a fear that holds some merit (imho), so them pushing it out in an unoptimized, buggy state wouldn't surprise me.

[Miasma]

A primary, legitimate concern is that cryengine hasn't been used on a game like this before.  The human model planetside, space station, hangar stuff will work fine because cryengine has a lot of solid work there but trying to use it as a basis for space ships is unproven territory.  The 3D models look great sitting on a hangar floor but actually flying is a whole other ball game.  If the game winds up being a completely broken piece of garbage I'm predicting it will mostly be because they weren't able to make cryengine work with very fast moving large objects in zero gravity.  My concern is that it will wind up being like the Hero engine in that it needs so much work and customization that it turns out to be an unworkable nightmare.

Quite frankly we will probably know within three months if the game will sink or swim.  The initial release of the dog fighting module next month will be really rough and if they can't fix it with patches the game is doomed.

That said they had no choice but to use someone else's graphic 3D engine.  Even with thirty million they wouldn't be able to code their own.

[Samprimary]

Honestly we can all be eye to eye on this more than we think. We like to discuss games, not the neverending telethon before the game

We have people on the lines right now to accept your call, it's your funding that make this possible, all this great programming, carl, just tell us about what these kickstarter funds make possible. Tell us what new backers get if they pledge at or above this level.

[Samwise]

That said they had no choice but to use someone else's graphic 3D engine.  Even with thirty million they wouldn't be able to code their own.

 

[Ironwood]

I have one you can have for five million.

It's in the Garage.

[rk47]

Good in the sense that it might replace Crysis as the benchmarking game du jour, something to drive up hardware sales and push some boundaries. PC gaming has been held back quite a bit by the need to be console portable,

PC gaming isn't held back by the graphics.
PC gaming is held back by the XBOX/PS controller.

[Quinton]

That said they had no choice but to use someone else's graphic 3D engine.  Even with thirty million they wouldn't be able to code their own.

As much as their endless fundraising by selling crazy-expensive exclusive ships strikes me as laughable, there are plenty of serious AAA projects with far larger budgets than 30M which did not build the 3d engine from scratch.  Building a modern, all-bells-and-whistles 3d engine from scratch, integrating a content pipeline, ensuring it works with a wide range of video cards, and performance tuning it is a huge undertaking.  Licensing an engine seems like a completely reasonable choice.  It does, as others point out, run the risk of the engine not being a good fit for what you're trying to build.

[Margalis]

I'd be surprised if CryEngine is an issue. If they aren't competent enough to make CryEngine work for this kind of game then they aren't competent enough to make their own engine either.

Space games are pretty easy to make in some ways - you have almost no level design, pathing is simple, rigid objects are easy to animate. If anything it seems like the problem they are having with CryEngine right now is "hey, CryEngine is pretty good at making FPS games, so why don't we put some FPS gameplay into this?"

I suspect the big problem is going to be that it just doesn't come together as a game. Right now they have all these fertile imaginations about how are all these different kinds of ships are going to work together to form an interesting and coherent space game, but there's a good chance that when they put everything together it's going to turn out that half the ships are useless or don't make sense in the context of the gameplay that emerges.

Rather than coming up with the basics of a game and saying "ground installations are very important in our implementation, so maybe we should make a couple of cool bombing ships that specialize in taking out structures" it's more "guys I had a cool idea for a bomber that I drew on a napkin, let's model it and imagine the fun you'll have bombing stuff!"

Creating a series of vertical slices and smushing them together is not a good way to make a game, nor is creating a bunch of ships before you have any real way to use them in-engine.

[Quinton]

I'd be surprised if CryEngine is an issue. If they aren't competent enough to make CryEngine work for this kind of game then they aren't competent enough to make their own engine either.

Space games are pretty easy to make in some ways - you have almost no level design, pathing is simple, rigid objects are easy to animate. If anything it seems like the problem they are having with CryEngine right now is "hey, CryEngine is pretty good at making FPS games, so why don't we put some FPS gameplay into this?"

That's a good point -- if your space game depends on some kind of huge static level design you're doing it wrong and if not it's not exactly rocket science to do some basic identification of local/relevant objects (in, as you point out, a generally very sparse "world") and letting the engine handle those -- after all a modern engine is going to have reasonable LOD and asset cache management.

[Mrbloodworth]

IIRCC, they never said "we can't make one" they said "Middle ware engines are able to do what we need now". The GDC game play was all in engine. So, it looks like they already have flight/space in and working way back then. I'm sure it needs refinement, he said as much.

It also does not hurt that for the most part, they have made this game before, with Freelancer. Microsoft cut it down though.

Rather than coming up with the basics of a game and saying "ground installations are very important in our implementation, so maybe we should make a couple of cool bombing ships that specialize in taking out structures" it's more "guys I had a cool idea for a bomber that I drew on a napkin, let's model it and imagine the fun you'll have bombing stuff!"

Bombers are for Capital ships. For the most part, they are following the Wing commander/freelancer system as far as ships go and what eats what, from what I can see.

[Azazel]

[UnSub]

I don't think this is a scam. I've got personal concerns with crowdfunding that pushes the financial risk off the people who will financially benefit from a successful outcome and onto its customer base. You need some crowdsourced money to finish a game you've been working on, ala FTL? Okay, that's good. You want to fund an entire AAA title using a system that actively rewards promoting the biggest scope and largest number of features you can possibly fit in a game by getting players to pay for pre-pre-pre-alpha? No, that rubs me the wrong way.

Plus SC has promised a lot of features. Players are rightly going to be pissed that they payed for a $150 space ship only to find that the feature set they thought they were getting are aren't there when SC launches. And we haven't even got into gameplay balancing or in-game economy or a host of other features that trip up every title.

Even assuming that it's cheaper for an indie to develop a title than a publisher-backed studio (due to less oversight, less management) SC is going to need to keep earning money from pre-selling things if they are going to develop based on their scope. They have 12 months of development left, including the launch. They are spending the money they are getting. So more things are going to be put up for sale all the way up to launch (unless SC starts taking on investors / publishers, which will just send SC backers INSANE).

But regardless, I'm watching SC as an example of a title that promises players the BEST GAME EVAH based on the reputation of a developer who hasn't done anything in games for a while now. F13 mocks MMOs for promising the Moon and eventually - late and over-budget - delivering a slightly damaged paper mache sphere. SC is on target to fall into the same traps.

[satael]

I haven't been following this too closely so I'm a bit unsure as to where all the ships you pay for now are used since it's not a MMO and contains a single player game (and privately hosted servers if I understood correctly)?

[calapine]

I haven't been following this too closely so I'm a bit unsure as to where all the ships you pay for now are used since it's not a MMO and contains a single player game (and privately hosted servers if I understood correctly)?

In Star Citizen itself - which is also refered to the as PU (persistent universe). I don't think they will be available in the single player part (dubbed Squadron 42) as that's a linear story based Wing Commander game.

[Malakili]

I haven't been following this too closely so I'm a bit unsure as to where all the ships you pay for now are used since it's not a MMO and contains a single player game (and privately hosted servers if I understood correctly)?

I've heard some stuff about privately hosted servers too, but can't find information about it.  Can anyone elaborate?

[Mrbloodworth]

I haven't been following this too closely so I'm a bit unsure as to where all the ships you pay for now are used since it's not a MMO and contains a single player game (and privately hosted servers if I understood correctly)?

I've heard some stuff about privately hosted servers too, but can't find information about it.  Can anyone elaborate?

You can host your own PU, and mod it. That's the only info at the moment.

[Margalis]

I don't think this is a scam.

I think few KS projects are outright scams but the majority of them play fast and loose with the money they get or are for projects they know in their heart-of-hearts are vastly over-promising. When you get a bunch of free money free from obligations at no point do you really have to consider things like budget, scope and schedule.

[Speedy Cerviche]

That said they had no choice but to use someone else's graphic 3D engine.  Even with thirty million they wouldn't be able to code their own.

These guys raised like $60,000 and are doing their own 3d graphic engine...

http://melee.org/

[jakonovski]

It's funny though, moral hazard is considered a big thing in economics and whatnot, yet in Kickstarter nobody even considers it. I guess it takes a Curt Schilling to pull the wool.

[Samwise]

I made a 3D game engine for an undergrad CS class.  It wasn't a whizbang Crytek engine or anything, but the basics are not hard any more.

Guns of Icarus got $35k from their kickstarter, and they have a working (and very fun) game with ships you can walk around on AND pilot with other people.  (They used Unity.)

[Merusk]

That said they had no choice but to use someone else's graphic 3D engine.  Even with thirty million they wouldn't be able to code their own.

These guys raised like $60,000 and are doing their own 3d graphic engine...

http://melee.org/

Hey now, nobody claimed it was 100% for financial reasons.  All those salesmen and artists cost a lot and no matter how much you pay them to code they can't do it.

[Stormwaltz]

That said they had no choice but to use someone else's graphic 3D engine.  Even with thirty million they wouldn't be able to code their own.

IIRCC, they never said "we can't make one" they said "Middle ware engines are able to do what we need now". The GDC game play was all in engine.

Off the top of my head I can name three full-fledged MMORPGs that shipped with budgets under $20 million, including original and proprietary client engines and server architectures. EDIT: And all "first-time projects" from their teams as well.

None of them are anywhere close to SC's level of graphical whizbangery, of course.

[Quinton]

Yeah, my assumption on "you'd almost certainly license an engine" was based on the assumption that they want super-fancy, shiny, pull-out-all-the-stops, latest-generation rendering features, and sufficient performance that you can get a reasonable number of those super-shiny ships on screen at once.  Certainly you could build a much simpler engine from scratch for less.

[Venkman]

Good in the sense that it might replace Crysis as the benchmarking game du jour, something to drive up hardware sales and push some boundaries. PC gaming has been held back quite a bit by the need to be console portable,

PC gaming isn't held back by the graphics.
PC gaming is held back by the XBOX/PS controller.

PC gaming is held back by business interests. A lot easier to have centralized QC when you control the hardware and the storefront. And when I say "easier", I mean "justify from a business POV". Because without benefitting from licensing residuals and software rev share, it's nigh impossible to get resources to centralize QC from management. Why should they care when it doesn't hit the top nor bottom line? This is whitespace from which ID and Blizzard and then Steam all entered... and then passed through on their way to consoles.

Apple did not develop from scratch a hardware / app store model the rest of the mobile industry picked up.

[Lucas]

Tomorrow there will be a 4-hour livestream for the last day of the LTI ships sale, starting at 11am PST (8pm CET). Beside the final sale of previously unreleased ships, apparently we're gonna get a first look at new stuff they've been working on. Fingers crossed for a preview of the dogfighting module in action (and who knows, maybe a tentative release date for it.).

The channel to watch should be this (the same they use for Wingman's Hangar):

http://it.twitch.tv/roberts_space_ind_ch_1

(maybe they'll add more, plus a Youtube video channel).

[calapine]

I think that sums up the debate: 




I made that myself! With the Photoshop Paint.Net!

[Paelos]

Needs to be surrounded by $$$$$$$$ on the borders.

[Stormwaltz]

I was ahead of the curve, and also retro.

Which makes me a double-hipster! 

[lamaros]

Hipsters are ahead of any curves?

[Mrbloodworth]

Hes a profit!

[schild]

:(

[Rasix]

Man, I hope that was intentional.

[Samprimary]

PLEASE NOTE: due to FinCEN regulations, packages over $1,000 can not be gifted. This includes the Idris-P corvette, so do not purchase one expecting to transfer it to another account! You can check for Idris availability here.

4 rounds of 100 Idris ships being sold across the day. Watch me be completely unsurprised when they sell out.

[Lucas]

Nice, I like the look of the "Merchantman" ship of the Banu race:


Hopefully, with the transition to the in-engine version, the ship will keep its coolness :)
----

Regarding how the official persistent universe server of SC will function, in case you missed it, here's a portion of a post quoted by Bloodworth about a year ago. I think it might be of some interest in order to post some more gameplay-oriented stuff. Mind you, it's not for the tl;dr crew :P. Of course things may have changed during the last twelve months, although I think the basic approach will stay the same.

But me being me, I wanted to combine things I like about the promise of a MMO, but avoid the aspects that I’m not so keen on like splintered player groups, griefing and grinding. I also was really impressed with how Demon’s Souls merged the single player experience with the multiplayer side.

 

All of this helped form my thinking on how Star Citizen is going to balance the difficult balancing act between multiplayer and single player.

All multiplayer games – whether they are a persistent world massively multiplayer game (MMO) like World of Warcraft or just an online multiplayer game like Battlefield 3 – have a limit to the number of players that can be active in anyone area or level. This number is usually inversely proportional to the amount of data that needs to go between the client and the server. For a game with complex physics and a fully destructible terrain, like Battlefield 3 the number of players that can active in an instance is less than a game with less real time fidelity like WoW, or Eve on Line.  But in all cases there are always more players than any one server instance can handle. For a persistent multiplayer world like WoW the solution is to split up the player base into more manageable groups called “shards”, which are a permanent instance of the universe that look after a certain amount of players.

One thing I don’t like about most MMO structures is the fragmentation of the player base between these “shards”. If you had joined much later than a friend of yours, there may not be room on his world instance anymore and you have to join another parallel one and so cannot play together. This is one of the nice things about the Eve Online design – everyone plays in the same universe.

In Star Citizen there is going to be one persistent universe server that everyone exists on. So you will never be separated from your friends, and if you want you’ll be able to join up and adventure together, you can.  Due to the fidelity of the dogfighting and physics simulation we can’t however handle thousands of players in the same area of space. Even if you had enough internet bandwidth to handle the data going back and forth and a super computer for the server there’s no PC, even with quad SLI that could render that many spaceships with Star Citizen’s fidelity.

So the “magic” of Star Citizen’s multiplayer design is how we combine a persistent universe with a more traditional (and easier to implement) temporary multiplayer “battle” instance.

The way it works is that the persistent universe server, which we’re calling the Galaxy Server, keeps track of all players’ assets, group relationships and locations inside the Star Citizen universe. As the Galaxy server isn’t handling any realtime action it can handle our complete player base, which right now would be about 45,000 players, but is designed to be able to scale to millions if need be. The other key thing the Galaxy Server does is dynamically place players based on their location, skill level, alignment and player versus player (PvP) preference into battle instances. Think of a “battle” instance like a Battlefield 3 multiplayer session or a World of Tanks Battle with the key difference that the selection of players is done transparently and is “in fiction”.

An illustration of how this would work is like this –

I start out planet side on New Pittsburg. I decide to buy a few tonnes of steel to fly to the shipyards of Terra. I’m currently in the hands of the galaxy server that communicates with my client and handles my purchases and interactions on the planet as these are not real time in the manner that the space action is. We render these in the manner of Freelancer, as detailed 3D environments where we see a third person view of our character in a location and we can click on Non Player Characters (NPCs) or terminals to buy / sell, upgrade your ship, get gossip, hear about a mission and so on.  You’ll also be able to interact with other players via a chat interface. We haven’t fully worked out the player avatar handling planet side but the bar or private clubs will be where you can meet / chat to other players. Besides populating the bar with NPCs, the game will also populate the bar with other players. If there are more players planet side than there are slots of avatars in the bar the ones visible to you will be based off your friends list and then it will be based on relevance to you – a player looking for a wingman, one from a similar group, or maybe someone that you’ve been given a mission to find or hunt down. You will also be able to see the full list of players in the room if there are more players than there are slots. Default would be a drop down list for this, but as I hate anything that breaks the immersion, we’ll probably come up with a better in fiction way of seeing the list of players – maybe you tell the bartender who you’re looking for, maybe you can look at the door list for the bar.

Having bought my cargo I launch into space. If there are players already in orbit there will be an orbit instance already created. If it’s not full then I will be placed into that. If it is full then a new one will be dynamically created. All orbit (and battle) instances reserve slots for friends and persons of interest (POI), which can be NPCs or other players, so if you’ve launched and there are multiple orbit instances and you have friends already in orbit you should be placed into that instance. This is also the dynamic that will be applied if you want to follow another player – you can “tag” them as a POI and then the game will do its best to place you in the same instance as your POI. For instance if you tagged someone planet side and they launch your PDA with its future version of Siri will notify you that your POI is leaving, giving you a window to launch into space too.

Once in orbit I can pull up my Navigation computer and set a course for my destination. If its several systems away like Terra, the nav computer will chart a course through the relevant jump points. You will be able to adjust this like on Google maps, so if you click a different jump point on the system map it will then re-route you on the shortest path to your destination with that jump point as the first “jump”

Once I’ve plotted my nav course I would then engage auto-pilot and head towards my first “way” point on the path to my destination (a jump point, an interim space feature, like an asteroid belt and so on). At this point I’ve been handed back to the Galaxy Server, which is determining whether I will encounter a hostile, someone that has tagged me as a POI, or a predetermined encounter on the way, or if I’m going to run across ongoing battle instance that is relevant to me (some members of the instance are aligned against or with me). These encounters could be with an NPC or a live player(s) and are sorted on skill level and also – which is important to all of you that like a more single player experience and don’t want to deal with griefers – based on your player versus player (PvP) preference. So if you’ve set your game settings to be low PvP and you’re in a relatively safe area, you’ll likely have an NPC (PvE) encounter as opposed to a PvP one. Of course your ranking and any reputation you earn won’t be the same with a PvE encounter versus a PvP. My hope for this dynamic is that it will allow people to first play Star Citizen in a safer more single player open world style, but as they grow in confidence and want to test their mettle against other real players they can take the training wheels off and get into battles with real players. There will also be areas of the universe that no matter what your PvP setting is, will be PvP. These will be systems that are on the fringes of the policed galaxy and will be notorious for pirate and other illegal activity. They will also be the most lucrative areas – if you can survive.

Now if you’re flying with your friends, who you can link to via the game POI “tagging” system, they will be with you when you’re pulled into a battle instance, whether it is against NPCs, real players or a combination of both.

Once the Galaxy Server has determined that you will have an encounter based on the above criteria it either dynamically creates a battle instance, or puts you in one if one already exists at the encounter point, and that instance has room for new players.  To exit this instance you either have to resolve the hostilities by defeating who’s targeting you, negotiating an exit or just outrunning them. Once in an instance you can put out a distress call to your friends. There are two ways people on your friends list (or squadron as we’re going to call it) can help. We save slots in all instances for friends to warp in to fight. To do this they need to be in the same system. If they are they can autopilot in to your rescue and will be dropped into the instance. If they’re not in the star system, if they can get to your system before the battle is over then they can join (but will only be able to join once they’ve reached your system). The second way for your friends to help out is by “dropping in” on your ship. This only works if it’s a multi person vehicle like the RSI Constellation. In this scenario they don’t need to be in your system, they just will drop in inside your ship and will be able to move around in first person, climbing into a turret to man it, or jumping in you P52 to fly it in combat while you fly the main ship (or they could fly your main ship and you pilot the fighter)

Once the hostilities or the event (sometimes you could be pulled into an instance because you came across a derelict ship or space station and we want to give the player a chance to explore) that triggered the drop out of auto-pilot has been resolved, you can hit auto-pilot again, get handed back to the Galaxy Server and go about your way on the nav course you’ve plotted.

You will always drop out at jump points and planets, where you will need to either make a jump to another system or land.

This process is continued until you reach your final destination, which in my example would be Terra, where I would use my comm system to negotiate a landing slot, which would take me down to the planet’s surface via an in-engine cinematic. Once planet side I’ll be able to sell my cargo, replenish my supplies and look for new opportunities via the third person planet side interface.

The advantage of this system is that is allows you to tailor your experience towards your preference – solo, co-op or full PvP. It also doesn’t partition you into different, parallel versions of the Star Citizen universe as everyone is kept on the persistent server. Because our battle or orbit / space instances are temporary, you’re never stuck

with one group over the long term and due to our heavy emphasis on friends and co-op, there will always be room for your friends to join you on your adventure; whether it’s against other players or NPCs.

The same instance system underpins the single player Squadron 42. If you’re playing off-line, your computer will be acting as the server and client, there will be no opportunities for friends to join and everyone will be an NPC. But if you play Squadron 42 through the Galaxy Server, even though your missions and space areas are pre-determined (you don’t get to pick where in the galaxy you are flying if you’re in the military) we will allow your friends to drop in / drop out to take over NPC wingmen and if you want extra skill ranking you can allow other players to drop in and take over enemy ace characters. This system is pretty similar to the Demon’s Souls setup where people could drop in as a Blue Phantom to help you kill a boss monster or fight off another invading player, or you could drop in as a Black Phantom to someone else’s world and try and kill them for XP and other gamerewards.

The key to all this is to allow player choice – you want to play alone you can, want your friends to join you in co-op we allow that and if you want to be challenged by other real players you can do that. The special part is that it can all happen in the same holistic universe.

I hope this helps in terms of understanding how we’re balancing the aspects of multiplayer as well as making the game fun.

Original post quoted by Bloodworth:
http://forums.f13.net/index.php?topic=22591.msg1132341#msg1132341

[Mrbloodworth]

Man, I hope that was intentional.