At The Gates is a 4X game that learns from RimWorld and Spelunky

Jon Shafer has joined Paradox to found and lead a new team and make a new strategy game, but work continues on At The Gates, the project Shafer has been developing independently for years now. It’s a 4X game that casts off many of the recognisable patterns of the genre. Instead of building an empire, you control a barbarian tribe trying to tear down an empire, or to co-exist alongside that empire. In this case, it’s the Roman empire. To survive, you’ll need to exploit resources, move with the seasons, and build a relationship with the world as well as your neighbours and companions.

At The Gates has been a long time in the making. Kickstarted for just over a $100,000 in 2013, it’s the first project of Shafer’s Conifer Games studio, founded following his work at Firaxis on Civ IV and V, and a stint at Stardock. Shafer believes that strategy design often requires iteration after iteration to find the best version of an idea, or even to find the right idea. With At The Gates this has involved finding the character of the tribes and their members, as well as bringing the map to life as a character in and of itself.

Seasonal change plays a big part in the game. Tactically, the alterations to the map that come with the passage of time can open up new options for attack or retreat as snows thaw or rivers freeze over, and the resources available shift with the weather. Since its announcement and Kickstarter, I’ve seen At The Gates as a response to Shafer’s work on Civ V; an intricate slice of history as dessert after Civ’s main course. I put that idea to Shafer at the recent PDXCON, where we also talked in-depth about his wider approach to design and his work on Civ V.

“Yes, exactly! At The Gates is my version of these ideas”, he replied. “And, yes, it’s something that I’m still working on and it’s definitely going to be finished and released. But it won’t be a Paradox game. Here at Paradox I’m ramping up, building a new team, starting a new project. We’re still in the idea phase. When I go home on weekends, and the rest of the time, I’m doing programming and iteration on At The Gates. It actually works out kinda nice because at Paradox I don’t need to be super hands-on with the programming or really in the thick of it, so I get that aspect of development from At The Gates. It’s almost like having a day job and a hobby, both of which are equally important.”

I’ve noticed while talking that Shafer has a habit of talking about one idea, and then following up with another idea that either builds on it or runs somewhat contrary to it. I ask him if his approach to game design involves formulating responses and reactions to the games he’s currently playing, or that he’s worked on in the past.

“That’s always going to be the case. Everything influences everything. For example, in At The Gates, the characters – the clans – are in a lot of ways inspired by games like Crusader Kings II and King of Dragon Pass where you have individuals that spice up the game. In part that was a response to playing the game as I’d been building it and realising that because you move your tribe around rather than staying in one place, you don’t get to build infrastructure and settlements that become your stamp on the map. You’re not leaving that trace of yourself on the map in the form of a kingdom or an empire.

“With that missing, I realised the game wasn’t all that much fun and I needed to add something else. That was the character system. So the characters are a direct response to a problem that I encountered in the design, but they are also inspired by other games I’d been playing. And of course I put my own twist on it all. Another aspect in At The Gates comes from Imperialism. So the surveying system, where you would uncover resources in that game was kind of micro-managey. It got kind of tedious, but the idea of searching the map trying to find something cool seemed like a good thing to grab onto.

“So in At The Gates, when you first find a resource out in the world, you don’t know what it is. I call that being ‘shrouded’. You then need to send a unit to identify what it is. It’s basically a multi-tiered exploration system. At first you might know that it’s a mineral, but then in the second layer you need to get someone close up to figure out what it is so you can use it. It’s like a second hit. The first one is the discovery, the second is the unveiling.”

This idea of a ‘second hit’ seems relevant to one of my main criticisms of strategy games as a whole and 4X games particularly. There are often rewards, doled out to keep the player invested in the game, but after a while they become part of a few repetitive loops. Fill the construction bar to improve the city, fill the settler slot to found a new city, break down the walls to hurt the enemy. By the end of a long game, you’ve been completing the same loops to get the same rewards over and over again.

“Yes, that is a huge thing and it’s something I think about a lot, not just in At The Gates but for all the future games that I work on. I’m going to be tackling that precise thing in a very direct way.”

His approach to this problem hasn’t been informed solely by strategy games though: “A lot of the games I play now are more roguelikes: The Binding of Isaac, Spelunky, RimWorld, Don’t Starve.”

“They have strategic elements to them but they’re very different when it comes to what you were talking about with the loops. You don’t unlock the same technology every time, or a building that does the same every time. I want to add more, as those games do – new things to discover so that you don’t hit the same beats at the same time on every playthrough.

“In At The Gates you might find a shrouded mineral but when you reveal it, you might just get an iron deposit that lets you build some weapons or tools, but it could be a huge gold deposit that completely changes your game. They’re rare, but they exist. It’s not always going to a basic mineral – there’s a bigger spread of things that can happen.”

Everything else has roguelike elements these days (except maybe sports games – give me a roguelike football game, please, thanks) so why not a hardcore historical strategy game? It seems like a sensible step, bringing the unexpected to the journey through history. There’s more to At The Gates than a set of slightly unusual inspirations though. It might also be the meeting point of Civilization and Crusader Kings, a game that is tied to 4X conventions but interweaves them with roleplay and interactions between characters. In our interview, Shafer said that the great opportunity for strategy games lay in their use of characters. After years in development, At The Gates might be primed to seize that opportunity.


  1. LewdPenguin says:

    A roguelike sports game you say? PES has (or at least had, been a few years since I played one) the Become a Legend mode, where you take on the role of a young player and work your way up from a hopeful in the lower tiers to superstar status, so that’s already part of the way there. Throw in the risk of career-ending injuries to have some sort of game over fail state and it’s done.

    • Augh_lord says:

      Call me when they add procedural stadiums, equipment crafting and have survival mechanisms.

  2. jamhov says:

    Not a single follow up question about the ridiculous delay in release, the utter lack of communication, or even a discussion of ATGs current status in development? He gives a hand wavy answer that amounts to “yeah I’m still working on it.” And there is no follow up?

    This reads like a promotional puff piece for a game that at this point should be considered vaporware.

  3. TillEulenspiegel says:

    “A lot of the games I play now are more roguelikes: The Binding of Isaac, Spelunky, RimWorld, Don’t Starve.”

    If we’re calling RimWorld a roguelike, I give up. The word is dead, unreclaimable, utterly devoid of any discernible meaning.

    • Premium User Badge

      Drib says:

      That’s an awfully roguelike statement to make.

    • Enkidum says:

      How about “procedurally generated with game-ending failstates”? That seems to be the thing that all things currently referred to as roguelikes have in common.

      • skeletortoise says:

        This is a good start, but a bit too vague, I think. Game ending fail states could be interpreted a number of ways which might cover the vast majority of games, and plenty of game genres tend to have procedurally generated elements or maps (at least, as far as I understand the term, which may not be totally technically correct). I think a key element would be having an unorthodox save system, or none at all.

        • Enkidum says:

          I guess “unorthodox save system” is what I meant by “game-ending”. That is, once you die (or all your colonists die, or whatever), you’re done, and you don’t reload. That plus lots of procedural generation seem to be the two defining characteristics of virtually everything that has every been described as a roguelike, from Rogue on up.

      • ulix says:

        “Unorthodox save system” in modern roguelikes (or roguelike-likes, or rogue-lites) usualky means only having one save that is automatically overwritten when you end the application. There’s no turning back.

        I do however usually get around that in don’t starve by quickly hitting alt-f4 when something bad happens. The autosave only comes a few seconds later, so by closing the game I can then continue from the last autosave.

        It is cheating, I know.

    • skeletortoise says:

      I originally read ‘word’ as ‘world’ and thought you were being a touch dramatic.

      • Captain Narol says:

        Funny coincidence, me too !

        • syndrome says:

          It wasn’t a coincidence, and it isn’t that dramatic.

          “The worLd is dead, unreclaimable, utterly devoid of any discernible meaning.” is what your subconscious is telling you.

          Welcome to Samsara. huehuehue

    • Turkey says:

      When will these plebs learn to follow the rules set by the Munich Declaration when defining what a rougelike is.

      • Junkmail says:

        The Berlin Interpretation is a stuffy and elitist document, but it ultimately comes closer to being a good definition of a roguelike than the modern one we use. Roguelikes aren’t actually about random generation or permadeath at the end of the day, they just enable surprising and funny interactions as well as weighty decision-making, which I would argue is the core of the roguelike, and that most modern ones (save spelunky) fail to do.

    • Snowskeeper says:

      Yeah, we all know you’re supposed to call them either roguelikelike, or, if you’re feeling really particular, roguelites. Because that matters, in the grand scheme of things. And the meanings of words never change.

  4. bramble says:

    Back in November 2015, Jon sent a lengthy and earnest Kickstarter backer update email where he discussed the pressures of indie game development, his failure to anticipate and plan for them, and the effect it’s had on his life. He talked about a lot of things, but almost as a post script to the whole thing he discussed how not only has he sold his house , car, and cashed out his retirement to keep working on the game, but that he suffers from schizoid personality disorder, which he said prevents him from having emotional reactions or ‘regular’ hobbies outside of working on the game.

    I remember reading all of this and feeling incredibly guilty that I was one of the Kickstarter backers exerting this invisible pressure on a sick man who sinking every conceivable financial and personal resource into a game that looked interesting to strategy enthusiasts, but I knew would have almost 0 mainstream appeal. He was throwing his life away for something that would never pay him back. I’m relieved Jon has managed to rebalance his life with a new job at paradox, and that AtG is a weekend passion project that it always should have been. Even though I kickstarted the game in 2013, I never wanted​ anyone to ruin their life because I threw him $20 5 years ago for an interesting idea. I look forward to playing AtG it and when it is ever complete, but mostly I hope Jon stays safe and happy and balanced as a human being as he pursues his career and passions as a game designer.

    • meelawsh says:

      I will wait for AtG as long as it takes. I will also accept a failure and get over the $20 I paid. But don’t tell us about suffering anxiety and then disappear for half a year, that brings on weird feelings.

      • Premium User Badge

        Iamblichos says:

        I agree… and I would also point out the surfeit of developers who develop said “anxiety” or supposed psych issues after taking a lot of people’s money and delivering no product. It may be completely legit in any given case, but it’s also a well-known technique used by con men for making their getaway. Speaking personally, when I enter into a financial transaction, I’m not looking for a deep personal connection with the person with whom I enter said transaction, I’m exchanging money for product. If the product is not forthcoming, I don’t need to hear about their childhood or dead aunt Linda or any of the rest of it… just give me my money back and off we both go to more profitably use the time we have left on Earth.

    • AceJohnny says:

      Reminds me of the saga of Limit Theory, whose developement updates were often featured here, until the developer suddenly fell off the face of the earth in early 2015.

      Turns out he had a similar kind of breakdown.

      He came back, and had to restructure his approach to work, but it’s another useful datapoint to the perils of creative development under KS-like pressure.

      Development continues, and I’m glad to see he’s resumed regular updates in 2017.

      • Hedgeclipper says:

        Seems to be an awful lot of kickstarter backers who think that $20 entitles them to not only be the boss but also the sort of nasty micromanaging sociopath boss everyone hates. “Smithers I want a twenty page progress report and novel length dissertation on why you’re a useless waste of oxygen on my desk in ten minutes or I’ll rant on the internet again.”

        • cpt_freakout says:

          You should see the updates and the forum of That Which Sleeps, another KS darling back in the day that’s been delayed for quite a while now but that’s produced the developer no end of anxiety due to people just being unable to let the man work.

          • teije says:

            Yeah, that’s been a mess, with a falling out between the 2 main guys, lots of angst, etc – and as result, most have declared it dead and gone. As a backer, I was very surprised to receive an update earlier this week, so it appears it has life in it yet. Sure hope so – great concept.

            My approach is now to view a KS pledge as a bet on something interesting being created. If it pays off in something enjoyable, great. If not, because of stress, life challenges, health – on behalf of the maker – then a twinge of regret for what might have been and move on. No point in internet raging at people I don’t even know.

          • TheOx129 says:

            As a backer of That Which Sleeps, I’ve pretty much written it off entirely at this point. Josh makes periodic updates, but they generally consists of 1) information about minuscule aesthetic changes (e.g., “I changed this border to be red instead of black to make it easier to see!”), 2) excuses for why a playable beta or even a gameplay video has been delayed for the umpteenth time, and 3) apologizing for how long the project has taken.

            While I haven’t followed it as closely as others, my understanding is that the project took a wrong turn a while back, as Josh became fixated on the idea of complex, deterministic AI: it essentially entailed redesigning the game from scratch as it went from “abstracted, board game-inspired 4X” to “insanely complex, agent-based simulation.” I’m no programmer, but from what I’ve read of comments by people who are, Josh would basically need a full-time team of programmers specializing in AI working exclusively on that for the development cycle to achieve even a fraction of what he wants.

            All of this drama has led to folks questioning if there was even anything playable in the first place, or if it was just a bunch of art assets that they were hoping to turn into a game. In any event, at least one frustrated backer got productive and is making his own cosmic horror sim – Shadows Behind the Throne – that he plans on kickstarting.

        • Landiss says:

          It’s interesting. First, the game developer does everything to make people emotional about the game, to have a kind of personal attachment to the product. This is used to get people’s money. Then, when it works, same developer gets a burden and stress that comes from people who feel emotional about his project. Is this really a surprise?

          Just to be clear, I don’t think constantly ranting in the internet about delays in kickstarted games is good. I also don’t want anyone to have psychological health issues because of that. I would just like if game devs were doing a more educated decision when they launch kickstarter and think about consequences before they do it. Hopefully everyone thinking about starting a kickstarter project is first reading all those stories of failed kickstarters and issues due to high pressure from the community.

          And seriously, don’t budget your game for $40k. I know, the guy is young, but seriously. If think you need that low amount of money, think twice. You either don’t need it and you are able to finish it on your own, or in reality you need much more than that. 40k per 3 people is 13.3k. Minus other costs and taxes. Even assuming no delays and sticking to the original plan, I doubt it would be more than 500 $ per month per person.

          Ah, but I admit I’m full of shit and it’s easy to criticize when you don’t take the risk yourself.

      • Smurph says:

        It bothers me a lot less when devs who are new to game development like the Limit Theory guy struggle. When someone advertises their big time commercial game experience to sell a Kickstarter they’re implying that they know how to get the thing done. So I’m a little more dissapointed when they flame out just like an inexperienced dev because they should have known better.

        • tlwest says:

          So I’m a little more dissapointed when they flame out just like an inexperienced dev because they should have known better.

          Um, you should be aware that a large number of professional games flame out after a few million dollars investment.

          Personally, I would like KickStarter to put up a dialog box when you pledge warning

          You are donating money for the *hope* that someday you just might see something that might possibly appeal to you. You are donating the money to someone’s artistic vision that may pan out, but more than likely, will end in tears. This is most emphatically *not* a pre-order service.

          Do you still want to give?

  5. brucethemoose says:

    Speaking of Paradox, that sounds like modded Stellaris to me.

    In the vanilla game, you find a 22-tile habitable planet on sensors, and you’re like “oh good, another one”. You colonize as part of that same loop.

    But with mods, you don’t know what the heck it’s gonna be. It could have precursor shipwrecks that turns your empire into a research meca. It could have a mysterious factory tile that gives you a production boost, only to have it dump hostile armies on your planet later. It could have giant spiders. It could be a paradise with like 5 happiness and food modifiers. Or it could start a crisis that puts the Unbidden to shame.

    Same with anomalies. Vanilla Stellaris varies from “Your science ship was destroyed” to “+3 minerals” to “here, have some unity”. Whereas one of the long mod chains could lead to a cruiser with a Titan laser on it. Or Cthulhu. Point is you just don’t know until you’ve already committed to the choice, like that dev said.

    • Someoldguy says:

      Is that supposed to be good? Don’t get me wrong, I played the crap out of Rogue back when it was an acceptable thing to do in a worktime lunch break, but most games of Rogue lasted about as long as it took me to eat a sandwich, drink a cup of coffee and smoke a cigarette so it was no big deal when the RNG pulled some nasty stunt and killed you. Like playing any of the arcade games of the day, it wasn’t about winning, it was about trying to get a high score before you lost. When games like Stellaris can last tens of hours, having the RNG unleash interstellar mongols on you for no particular reason doesn’t seem like a lot of fun to me.

      You also do vanilla Stellaris a disservice to suggest it had no interesting events. I’ve had some pretty useful stuff from anomolies and followed some long event chains. Just not stuff that’s going to trivialise your last 2 hours of effort in one dice roll.

      • brucethemoose says:

        I was being a bit dramatic. Most of the really bad events (eg the LEX mod) are progress gated (either a giant fleet guarding the system, or doesn’t spawn till late in game), and even if some enormous purifier nation takes that fleet out, it’s still a VERY ominous “you REALLY shouldn’t touch this” artifact. Worst I’ve got from colonization is a destroyed planet or a hostile army holding it, never a huge crisis fleet or anything.

        My Stellaris games tend to have big galaxies with lots of planets and a wide empire, so the effect of a really awesome planets/resources or a nuked colony ultimately average out. There’s also a decent amount of rubberbanding in the vanilla game, like salvage research, the vassal system, opinion modifiers etc.

        However, I will say that I don’t play on ironman mode, just in case…

      • brucethemoose says:

        Actually that is a lie. There is 1 system guarded by a big fleet with a 25-tile Gaia world that… well, VERY bad things happen if you colonize it, and it’s not 100% obvious.

        • Someoldguy says:

          I’ve learned to stay away from 25 Gaia planets. Once you have a bit of tech, all those 23 tile 80% habitable worlds are plenty good enough.

  6. zulnam says:

    Kickstarted in 2013, sweet christ! I’d hate to be one of the backers right about now.

    • SanguineAngel says:

      I am one of those backers and I say “Eh”. Honestly, Kickstarter has always been about giving to people who have a dream, not about pre-ordering. The only negative feelings I personally have about kickstarting AtG is that I feel bad for john as he piled all the pressure on himself.

      For me, and I suspect for many (I’d hope most), we’re giving to provide someone the chance to follow their aspirations. If it pans out then fantastic, and we get a little something. If it doesn’t then no big deal. It was a gift, not a purchase.

  7. NailBombed says:

    On the offside of all this – the only thing I think of when I see the words At The Gates…… is this.

  8. NelsonMinar says:

    This game took $100,000 in pre-orders for a June 2014 release date.