Game Design Zen #1 lessons for High Frontier

I've just (finally!) gotten around to listening to @Gigiwoo's first Game Design Zen podcast: How much is enough? The Paradox of Choice for Game Design. It raises some really good issues; things I knew but hadn't thought about in a while. It's made me realize I need to make some changes to High Frontier, and I thought I'd share here, in case the lessons are useful for anyone else.

First, when designing a colony, there are currently a lot of parts to choose from, and a lot of ways to configure them. A typical colony only consists of one "main" part, and there are limited choices for those (sphere, torus, barbell, cylinder), so that's OK; but there are lots of add-ons and accessories, and of course a new player probably perceives it all as even more complicated than it really is. Right now everything is in sandbox mode, which means all parts are available from the beginning; but we should get "career" mode in there ASAP, and when we do, aggressively limit the parts available in the beginning, unlocking new parts gradually as the game unfolds.

When a new part is added to your design, we do have reasonably a reasonably sensible configuration in most ways... except spin. We start parts with no spin because the correct spin depends on the size, and also because I had some vague notion that discovering the spin and how that "fixes" an uninhabitable design would be rewarding. I now think that is probably a mistake; parts should be set to spin at 1G by default (though if you change the radius, you will have to change the spin too). I've already heard from a number of players who failed to discover this knob, and couldn't understand why their colony was not livable. So that should help with that.

Then you get to the build screen, which currently looks like this:

That's six choices. Too many, according to the podcast. But wait, it gets worse: we're currently working on expanding this to orbits throughout the solar system. We don't have a complete list yet, but it's going to be dozens of choices.

So, what to do? In career mode, we will certainly lock (or maybe not even reveal) all of those orbits at once. Your first colony should probably be limited to LEO or HEO, which is an interesting choice in itself. Then we unlock the lunar orbits... but maybe we could combine L4 and L5, which are functionally equivalent. And then we open up the inner solar system (just a few more choices around Mercury, Venus, and near-Earth asteroids), and then the middle solar system, and finally the outer planets and Kuiper belt. And, of course, we should make sure that a sensible default is selected as soon as you enter this screen.

Finally, once you get into city management mode, the player is faced with a SimCity-style task with lots of power to shape the land, zone for this or that, and buy and place municipal buildings. We can (and should, I now see) limit the choices somewhat by unlocking municipal buildings, and perhaps some of the other tools, when certain criteria (e.g. population size) are met. There is also much we could still do to make it clearer what is needed.

I wonder if we should go so far as to pop up "demand" dialogs that give the player specific tasks? For example, as soon as you enter the empty habitat, a box might pop up that says:

RESIDENTIAL ZONES NEEDED
Developers are demanding to know
where to build new residential buildings!
Use the Zone tools to paint
some blue residential areas.​

We already have a "Squawker" feed that provides some feedback of this sort, but this would be more persistent, popping right up somewhere it can be easily seen, and then hanging out somewhere to provide specific task-related feedback (a progress bar as you work towards the goal, and happiness-inducing audiovisuals when you complete it).

I guess that's getting into the design space of "flow" and providing clear tasks and feedback. But it seems to me that it relates directly to decisions, too. My simulation is suffering from overwhelming decisions, and this would simplify it considerably: you just have to decide, am I going to pursue this task, or ignore it? And once you decide to pursue it, you're just looking for: what's the best way to do that? In the example above, my decision space has gone from "what do I do with all these tools in this giant empty plain" to "where shall I put my first residential zone?"

I highly recommend the podcast if you haven't heard it already, and if anybody has any thoughts or comments on the above, I'd be thankful to hear it!

 

Well, DAANNNGGGG @JoeStrout ! I couldn't of said it better . Your discussion totally rocked - thanks for sharing!

Gigi

 

If you implemented career mode where things were progressively unlocked so it was also one long tutorial at the same time you could possible leave the 'hardcore, everything is unlocked & sandpit' for the hard core players or those that have finished the career mode so they can experiment with trying things in different orders. This lets them find out if there is a lean & dirty way to get to a situation to then start their expansion or if there is a need to do some things in certain orders.

Limiting choices when starting & learning the game is ideal for nearly all players but with simulations I've noticed (from friends & forums) that there is always that dedicated bunch of risk takers/number crunchers that really like to explore all options/combinations to push the envelope & find new gameplay after they have grasped how it all works.

 

Yes, I think you're right. We'll keep sandbox mode as an option, and when you start a new game, that'll be your first choice (sandbox vs. career). But we'll set it up so "career" is selected by default, and it requires an extra click to switch to sandbox mode!

 

One thing you might want to think on is how many choices are serious versus how many are cosmetic, and how far apart these are from each other. More than just saying there are too many choices, I would say there are too many choices that compound on each other along a chain. Before you even get to the point where you decide what should be in a module, you've already defined a ton of properties that could easily have no purpose, or even be unwanted.

Some of this might just be workflow. Starting with module roles, then moving to properties, would be cleaner because you can then suggest values and present metrics like how many people would live in an area or production values for a factory, essentially helping people make more informed decisions.

 

KSP and Civ are both good examples of how to slowly introduce high complexity. Civ does it better, I'm still referring to google a lot for KSP.

Edit: But I do wonder what civ would look like if it was released today in a post google world.

 

How could this be further refined? There's now 2 simple options, with a nice default. It's a good start. Next, we look at what people interpret with these two options. "Sandbox" vs "Career" - is obtuse. If you kept those names, I would have 'Career' be disabled, until it is unlocked in game. That gives people something to work toward, plus it allows the illusion of choice, while also triggering the zeigarnik effect - so they want to find out what that means later! Plus, if 99% of your players should chose the 'default' option, then take it one step further.

Love your progression!

Gigi

 

What does Civilization do wrong? The obvious mistake Civ makes is that many of hte versions let you pick any civ you want. This only works because Civ is a long-running franchise, whose players generally know what they're getting. For newer IP, this would not work. That's why in Destiny, they start with 3 options - one's more brute like, one's slightly more range oriented, and one sounds like a mage. All of them sound good.

Gigi

 

I really like this. I also really like the tone you use in the description "Developers are demanding to know" the way you word this also helps to solidify the player in his role and helps to provide material to the players imagination.

I'm not sure how much you're using this kind of tone, but I think that it's especially important in a more abstract game like a space building game, to have these kinds of reminders that there are little imaginary people involved. I imagine most of your actions directly impact them and they react passively, on the flip side, in the way you present this choice, you're also making them more active. The idea that they're demanding something gives the impression of agency in the little simulated population. That there are hints at the population also being active can help make the whole thing really feel alive.

That they help offer the player goals, highlights decision points, and helps with the narrative - all in one shot, means that this is likely a winning design choice.

 

I assume you meant to say "Sandbox" would be disabled, until it's unlocked in Career mode.

And yeah, this is something I'm considering too. Certainly Sandbox would not be recommended for new players, so I would at least label it as such. But as you said, why not take it one step further, and just disable it until some milestone is reached in career mode?

There are two reasons why I might not want to do that:

1. The user might be an experienced player who has simply installed the game on a new machine, and doesn't care about his old game(s) enough to dig up and move the data files. He just wants to get back into sandbox mode and continue creating crazy cities, without having to work his way back up through the ranks.
2. The user might not be a gamer at all, but might be a space enthusiast (or even professional) who wants to use my game more as a space station CAD program than as a game. Yes, I have users like that.

However, I'm thinking about doing it anyway, but with a back door: in the Options screen, there would be a "Sandbox Mode Locked" option tucked away somewhere, checked by default until you complete the game in career mode. But you can go in and uncheck this any time you want. User 1 above will know about this, because they've been around long enough to have poked through the options. User 2 may not, but will either find it upon a bit of poking around, or will learn about it in some "How to use High Frontier as a CAD program" document that circulates in the space community.

And yet I'm still on the fence, because now it seems like a very arbitrary hoop I'm making the player jump through. I'm worried a player will find it and say, "what, there's a sandbox mode, but to use it I gotta go to the options screen, unlock it, then go back to the main menu? That's stupid!" And I prefer not to have that exclamation directed at my software whenever I can avoid it.

 

I think the convention in these kinds of games is to have 3 game start options:
- Campaign
- Scenario
- Sandbox

Scenario is probably the least popular, but it allows for really specific highlights of certain mechanics or situations that might not often naturally arise and are fun or interesting. Personally though, in games like this, I always just play sandbox.

EDIT:
I just sort of thought through some of my experiences with building/sim games. Especially when considering a limited time budget and resources, I think that focusing on a scenario system, entirely in leu of a campaign/career mode might make a lot of sense. Career mode would be cool, but it may also be a lot of additional work to implement well.

Scenarios on the other hand can be used to solve many problems at once.

#1 - you can do a very simple progression where you unlock later scenarios.
#2 - you can very deliberately limit choices or highlight mechanics / tutorial points.
#3 - you can provide explicit and changing goals or explore a specific subsystem (tourism, industry, total income, etc).

I believe this is how the roller coaster tycoon series worked IIRC.

 

Good point. And I do like "Campaign" better than "Career" now that you mention it.

I don't know if Scenario is worth the trouble, but I could see putting Tutorial there. This makes it really easy to go through the tutorial again whenever you want to, without having to go through it every time you start a new game. The tutorial would be just a mini-game with a very clearly defined (and quickly reachable) goal, and some extra UI to give the player very specific sub-goals and extra help in reaching them.

Of course many players immediately skip right past the tutorial... and that's OK. It should be easy enough to pick up in campaign mode too.

Sandbox, of course, is overwhelming to a new player, as I've already found. @frosted, do you think Sandbox mode should be hidden or locked until the player completes campaign mode? And if so, how do we provide an end-run around this for users who have a good reason to skip the campaign?

 

I actually just edited in some thoughts into the post above.

Originally when I thought of scenario, I was thinking of SimCity style scenarios "there's a hurricane that destroys your city can you rebuild?" - but I think the roller coaster tycoon model of scenario may really be a good choice.

http://rct.wikia.com/wiki/The_Complete_Scenario_List

These can be unlocked to provide a meta progression and can act as both a tutorial system and help to highlight different kinds of tradeoffs.

I am not in favor of locking sandbox. Let's be honest, this game is not targeting more lightweight gamers. Your target is a cross section of space enthusiasts (often long time gamers) and building gamers (a pretty hardcore group). If these people want to sandbox, let them sandbox.

 

Well, I'm not willing at this point to forgo campaign mode, because it's going to be epic (and epically fun! — and shut up, spellchecker, that is so a word!). You're going to start out building tiny little habitats in low-Earth orbit, and then build your way up to giant cities around Saturn, Neptune, and out in the Kuiper belt.

A campaign mode is preferable to a series of scenarios because (1) you'll have a lot of freedom by that point to choose where you go and what you do, and (2) how the development of the solar system unfolds will be impacted largely by those choices. A series of scenarios is too restrictive. I want people to look back at the solar system they have developed over the course of a year or two, and feel the pride of having created something unique, not like anybody else's.

In fact now I'm having second thoughts (sigh) about the term "campaign," because from my experience, that is often used to describe a linear series of scenarios. And that's not what we've got going on here at all. I don't know if my experience is representative, though; I need to research the standard uses of "career" vs. "campaign" more.

But that Roller Coaster Tycoon model is worth consideration, and that scenario list is really helpful. I will look it over and think about how its lessons might apply to High Frontier.

 

Just keep in mind that even though your demographic is pretty hard core, only a small fraction of your players in a small percentage of games will get to this point. For most players, having strong limits like @Gigiwoo suggests will probably give them a better time, and help them learn to explore the game to its entirety.

 

May I push? I like your idea, the stories you've shared, and this discussion. And, because you're an advanced developer, I feel the urge to keep pushing. So, let's look at these 2 cases.

1) Experienced user, reinstalls. - Why would you support this case? I've fallen for this so many times, I've stopped counting. Until finally, I realized that this is amateur'ish thinking. Professionals are NOT doing this. First, it impacts the first-time user, which is a terrible, NO NO. Second, anyone who loved the game so much that they were willing to reinstall it, will probably have a dandy time working through the campaign to unlock those features again! Looking at other games, there's a reason why it's rarely done.

2) Space enthusiast - Why would this matter? Just because they love sci-fi, doesn't they'll be masters of your game. My wife doesn't start a book on chapter 6, just because she's a history buff. In short, the early experience matters JUST as much to a professional space enthusiast as it does to an average gamer. Whether they know it, or not.

/Cue the inspiration, rocky-climbs-stairs music

You are awesome! You are making great strides along your journey. You are trying, improving, and repeating! Your knowledge is incomplete, just like the rest of us. So, design the best experience you can, with what you believe now. Which means, deciding whether the paradox of choice matters, or not. Pick one.

/End music

Gigi

 

Will there be online multiplayer so players can work together or against each other to colonise space? This could alter some of the options available or at least how you structure them.
Is it possible to break up & group the choices into chunks small enough that a player can play a quick scenario from a starting base to then pick some options & in an accelerated time span see what happens, then reset to the same start & try different combinations? This may not give them a thorough tutorial like understanding but it lets them experiment without being a slow guidance type tutorial & it's quick enough that if they choose to just start playing then they can jump across to do one or two when they are stuck in that situation in the game.
Or provide a scenario play out like Frozen Synapse. In that game you make your choices & can play to see how it could play out. What you see happen is rarely exactly what happens when you say you've finished your turn but at least you had the option to see what one of the possible scenarios would look like & you can keep adjusting until you are happy with the probability. Depending on your time scale you could just have something like let them pick their options & then show them one of the possible states in 12mths & let them change some things & chck again before they choose what they want & press continue.

 

It's strictly a single-player game.

But those are some interesting ideas, thank you — I'll ponder them some more.

And @Gigiwoo, it's always fine to push! I appreciate that you care enough. In the end, as you say, I'll have to decide myself, and I can't let my self be paralyzed by the choice. But I always appreciate the input.

 

Nice podcast @Gigiwoo! Just got through #2 today on the commute to work and really enjoy it so far.

Thanks for pointing it out @JoeStrout.

Going through similar efforts to simplify the interface in my game, after my wife pointed out: "I don't want to choose how to play it, I just want to play it".

 

I love everything about this thread.