Design challenge: finding fun in a realistic simulation

From the title, I bet you think this is about High Frontier. But it's not.

I was thinking this morning about orbital rendezvous, which is one of the hardest things to do in KSP. Frankly, I rather suck at it. But my older son Kevin is quite good at it, proving that this is a skill that can be learned. I suck at it mainly because I don't play KSP that much, and when I do, that's only one tiny and rare part of the time spent in the game, for a grand total of almost zero practice (and thus when I do need it, I usually fail).

So, I thought, why not make a game out of just that? Each level would start you out in a small craft with a certain position and velocity relative to the target (larger craft or space station). You have to pilot your craft to match position and velocity (a six-dimensional state!) with the target. This is much harder than it sounds, because maneuvers in orbit are rather unintuitive at first, but we would start out with easy scenarios (e.g., locked to a 2D plane), and progressively advance to harder ones (full 3D matching, limited fuel or control authority, etc.).

​

OK, so there's a challenge that is easy to scale, educational, and has some intrinsic value (play me, and the skills will transfer to your future job as an astronaut, or at least to your KSP game!). But difficult != fun, and I've been watching/reading a lot of Nintendo game design stuff lately, which leads me to:

How do you add fun to a challenge like this?

Nintendo's answer would probably be not to do it at all; their games tend to be more whimsical than realistic. But let's not give up so easily. What if it were Mario in that little craft? Would there be coins in orbit with you that you can collect? Power-ups that give you special abilities? Enemies of some sort? And does all that actually add to the fun, or is it mere scope creep?

 

Reminds me of Lunar Lander style games. I found an excellent one on PS3 that I used to play a lot. You had to rescue astronauts and get fuel in that one.

EDIT: Here is the game. Not exactly what you are describing but the mechanics seem similar so maybe you can get some ideas/inspiration from it.

It is strangely addicting. Haven't played it in a while but I used to.

 

The first thing I would think about isn't mechanics, it's narrative. Who are you, why are you doing this? I haven't played KSP so I'm not really familiar with the sequence. But here's where the kind of idea I would start looking at:

You're a "Space Trucker", you might wear overalls and be covered in coffee stains, but you're the best space trucker in the business. You're called on to deliver cargo in the most difficult of rendezvous. "When nobody else can dock the craft, call on Space Trucker"

The different scenarios you'd encounter would follow this narrative, and the mechanics would follow the narrative needs of each scenario.

A "Solar Pepsi" station ran out of fuel in an asteroid field.
A "Moms Robots" ship had an engine failure and is in an orbital spin into the sun, you dock and tugboat them to safety.
A "Robo Miner" went haywire and might try to mine you while you dock with it to replace its AI Megatransconductor.

 

@GarBenjamin, you're right, the analogy with Lunar Lander type games occurred to me too.

@frosted, those are some great ideas with regards to narrative. And narrative certainly can deepen the fun. But only so much, I think... Super Mario Bros 3 would have been nearly as fun without the backstory about Princess Peach (which I only vaguely recall), but the best story in the world wouldn't have made the game interesting if the mechanics had sucked.

But yeah, putting a bit of story to it is certainly a good idea!

 

I guess the point is that - the best mechanics are also narrative tools. The asteroids aren't just things to avoid, they also tell a story about why you need to be there, and they help to add a feeling of 'immediacy' to the situation.

Coins might break immersion (or mix metaphor too much), but maybe escape pods would add to immersion. Perhaps you also need to approach escape pods in a very specific way.

Maybe instead of escape pods, there is floating cargo or debris or something. Maybe both.

 

Well, that's the thing. You're trying to make it make sense. That's what I've usually done too.

But if you look at some of the greatest (as in "most fun") games, they don't make any sense at all. Mario eats a mushroom and gets huge? But other mushrooms have eyes and feet and kill him? And that's a turtle... except with little wings, so it can jump and fly a bit? And why exactly do some blocks ejaculate coins when you hit them?

That was all from Mario obviously, but to take a more modern example, let's see, how about Angry Birds. You're flinging birds (which apparently can't fly on their own) out of slingshots? To kill pigs that are waiting for you among towers of blocks?!?

None of it makes any sense at all, and that's OK. When I design (most) games with my boys, rule #1 is: it doesn't have to make sense, it only has to be fun.

The exception, of course, is games where being a realistic simulation is the whole point (like my High Frontier game). In that case, it does need to make sense. And sure, that would be the easy way to go with an orbital rendezvous game too.

But for the sake of this thread, I'm explicitly declaring that sense is out the window. The physics can be realistic, while the rest of it is sheer nonsense (as long as it's fun)! The point is to unshackle our creativity.

WWMD (what would Miyamoto do)?

 

Level based design and space have not had the best chemistry. I have a feeling it would probably be too restrictive. It would be too easy to make each level have only one right answer, that only adds one degree of complexity off of the previous levels, and ends up being educational, not playful. KSP being the opposite, where in your play, you are likely to turn to other sources (like Scott Manley) for learning. Orbital rendezvous is pretty easy (not including docking) once you know how to use the instruments to make it work, but since there isn't any real instructions on it (since it's something you just decided to do one day), it becomes a lot harder.

I don't know, space and fun just don't mix easily.

 

You could make that it all start working, almost everything automated, but in a very basic stage.

For example, you're from a barbarian alien race that barely know how to roll a wheel, and find a basic small spacecraft from an advanced alien race.

You pilot it, everything is automated, and as a good barbarian, you go full out space pirate, to loot passing by more advanced space crafts. (The advanced alien race could be super evil to make our barbarian more empathetic.) Each space craft you loot, you get a component to substitute in your ship, but with one caveat: you have to make it work, and it's controls are not automated like the simple and weak one you had, so it's like bite-size manual snippets each time, as in more difficulty brings more power.

 

The fun exists in a simulation of understanding the underlying principals (at a basic level at the least) and then encountering something unexpected, compensating for it, and either succeeding and feeling good about yourself or failing and watching the spectacle of your own failings.

For KSP it is that moment of "oh dear, I'm coming in too fast, need to adjust.... maybe... maybe... *CRASH*" and as your play you learn how to avoid that last part. Since you can change (via procedural generation) the specifics of the scenario each time, it is always a challenge to overcome the forces that act against you (mostly yourself).

However, if you take a simulation you do NOT understand the basic level of, this entire process completely falls apart. For example, movement through 4 dimensional space. What's it like? I don't know, neither do you. All we know is what shape the shadow of a 4D cube casts, beyond that we know nothing. We can even use more real world models:

A simulation on fixing your microwave. Turns out it isn't fun because we lack the core understanding of how to do it in the first place Flying a space ship, however, that's just four forces at work! (Lift, Thurst, Drag, and weight/gravity) - A simple formula with a wide variety of possible outcomes.

In theory if you can find anything we do in life that has a simple set of rules with a wide variety of outcomes you should be able to make a fun sim out of it

 

You are confusing the fun in a simulation with the fun in a fast paced action game. It's two different player bases, playing for different reasons.

The fun in heavy simulations comes with the complexity. Games like KSP and the Civ series have this in spades. This is the 'series of interesting decisions' player. This player will happily spend hours searching tables of numbers, reading the games wiki, and actively looking for real world techniques to use in the simulation.

Simulation games often don't have or need an explicit narrative, the narrative is driven by the player. The player is naturally invested in the success or failure of a docking manuever because they have spent weeks playing the game getting to the point where docking is a possibility. Jed has no particular traits that make him special. Yet I've reset hours of gameplay just to save the life of the pilot that flew so many early missions for me.

These sim games are also slow. In both KSP and Civ, players have effectively infinite time between decisions. The games don't reward fast gameplay. They reward the ability to process large amounts of complex data and mechanics. No matter how long analysis takes.

Contrast this with the fun found in a Nintendo action game.

TL;DR - Accurate simulation games are fun. For a totally different audience and reasons then action games.

 

My take:

Using the Mario arcade/fun/cartoon/unrealistic thing you could have the player be a space scavenger/treasure hunter that needs to dock with objects to see what they find. Some may be parts they can store for later upgrade to their ship, some may be partially damaged that need funds to repair, & all can be sold for $ to use for upgrades or aesthetic changes like colour, flags etc. The game itself is unrealistic, & looks it as well, but the actual docking mechanics would be fairly realistic without the player knowing it - they'd just think it was the increasing difficulty.
For grading it it would start simple & unrealistic for physics but they would find items to help upgrade craft & could sell salvage to buy extra bits. The shop would describe what each bit does in a way they'd understand (you travel faster but need to time landing/match speeds better to reduce damage as there is a higher risk of coming in fast & hitting with considerable force. Benefit is you can match speed with faster objects that you wouldn't have been able to salvage earlier ..... Etc) so the player would be knowingly choosing what aspects they wish to make difficult first until eventually they have all upgrades & are playing the most (physics) realistic version of the game.

 

I haven't read the other posts so far as I'm in a bit of a rush. I think a big factor is whether or not you would be able to pace out the dificulty curve or if it is a brickwall. I'll talk about ARMA, because that is something I know more about than most other simulation genres. At it's peak it is a rather complex combined arms multiplayer combat sim with both infantry and vehicles and realistic ballistics on top of that. That's a hard game to get into because it's easy to walk 2 miles and then just get sniped before you saw a single enemy. But the varied aspects can be broken down and learned in isolation. You can do singleplayer training missions to practice at a target range, learn how bullet drop over distance works, then learn how to lead a moving target at distance etc.. You can learn to drive without being shot at, learn to use AT weapons etc.. In the end you can approach the multiplayer experience and learn the tactics with an established familiarity to many sub systems that makes the whole experience less overwelming than for example Dayz standalone. That game is also rather punishing towards noobs, but does not provide the same isolated learning experiences to get used to the weapons etc..
Summed up I think it is important to have the complexity in a way that can be learned piece by piece instead of "this is a realistic space flight sim, unless you read a book on rocket physics you have no chance of even getting into orbit without blowing yourself up". It also helps if the skills you learn are somewhat grounded in reality and are not unique to the game. For example I see a point in learning to navigate using the stars to do it in Dayz if you don't have a compass found yet, but I see no point in learning the obscure knowledge you need to be really good at DOTA.

 

Thanks everyone, you've all raised some really interesting points!

Rather than point-by-point responses, here are some general thoughts in response.

I'm thinking of this game as neither an action game nor a simulation game — more like a puzzle game. Think of Cut the Rope, perhaps, which has realistic(ish) physics, but each level is independent, and takes you maybe a minute or two to figure out. Orbital docking is a bit like that: within a minute or two, either you've got it or you've hopelessly stuffed it up and need to start over.

Like most puzzle games, this would be something you do on your phone or tablet while waiting in line at the DMV or riding the bus or whatever.

But I'm still intrigued by the notion of making it more whimsical and fun than a straight-up spacecraft simulation. I think that's doable (and there have been some good ideas in this thread as to how).

 

How about a hardware "hacking" game, where real life electronics rules apply and you have to get through a series of puzzle challenges with some cool heist backstory.

 

I have no idea if it is realistic in any way whatsoever, and it is a mobile game, but F-Sim Space Shuttle seems to be all about simply landing one of the most un-aerodynamic objects to ever leave the ground, and has hundreds of thousands of downloads at a price of $5.99. Maybe some pointers to be found there.

Every time I've tried to figure out how to make a game fun, it has come back to the same basic principles: clear task, responsive controls and strong feedback. The shape/form that these principles take doesn't really matter IMO.

 

Years ago, the Navy asked me to help improve their sailors. They needed recruits to learn to navigate a ship, talk on a phone, and open/close doors. NOTHING is as fundamentally boring as opening/closing doors. And yet, the game was so effective, that it's now standard curriculum for 40k recruits a year. Fast forward to today, and I've just finished the Safe Surgery Trainer, where our only goal is to teach communication.

Here's a few things to consider for learning games:

1. A game is: Goals + Mechanics + Feedback.
2. All gameplay should focus on the Learning Objectives
3. Put the learning, in the mechanics.
4. Story is optional

@JoeStrout - Where I see you struggling is that the docking process is quite long. Which means, you have 1 core objective, with a lot of trial/error/frustration along the way, with minimal real feedback. The reason people don't like space flying games is that the feedback seems disconnected from what we're doing. It's slow, complicated, and unintuitive - it feels either too hard, or too boring (i.e. flow).

So, borrow from AAA. In GTA5, the missions occur all over the city, with huge distances between points of interest. Rather than provide teleports to skip between these locations, the designers make DRIVING BETWEEN MISSIONS fun. Every intersection is cleverly designed to have a very minor challenge. Sometimes heavy traffic, sometimes light. Some green, some red. Sometimes cars stop unexpectedly, sometimes they accelerate. Each and every intersection, is a minor challenge to be overcome. The feedback+goals+mechanics work extremely well together - millisecond, to millisecond!

Steal from GTA5. Create a compelling goal+mechanic+feedback loop, at a micro level - second to second. It's what I did in the Navy's Damage Control Trainer. It's what I did in the Safe Surgery Trainer. It's why I've won national awards. And that is a quick summary of what I teach in my workshops for educational gaming. You may also enjoy this article.

Gigi

PS - Loved your ted talk!

 

Hey Gigi, thanks very much for that thoughtful response! Yes, I think you've hit the nail right on the head.

Off the top of my head, this seems to reinforce my instinct to add pickups. They give you intermediate challenges and quicker rewards. They also turn the pass/fail nature of docking into something much more graded: "I didn't dock, but I did get 80% of the pickups, which is my best yet on this level!"

But there's something lacking. Your state in space is a 6-dimensional vector — 3D position and velocity. (Plus three more if you consider rotation, but let's ignore that for now.) You could easily grab a pickup by getting your position in the right place, while having a completely incorrect velocity that makes it impossible to accomplish anything else. I think this is why beginning space flyers feel that feedback is disconnected from what they're doing; they aren't perceiving (perhaps because the sim doesn't make it obvious) the velocity vector.

So! If we were going to make a light and fun Nintendo-style game, that nonetheless had realistic physics, we would need to make the whole state vector obvious. And not in a numeric-readouts kind of way! Perhaps the velocity vector could appear as a graphical arrow that extends from your craft with appropriate length and direction.

Then, could you tie this to the pickups in a useful way? Perhaps the pickups have not only position, but also a target velocity that you must match (within some slop) in order to pick it up? That could certainly lead the user through the "approved" approach to docking, but when I imagine playing it, I hate it. It seems far too constraining — I want the freedom to come up with some crazy path of my own to collect the pickups on the way to the docking.

OK, maybe not pickups, but maybe there's some other way to provide intuitive, graphical feedback about where you are (and how this is changing) relative to the goal. A big translucent arrow or something. I'll give it some thought.

What's really interesting to me is that these design principles — clear goals, intuitive mechanics, clear feedback — are exactly the same principles used to design a good UI for anything. In pursuing a game like this, we might accidentally design a better UI for docking that might be used on real spaceships someday (for those cases where the ships don't just dock themselves for some reason!). It's a neat thought.

 

How about adding some sort of 3D path UI that overlays the game reality, sort of like the blue curves on the road in Gran Turismo for learning cornering, along with a ghost ship that you have to match path and rotation with:



Add some sound (voice of mission control?) to give you feedback and compliments if you are on track. I think mission control voices are a great thing to leverage (which interestingly seems to be the case in the game I mentioned above), since it makes the player feel as if the eyes of the world are on them and pushes them to try to get positive feedback.

 

Could you drop the pickups? For learning games, I recommend sticking to the learning goal - make the ACTUAL BEHAVIOR FUN! Here's a quote that's still on my white board:

That was a guiding challenge between iteration 1 and iteration 2 (4 months). We refined the core conversation mechanics to be pithy, light, and fast. Then, we added a mini-game - a emotional change, that went straight to our objectives, from another perspective. That's how we got a surgery conversation game that is fun enough to get a 32% improvement in knowledge, and a 300% improvement in behavior, in a 30 minute experience.

Trust in the fundamentals

• Flow - 1) Clear Goals, 2) Immediate Feedback, 3) No Distractions, 4) Balanced Difficulty
• Simplicity - 1) Core, 2) Limited Choice, 3) Intuitive, 4) Player's Perspective
• Games - 1) Goals, 2) Mechanics, 3) Feedback

Keep digging into your core objectives until you find the overlap of flow, simplicity, and games. Flying a space ship SHOULD be fun.

Gigi

 

@Billy4184 - Good thinking. I'm imagining that they are shooting for little mini-goals. Then, depending on how close they got, they get a intermediate score. And immediately, the next objective pops up, with auto-correction, based on how far off they were. So, they have little mini-objectives, at every step. This also fits nicely into the emotional interest curve (see episode5).



Gigi

 

If you did want pick ups you could possibly combine it with the path suggestion by making the pickups on planes so they are only seen properly when you are in the correct position. Come in from behind & they are invisible, wrong angle & they are fuzzy...............
You could try to calculate the angle they hit them at & award the pickup bonus as a % of the correct angle/rotation then provide the stats like you got 80% of the pickups available but achieved 75% of the pickup value for those items, & only got 60% of the value for everything that was available overall. This would be more complex for you to calculate but provides finer control over the rewards mechanism you use. Players can try for every pickup but might then want to keep refining to try & increase the value they get from all those items.

 

Agreed about dropping pickups, IMO it just doesn't fit here and muddles the task at hand. But I think continuous feedback is probably better than mini-goals here... or maybe make positive feedback as a regular mini-goal.

Overall I think it would be best to leverage the drama and excitement that is unique to this specific situation, which means ... cool, calm mission control voices.

 

Pickup use will depend on the target. Younger players not understanding the realism being shown in the behaviour are used to/expect power ups so they could be a way to subtly introduce the higher level concepts so they learn without realising they are learning. Without those things that kids associate with fun they will likely not play for long. If the target is for harder core space enthusiasts then yes, power ups may not be the most appropriate but neither will gamefying it to much as it will detract from the 'realism'.