Share this project


Share this project

Lead a barbarian kingdom to glory during the final days of the Roman Empire. An empire builder by Jon Shafer, designer of Civ 5.
Lead a barbarian kingdom to glory during the final days of the Roman Empire. An empire builder by Jon Shafer, designer of Civ 5.
3,009 backers pledged $106,283 to help bring this project to life.

July 2013 Update - Alpha Testing & Resource Appearance

Hey all, fairly short update this time - news I'm sure many of you are rejoicing over!


Alpha Testing

The plan is to kick things off in late August or early September. I'll be wrapping up a couple more big features next week, but before spending a lot of time on playtesting I want the AI to be capable of the basics: defending itself, claiming resource deposits, taking out hostile tribes, etc. We have a good idea of how it will do this, and now it's time to actually put that in code. AtG is very much playable right now, but ultimately, until the AI is competing with you it's impossible to get a good feel for what state a strategy game is in.


Resource Distribution

Our second topic is those big features I alluded to above.

As you all know, in AtG it's important to be thinking about the opportunities in front of you, be it migrating, claiming a valuable resource that just appeared, or taking advantage of a surprise diplomatic request. Resources in particular are a major factor which drives the game, and lately I've been working on how they're distributed across the map, and the rate and manner in which new ones appear.

One of my goals is for different regions of the map to have character, instead of everything feeling like one homogeneous mass. To help achieve this, inside the random map generator I added the ability to produce clusters of items. These can be resource deposits, neutral-owned improvements, goody huts, you name it. I call these "Region Traits," and down the road I might flesh them out even further and allow for properties that aren't simply objects which occupy tiles.

I've had to make several tweaks in order to achieve game balance, and no doubt many more will be necessary. Fortunately, I knew this ahead of time and the system is very flexible. You can throw down basically any type of object with it, define what terrain must (not) be present, which other traits must (not) be present, etc. The old, scattershot distribution logic is still in there and everything is in XML, so modders should have a lot of fun playing around with all the knobs.


Resource Appearance

The other resource-related feature I've been working on is resource appearance. At the start of a game a small percentage of them are visible, and new ones pop up over time. I've thought a lot about the best way of doing this... should the appearance of new deposits be completely random? Should you have to send a unit off to find them? Is there some kind of abstract investment players can make to speed up the rate? Unfortunately, there are problems with each of these.

Units manually checking each tile is nothing but tedious busywork, and wasn't something I particularly enjoyed in SSI's Imperialism. Overall I liked that game, but there really wasn't any strategy to finding new resources. Off-map investment is too abstract, and doesn't fit thematically in late antiquity. And while deposits just popping up out of nowhere works, it's a shot squarely across the bow of suspension of reality: if everything else in the game requires sending units out to do it, how are you coming across this perfect geographic information?

The design I've settled on is a mix of the best elements of those problematic ideas, joined up with a couple new ones. When you first encounter a new resource deposit it's "shrouded." You can tell if it's a rock, a plant or an animal, but not exactly what it is, nor can you use it. To resolve that, a "surveyor" unit has to go out and identify the deposit. New resources do appear somewhat randomly over time, but instead of seeing a "large iron deposit" appear in the fog, the information you're getting is basically just hearsay that needs to be verified. The deposit may not even exist at all!

There are several reasons why this is the approach I've settled on.

First, it dangles a carrot out there for players, and gives them something to get excited about and plan around. It also requires active involvement that can be prioritized, shifted, etc. I didn't want a system where you just waited around, hoping that at some point in the near future the game decides to be nice and serve you dinner.

And, gameplay aside, what I really like is that it makes sense. While a scout is off running around he might come across a shiny metal in a pile of rocks or find animal dung on the ground, and you should at least have some information about what you just found. Or rumors about some metal to the east might circulate, but there's no way to know for sure. In the end, drawing accurate conclusions requires someone with experience taking a look.

This new feature will be one of our focal points with the upcoming alpha test. As always, there's a chance it's just plain unfun or has other major problems, and we have to go back to the drawing board - but hey, that's game development!

- Jon

AtG Website | Conifer Forums | 'Like' us on Facebook | Conifer Twitter


Only backers can post comments. Log In
    1. Missing avatar

      Joseph Miller on

      I like the idea of shrouded resources. I assume they resources appear as soon as they are "seen" by the scout's sight range. However, I do wonder if you might just want to give the scout a "survey" action and cut out the middleman. This makes for an interesting choice for the player. Do I continue scouting more lands? Or do I take the time to survey the resources I find? Do I build another scout just to survey what I find?

      Of course, you could do something along the lines that scouts can identify animal and plant resources, but more advanced resources like stone and metal ones require a special surveyor unit.

    2. Missing avatar

      Juhani Nurminen on

      Seems like a nice model for handling unknown and newly appearing resources. Hopefully the surveying costs something to deter the player from just surveying everything everywhere, but rather that you'll have to think about if you actually want to expend effort in surveying them.

      I'm more interested about the Region Traits part. You didn't specifically mention that the terrain tiles would be any different in them, and to me, that is the most important part in a heterogeneous landscape(dishearteningly, you mentioned the old scattershot terrain method still being in there). Few games have random map generation algorithms that provide varied, but more importantly, interesting maps. I mean, if it just splatters deserts and mountains roughly equally on the map, that's boring. If the Region Traits also include terrain, it would be great. I mean, it could set a 15x15 tile area as being, say, "Swampy", so it contains swamps or bogs, some grasslands and some forests, but few mountainous or hilly areas. Likewise, a "Mountainous" area would have a much higher change of mountain clusters and fewer swamps and such. I want each particular part of the map to have it's own theme or feeling to it. I feel more or less the only game to have really done this justice is the Civ4 mod Fall from heaven 2. It had a great random map generation, with huge deserts, jungles, valleys, and so on. Really gave it both a cool thematic feel, as well as a meaningful strategic decision feel. It would be double so in a game about migration. Do you choose to go to the fertile valley or to the possibly iron rich swamps or mountains? The deserts might have rich trading cities, etc.

      In any case, if the map stuff can be modded as you mentioned, perhaps that will have to do if you have no plans for it.

    3. Missing avatar

      Emaze on

      Lol! Jon's KS project trolling him. Great idea for resource discovery. The random and arbitary appearance of resources in the civ games always bugged me a little. I wonder how you'll connect resource appearance with region traits. I imagine as you find an iron ore vein in mountainess terrrain that there could be a regional trait active where there is a larger chance of finding another vein than if a vein was found in a forrested area. Also logically the longer you spend and deeper you mine in a mountain the likier it is you find another vein. Not that there shouldn't be a diminshing return in both areas before they are depleted, but it should happen sooner in less obvious places. Keep up the good work and thanks for the update.

    4. Jon Shafer Creator on

      The big advantage of being indie is that I can set our priorities to be whatever I want them to be: and our only goal is to make an amazing strategy game. If that means we have to delay things then so be it. It's not ideal, but I have no doubt people would rather have a great game late than a mediocre game on time.

      - Jon

    5. Mr Understanding on

      I was thinking about how resources should appear over time in strategy games (yes, I have a very exciting life) and you've come up with a great idea.

      And don't worry too much about holding off with the alpha, I imagine most of us are still playing Brave New World. :P

    6. Jon Shafer Creator on

      It doesn't say "Jon Shafer's At the Gates by Jon Shafer"? :P Every time I read that I feel like Kickstarter is trolling me.

      - Jon

    7. Squishymaster on

      I love your updates. 1) it means we're getting more intel on a great upcoming game. 2) The e-mail I get says "Jon Shafer's At the Gates". Aww Jon's attacking!!!!