Try to describe A Valley Without Wind, this open world dungeon crawler turned on its head, and you’re bound to forget something.
For anyone looking for a truly unique take on an old standard, try surviving in a world where the game plays against your strengths, taking on randomly generated worlds, and beating the game is only the beginning. Instead of your traditional hack-and-slash, you rely on magic to take on your foes in this 2D post-apocalyptic world where Iron Age collides with the world of technology.
See what I mean? I still haven’t mentioned that you have to build your own civilization or when you die to come back as someone else. Every area presents a new challenge or even a new spell in your ever-growing inventory. Even after hours of playing through it, I still have yet to scratch the surface of this game.
Fortunately for you, I’ll let one of the creators of one of the most ambitious games you’ll play all year, Chris Park, tell you about what it takes to create A Valley Without Wind.
Where do you even begin by creating such a unique world?
A lot of it just comes from mashing all of my favorite things together. Since I was a kid I really got hooked on the idea of magic and technology co-existing side by side. I think it was probably Final Fantasy VI that really pushed me over the edge into really adoring the concept, but I'd encountered it even before then. With AVWW, at some point there was the thought "what would it be like if there was a world where magic had always existed, and was considered just a regular ordinary tool that everybody uses?"
That kind of combined with one of my other favorite tropes, time travel: Chrono Trigger is another favorite of mine, and I can probably recite the Back to the Future movies from memory. The thing is, time travel gets done a lot and the actual time traversal mechanics often aren't that interesting these days -- it's been done. I read a fair bit of sci-fi, and another interesting concept along those lines is the idea of Time being the 4th dimension, and about how it can be thought of in some senses as a spatial way. Flatland is a great primer on that, dating back to almost a century ago at this point.
If you break an object in three dimensions, it's obvious what happens: you get pieces of that object. You can then glue them back together if you want. I thought: what happens if you break an entire world in four dimensions, and then have it all recombine? You'd wind up with both spatial and temporal shards that got jumbled together, and suddenly you don't need time travel to see the whole history -- past and future -- of this world all laid out in one plane of existence. Plus you now have the interesting situation of Bronze Age folks getting introduced to new technology, and so forth.
The story is understated in AVWW to say the least, but it was these narrative ideas that drove the creation of that world. And we've started pulling at some of those narrative threads with the mysteries that you can find and unlock. Right now there are two mysteries with a total of around 28 clues you can use to solve them and find out more about just what the heck happened to this world of Environ (which is meant to be a clever substitute for "Earth," not something generic-sounding). As we proceed with post-release content there will be increasing numbers of mysteries to solve to find out just what is going on; but frankly we didn't just want to reveal everything right away!
While designing the game, you've mentioned that you listened to your public testers throughout the process. How did that change the direction or the way that you built the game?
Certainly their feedback was invaluable in making the game more accessible; developers tend to be far too close to their own games to be able to judge that sort of thing accurately for themselves. But even moreso, players are great at identifying things that aren't fun or which just feel out of kilter with the rest of the game. Sometimes as a developer, there's a feature that is bugging you below the level of conscious thought, but when a player raises the issue you instantly know you agree and wonder why you hadn't thought of it yourself.
Other times it was less obvious that the players were identifying a real problem, but after a lot of forum discussion and group brainstorming, a better solution that satisfied all parties could be arrived at. The entire concept of continents came up from the beta process -- before that it was just all one landmass. The tier system was arrived at during beta as a replacement for a more traditional level/EXP system that you generally see in RPGs (which was super problematic for this game for a lot of complicate reasons). The way that monsters spawn was even heavily changed during beta, as we learned what players found fun and satisfying and what they didn't.
Oh, and the systems for healing, fast travel, and even the entire concept of missions -- central to the game at this point -- resulted during the beta process. I just can't even express how important the players were during beta. We don't "crowdsource" per se, but we are very experimental with how we prototype new features. And since we're so experimental, we can't just say "oh, we know this works because it worked in other game X." Usually, for better or worse, there's no other game doing things the way that we are, and so the only thing we have to go by is our own gut instincts and the feedback from players. Gut instincts alone are not remotely enough, I can tell you.
One of the great things you worked into the game is a system where the game learns from the way you play. How hard was that to pull off?
With a traditional linear game, level designers handle all this simply as part of their level designs. Early levels have easier enemies, later levels have harder ones, and there you go. That's hard to tune exactly right, but it's conceptually really simple. In our case, since there is no linear path and we have no idea what the player will do first, we couldn't do anything remotely like that.
So we looked at it from the other way around: okay, so we have Feature X that is too complicated or hard to have available to the player right at the start of the game. What actions of the player can we measure that show us that they are ready for Feature X and in fact probably relishing the opportunity for something new and more interesting? We want that new feature to be hitting just as they're getting bored with whatever they've been up to prior to that, instilling new energy into the game.
We thus came up with an Unlockables system, of which I think there are about 96 at the moment but we add new ones all the time. The basic idea is that when you complete some action (killing some number of enemies of type A, reaching the second continent, beating a certain kind of mission, etc), then you unlock some new part of the game. We call this "content gating," and it's the very same concept that is used by a level designer working on a traditional game.
It's not meant to keep you from getting to the good stuff early, it's meant to keep you from getting overwhelmed. The difference here is that rather than pre-scripting out the experience that everyone will have, the unlockables are based on your own actions and thus happen in arbitrary order that you choose. That makes the experience more custom, and brings in that adaptive element that you mentioned. Surprise, surprise -- this was another innovation that came about during beta, when we were trying to figure out ways to keep the game deep and open-ended without just throwing new players into the deep end.
Are there plans of adding anything to the game post-launch?
Definitely! As noted already, we've done ten post-release updates just in the last two weeks, and those have added a variety of things. Among them have been yet more polish and clarity to a few mechanics, several revamped missions based on player feedback, two new enemies, twelve new "elite" forms of enemies, revised upgrade stone mechanics (to make those less grindy), revised windstorm mechanics, and so on.
Perhaps one of the most exciting post-release additions for players has been a new store at which they can spend consciousness shards to get specific supplies or enchants that they would otherwise have to acquire through exploration. These shards are found both through exploration and from killing trash mobs, and thus this creates a whole new feedback loop in the game for players to have yet more flexibility in how they go about their business. Those that are more combat-oriented rather than more exploration-oriented now have some powerful new options, in other words.
With our first game AI War, it is now on version 5.0 and has three expansions out. The expansions are optional paid bulk content bundles, but alongside that we've put out enormous amounts of free updates over a span of three years. Our plan, assuming that player support remains at all as enthusiastic as it has been thus far, is to do the same sort of thing for AVWW. My dearest hope is that, three years from now, we'll all be able to look back and think "oh, how quaint! Can you believe how small AVWW was back then when it only had 40+ hours of content?"
AI War 1.0 was massive in its own right; AI War 5.0 is one of the largest strategy games ever made, I feel safe in saying. If the player support is there, that's exactly the sort of arc I want to see for AVWW as well.