next »

[LuisAnton]

No matter if you are an iOS player or not, you probably know about these games... Where's my water (

), Cut the Rope (

), Greedy Spiders (http://youtu.be/qv0X-m1pbCw) and Angry creatures in general.

They have one thing in common: a usually one-screen puzzle that sometimes is solved in a matter of seconds, sometimes they take you minutes to figure out, or even more.  But it's not ten, or twenty puzzles, it's hundreds of them! (or so it seems)

We recently released an action title with just 16 levels and we spent a few days balancing gameplay. How do designers come up with so many levels for these games?

I could imagine a procedural puzzle generator based on graph theory for Greedy Spiders, and even a random structure creator for  Angry Birds... but I fear they really require an army of people just devising new content, is that so?

Sometimes I feel it shouldn't be that complex. You have some building blocks (whatever they are) and you place them in a certain way so a goal is achieved. Then you remove some pieces and show it to the player, who must recreate your puzzle... But I'm sure that best puzzles aren't created simply by removing pieces...

Have you created something similar? How did you handled the puzzle creation issue?

[Bishop]

I think when you look at the difficulty curve, those games it's incredibly slow increase across the levels. At some point it'll introduce a new element and start again at the bottom, later mixing elements together. I do agree though, hundreds in pretty impressive.

[1982]

You must be intelligent to design puzzles, I am not, so I never design puzzles.

[st33d]

Just released a physics puzzler today.

It's really just a matter of stacking things up and having them topple over in the right order / direction.

Then you introduce elements that complicate that premise.

[rek]

I'm pretty sure the Angry Birds people just throw in random blocks and, if someone in the studio can beat it even once, it gets shipped as-is. That's why you'll find an incredibly easy level next to one that will take days of trial and error to get just one star on.

[gimymblert]

Ask scott kim (and pajitnov): http://www.scottkim.com/thinkinggames/index.html

http://www.scottkim.com/thinkinggames/GDC00/index.html start here

[J-Snake]

Gold-Tip:

There is actually a method how to get creative. If you want a great variety of puzzles then your best bet is to design the rule-set with as many complementary parts as possible. That way you will have great control over level-design. Each rule in your set might represent a certain number of according possible gameplay-concepts/situations. Try to filter as many as possible out by simply playing and thinking what you are actually doing. With a concept I mean a certain requirement how to solve a certain situation/s. I also refer to it as "tools". Because you use an intellectual property (idea/concept) to solve a situation.
Perhaps this vid helps to make it clear (you can skip the first minute if you have no patience):




So in the end you are left with a large amount of that concepts. Then it is just about how and how many of them you want to mix into each level.

[s0]

You must be intelligent to design puzzles, I am not, so I never design puzzles.

i'm in the same boat. dumb people unite! 

[TaintedFork]

I think when you look at the difficulty curve, those games it's incredibly slow increase across the levels. At some point it'll introduce a new element and start again at the bottom, later mixing elements together. I do agree though, hundreds in pretty impressive.

This is exactly right.

If you play these, you'll find 2-3 levels teach the exact same thing. "Pure" or "true" or whatever puzzle games tend not to do this. They introduce something, teach it, and then continue to expand upon it, level-by-level.

From what I can discern, the games you mention teach it, use it a few times, expand upon it, use it a few times, and continue.

These games are more repetitive and perhaps monotonous, so they contain more levels.

I have no basis for this statement (accusation?), but I think that the people behind these games assume that their players are going to be dim-witted and require more hand-holding.

Honestly, when you ask me, I prefer quality over quantity for puzzle games. I think I find that fewer, but more well-thought puzzles will take longer to solve than a large collection of simple, mindless ones.

I just made a puzzle game for Ludum Dare and could have made more levels, but decided to remove them because they were too simple or didn't bring anything new to the table. I think these designers keep these levels [even if they shouldn't].

[J-Snake]

I have no basis for this statement (accusation?), but I think that the people behind these games assume that their players are going to be dim-witted and require more hand-holding.

Honestly, when you ask me, I prefer quality over quantity for puzzle games. I think I find that fewer, but more well-thought puzzles will take longer to solve than a large collection of simple, mindless ones.

I have the same philosophy like you here. However it seems that most agree that one should lead the player through progressively harder getting levels. So it happens that alot of them are easy, when you look back.

What do you think about following that route but avoid being too repetitive?

If you like a challenging puzzler you should definitely keep TrapThem in mind.

[st33d]

There's a wider audience for physics puzzle games, hence the casual entry level.

As for level quantity: Time spent in a puzzle game is extremely variable. That's why you offer a lot of levels. Some people will whip through it all in minutes and some will take ages to solve something simple.

It's not easy coming up with a large amount of interesting levels, and not everybody succeeds at it, but you have to remember that puzzles rarely have any replay value and even less story. You shouldn't just cut levels, you should make better ones to replace them.

This doesn't hold for all genres but I think that puzzle gamers in my experience have a larger appetite for levels.

[stevesan]

I haven't played those particular games, but I got the same impression with Crayon Physics and World of Goo.

In my mind, games that have tons of quantity typically sacrifice quality. Can you really say that many of those hundreds of puzzles are memorable? I would guess not. Whereas, I would say most of Braid's puzzles are quite memorable, since each one offers almost a narrative about a unique mechanic. I can recall so many of the "ah-ha!" moments in Braid. This is true to a lesser extent in Portal, but still true for many of its puzzles.

[J-Snake]

but you have to remember that puzzles rarely have any replay value

In my case they often have replay-value because you can challenge yourself to beat it in less steps. I think it is a great idea:
Look here, there is a fundamentally different idea hidden in this level that allows you to solve it much less tedious than what you see here. What you think about that?

[st33d]

Personally I thought that World of Goo was borderline perfect, and pretty much everyone else who's played it I've met has reported the same (the box stacking level aside). That had a lot of content. And I remember most of it.

Bringing up Braid is a weird one, since he reused levels several times in a fashion that felt like padding and often the puzzles involved some pixel perfect jump doing something illogical. Not really a puzzle. More like a riddle. I like puzzles, I don't like riddles.

Simply saying, "there's more, so it's going to be crap", seems a bit self-defeating really. Why not create more, and do it properly? That's essentially what I'm arguing - that doing more is a very difficult undertaking, but that doesn't mean you shouldn't try.

[J-Snake]

There will be 100 levels+ in total in TrapThem. But one level can have up to 3 different versions. So it seems like only 30 of them are unique at first. But a harder version of a level will require new ideas in order to be solved.

What do you think about that?

[moi]

I'M PUMPED UP ABOUT TRAP THEM
 
TRAP THEM TRAP THEM TRAP THEM TRAP THEM TRAP THEM TRAP THEM TRAP THEM TRAP THEM TRAP THEM TRAP THEM TRAP THEM TRAP THEM TRAP THEM TRAP THEM TRAP THEM TRAP THEM TRAP THEM TRAP THEM TRAP THEM

THEM TRAP!!!!!1111

[J-Snake]

I know, I know, son. Just you wait!

[stevesan]

Personally I thought that World of Goo was borderline perfect, and pretty much everyone else who's played it I've met has reported the same (the box stacking level aside). That had a lot of content. And I remember most of it.

Bringing up Braid is a weird one, since he reused levels several times in a fashion that felt like padding and often the puzzles involved some pixel perfect jump doing something illogical. Not really a puzzle. More like a riddle. I like puzzles, I don't like riddles.

Simply saying, "there's more, so it's going to be crap", seems a bit self-defeating really. Why not create more, and do it properly? That's essentially what I'm arguing - that doing more is a very difficult undertaking, but that doesn't mean you shouldn't try.

Wow... OK I think you and I just have very different taste in puzzles/riddles