Roberts Space Industries - Star Citizen

[Topperfalkon]

well first of all star citizens hype is managed by the baby universe releases with current version at 2.5 in ptu test phase soon to go live, it gives you an idea of what game would look and play like. so hype meters remain in sync with the progressive releases of the game.

I’m sorry, but that’s blatantly false. CIG/RSI have been pushing the hypewagon since day 1 (or at least day 1 after the KS), as it’s their revenue stream. The problem they face is that they have to deliver on everything before they run out of money or they’ll face an almighty backlash. If they’re not careful they’ll end up like NMS, but unlike NMS, it’ll be entirely their own fault.

planetary atmospheric clouds etc which they have done and shown though for planetary clouds they are hiding it for the event.

Iirc, the Inovae engine has volumetric clouds, but it’s not performing well enough to make it into the game yet, but I’m pretty sure I’ve seen screenshots from @Hutchings using it

flying on an empty planet gets old quick no matter how seamless it is. You need a filler gameplay to make it interesting.

Sure, the Battlescape planets are already not empty, and there’ll be more gameplay around them. SC suffers the opposite problem though, imo, in that they’re trying to cater for too many types of gameplay, and that they’ll struggle to make any particular one of those enjoyable

Because keep in mind, if disasters hidden in time dont popup as we progress, then star citizen by then would be way ahead in its quest for dream space game.

SC’s biggest barrier to it’s completion is RSI’s project management.

also we have yet to see how seamless star citizen would be with a single solar system and especially planets with large cities, based on discussion on some reddit, I am totally talking about community rumours, that in star citizen the planets with large cities would be restricted vs planets with some stations or landing zones. so they have yet to crack the tech of rendering large detailed cities on planets. lets see how it goes.

It might help if their planets weren’t laughably small. It would be interesting to finally see a game with an actual large city in it though

you can redesign an existing code into something that you want so cry engine code didnt put a hard limit that you can only create an fps game, especially when you have the people who created cry engine in the first place, core engine programmers.

Sure, but if they’d picked a more suitable engine from the start, or better yet, designed their own, they wouldn’t have had to hire half the developers responsible for building the engine to hack it to support types of games it was never intended for. CryEngine was an FPS engine designed around primarily FPS games operating on a relatively large map for a ground-based FPS, but a tiny map compared to a solar system or galaxy. It took them the best part of a year or more to fix the engine when they got to the point where engine limitations were stalling further development, and it was expensive.

There are other distractions, of course, which include the media spotlight. People can get seduced by that stuff pretty easily, and just saying “Yeah, we’re looking into that” can be enough to fan the flames of the press and bring ever more adulation.

And the shit really hits the fan when your executive producer is from the Molyneux school of media relations.

Lets just be clear here, everything in that wikia is based on loose estimates given by the devs prior to some significant delays. Having been delayed, they are no longer facts but are simply a record of intent. Dev scope can and will change so it’s fallacious to be so single minded about these estimates.

I struggle to see how CIG/RSI failed to predict the delays they were going to encounter. It smells more of bad planning to me, or at the very least extremely poor/optimistic estimation.

[INovaeHutchings]

Iirc, the Inovae engine has volumetric clouds, but it’s not performing well enough to make it into the game yet, but I’m pretty sure I’ve seen screenshots from @Hutchings using it

A 2Dimensional Clould layer does exist in engine atm, as viewed in some of the screenshots I’ve posted in the past, however we do not have a volumetric solution implemented at this time. The 2D Cloud layer is quite limited / simplistic as well, and only looks good from high orbit or some places on the surface.

[Topperfalkon]

I stand corrected, thank you

[innociv]

The 2D clouds do look really bad from far away at sharp angles, tbh. The worst part about any screenshot I’ve seen, generally.
Are they even parallax or anything?

The volumetric will be done in the final render pass on screen space, right? I’m looking forward to those, especially if they’re on gas giants as well. (Plus, good dramatic marketing image there)
Looking forward to seeing volumetric rings dust as well, that’s the other low point of the graphics.

Everything else is pretty awesome though.

[INovaeKeith]

For the record clouds were a stretch goal we didn’t hit. They’ll only get added later on if the game does well enough at retail.

[JonnyRedHed]

And also if the game funding picks up enough during the next year or two leading up to launch? I’m doing my bit to promote the game and bring in new pledges.

[INovaeKeith]

Yeah if we raised somewhere around $200k then we could hire an extra programmer. We know, we greatly appreciate it! =)

Hopefully once we release Alpha and there’s a lot more game to show off we’ll see a big uptick in pledging.

[innociv]

I thought the combination of the two funding sources put it there? Oh well, I guess.

Personally, I think it’s an odd way to look at it, as it’s a system that goes “on top” of other things and shouldn’t really interfere with other systems like adding some others can. In short, it’s not the riskiest thing to add. While the look of it would be a lot more impressive to get actual sales instead of “funding”. A big ship partially exposing itself out of some clouds if something very marketable, eh?

This is a bad example, but a lot of Kickstarter games fail post-launch in getting more sales and being successful past their initial funding because they focused on ticking the check boxes on what they said they’d do on their campaign page, but didn’t put more attention in making a game actually good, and fun, that’d sell once it’s gone “gold”.

I’m not really saying you guys are doing that, you’re doing a lot of things right, and I know that really good volumetric clouds isn’t something you do in a week, but I still feel I need to leave that bit of thought. Especially since the marketing has been the lowest point here, and I really think you need more of a thought in what you do in how marketable it’ll be for greater success.

[INovaeKeith]

I think you severely underestimate the amount of effort it will take to build convincing fully volumetric clouds on a planetary scale. While I agree with you that the result would be spectacular and would provide for some excellent marketing material we have extremely limited resources and still a lot of game left to build. $330k is not much money for a team of 5 people, we’re barely scraping by and 2 of us (myself included) aren’t even getting paid.

We’ve put a lot of thought into it - you have no idea how bad we want to do it (along with vegetation) - we just don’t have the time/money right now. In a perfect world we’d have 15-20 people working on this game but that would cost millions of dollars.

[Zen]

I agree. Besides, best not to get caught up in over promising and under delivering which is a big problem for many developers, especially SC. And I feel bad you and the other DEV are volunteering your time - actually I feel a bit guilty. But take heart, I know once Battlescape goes Alpha it will be a different story! And if it isn’t, come to Japan and we’ll set you up for a week and show you the town. Maybe catch a few Maiko.

[INovaeHutchings]

Japan you say? I should be over there next spring with members of my local Judo club. It would be rather excellent to meet, schedule and location permitting of course!

[innociv]

Na. I don’t underestimate it at all. I said myself I know it’s not something you do in a week. I’m just saying it’s easier than, well easier than a lot of the things RSI tries to fangle into their engine that it wasn’t made for. Like easier than adding FPS people on a multicrewed ship smoothly moving around in foot inside while the ship is swooshing through space, you know?

And I’m saying it’s something you might find worth it.
Basically, a lot of the problem with Infinity is that people look at it, and they’re bored.
Yes it’s beautiful. But it’s serene.
Like look at your 8chan ad campaign. Most ads were boring. Some were so super dark you could hardly see anything. The concept art one with explosions got as much revenue as like 3 of the boring ads combined.
Volumetric clouds, especially with some fluid dynamics, could more for an absurdly more dramatic look if executed correctly. That means money in return.

You can spend $25,000 on marketing that maybe gives you 100k to look at the game at most, who might not be convinced to be consumers.
$25k of programmer time though to make something like that might get 100k impressions just from how awesome it is, on top of it being something added to the game instead of “wasted” on advertisements. It’s something that can be spread more virally, as well as making other advertisements more effective, etc.
But yes, obviously it’s hard to bring a programmer on for 2-4 months. They want more work than that.

If you can’t, you can’t. I won’t be upset since it’s not like it was promised. I’m just saying that it’s the thing that’s really lacking graphically.
I’m saying this not to bitch, but I want to see Infinity get more of the success it deserves. But to be honest only because I love you guys, you guys often seem to naively make decisions that prevent success. Hey, maybe I’m wrong, and I’m still happy with what you’re doing, but that’s how it appears sometimes.
You have beautiful terrain, ships, effects, realistic HDR, lighting, that ralleigh scattering type stuff. Then you have those flat, ugly clouds. Which is especially sad since volumetric clouds can add to gameplay and not just looks, too.

I’m just saying I hope it somehow makes it by late beta, before release.

You could use Indie Fund if it’s so much a matter of needing the funding, but, ya’ know, then you have to pay it back 2x. But I don’t think you have to do that.

[Zen]

Well, if you are in Kyoto, let me know.

[ThornEel]

I think you severely underestimate the amount of effort it will take to build convincing fully volumetric clouds on a planetary scale.

For those of us who never really thought about it, how much/little would it save, compared to volumetric clouds, if you tried to make simpler “good enough” sprite-based clouds?

[satanfreze]

Well, they planned everything through. Just imagine you get better improved clouds with parallax effect (maybe even done with some kind of cheap particle system). It still requires time to do. Just imagine three weeks for a single guy to do. This would mean: there is a chance that the release of the game gets two or more weeks postponed!

I agree, realistic clouds and vegetation would add so much more but people need to eat and need to pay rent and bills. It’s a hard business, not just some fun thing to do. Spread the word and tell others to give some money. Maybe we get those nice clouds in return.

[INovaeKeith]

You could use Indie Fund if it’s so much a matter of needing the funding, but, ya’ know, then you have to pay it back 2x. But I don’t think you have to do that.

We’re building our own funding platform so that we only have to pay the payment processing fees. Its debut will be the pledge upgrade system.

For those of us who never really thought about it, how much/little would it save, compared to volumetric clouds, if you tried to make simpler “good enough” sprite-based clouds?

That really depends on how bad we want to make the sprites look. It would still be a non-trivial effort. Everything with planets is harder - we constantly run into unexpected problems. That’s why it took us so long to launch our KS in the first place.

[innociv]

Isn’t Infinity using a deferred renderer? Or is it forward+ or hybrid deferred?

For deferred, generally you have an extra pass for each level of transparency so you greatly degrade transparency for each level of overlapping there of the transparent/translucent sprites. That’s lots of extra render calls, plus the CPU work of figuring out the optimal number of render layers (how many overlaps, in order)

You can do a lot more if you have a mesh volume with a volumetric fog like shader inside it (repetitive but I don’t know how else to say it) for the clouds. Sort of like this: http://www.3dartistonline.com/news/2014/01/how-to-create-clouds-and-candy-floss-in-modo/
And plus the added benefit of being able to manipulate those volumes sanely instead of trying to simulate particles, being able to generate them with 3d noise, etc.

So… I think it’s kind of pointless to even think about 2d sprites.
Not only does it look worse, and unconvincing (especially as you circle around them), but it’s harder to blend the look from space to have them transition cleanly from space to up close.
With a mesh, well it’s a similar concept to the generation of terrain, just in 3d with a bottom to them. You can have LOD steps on them just the same, and translate the 3d noise to 2d noise when you’re just looking down on them from space.

Not to mention that modern rendering pipelines tend to make volumetrics cheap, no? Deferred fog is super cheap and a fog volume is not much more of a heavy calculation, is it? Well, then you add lighting and shadows, but still.

[selbie]

Stream is up for the Gamescom event. The main presentation is on Friday.
https://robertsspaceindustries.com/gamescom

[Red_Syns]

Japan, huh? @Hutchings, any idea when you’d be there exactly? I’m headed over at the start of April for 3 years, I’ll be stationed in Yokosuka.

[kamui]

Does anybody know if there’s some big news coming ?

Fourth Annual Gamescom Presentation starts in 6 hours…