Planned vs. Practiced vs. Improvised - A theory of types of play

Made sense to me. Thoughts?

 

Brilliant! And, right on topic for our new forum. Thank you for sharing it!

(I'm going to let it sink in a bit more before offering any substantial comments, but thought it deserved some immediate thanks.)

 

Agreed. And also just want to say I am very pleased with the content and discussion so far in our new forum. Sure there were a couple minor bumps in the road but by far the majority of the discussion has been very focused on game design. In fact, I think this is quickly becoming my favorite forum. So... a big thanks to you all!

 

I think the Plan, Practice, Improvise classifications are good. I had no objections to anything covered in the video. Of course, these are very high level... kind of a guiding design philosophy for a game or certain parts/areas of a game.

 

Glad someone posted this episode. Rather insightful this time round, or at least not something I've stewed over before. Pretty sure I've never heard this classification before, which probably means it's just something James has been concocting for a while.

 

So here's how I look at it, for whatever that's worth.

Fun is fundamentally based on two things: (1) biological imperatives, and (2) learning. Fun type 1 is the fun that pretty much never gets old, but as this is a family-friendly forum, let's put that aside for now. Fun type 2, however, is based on a psychological principle that's just as deeply rooted: we enjoy learning things, and when we're not learning anymore, we get bored.

So, let's relate this back to the classifications in the video:

• Improvise: in these types of games, you're learning a skill. In pretty much any real-time game where you control an avatar (Mario, Prince of Persia, any fighting game, etc.), this is mostly a manual skill, learning to make your avatar do what you want. Levels in these games (whether randomly generated or not) are generally just tests of how well you can do that. Turn-based games completely take away the manual skill; here you're learning to make the right choices, which usually boils down to correctly predicting what will happen if you do this or that. You're improvising because, in general, you never see the exact same situation twice, so you can't just remember what will happen; you have to learn an accurate mental model.
• Practice: these games present the same situation multiple times, so you can learn exactly what will happen. This is usually combined with learning a manual skill as above. In a turn-based game, this sort of repeated level design would seem pretty dull, because you don't even have to learn a mental model of what will happen; you can just remember what happened last time, and make different choices this time. Unless, of course, the decision tree is so deep that this becomes impractical (like Chess or Go, even without any randomness in the AI opponent). In that case you're back to learning a mental model of the game and predicting outcomes.
• Plan: in these games, you're learning both a model of what will (or may) happen, as well as what you can do ahead of time to affect those outcomes in your favor. Then, as he points out, you usually have to improvise to adapt your plan to contact with the enemy. (But that's not always true: a game like RoboWar is all planning, and once you've built and programmed your robot, you are a mere spectator as the plan unfolds.)

[Editing because I bumped the Post shortcut... need to learn a better manual skill there...]

So, I think the Improvise/Practice/Plan classification is useful, but maybe not as useful as something based on what the player is learning. Is it a manual skill? A mental model? A set of useful heuristics? Or some combination of these?

 

I would be more inclined to think of this as what skills the player is reliant on at the moment. Practice is the skillset the player has built up, be it through pattern recognition or physical skill; improv is purely reacting to the current situation, without needing to pay a ton of attention; while planning is about looking ahead and preparing. In essence, it's past, present, and future.

 

I'd never seen this video! And without this new forum, I might never have. I concur with @JoeStrout, @Asvarduil, and @RockoDyne that this was a great contribution to the design forum. And, also with @GarBenjamin. The community is creating something awesome together - it's quickly becoming my favorite forum!

Gigi

 

It's a good thing you've never seen this video - it's one that came up today. If you'd somehow seen it before, either you work with the EC team, or you've got precognition.

 

After rewatching the video, and giving it some lunchtime thought, I agree on all points with the video. But, I think the most effective games utilize all three types of fun.

Think about The Elder Scrolls V: Skyrim. While the start of the game doesn't lend itself to being much more than a well-thought tutorial level (it gives introductions to basic game systems more than anything else), the meat of the game is a case study in all three coexisting. You start by improvising after escaping Helgen - you can either go with your guy to Riverwood...or you can check out those mountains over there...hey! There's an abandoned hut...oh cool, ruins up the mountain! Nothing is guiding you, and you have no information with which to come up with a plan.

Of course, what's there will change that. The tutorial introduced the player character to fighting, possibly sneaking and archery, for certain. The good news is that the enemies in the game scale to your current level. To get to anything interesting, you're going to have to rough up some bandits (or doomsday cultists, or vampires, or random hostile wildlife.) While due to the action RPG nature of the game you'll likely do some improvising, I feel that the combat shifts to a balance of planning and improvisation if something goes wrong. "Ok, I see three bandits; one has a huge axe, the other looks like a mage, and another has a one-handed sword. I'll take down the mage first, while the two melees charge me, then I'll wing it from there." That's a plan.

Whatever the first encounter is, practice is baked into the game via every skill having a levelup meter of its own, in parallel to your character's core XP meter. Not only is it necessary for the player character to practice skills that will come in handy - for instance, one-handed weapons, illusion magic, sneaking, lockpicking, whatever - it's necessary for the player holding the controller to practice too. There are cases where certain things work just a bit better (for instance, try using Destruction/Frost magic on Draugr inside the many ruins dotting Skyrim. Try again using Destruction/Fire magic.) Skyrim doesn't hold many punches, but by dying, you achieve this 'practical' fun.

That's just one game, though. There's countless other examples out there. I think the Metroidvania genre has this concept of a balance of the three types of fun baked into it, thus why it's a perennial genre.

 

I downvote all extra credits content. James is a weirdo attention addict.

 

Not all EC content is bad. Some of it is rather out there (the bit about how the Blue Shell is a catch up mechanic, I heartily disagree with, it's actually a trolling tool for the guy in last place, so that that player can have a little fun at least.)

I have to admit, he does actually sound a bit like a stoner. More than a bit. The "James Recommends" stuff doesn't ever get a watch from me, though, because it's usually just not well-done. He tries to do a stream of conciousness thing which conveys very little in the way of reasons to try whatever game he's talking about. He would be better to just pre-write a list of talking points before doing that stuff, and it'd work a lot better.

 

He is getting better at them, which I get the feeling was probably the point, so that he could get better at extemporaneous speech.

 

Exactly... he rambles like a madman, he goes on and on about how wonderful whatever game he wants to talk about is, he is just something we're going to have in the future, "rockstar" style game devs... who know everything about game design at 25 years of age, and are rarely if ever questioned. Games are the new rock n roll, everybody wants to learn guitar.

 

Extra credit is the best!!!