I've been very interested in designing more interesting, satisfying puzzles lately, in the hope of turning my projects from 'interactive stories' into actual games. You know, ones that make you think and stuff, instead of just passing a few minutes.
However, my method of designing a puzzle is generally primitive at best. Usually I just look at the story, find where the character needs to be next and throw an obstacle in their path. In order to create the obstacle, I usually create an environment or character, and then think to myself how I can use this environment or character to create gameplay. Sometimes I actually think of the puzzle first and design the gameplay around it, but this is rare. Mostly it is me drawing scribbles in a battered notebook trying to figure out how to keep you guys busy for a while.
So, I thought it might be helpful to define the stages of solving a puzzle, because if I can keep in mind the stages one goes through to solve a puzzle, surely I can apply this to designing each puzzle, and therefore end up with more enjoyable gameplay. To do this, I turned to scribbling in a battered notebook, naturally, and ended up with the following steps:
Introduction: This is when the player initially realizes that a puzzle exists - for example, coming across a locked door. This is the most simple stage of puzzle completion; realizing that you are going to have to do something in order to progress further.
Observation: The next step is looking around the environment and taking note of everything contained within it, in order to find something that might help us bypass the obstacle. For our example, we could have a man standing guard nearby, who pauses to stare at something every 30 seconds.
Correlation: I like to think of this as the 'eureka' moment - that is, when the player realizes how the elements can be combined in order to bypass the obstacle. In our case, it could be when the player sees the keys on the guard's belt and realizes that if they grab them when the guard looks away, they can use them to unlock the door.
Implementation: Often the least tricky part of solving adventure game puzzles, but not always. This is the actual process of combining the observed elements - in our example, waiting for the guard to look away, taking the keys, taking the keys to the door and using the keys to unlock the door. This concludes the puzzle and the player is free to move forward.
With these four points, we can look at the various ways in which puzzles can be improved and what problems the player may encounter forcing them to be stuck on a puzzle (and how we can try and avoid this).
If a puzzle is not clearly introduced, the player is left directionless and has nothing to tie their obervation stage to. If you don't tell the player that a door is locked, they're not going to care about unlocking it, or be unsure as to why it is even there (is it just decoration?). However, if you can somehow make it require some extra investigation to indicate that a door is locked, you can extend the puzzle. Perhaps the door is a secret door, which requires certain glasses to see or some such further complication.
The observation stage is sometimes a frustrating weak point in puzzles - giving rise to the term 'pixel hunt'. If an exit is not clearly shown or an item too small to be reasonably noticed, it is going to be more frustrating than enjoyable to solve the puzzle. However, this is not the only way one can extend this stage - perhaps you have to overhear a conversation in order to understand the true purpose of something or find that a certain brick has something hidden behind it.
Correlation is my favourite part of puzzles. When you succesful interpret the clues given to you and see how they apply to the situation you're in, there's a real feeling of 'cleverness'. However, if the solution relates to obscure knowledge or is not fully logical (You all know the monkey wrench puzzle, I am sure) then the correlation is much less likely to happen. This is another common frustration - and when the solution is finally obtained, the player feels more irritated than clever. However if you can leave clues subtle yet clear enough, the player is bound to find great pleasure in putting all the elements together. Perhaps they managed to understand how the picture in the painting relates to their environment all of a sudden (Hallway of Adventures, great puzzle!). The joy of this moment of realization is a key element in encouraging your audience to follow your story.
Implementation is rarely a stumbling block in puzzle solving, but when it is, the frustration is further increased by the fact that the player knows exactly what they have to do, they just cannot do it. Perhaps you've been a bit enthusiastic with timed limits, and the player just can't walk that distance in such a short time, or your interface makes something more fiddly than it should be (think of Vince XII's comment on Flux World with the mirror inventory item). However, if you make a puzzle with a timer that does take a couple of goes to get, it will still be fun for the player to solve. Possibly even more fun...
Well, this is how far I've gotten with my ideas so far. I'm interested to hear any thoughts you might have to add to these, and hope to expand on these. Hopefully by focusing on each of these stages when designing a puzzle,puzzles can be made more interesting to solve.
6 comments:
The amount of thought that you have put into just thinking about puzzles has given me a great insight into why the puzzles in Shifter's Box are so devious.
You have put more effort into decoding the field of science that is "puzzle-ology" than I ever put into any of my actual puzzles. (Which explains why my puzzles are so tragically lame).
Well done, Mr. 304!
Interesting read, Ben. Cheers.
To add to that, I think games can get boring if the puzzles are just a series of obstacles, one after the other.
Take your locked door for instance. So the guard looks away every 30 secs, giving you window of opportunity to grab the keys. Great. You could solve that puzzle, open the door and move onto the next puzzle in line OR you could throw another spanner in the works.
What if the guard still catches you, even when he's looking away? How do you get him to look away for longer... perhaps move the object he is looking at, so that he has to turn his head further, giving you that split second extra to grab the keys.
Then what if you realise his keys are on a chain... you need something to cut them with (which could be a case of finding a pair of wire cutters lying around, but wouldn't it be nicer if getting hold of some wire cutters was more of a puzzle aswell).
These are all single puzzles in their own right, but they all basically blend together into one main puzzle, which is the locked door, which in itself could be part of a larger task.
I steal a lot of my game design theory from Valve, and one of the thing's they're very big on is training the player - i.e. they introduce the player to a simple concept in a safe environment, then they use the same concept while the player's under fire, then later they might add a twist to it. You can introducing basic concepts in a simple puzzle before you use them in a more complicated one. This way you can make difficult puzzles that will still be solvable and make logical sense in the game world.
Example: in Toonstruck there was a puzzle where you had to break out of a jail cell. There was a dusty welcome mat in front of the cell, and the key was under it, but there was a guard in the way. Talking to him reveals he has allergies to lots of things but especially dust. Take the mat and bang it against the bars to make him sneeze himself into the wall and knock himself unconscious. Later, if you get caught again, he's back guarding your cell, but this time he's wearing a gas mask. Get caught again and he's hoovering up the dust. Each time the basic solution is the same, make him sneeze on the dust, but the method is different.
It's also a good way of building up a world. In Myst the game teaches players that they must use the tower to get to different ages. The actual way of getting to each age differs, but using the common starting point of 'go to the tower to see where the linking book is, and get a clue of how to open it' means the player has something to focus on.
I also like to break puzzles up into objectives. The player needs to get to a room, but there's a locked door between them. Their overall target is to get through the door. They see a guard with the keys - their objective is to get the keys. They try to take the keys but the guard sees them. They're now trying to stop the guard seeing them. By rewarding each little step in the right direction - telling the player "Yes, that's part of the solution but you need to solve this problem as well" you can lead players through the puzzle without them realising. Another way to approach it is to think about what the player will try to do, and how you can tell them why that won't work while subtly pointing in the right direction. Saying "That doesn't work" and not making it clear why a solution isn't working just leads to frustrated players.
(Sorry for the essay!)
That was a very nice read.
Personally the environmental puzzles are the most satisfying. I think it shows depth and realism. And the more logical and realistic a puzzle is, while still being challenging, the more satisfaction you get from solving it.
I wish every person who decided to make an adventure game reads this, cause surely it will make games 1000 x better.
I particularly enjoy adventure games if they are not so much about puzzle solving, but if puzzles keep me from advancing within a well told story.
I liked Shifter's Box a lot, and in this case being able to see a new colourful background (and possibly listen to a new tune, if I may be so bold to claim this) made up for all the effort of "solving one level" - but it was strictly a puzzle game in my opinion.
Puzzles that really blend with the story can also help the story gain momentum in some moments or distract the player from his main objective and punish him for it, or they can even change the story depending on their solution - which is awfully hard to script and to think of, of course.
Good thoughts on puzzle design from Stu and OneDollar, too - you basically said the same thing: It is very nice if a puzzle is established, the rules are laid out somehow, and then another twist is added, this add a great deal of immersion, I imagine.
Ponch: A lot of the 'thought' put into Shifter's Box puzzles was actually just lying on the floor staring at the ceiling demanding that my brain think of a new puzzle. I got to know the ceilings of my house suprisingly well making that game. If staring at ceilings equals puzzle science, then I am a puzzle scientist! Otherwise I'm just a weirdo ;)
Stu: Yep, I agree that if you continually throw samey puzzles at the player it becomes tedious, just like action games become tedious if you face the same enemy over and over again. Making lots of little puzzles turn into one big puzzle is a very nice way to keep the player active for a while, but it does have the downside of slowing the pace of the game down a bit, so it's the sort of thing I personally would use sparingly.
OneDollar: Very nice ideas there, that I have seen before in games and really think they are good ideas. The only problem for me is teaching myself to incorporate these sorts of methods into my games. Don't apologize, I am glad you took the time to write it!
Indie: Thanks. I agree that puzzles that see the player work with the environment are the most satisfying, even if they're not the simplest to do!
Sebastian: I personally think of Shifter's Box as a puzzle game as well, but that was the intent the whole time. Working puzzles into story in a seamless manner is something I hope to try at some point when I can write good stories. And, of course, multiple story branches is something I must try as well, when I am less lazy and more motivated!
Post a Comment