next »

[Graham.]

Here is a follow through on some of these ideas: some design notes. To jump to the part on expression/creation search for this string:

1.2.2 expression

This stuff is more technical. My previous posts were more like diary entries. This stuff reads more like design theory.

[J-Snake]

Anyway, I think creation is similar to puzzle solving. But puzzle solving sucks because:
 * The solutions are thought up in advance. No exploration possible.
 * Sometimes the given solution is worse than what you would've imagined, like in adventure games.
 * There is often only one right solution, and it stifles creativity.

There are many bad puzzles but in general what you are saying is wrong. Narrowing down solutions enforces creativity because you need to discover a way to overcome your limitations. That's where true thinking kicks in, unlike in plain sandboxes. I consider the ultimate mastery is to construct challenging puzzles out of a sandbox with no cheap exploits in them.

[Graham.]

Yeah.

[Chromanoid]

I like your text about expression (and the other notes too).

One thing that came to my mind: By externalizing implicit expressions you make them to external expressions.

In my opinion this is often a bad idea. Maybe a player has a very aggressive play style, but doesn't like the representation of his play style in the game. By tying a specific extra outcome (an external expression) to an implicit expression you limit this expression instead of broadening it. This might be fun in some cases, but it can also lead to frustration. E.g. in Black & White (similar to Fable) you change the outlook of things by doing evil deeds, this frustrated me because I didn't want to commit evil things to make my creature look "cool" (in my eyes).

[Graham.]

Right, but you're just talking about a mismatch, and bad design.

A lot of games want to give meaning to player actions, so they create interpretations that have nothing to do with what the player is actually doing. "Having fun" can translate to "evil." "Good" may require "monotonous bull shit," etc.

The idea is that you don't judge the player, and you don't give him too much insight into himself at once. For example, if someone says something stupid, it may be prudent not to tell him right away. Maybe you should wait. This same kind of prudence translates to design.

[Chromanoid]

I think the player should always be able to anticipate the feedback (by introduction through the game or by common sense). Only in this cases the feedback can not surprise the player in a bad way. Even with anticipation the feedback will limit the freedom of the player in regard to the action (that causes the feedback). The player will want to align her actions to get the desired feedback.

Of course I don't have a problem with a game that gives external extra feedback for implicit expression, but the above mentioned is something the designer should keep in mind when implementing such sometimes "easter eggish" functions.

[Graham.]

No easter eggs.

Consider Mario. What if he had Luigi by his side, controlled by an AI. If Mario plays sloppy, Luigi reacts, showing expectation of mistakes, annoyance etc. If you don't want the player to avoid sloppiness then don't make Luigi's reactions have any mechanical value. Have them just be for show. If you do want the player to avoid sloppiness then make Luigi's reactions matter. Like have them affect how useful Luigi is as a partner, or in what way he contributes.

Games are _constantly_ providing feedback to the player. Score is feedback about implicit expression if the player doesn't know the ins and outs of the scoring system. When you complete a level in an action game and get feedback on how many headshots you got, how accurate you were, how many guys you killed relative to the other players, that is all feedback of stuff you did implicitly. I'm only talking about bringing out more interesting observations, such as those that are more personal to a player, that are more likely to be noticed by a human than a machine.

I think it's interesting you went the easter egg route. Definitely not what I'm talking about.

Here is a killer example. How many players of Mass Effect and Fable have wished for what they considered to be naturally good/evil decisions in-game - not through contrived story decisions - to have the appropriate impact? Or Fallout 3? Why isn't there a system in that game to link what I do out in the wastelands with how other people perceive me? These kinds of systems are not inherently flawed at all, not even remotely. What they are is a pool of design challenges. That's all. Nothing more.

In Fable players _do_ align their actions to get desired feedback. Mass Effect players do renegade/paragon runs. They do one run one way, then another the other way. Why? Because the relationship between in-game decision and response is broken. It is unnatural. What the player does and how the world responds is a contrived relationship. .... What if players could act naturally, the way I tell you how I feel about you to your face, and let my emotions pour through me, then expect the game world to react properly?

Creating a system that is consistent is hard, but if you do have a consistent system then players will be far less likely to manipulate their own actions to produce a certain result. Even if they wanted to they would be hard pressed to do it. Consistency is all that matters. Have the relationship between player action and world response have a consistent logic. Then players will enjoy exploring it. The system will also be a lot harder to cheat.

[Chromanoid]

Consistency is all that matters. Have the relationship between player action and world response have a consistent logic.

Yes, this is what I mean with the ability to anticipate the feedback.

What if players could act naturally, the way I tell you how I feel about you to your face, and let my emotions pour through me, then expect the game world to react properly?

I know what you mean, I just say it's not hard but very hard to do right.

Also: while additional feedback might add something to the game it can also remove something - a degree of freedom and creativity that is born out of no extra feedback at all.

Imagine you want to draw a picture with a pencil, you have one that always paints black and one that changes the color according to your mood. While the second one can draw colorful pictures it also limits you in your expression in many situations (especially when everybody knows the meaning of the color).

A more real example would be the comparison of face-to-face communication and communication through a chat room. The 1st one gives you better feedback the second one allows you to communicate without caring for your instinctive body language etc.

For example I think Minecraft is so successful, because there is nearly no in-game (extra) feedback for your actions in general. This game is an empty canvas.

[Graham.]

Incorrect. Minecraft is a world of external feedback of implicit expression. You're getting very attached to a particular idea here.

When I create something in Minecraft I am telling a story about how I think through the world. When I am done I have a reflection of my journey. That is a natural design. The journey is reflected by the product.

In Mario this does not happen. When you and I finish a level the only differences are our lives, power-ups, coin counts, and whether we have Yoshi. That is semi reflective of our experiences.

Your coloring example is balls... because that would be an awful design. Imagine I made a swimming pool inside your bedroom. You'd have to swim to get out of bed. How annoying is that? Swimming pools are bad! This is your argument.

A better example is me drawing a picture - oh wait, _drawing_ is implicitly a task that reflects back the process, hmm....

Think more like this. Make a game more like drawing. Highscores are feedback. Are those bad? Replays are feedback. Are those bad? Every single thing people post on youtube from gaming sessions are attempts at communicating what was going on in the player's head at the time of playing. When you take that information and feed it back to the player directly you win. When you get an award for being a cowardly worm in Worms you love it.

Back to the drawing example. A better idea is have a tutor beside you commenting on your work. "You're getting too mundane here, and too aggressive here. Take more care here. I like this idea. It shows inspiration and courage."

Don't we all want AI partners in-game commenting on how we play. Isn't the point of a cooperative experience to have exactly what we are talking about, a feedback mechanism for the personal elements of your experience (relayed through your partner)?

New games are made with new ideas. ... Who would prefer a chat room to face-to-face? Is that what you are suggesting? Face-to-face is infinitely better. Real life intimacy is infinitely better than online. Proving my point.... The richer the pipeline of communication between game and player the better the potential experience. Design is always an issue, so just don't fuck it up and you're good.

[Chromanoid]

Incorrect. Minecraft is a world of external feedback of implicit expression. You're getting very attached to a particular idea here.

Do you think Minecraft would be more successful if there would be an AI (and I mean an AI that is possible today) that comments on what you build and what you do (maybe like in Bastion)? Minecraft features rich possibilities for expression but no rating system on top of that and that is one of its key features to me.

Highscores are feedback. Are those bad?

Yes, sometimes absolutely.

Every single thing people post on youtube from gaming sessions are attempts at communicating what was going on in the player's head at the time of playing. When you take that information and feed it back to the player directly you win.

Recording a game session is fundamentally different to a virtual Luigi that laughs at my poor game style.

Who would prefer a chat room to face-to-face? Is that what you are suggesting? Face-to-face is infinitely better. Real life intimacy is infinitely better than online. Proving my point.... The richer the pipeline of communication between game and player the better the potential experience. Design is always an issue, so just don't fuck it up and you're good.

Yes, sometimes I prefer chat over face-to-face. Often I prefer reading a book instead of hearing a lecture etc. You talk about a specific kind of games and a specific taste.

Don't we all want AI partners in-game commenting on how we play. Isn't the point of a cooperative experience to have exactly what we are talking about, a feedback mechanism for the personal elements of your experience (relayed through your partner)?

To me the point of coop is a shared experience. I can hardly imagine an AI that talks with me how cool that one boss was to fight. And even if there was such an AI I would not want this AI to be part of a specific game.

Can't we just agree that more feedback is not always better and that any feedback you give a player can influence her experience in a fundamental way and should be thought through?

[Graham.]

No, what I mean is that Minecraft already has feedback. I'm saying that the Minecraft world is designed in such a way so that the feedback mechanism is a part of the world. It is an excellent example of invisible feedback systems. You think and play, then your creations are a reflection of what you did. The design is so simple and elegant it is easy to miss what's going on with it.

Recording a game session and having a commenting Luigi are the same thing. I'm drawing an umbrella called "shit that gives feedback on things the player did." Both of those examples fall under that umbrella. I am showing you how wide that umbrella is.

You prefer books to lectures sometimes because they focus on a different part of your mind. Notice that a book has a much higher ratio of author effort to reader effort than a conversation does (between speaker and listener). So the comparison doesn't work.

So in a sense the book opens up a "richer pipeline" in a different way. The book increases the content the author can pump into you by giving him a lot of time to prepare what he says. Face-to-face increases the pipeline in a different way. In both cases you prefer the large pipeline.

"Shared" experience is an abstract term. For example, consider this juiciness talk, that I've seen on this forum before. The guys put a face and eyes on a breakout paddle, changing how the game feels dramatically. They put personality into the feedback system. Bonus!

Half-Life 2. My buddy loved the guys you get who act as your fire team in the latter part of the game. They shout things out, tell you to reload and stuff. That stuff is small but very effective. The dog in Secret of Evermore follows you where ever you go. He also attacks, and lies down when he dies. Small feedback that is, but very powerful. The lying down dog makes you really feel the desperation of your situation.

Bubsy tapping the screen when you take to long to interact with him; the Skies of Arcadia characters breathing hard at the end of a battle saying, "that was close;" having the cheer squad roar when you get a killer smash in Smash Bros; all of these things are small reflections on what the player is doing. They don't change the mechanics. They just make a statement about what the player is going through mentally, and the effect is enormous.

I think we always agreed. There's no argument here. This is about language I think. But for the purpose of discussion I'll dig wherever. Yeah, feedback should be thought through carefully. You have to be very careful with what you say. Think of the player as a person on the edge of a cliff, suicidal. What do you say? You choose your words carefully. That's how it has to be.

I think more feedback is always better though. You always want more feedback from your friends, parents, g/f, teachers etc. Games are about feedback. They are movies with feedback. Why would you not want to increase that? You just have to do it smartly. That's the key idea: smartly.

Sometimes you need to design more game to give more feedback. You'll have to create new contexts to support the feedback. But more feedback is always a goal.

(phew. nice convo)