I need something to keep me caring about this game

Battlescape
41

sorry for the ambiguity, the arena will be “20 kilometers cubed” to quote CR’s exact words.

Haha yes, if we were talking about real scales compared to the size of a solar system, it’s not. And as far as I know, SC will not be using real scaled planets even.

But if we’re talking about the size of an arena in regards to the engine’s capabilities, I am just saying that their modified Cryengine is capable of some pretty incredible things including battles between capital ships. I can definitely say though, the battles will not be equal in size to Eve for example. This is because Eve doesn’t have to compute weapon hits and real-time maneuvers, ect.

42

Well, either that or they simply defined 1 meter to something else than it would be in a CryEngine FPS. Which makes perfect sense: space ships tend to be a fair bit larger to people (at least that’s the case for spaceships containing people) so you don’t need as high precision on locations. If you simply define 1m to use the same number of bits as 10cm would in a FPS, bam!, you’ve got a map-limit 10 times greater.

Having people visible walking around inside other spaceships will probably require jumping through some hoops or just be slightly less smooth than if using an unmodified CryEngine 3. To be fair, “less smooth than CryEngine” can still be pretty smooth.

As for this, though, this is nothing more than an excuse.

PlanetSide does weapon hits (albeit client side) and real-time maneuvers for hundreds of players. I’m talking the original here, which is over 10 years old now. PlanetSide 2 does it, with bullet drop and everything, still client side though. I:B is planning to do it (OK, only for 256 players per server, but then their map/play-volume-per-insanced-area is 4 billion times larger in each direction).

I don’t know why SC doesn’t have an interest in allowing large numbers of players to fight each other at the same time. Maybe it’d be too expensive - server-side bandwidth costs tend to grow exponentially if you don’t limit how much info clients get on each other. Maybe their engine just can’t do it (entirely plausible). Maybe squeezing several hundred players into a box of space only 20km per side just wouldn’t make for a fun game.

By the way, how would battles between capital ships work if the box is only 20km a side? I mean, if you think about it that’s only like 60 ship-lengths for a large sea-going ship, and it’s hardly scifi if everything isn’t bigger and better (with the exceptions of electronic gadgets which should be smaller and better). Not that I really think of it as a battle if there’s only a few dozen people involved, MOBA-genre notwithstanding.

43

It’s not made by the same guys who made the original. Julian is currently working on his own project, a sequel to Chaos. I’m really looking forward to seeing it: http://www.gollopgames.com/

44

Well from a stylistic perspective, I know that CR likes small skill intensive and fast paced battles. From a practical hardware standpoint, nobody’s computer would be able to handle rendering the number of polys in the game assets. A fighter is 300,000 polys and a cap ship like the Bengal Carrier is 18 million. If you have hundreds of ships with this kind of detail on your screen, even with LOD, you will crash your GTX Titans. If I’m not mistaken, we don’t have a solid number on instance sizes yet in SC so it’s hard to say exactly what seems reasonable for the game engine and client/server networking to handle. My guess is anywhere between 30 and a hundred but let me know if you guys know anything different. From the networking side, it seems like the number of things happening and the precision with which they’re happening place a limit on the number of ships in a battle. The weapon strike detection seems like it will be a fidelity equal to an fps, except across the entire ship, a larger entity than a humanoid character. And the entities will be moving at speeds far greater than an fps. Then you need to keep track of all of the characters running around the ships and manning the systems. You have to keep track of whether the laser pistol blast from the enemy boarder, moving through the ship, hits the security officer while the ship itself is also moving, and itself shooting lasers at another ship. It just doesn’t seem far fetched that there is a technological limitation here as opposed to a hidden agenda.

46

[quote=“INovaeKeith, post:45, topic:205”] in fact it looks like his high-poly ships have that many and he’s using them to bake normal maps for low poly ships
[/quote]
Interesting, so for a layperson, what does this mean?

47

Alright Flavien had me delete my post, even though it’s my opinion, just in case I’m wrong. If I’m right what that means for the lay person is not much. It’ll look good either way. Raw polygons look better but it’s unlikely (with rigid bodies) a lay person will really notice as long as they do a good job with their normal maps.

48

Sure, I just don’t literally know what it means to “bake” a normal map vs using raw polygons. I will try to Google it.

It does seem like from a performance perspective that the outcome of large numbers of these ships loaded into the program completely destroy the computer. I’ve seen people put up videos of trying to pull the assets into CryEngine and 18 bengal carries loaded at once completely crashed SLI 670’s. And I’m not even talking about the running game, literally just loading them in the editor.

49

Here you go: polycount

1 Like
50

I wasn’t going to say anything, as I don’t know the technical details involved in using normal maps and the like, but from what I understand “baking” involves building a high or ultra-high polygon model and saving the information about the angles of those polygon faces in a texture map. That texture map is then rendered by your graphics card in such a way that reproduces the high polygon details without having to actually render those polygons.

Edit: Beaten to the punch by WD.

51

u sayed I:BS can run smoothly on previous generation AMD. how do you know than, and where can i find system req. if they are even configured yet

52

It’s not like they’re not testing everything they make. To make a game you have to be able to run it, after all.

53

But then that’s what LOD is. A system to make it so you wont have hundreds of ships with that kind of detail on your screen. In fact I’m pretty sure I heard CR talking about how the reason they’re able to do 18 million poly carriers is because 95% of it will be lower Level Of Detail or not rendered at all. That’s rather heavily paraphrased.

As far as the idea that hit detection for people and parts moving relative to ships, certainly that would be extremely hard to brute force - which is why methods have been invented so you wont have to. The most popular one basically boils down to making the interior of a ship it’s own mini-zone with position measured relative to the ship rather than the space around it, and then just translate hits on the exterior of the ship to a route through the interior (which can be done with basic arithmetic operations).

Oh and again the LOD thing comes in; you wont have to keep track of people running around inside a ship unless you’re close enough to actually see them.

Having said that, while 3-digit-player battles are entirely possible, technologically, they would cost quite a bit of bandwidth, which of course translates to money. That might be why they aren’t doing it. Or maybe CryEngine 3 just can’t handle it, modifications or no modifications. Or maybe trying to squeeze several hundred players, in spaceships, into a 20km a side cube would just get really rather crowded.

54

Correct, and it is still a lot of polys that in other games would be rendered down even further. Are you a Star Citizen pledger? If you don’t believe me, go try to run the hangar module and tell me your fps. R9 290x’s are barely keeping up with screen refresh rates. On one hand it’s not optimized yet, on the other you are literally sitting in a room with nothing moving and only a few ships. There are few or no specular effects happening and no particles.

As far as netcode, anyone inside the ship is rendering everything outside which is why you are able to look out the window and see the battle. It is also rendered in the opposite direction as well so that you can look through their cockpit window and see how many crew they may have before you board.

Chris Roberts specifically said the reason why they won’t have Eve-like battles is because of the hit calculations required in this game vs Eve. Off the top of my head, that is the only thing I know for certain they’ve said. Now I assume they are prioritizing making it server-side to prevent hacking. If it was client side, it would be easier of course. Do you think he is lying to or misleading pledgers with that statement?

55

Yes.

Could explain why and how, but honestly can’t be bothered. You’re free to believe whatever you want.

And no, I haven’t pledged to Star Citizen, nor do I have any intention of doing so. I find their business model distasteful, in fact I’d describe it as pay to win if combined with even a basic level of applied intelligence. Not that that’s likely to happen with the average gamer, admittedly. Oh yeah and there’s the whole thing about using the right engine for the wrong game - I will clarify that Chris Roberts probably didn’t lie about EVE-like battles being impossible in CryEngine 3.

I’d like nothing better than to be wrong. It looks like it could be an awesome game, and if it isn’t pay to win then I’ll be able to spend several hundred dollars on whatever I want before then buying a really nice game at likely a lower cost than it would be if people hadn’t spent hundreds of dollars changing the colour of their spaceship.

56

Chris Roberts hasn’t lied to this community once. The development has been extremely open up until this point. It’s been about as open as Inovae has. There is literally no evidence that his character would take that turn. All of the evidence points to the contrary. I will believe where the evidence leads and it certainly doesn’t lead down the path you’re saying.

Saying SC is pay to win is like saying piracy is theft or that laying face down on the ground is resisting arrest. The definition of pay to win and the mechanics in SC don’t match. In a p2w game, the payers have the ability to obtain items that give a gameplay advantage through money that are otherwise unobtainable if you don’t use cash. (This excludes purely cosmetic items). This is not the case in SC. Everything in the game that you can buy is obtainable through play.

What SC definitely is during the dev cycle, is pay-to-get-things-faster. And at launch it will then turn into pay-to-get-credits-at-a-limit-per-week. Neither are pay to win. CR believes that people that don’t have as much time as a 12 year old on summer vacation should be able to spend a limited amount of money on the game to make up for some of the time they lose relative to them. In addition to making the game more accessible for busy people, this system will continually fund the game which will release new free content every week to everyone in the 'verse whether they pay for credits or have never bought anything but the 30$ pledge package.

The most important part of this business model is that it is able to raise the funds that are required for development ie that it’s effective, and that it’s sustainable. You need not look further than the funding progress for proof of the former. Also with both the matchmaking system, and the rock-paper-scissors style of gameplay, encounters with large pledgers that wouldn’t have happened if they weren’t able to buy their ship would either be minimized, or would result in a fair battle, or even a battle to your advantage. To put it in perspective, I have a 350r which cost me 125 bucks. Some have a Connie which costs 250+. My ship is the fastest weapon platform in the game. Theirs is battle freighter with big guns, big shields, and thick armor. I am looking forward to taking them out by flying circles around them. I don’t even see a problem. If I encounter a hornet pledger though, cheaper than my ship, they will most likely kick my ass regardless of the fact I pledged more than they did. They can keep up with my maneuvers and have far more shields/armor/weapons than I do. But I can at least run. And this is all the worst case scenario. More than likely, the persistent universe would limit a battle far far beyond my means. They just wouldn’t happen often.

Basically the only point at which the pledging makes a negative, possibly unfair, difference to my gameplay experience is when the persistent universe AI matches me up with someone beyond my abilities, and my ship is just generally bad vs their ship and they pledged more money than I did. This type of encounter is rare. It still happens of course because it’s a dangerous universe out there. That’s the way I like it and I think the small numbers of situations like this, are worth getting a game that’s completely revolutionary, changes the way development happens in the industry as a whole, and sticks the middle finger at consoles.

Edit*- As far as CryEngine 3 being a bad choice to base SC off of, With one of the most experienced game developers in the world at the head of this project, do you think he would have made such a terrible blunder on the choice of engine? And then lied to his community about it? I don’t think you have any evidence to support that, especially because this level of fidelity and graphics has never been attempted before. You couldn’t point to other games to compare to SC because there is no other game to compare to SC that makes sense.

Now if Battlescape takes off and Inovae develops an engine which ends up having graphics better than CryEngine, can render damage/flight/in-ship computers/ ect ect better than modified CryEngine, and can have battles of 200 people without lagging and without dropping fps to 5, then I will literally go onto the Star Citizen forum and email Wingman’s Hangar and spam them with media demonstrating this engine until they adopt Inovae’s software for Star Citizen 2. Or if you can point to another engine that has already done what they’re doing as an example… Otherwise, there’s just no evidence for what you’re saying.

57

There seem to be some drastic misconceptions about the game being expressed here… but being the competition, I guess that’s to be expected. (IE - pay to win in this case, which’s been addressed plenty over at RSI, by devs & CR himself, in both print and video form) The game’s current business model is a huge part of why and how it was actually able to take off so well too. I think more developers should pay attention to such.

58

Reading all the words help.

Getting ships/classes/heroes faster = learning them faster = learning to counter them faster = learning to win faster. If combined with a basic level of applied intelligence. Doesn’t help to be a genius if you don’t use it.

As far as cosmetics go, can you explain to me why most modern military organizations use some sort of camo pattern for their combat uniform rather than bright red with a white belt across the chest? Sure, it’s not a big advantage, but it can be a decisive one.

He merely phrased himself in a way that caused Saturday to assume that such battles aren’t possible in any engine. I do believe your question was “lying or misleading”?

59

Reading all the words helps.

He didn’t specify CryEngine 3. He simply said they are not possible with the level of fidelity. And this goes back to your original statement that I was responding to which was your claim that Modified CryEngine is the bottleneck. And my point has always been that it is the client’s hardware and internet connection, not the engine.

They are specifically building their own netcode to handle the game, they talked about it in CR’s dogfighting update last week. The whole dogfighting module is being delayed so that they can get the proper netcode integrated. But even with their own netcode, they can’t have hundreds of people in an instance. They specifically said they are not using the original CryEngine’s netcode. This fact leads me to discount your claim that other engines are capable of handling more players in a battle with the same circumstances as Star Citizen. Cloud Imperium are able to have the backend of any system they want because they’re building it however they want, and throwing a shitload of money at it if they wanted to. In essence, they can have any piece of software in the industry they want, and they don’t have it. This leads me to believe that it doesn’t exist in the industry yet. But hey, I digress again, if Keith/Flavien build what they’ve been dreaming of which includes massive battles and great graphics with fps-equivalent hit detection, ect, then you guys should sell that shizit at a premium and get our funding for Quest for Earth.

That last part is speculative but as I said, I’m totally open to seeing examples of any other game engine that could handle it. You say they exist, I say they don’t. And if someone like Inovae or Sony Online has it, they could make a lot of money selling it to Cloud Imperium. That hasn’t happened and yet we know that CI is all for using other people’s code and have the resources to spend on it.

[quote=“Runiat, post:58, topic:205”]As far as cosmetics go, can you explain to me why most modern military organizations use some sort of camo pattern for their combat uniform rather than bright red with a white belt across the chest? Sure, it’s not a big advantage, but it can be a decisive one.
[/quote]
In theory you are correct, but in practice that almost never happens. If you look at Team Fortress or Dota, the items only get flashier the more popular they are.

60

About what is this topic again? I can’t really follow …

61

[quote=“INovaeFlavien, post:5, topic:205”]
Neither Star Citizen nor Elite offer seamless procedural planets. Neither are massively multiplayer.
…
What we want to do with Battlescape is to offer players massive multiplayer battles, both in space and closer to the planets. All in a seamless solar system.

The keyword here should be massive. It’s not a linear, scenarized game made for a low amount of players. Our aim is to recreate the frantic, massive, chaotic battles you’ve seen in movies such as Star Wars, with up to 256 players per battle, in a semi-persistent game.[/quote]
Apart from 256 vs 32 players, I don’t see any real difference between what you WANT to achieve and what Elite: Dangerous WANTS to achieve. (E: D is not a “linear, scenarized” game.) And in fact Elite: Dangerous will offer an entire galaxy from the start, where-as you will only offer a single solar system.

As a very approximate guess, and assuming Infinity: Battlescape only takes a year from Kickstarter to being released (a big if), then I expect that Elite: Dangerous will get Seamless Planetary Landings around the same time that Infinity: Battlescape is released (give or take 6 months or so).

You really should avoid suggesting other developers & their projects will be massively delayed & even split into a separate project, when that’s exactly what has happened to your own project (Infinity:The Quest For Earth). Infinity is already delayed by over half a decade (I have screenshots saved in January 2007), and the ‘full’ game has been split-off into a separate much-later project which will have to be bought separately, such that we’re only going to get a single solar system initially (Infinity: Battlescape).

You also shouldn’t criticise other developers & spread FUD (Fear Uncertainty Doubt), when you also apparently don’t even read what they say they’re planning. Since the Kickstarter a year ago, Frontier have always said that E: D will add Seamless Planetary Landings as a paid expansion (DLC if you will) a year after the first game is released.

Given they’ve tended to under-promise & over-deliver (so far), I have high hopes that will actually happen (without too much delay). Maybe the planets will not be as pretty as yours (I do love your atmospheric scatting), but I’m sure they’ll be nice enough.

In short: Please refrain from saying bad things about competitor projects, and instead focus on the good things about your own project. Otherwise you risk upsetting people, and not being taken seriously.

2 Likes