Jump to content

Fireside Chats (with a video game designer) - June 27, 2014


Recommended Posts

  • Developer

Hello again! Not much has changed in the development of Don’t Starve Together over the past week. Our programmers are still focused on shoring up the server tech. They’re also beginning to look into modifying the world gen code to allow for generating much larger worlds (to support higher numbers of players). I’m continuing with the ghost prototype; making haunt reactions for the many different items and creatures. I’ve also started to think about how to push the loss of progress that happens when you die in Don’t Starve (lost resources, lost structures and lost world) more onto the characters themselves to go along with the persistence that comes with multiplayer. The looming threat of lowered maximum health with the ghost mechanic is part of this, but the level of your various UI meters is more fluid and less attached to your character’s “progress” than the accrual of stuff.

 

So, for this post, I wanted to take a look at our design process using Reign of Giants as a case study. Fair warning that it’s long and fairly technical/close inspection type stuff. Because it’s on the long side, I’m going to spoiler tag the meatier sections. Expand them to see my long winded word vomit at your own risk.

 

Designing Reign of Giants

As I outlined in my previous post, after we finished our six months of regular updates for Don’t Starve, we had a moment to stop and consider the state of the game. As with any project, you want a set of goals to work towards and a timeframe to accomplish them in. For Reign of Giants, we wanted the expansion to help keep Don’t Starve fresh and interesting over long periods of time, both within a given playthrough and across playthroughs. While this goal is implicitly veteran-player-facing, it was important that we keep Don’t Starve’s difficulty and discovery curves intact, so as not to alienate newer players. It was a pretty general goal, so we needed to break it down into parseable and actionable sub-goals. This doesn’t quite get us there, but we start by breaking this meta goal into “things we can improve to increase longevity” and “good things we want to replicate or expand on to increase longevity”.

 

The two main items we identified as things that we wanted to improve with our expansion were:

  1. When you start a game of Don’t Starve, the first handful of days of your game will be spent in a very similar fashion to the first handful of days of nearly every other game of Don’t Starve you’ve played. At best, this is tedious. At worst, it creates an artificial barrier to get to the “interesting” gameplay for a more veteran player. I’d actually argue that the former is actually in service of the aesthetic experience of Don’t Starve: survival can be tedious! However, with relatively few viable early-game survival strategies, that often meant that the tedium became boring, which is not what we want.

  2. After surviving your first Winter in Don’t Starve, you reach a game state that I call “the plateau of stability” (note: this isn’t some fancy design term or anything like that, just a shorthand phrase that I use to describe this game state specifically in Don’t Starve). Basically, the idea is that once you’ve gotten that far, you can survive indefinitely. There’s nothing to prepare for. You just have to maintain the status quo: hang out in your well-established base while staving off hunger, dealing with the occasional hound wave, and making sure you don’t whittle away all of Summer without getting a new winter hat. If you don’t put yourself in a dangerous situation, you’re unlikely to die.

 

But Reign of Giants wasn’t ever going to be all about “fixing” Don’t Starve (after all, we were pretty happy with what we had)! These were the things that we felt were working really well in Don’t Starve and we wanted to push further:

  1. Individual entities or events that create high-drama moments (and panic). The empirical example here is Deerclops.

  2. Systems interacting in unexpected or multiplicative ways to create high-drama moments or interesting/funny results. Examples from the base game: a hound wave coming at the same time as Deerclops or a frog stealing an item from Krampus (which Krampus just stole from you).

  3. Discovery of the world and its mechanics. A huge part of the Don’t Starve experience is seeing something new and poking it to see what happens.

  4. Players have a constant need to manage several different threats to their wellbeing. In most cases, taking action to mitigate a given threat precludes taking action to deal with other threats. Don’t Starve’s difficulty is largely derived from scarcity of time to gather all the resources necessary, rather than from scarcity of the resources themselves. The diversity of demands on the player means that they have to constantly plan and prepare while also being ready to respond to new threats.

 

The final guiding element in designing the expansion was that we knew we wanted to stay focused on the overworld. The caves are great, but we wanted the new content to be an integral part of the Don’t Starve experience and for it to be very easily encountered (i.e. during the course of normal play and not hidden behind a load screen).

 

All this abstract hemming and hawing about what types of things should go into the expansion is a great place to start, but we needed to make some concrete plans about what we would be implementing, so we could start implementing it.

 

Making the early game less tedious

We needed more variety in the early game. The first thing here was to introduce more options. Specifically, this meant more ways to get your basic resources and foodstuffs. We toyed with the idea of adding alternate types of basic resources, but that had pretty serious knock-on effects: we would need to change or add a lot of recipes or we would need to fundamentally change the way the crafting system works, neither of which were viable options. So, we stayed within the existing resources and decided to add the moleworm and tumbleweeds. In addition to being a different creature and “plant” from rabbits and saplings/grass, it was important that the ways that you interact with these things be meaningfully different from the existing sources of morsels, twigs and grass. Moleworms are actively but easily hunted, whereas rabbits have to be cornered to be hunted and are much more effectively caught with traps (which we decided wouldn’t work at all on moleworms to further differentiate the two). Similarly, we wanted tumbleweeds to require a bit more active participation. To justify the additional time and effort, we made it so the payoff for tracking down a tumbleweed would be larger than for picking a sapling for tuft of grass. We really liked the more active hunting and gathering and ended up using that idea with the buzzard as well. Overall, we were looking to robustly support more active playstyles in the early game.

 

Tumbleweeds were an immediate success! Moleworms, on the other hand, needed some more work. We made a couple tweaks to the moleworms themselves (made them nocturnal, made them re-dig their burrow if it got dug up), but the big win in making the moleworms more useful and important (and different from the rabbits) actually came from changing things about other creatures and systems. It’s often the case in game design that to solve a perceived problem, you need to change something about the game outside of the immediate problem area. First, we made it so that rabbit holes would collapse during the rainy Spring season, making rabbits all but unavailable during that period. Second, we made it so that when you start a game on default settings, you’ll have a 50% chance to start in Autumn and a 50% chance to start in Spring (more on the new seasons below). This second change was hugely impactful: upon starting a game, you now had to spend the first few minutes examining your surroundings to determine which season you were in and then adjust your plans accordingly as you prepared for the appropriate coming harsh season. With respect to the moleworms, it had the side effect of making them a relatively important early game food source for Spring starts.

 

Towards the end of development, we added customization options for starting a game with a random character, in a random season and with random season lengths. These are opt-in features that are likely primarily used by more veteran players, but they do a good job of increasing the variation from game to game when used. All in all, I think we hit our target for making the early game less tedious. At the very least, there are more (and meaningfully different) options available early on, should you want to mix it up.

 

Extending the time it takes to get to the plateau of stability

We wanted the game to stay fresh, offering new challenges for a longer period of time within a given playthrough, as opposed to the then-current pattern of surviving your first Winter and subsequently having survival become trivial. Our two initial ideas were a new tier of tech and the new seasons, only the latter of which made it to the shipped expansion. The new tech tier was largely made up of mad science devices. It was intended for late game and even had a biome associated with it: the Tesla biome (working name), which some of you may recall from some art of a Tesla tree that got left in the game directory despite the biome being cut. In fact, mad science was an idea that was initially very central to our plans for the expansion. We had long meetings discussing how a system of powered tools (and harvesting the power) would work. We talked about the population of the mad science biome. During one meeting, someone threw out the idea to have an electric goat with Jacob’s ladder horns. Kelly (one of our artists and animators) bleated and the whole room erupted in laughter. We had a big list of things we wanted to put into the expansion and at the time, it seemed doable.

 

After working for a month or so on some of the basics needed for the new content, the whole team sat down, looked at that list of ideas, and updated our estimates for how long it would all take.  We looked at our timeline. It simply wasn’t all going to fit. As much as we liked the ideas percolating, we didn’t have time to prototype and iterate on things like a new system for powered devices while still making enough content in other areas to deliver a meaningfully large expansion. Not wanting to ditch it all, we reshuffled some things into other areas of the game (like the Volt Goat--we couldn’t say no to getting Kelly’s bleat in the game) and re-kitted other items into a very light version of our mad science ideas, which manifested as a sort of new high tier of items that lived within the existing second tier of science, all based around the Electrical Doodad. We stripped mad science down to a point where it included a couple luxury items and a handful of items to help with the new seasons. Without a full-fledged mad science system, the seasons became increasingly important to get right.

 

Additional seasons fit very naturally into the existing framework of Don’t Starve without introducing any systems that didn’t have an obvious internal logic to them. Even in the base game, it was the seasonal cycle that pushed out the time it takes to get to the plateau of stability as far as 35 in-game days. By adding Spring and new Summer (Summer from the base game became Autumn, keeping it as a relatively easy time of year--appropriately morose for Don’t Starve) and making each of the new seasons present a new and unique challenge appropriate to the seasonal element, we could ensure that players would need to prepare for coming threats for three full seasons. Inevitably, skilled and knowledgeable players will still make it through a full year cycle, but with the new seasons, they would have twice as much game time in a given playthrough before they can settle into a routine of stability. Without some adaptive system that is constantly reacting to the player’s actions (similar to Left 4 Dead’s Director), it’s difficult to design a system that stays challenging no matter what. Given that, doubling the time to reach the plateau of stability seemed like a pretty good solution.

 

Creating drama with individual entities/events

Simple. The eponymous Giants. We already had Deerclops in the game and has seen him repeatedly create those wonderful blood-drains-out-of-your-face “oh no…” moments. We brainstormed ideas for giants to associate with each of the other seasons. They needed to be similarly threatening but meaningfully different. We wanted each giant to target a different aspect of your livelihood, to have different combat behaviors and for there to be multiple viable ways to deal with a giant. Deerclops was mostly concerned with smashing structures, prioritizing that above all else. The Bearger would target your stockpiled food and became massively destructive to anything nearby if angered. But he can also be satisfied with offerings of honey or led away from your camp by laying down a trail of food. The confusingly named Spring giant is not terribly threatening on its own, but it lays an egg that spawns several, well, spawn. The Mosslings will eat food left out but are actually quite passive creatures. Players could choose to live in harmony, relocate for the season, or engage in a difficult combat against an army of Mosslings (or, as we saw in the beta, kill the mother before she could lay an egg, thus missing out on a substantial amount of “boss loot”). The Dragonfly targets your sources of fuel as well as your structures. Also extremely dangerous if angered due to his fiery aura, the Dragonfly is best handled by laying down a trail of flammables or ash, the latter of which he likes to vacuum up until he is full. We also added some weather/elemental systems to the seasons that could create high-drama events, like the frog rain and wildfires. But more on that in the next section.

 

Systems to create drama

As I mentioned above, we added wildfires and frog rain to survival mode. These two systems are very effective in creating drama on their own: a plague of frogs or a sudden fire in your base will always be alarming. But systems that are part of the simulation like these are more than the sum of their parts. With more systems and cycles, we increase the likelihood of overlaps. Frog rain, wildfires, full moons, rain (which affects sanity, temperature, resource quality and comes with now-dangerous-to-the-player lightning), and differences in creatures’ seasonal behavior all create drama on their own, but create exponentially more drama when they occur in concert. Such is the nature of a simulation game. It’s a huge strength and it was clear we needed to play into that. By authoring systems to create drama, we’re also enabling unpredictable disruptions once players do reach the plateau of stability: these systems can create challenges for players with large bases and farms. While Don’t Starve is a very systems-driven game, sometimes we want to author bespoke content and interactions within those systems to make them appear richer or have more surprising results. Functionally, this means adding special cases for answers to questions like “what types of things should the moleworm want to snag off the ground?” and “does anything special happen when the moleworm does grab a given item?” Rather than spoil one of my favorite interactions in the game, I’ll leave it at that.

 

Discovery of the world and its mechanics/rules

This will happen naturally by introducing a large swathe of content at once (new seasons, biomes, characters, creatures, plants, craftables, etc etc). Dropping a lot of unfamiliar entities and mechanics on a player at once will inherently “send them back to the stone age” (i.e. a time when she did not have mastery of Don’t Starve). Even when it was in beta, we saw a lot of posts on the forums saying things like “it feels like a whole new game!” Success.

 

Managing multiple conflicting demands and threats

Adding more systems and cycles (or, in some cases, adding teeth to existing ones) is again the answer here. Static cycles (i.e. seasons) are important so that you can anticipate and plan. Dynamic systems (i.e. frog rain) are important because they can throw wrenches in your plans. Taking existing systems and entities and giving them teeth (i.e. new effects from rain, Deerclops’ freezing attack) further diversifies your concerns at any given time, which is all we’re really trying to do here. Though it’s not necessarily presented in this way, time is actually the most precious resource in Don’t Starve. Yes, it’s true that this is only the case because of other time-based mechanics (i.e. constantly dropping hunger, impending season changes), but it still boils down to needing time to attend to everything. Don’t Starve is difficult because you need to manage these demands within a fixed amount of time (i.e. before nightfall or before the season change). The thing to watch out for here is letting things get out of control: ideally, the player always feels pressure from the multiple demands, but can address their needs within a reasonable timeframe. Our way of trying to avoid this was to keep it to one general seasonal threat (i.e. overheating) and one dynamic seasonal threat (i.e. wildfires). We also tried to encourage movement around the map during the different seasons by making the various biomes more and less dangerous in the different seasons. A couple quick examples of this: the savannah is quite safe, except for Spring, when rabbit holes collapse and beefalo are permanently aggressive. On the other hand, the badlands are quite dangerous but might be worth hanging out in during the Summer due to the lower vegetation density and resulting lower risk of wildfires. This migration isn’t necessary for survival, but it does add another layer of variables to consider. Overall, I think Reign of Giants has a diverse set of threats and demands with a generally appropriate level of concurrence.

 

How Did We Do?

Pretty good, I think! I’m admittedly not the best judge since I’m so close to the thing, but from reading forums, reviews and various anecdotes, it seems like people are enjoying Reign of Giants, which is incredibly gratifying. Now that I’ve kind of laid bare what our intentions were, I’d love to hear if you think we succeeded (or, similarly, if you think we totally missed our target but enjoy the expansion for some other reason that I haven’t talked about)! Either way, it’s June of 2014 and we’ve got a great single player game and a robust expansion for that game. We’re really proud of both. On to the next chapter!

 

Applying This to Don’t Starve Together

 

To pull it back for a moment (and to close out the post), let’s see how some of these ideas apply Don’t Starve Together. To start, the changes we need to make to turn Don’t Starve into a compelling multiplayer game are something of an unknown (as compared to expanding the existing single player game). Functionally, this means that we have to experiment with ideas that seem like they might have legs by building prototypes that are representative of the complete idea and playing with those prototypes. But we’re not doing so blindly: we’re still trying to answer questions when we build these prototypes. The overarching question is “How do we make multiplayer Don’t Starve?” That’s our goal. The specific questions that we’ve talked about before are the sub-goals: the actionable items. To rehash a few of them quickly: How does persistence and day count work, given the fact that not all players in a group will necessarily be online at the same time? What happens if one player in a group dies and the rest are alive? How do we maximize the ease of playing with others while maintaining the tension and sense of loss that is so central to Don’t Starve’s death mechanic? What does progress on a character mean, given that a world is no longer solely yours? Are players able to communicate with each other? If so, how? As we start to answer these questions, it’s a matter of identifying what is technically feasible and then iterating on those answers, which will inevitably lead to additional questions.

 

Wrapping Up

 

Next Friday, a week from today, we’re planning on doing our first internal playtest in a while. I’ll try to grab some video during that and post some clips along with news and thoughts from the playtest afterwards (i.e. sometime the following week). Lastly, I’ll leave you with this, ripped right from our brainstorm document:

Name Ideas!

  • Kings and Queens

  • (The) Great Beasts

  • (The) Great Old Beasts

  • The Magnificent Ancient Ones/Beasts

  • In the Seasons of Madness

  • Elder Beasts

  • The Season Out of Time

  • The Meatiocre Old Ones

  • Inner Gods

  • The Fantastic Aged Ones

  • The Considerable Old Ones

  • The Massive Old Beasts

  • The Colossal Old Beasts

  • Giants

  • Season’s Meetings

  • Season’s Meatings

  • Season’s Beatings

  • Seasons Fleeting

  • Mega Beest

  • The Unfriendly Giants

  • New Beasts on the Block

  • (The) Stupendous Beasts

  • Season’s Beasts

  • Year of the Mad Scientist

  • A Science for all Seasons

  • Plight of the Gentleman Scientist

  • Disaster Season / Seasons & Disasters

  • Beast Feast

  • Beasts of the East

  • The Seasonal Royalty

  • Seasons of Discontent

  • Seasoned Science

  • Giant Season

  • Season of Giants

  • Reign of Giants  - This ONE!

Link to comment
Share on other sites

 

 

  • Kings and Queens

  • (The) Great Beasts

  • (The) Great Old Beasts

  • The Magnificent Ancient Ones/Beasts

  • In the Seasons of Madness

  • Elder Beasts

  • The Season Out of Time

  • The Meatiocre Old Ones

  • Inner Gods

  • The Fantastic Aged Ones

  • The Considerable Old Ones

  • The Massive Old Beasts

  • The Colossal Old Beasts

  • Giants

  • Season’s Meetings

  • Season’s Meatings

  • Season’s Beatings

  • Seasons Fleeting

  • Mega Beest

  • The Unfriendly Giants

  • New Beasts on the Block

  • (The) Stupendous Beasts

  • Season’s Beasts

  • Year of the Mad Scientist

  • A Science for all Seasons

  • Plight of the Gentleman Scientist

  • Disaster Season / Seasons & Disasters

  • Beast Feast

  • Beasts of the East

  • The Seasonal Royalty

  • Seasons of Discontent

  • Seasoned Science

  • Giant Season

 

This looks intresting!  :victorious:  :encouragement:

Link to comment
Share on other sites

I think after Multiplayer when things settle down you guys should 100% make another DLC with the intention of the Mad Science Tech Tier. It sounded awesome and I think I'm not the only one who would get it.

 

Edit: Also with Reign of Giants, You guys were spot on; its amazing.

Link to comment
Share on other sites

@SethR thanks for all that precious info! It's really interesting to have some insight on the developpent of this game, to see how developpers came with RoG as it is right now and how they could eventually make DST! It's appreciated that you take a lot of your time just to keep this community informed, especially considering the other responsabilities you have towards the developpement of DST! Cheers! =)

Link to comment
Share on other sites

I'm loving these dev blogs so much! Can't wait for more!

 

I'm not sure if you're taking requests, considering you probably have some sort of overall plan for these dev blogs, but it would also be nice to have something about the music of RoG from the sound team, if it's possible to squeeze it into next week's. The new tracks in RoG (and Don't Starve as a whole) are really awesome and I would just love to hear them talked about a little.

Link to comment
Share on other sites

Name Ideas!

  • Kings and Queens

  • (The) Great Beasts

  • (The) Great Old Beasts

  • The Magnificent Ancient Ones/Beasts

  • In the Seasons of Madness

  • Elder Beasts

  • The Season Out of Time

  • The Meatiocre Old Ones

  • Inner Gods

  • The Fantastic Aged Ones

  • The Considerable Old Ones

  • The Massive Old Beasts

  • The Colossal Old Beasts

  • Giants

  • Season’s Meetings

  • Season’s Meatings

  • Season’s Beatings

  • Seasons Fleeting

  • Mega Beest

  • The Unfriendly Giants

  • New Beasts on the Block

  • (The) Stupendous Beasts

  • Season’s Beasts

  • Year of the Mad Scientist

  • A Science for all Seasons

  • Plight of the Gentleman Scientist

  • Disaster Season / Seasons & Disasters

  • Beast Feast

  • Beasts of the East

  • The Seasonal Royalty

  • Seasons of Discontent

  • Seasoned Science

  • Giant Season

  • Season of Giants

  • Reign of Giants  - This ONE!

Oh wow. All the names...

Link to comment
Share on other sites

Thanks for this - loved getting an insight into the development of a game I still find totally addictive over a year later.  For me you got it just right.

 

Also, what is this 'plateau of stability' of which you speak?  I look forward to getting there some day, when I learn to stop taking unnecessary risks to get that oh-so-tempting loot.

Link to comment
Share on other sites

A random thought hit me, I know people have been debating how sleep will work in DS:T. How about set it, so that way activating a sleeping item will do an animation like the bush hat or the snurtle shell, and instead of forcing the time forward, you sit there in real time "sleeping" which causes healing and sanity gain, and I guess slightly sped up hunger loss (not sure how that makes realistic sense but it's a balancing thing)

Link to comment
Share on other sites

A random thought hit me, I know people have been debating how sleep will work in DS:T. How about set it, so that way activating a sleeping item will do an animation like the bush hat or the snurtle shell, and instead of forcing the time forward, you sit there in real time "sleeping" which causes healing and sanity gain, and I guess slightly sped up hunger loss (not sure how that makes realistic sense but it's a balancing thing)

I, as well as others already, agree with this method. It seems like the only way.  The current way certainly will be thrown out, that's for sure.

Link to comment
Share on other sites

I'm one of the people that was sad when DST was announced. I really like the idea of DS being a single player survival game at the core. With that being said, these Fireside chats are awesome. I thoroughly enjoy reading about the development of RoG and as a side affect, I'm starting to come around to DST.

 

I told my 8 yr old son (who loves this game almost as much as me) that you guys were developing a multi-player and he was all smiles, asking me a million questions that I had no idea the answer to... but when he asked me if we could play together and I told him yes, his first response was, I will destroy your base!! Lol. So now I'm getting excited about DST just so I can grief him every chance I get!!

Link to comment
Share on other sites

I told my 8 yr old son (who loves this game almost as much as me) that you guys were developing a multi-player and he was all smiles, asking me a million questions that I had no idea the answer to... but when he asked me if we could play together and I told him yes, his first response was, I will destroy your base!! Lol. So now I'm getting excited about DST just so I can grief him every chance I get!!

 

That's the most adorably cruel thing I've ever heard.

Link to comment
Share on other sites

I'm one of the people that was sad when DST was announced. I really like the idea of DS being a single player survival game at the core. With that being said, these Fireside chats are awesome. I thoroughly enjoy reading about the development of RoG and as a side affect, I'm starting to come around to DST.

 

I told my 8 yr old son (who loves this game almost as much as me) that you guys were developing a multi-player and he was all smiles, asking me a million questions that I had no idea the answer to... but when he asked me if we could play together and I told him yes, his first response was, I will destroy your base!! Lol. So now I'm getting excited about DST just so I can grief him every chance I get!!

Now that's a healthy father son relationship!

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Please be aware that the content of this thread may be outdated and no longer applicable.

×
×
  • Create New...