We chat to lead designer Paul Boyer as Stardock's space 4X behemoth prepares to ease out of Early Access

Galactic Civilizations 3 launched on Steam Early Access last August, and Stardock has been steadily adding fleshy content to its space strategy bones ever since. As the studio prepares for the game’s grand voyage out of Beta and into full release, I tracked down lead designer Paul Boyer for a chat about the game’s development, and its short and long-term future.

GameWatcher: So you’ve released the Beta 6 update of GalCiv 3 recently, and the game’s moving out of Early Access next week. Do you feel happy with the state of the game right now, does it feel like it’s come together?

Paul Boyer: Oh yeah, there’s always little things that we’re trying to track down and find – there’s a very odd bug with certain video cards that’s haunting us right now that makes planets explode. So there’s always adventures up to the last minute, but we’re confident the game is fun, nice and addictive, and we just need to make sure that nothing goes wrong, and then of course once we’re out we start obsessing over how to make it better.

GameWatcher: One of the things the last update was intended to fix was memory usage slowing things down – is the game running at a decent pace now?

Paul Boyer: Vastly. By orders of magnitude. At one point in Beta 5 there were games taking 14GB or something like that. The problem that we had is that the matrix of players, and the possible maps sizes, you just have to track so much data. It’s not objects on the map, or graphics or anything, just the sheer amount of data. So we’ve had to find lots of clever ways to simplify that, it’s been a case of it going up and down, up and down. Right now I think we’ve got it to the point where it’s going to stay at a reasonable level. Although I’m sure by the time we get to the final expansion we’ll have found a hundred other ways to reduce memory usage. Of course we’ll have added a bunch of other stuff too, so it never really ends.

GameWatcher: How does the engine handle those massive new map sizes GalCiv 3 introduces?

Paul Boyer: Well, you know, it’ll always take longer in those big, 1500 star system games. Not really so much if you’ve got a couple of players, but if you’re playing a ten-person game and they’ve all got ten planets, they’re all fighting their own wars, that’s obviously going to build up and build up. The AI actually does its stuff very quickly. We’ve found a couple of interesting things that we’re probably not going to rush in for release, but there are a couple of other little optimisations we might add in further down the line. Right now we’re happy with how it works, it’s amazingly fast compared to other games in this genre, especially considering the scale. We build and build, and then we optimise.

I think if you spend all your time optimising every little thing, you’ll never be able to get things done and I think that’s just part of development – you make something work, and then you optimise it. We do put in a warning for the larger maps if players only have four gigs or so of memory, that’s not really for the gameplay it’s for performance. You’re going to eat up a lot of memory on lower-end PCs, laptops and such.

GameWatcher: AI’s always something of a sticking point for 4X fans. Do you feel comfortable with the level it’s at right now?

Paul Boyer: Yeah, Brad Wardell (CEO of Stardock) came in to help us out, and he really beefed it up a lot. It’s a lot smarter now. We were always going, “oh, let’s have it do this, let’s have it do that”, but at some point you just have to go with it, and then we can keep adding to it. The crazy thing with AI is that you make it super smart, and then people think you’re cheating. You have to go in and basically make it a little dumb, make it make random decisions here and there. Because that’s what players do. Nobody makes the right decision every time, whereas an AI will do exactly the right thing every single time if you design it right. We have to go in and mess with that a little. So a lot of the time, the higher difficulty levels are just us going in and removing those blocks.

GameWatcher: Windows 10 is on the way this year, and Direct X 12 is set to be included with it – are there any plans to include DX12 support for GalCiv 3 in the future?

Paul Boyer: Well GalCiv 3, at least current GalCiv 3, won’t be using DirectX 12, obviously. Ashes of the Singularity, our other big game, is DX12, and we will of course migrate over to the new tech at some point. Right now it’s just a case of making sure the game works on a variety of different systems. It’s one of those things when it does make you excited for GalCiv4 or whatever, and it’s very interesting what you can do with it. Ashes makes really heavy use of it, they can do some crazy stuff with the new low-end stuff in DX12. GalCiv 3 we’re going to keep running on current tech, and the beauty of that is that it’ll just keep getting faster over time. I remember people complaining about the turn times in GalCiv 2, and now you can just ding through them no problem.

GameWatcher: Was the plan always to come out of Early Access now? You’ve been pretty vocal about wanting to continue development on the game, so was there a temptation to extend your stay?

Paul Boyer: Oh no, for us Early Access was just another way of doing what we’ve always done, which is; release a game in beta and get feedback from our fans. The main difference is that we got way more feedback in Early Access than we normally would. I think it really paid off. We did also get a lot of ‘casual feedback’, which is less useful – people buying the game and going “this isn’t finished”, and we’d have to say “yeah, we know, it’s not”. But yes, we’ve always done it this way, we just never had Steam to do it in the past. A lot of people, a lot of developers treat Early Access as this big new thing, but for us giving people access to early Betas and looking for feedback is something we’ve always done. But yes, having more people help us out, not so much with bugs but more with gameplay balance and features, that was really helpful. As for using the same model in the future, I can’t speak for the higher-ups at the studio, but I’m really happy we did it.

GameWatcher: I was reading Brad Wardell’s post on Steam about the next six months or so of GalCiv 3’s lifespan, and it sounds like you guys have a pretty concrete schedule in mind for that post-launch content.

Paul Boyer: Things always change, but yes we’ll have version 1.1 right off the bat, which will add in modding support, all the tools you need. I’ll probably start releasing some of the editors I use, so that people can start working on very elaborate mods. Then I think we have some DLC planned, so we’ll keep tricking out some stuff that’s not necessary to the story, but is just fun to have. A couple of substantial features might come in for free, things that we wanted to do but didn’t have time. I don’t know exactly what it will consists of yet, but we have a tradition of making the 1.1 patch a kind of mini expansion. Then we have a full-on expansion scheduled in for next year.

GameWatcher: What mod tools will the game ship with?

Paul Boyer: We’ll start with ships and factions, and then we’ll move on from there depending on how things get picked up. We really want to do a lot. I would like to be able to share whole ship styles, scenarios, things like that. Those are pretty complicated, so we might have to think about releasing them bit by bit, but we’ll have ship editing right away, and I’m pretty sure factions right away, too.

GameWatcher: While we’re on the subject, I wondered what you thought about the recent ‘paid mods’ kerfuffle. Stardock games are often pretty heavily modded, how would you feel about modders charging for their work?

Paul Boyer: You know, I didn’t really get involved, but I will say that I think Skyrim was a poor choice to promote that idea. It really needed to happen with a new game, one which can then grow with that model. But if you choose a game where people have been getting free stuff for years, and then suddenly you start charging, people will get upset regardless of any good arguments you might make in its favour. There were a lot of good arguments on either side, but I’m not surprised that it fizzled out. Actually, around 15 or so years ago, I got connected to Stardock because I did icon packs. That’s how I started here. Stardock paid me a little money to make an icon pack to sell, and I made some royalty money off of it - and man did we get yelled at for that! “Skinning should be free, icons should be free,” but then eventually it worked out. There’s a tonne of free packs still, but some people make a bit of money off it. So I don’t know.

Eventually I think some kind of payment model would work for modders and actually would be great for them, great for the game industry. You could get these people like Derek Paxton, who did the Fall From Heaven mod (for Civilization), he did that for free, but he became a really great designer and eventually got a job. If you can make money from modding, you can spend more time doing it, and then eventually you can start making your own games. I think there’s some strength to it, but I don’t blame people for getting mad about it, especially when it was floated with a game like Skyrim that has such a vibrant community around it already.

GameWatcher: I realise you’re probably pretty fixated on getting the actual game out, but do you have ideas in mind for future expansions?

Paul Boyer: Oh I have a vast list of ideas I want to put in, whether half of those ideas make it into the expansion is the question. We’re hopefully planning on at least three expansions, so I’m sure it will all make its way in there in some form or another. A lot of it ends up going with the theme of the campaign that’s included with it, so the next one is the ‘Crusade’ campaign, where the Terrans get out of Earth and basically wreck the place. So there’s going to be a lot of interesting things involving weapons, artefacts and so on, and I also have an idea for a commander structure. But none of this is certain, I have a lot of ideas that didn’t make it in to the main game, but will maybe work for the expansions.

GameWatcher: Do you find your player-base gets a lot of use out of your story campaigns? I always imagine 4X players head straight for the sandbox mode. How do you hook them in?

Paul Boyer: You know, there’s a lot of people, hardcore fans who care about the lore and enjoy the campaign, but there’s no getting around the fact that sandbox is the bread and butter for a lot of our players. As a matter of fact I’m making sure that the campaign in GalCiv 3 feels very much like the sandbox. Each of the missions starts of with some objectives, some choices, and then it lets you loose. We kind of drop you in mid-game, almost as if you’ve played half the game. Then you have to adapt just like you would in a regular game. That’s what we want, we want to push you down the storyline, but the campaign isn’t meant to turn GalCiv into an RPG. Even the tutorial is very much a sandbox, if you want to turtle you can, if you wanted to rush you can.

---

Many thanks to Paul for chatting to me, especially because he’s presumably going hell for leather putting the final touches on Galactic Civilizations 3 – the game launches out of Early Access on May 14, and we’ll have a full review for you lovely chaps next week.