DT: Looking Forward

Hey everyone, we’ve missed you! Though we spent all of last week down in Los Angeles for a company conference, we are now all back in the saddle and jamming hard on Alpha 18, which will focus on building and various toothy, complicated performance optimizations. Before we get into those, though, I’d like to spend an extra-long Desktop Tuesday talking more about some of the high-level things that came up in our 200th Twitch Stream: in particular, about where we’re going next together with Stonehearth as a whole.

Looking Forward

Recap/transcript from the video:

Looking back

Lots of people have asked for a Desktop Tuesday that takes a break from focusing on features and detailed development, and that instead focuses on the state of Stonehearth–where we’ve been, and where we’re going. In our 200th Twitch Stream, which you can find archived on our Twitch channel, we did a lot of the “where we’ve been”: the graphics test, roofless alpha 1, to trappers, weavers, and multiple stories in houses. We looked at the desert and Rayya’s Children, the introduction of save compatibility, and finally all the new combat improvements from Alphas 14 and 16. Woo!

But how do we even look forward?

It’s a bit hard to talk about where we’re going because Stonehearth’s strength as a game comes from our close, iterative development conversations and interaction with all of you. Our alphas are designed around delivering you either one significant, fun feature–like a new class, or around relieving a pain point, like fixing bad performance or incomplete buildings. We meet every Monday as a team to realign our plans for the week and to make sure we are always working on the most important thing. It’s very common for us to change our weekly tasks based on bugs you’ve reported and features you tell us would make your experience of the game better. The result is that you get consistent, regular builds that actually work. However, it makes it very hard to talk, with confidence, about a schedule–when you might all get the things you’re looking for. Details change every day, features change dramatically every month.

So, if we can’t talk about when we’re getting features, or what those features are in detail until we actually implement them and iterate on them with you, what IS a useful way to think about where we’re going? Well, one thing we know for sure is that we will continue to focus on finishing single player, PC Stonehearth before we look at multiplayer or Mac and Linux support. Multiplayer and Mac/Linux are definitely still planned, but multiplayer, in particular, is exponentially more difficult than single player, from a gameplay, technical AND social perspective. Therefore, it’s really important that we finish the foundational elements of the game first, and that means a completely satisfying single-player experience. That said, where ARE we going within the scope of single player, PC stonehearth? Well, one way to look at the future is to speak broadly about the three largest recurring challenges our team is currently facing, since meeting them head-on will be a recurring theme for the weeks and months to come.  

Challenge #1: The design loop at the heart of our game.

A lot of the really good feedback you’ve given us this year–about combat, about the trapper, about building gameplay–or our lack thereof–has made us realize that our existing simulations don’t work together nearly as well as we’d like them to. Really good games have an ever-expanding loop: they teach you something, reward you for doing that, and then broaden your scope of capabilities in an ever-expanding tornado until you are a legitimate master of the system. We need to tune Stonehearth’s design tornado, and make sure it links all our existing systems together.

Challenge #2: Performance

Are your hearthlings idle? Are they doing something dumb? Probably, it’s because your CPU is maxed out, and THAT is because the simulations we write are computationally expensive. Every time we write a new feature, we can’t be absolutely certain of its performance impact, especially since the goal of most of our features is to create emergent, unexpected behavior. When you do find issues, and report them back to us, it can take hours to reproduce and triage the issue, especially if it involves hardware or hardware combinations we don’t have in our tiny testing environment. We have to go back and figure out how to tackle our performance issues from base principles, instead of after the fact as bugfixes.

Challenge #3: Big outstanding art and design questions for upcoming features

For example, when you clear the trees, there are giant empty swaths of green. We can do better than that–those spaces should be filled with interesting details, but what should they be? How will they enhance your game, as well as the visual appeal of the world? When we add a new piece of furniture, or a new class, or meet as a team to decide what Titans should be, or how seasons/weather should work, or what should happen at the edge of the world, we need to make our decisions based on a unified, coherent sense of what it eventually SHOULD look like. As tempting as it has always been to put in whatever’s coolest, we’ve been putting off a lot of these questions and we need to take some time to figure out where we’re going, so we can get there in an efficient way, and with less backtracking.

Our challenges stem from a common root: Diffuse Vision

When we started to look at how to address these challenges, one thing we’ve increasingly realized is that they’ve been piling up slowly, and that addressing them individually is like adding bandaids to a leaky boat: lots of effort, with small, incremental rewards. Is there a better way? Do our challenges, in fact, stem from a common root?

After much analysis, we have, in fact come to believe that this is the case. One central problem is this: Stonehearth, as a game, is incredibly difficult to describe to people. I mean, have you ever tried it? We try all the time: in our kickstarter, at conventions, to our parents and neighbors and friends. It’s a building game with combat and RTS and RPG elements! It’s fully moddable! It’s like a cute Dwarf Fortress, which, in turn, has everything, from individual body part damage to dwarf poetry!

So, if the game is really difficult to describe, if its vision encompasses everything, well, that’s actually quite a large problem. It means that every design choice–rts or rpg style combat? D&D or FF Tactics style leveling?–is right choice. It means that every kind of class could theoretically belong. It means every feature is just as important as every other feature. And what’s really true is that we need more focus. Don’t get me wrong: Stonehearth should be inclusive, and one of it’s strengths is that via mods you should be able to add whatever you want–but at its core, it must focus on a single vision, in order to be any fun as a game.

So this is the vision: Stonehearth is a townbuilding/ant farm sim set in an SNES style fantasy world

So we did some soul searching, looking for proper visions for Stonehearth, ones that encompass the best of what we have but that also keep us honest to what we still want to build. We also talked a lot with you–we’ve been asking you what your ideal Stonehearth would be, and we really took your answers to heart, even when they surprised us. In the end, we believe that that Stonehearth is this: a townbuilding/ant farm sim set in a Super Nintendo-style high fantasy world. Put another way: have you ever wondered what it would have been like to build Hyrule’s Kakariko Village? Or set up ChronoTrigger’s Millennial Fair? Or Final Fantasy’s Figaro Castle and Town? And then live (really live!) there? This, we believe, is what we’re aiming for, in Stonehearth. To make a game where you guide a small group of hearthlings as they build a fantasy town to your exact specifications inside a living world full of monsters, mysteries, and looming disasters.

Ok, so how is this useful? Well, from this vision, a few strong design pillars follow:

  1. The world must be alive. Part of the appeal of the game is that the hearthlings and monsters and animals and pets all have their own lives. You can tell them at a high level what you want them to do, but they’ll do it their way. Furthermore, the world must be full of systems that interact with you as you interact with them: weather, terrain, water. And finally, the world should be full of ruins and mysteries and monsters, big and small, that are doing their own things, and whose lives interact with yours in an organic fashion.  
  2. The settlement you create must be yours to customize. We’ve already invested a ton in the building editor: what more can we do to make sure that your town and your townsfolk look exactly as you wish, within our fantasy aesthetic?
  3. Creative experiences are best shared. We want you to be able to share your towns, tell epic stories about your hearthlings, and eventually, build kingdoms together.

In the months to come, we look forward to applying this vision and these principles to the three problems I mentioned earlier, to clarify and enhance Stonehearth as a whole. For example, when we try to knit together our food, building, and happiness systems, or when we need to make a choice about how a simulation should work, we’ll pick the thing that will best allow for you the player to make a meaningful decision and then get to watch as something interesting plays out. When we look at performance, and try to figure out what avenues we can streamline, and what systems must be more robust, we’ll focus on allowing hearthlings to express more individuality. When we examine the UX of the building editor, when we decide how much to allow you to customize your hearthlings’ crafting and gear, when we try to figure out how endgame cities should look and play, we will prioritize your ability to build and place what you want.

So, what does this mean for Stonehearth? 

It is likely that prioritizing and strengthening core systems will mean that we’ll all have to wait longer for some the bells and whistles we haven’t started yet–Titans, weather, festivals, the northern alliance, and more. Alpha 18, for example, is chock full of performance optimizations and a few more building features and high level artistic projections, instead of new biomes or classes, because we believe we cannot get a good handle on what features we should add until adding features has less performance overhead. We think that this investment now will make for a better game overall, and we look forward to sharing our progress with you–in Desktop Tuesdays and streams and regular game updates, as we always have for the last three years.

So where does that leave us? Well, Stonehearth now, is an an ambitious, adorable simulation with some elements–block by block building, for starters–that are unique in the whole gaming field. This is a pretty big achievement for a team comprised mostly of people who’ve never made a game before. We credit you all with how far we’ve come, and it is because of you that we continue to push our boundaries, to dream of a Stonehearth that is beautiful and functional and amazingly enjoyable to play. We look forward to sharing this process of refinement and enhancement with you as we go, to continuing to deliver you awesome features and classes and simulations, and to giving you faster, stronger, and more enjoyable versions of things you already have. And of course, we want and need your your feedback, always, to keep us on the right track.

Other Announcements

Streams resume as usual this week and will run on our regular T/W/Th schedule until the weekend of Sept 1st, when we will be off to show Stonehearth at PAX West, in Seattle. See you online in a few hours!