Whats makes a good puzzle?
I personally think an overlooked element of Portal 2 maps is having most (if not all) of the elements in the one room.
Of course there are exceptions, especially in Old Aperture, where there are a series of small puzzles joined together.
But have a replay of the main game, you'll find that even as the puzzles get more difficult, they still remain in the one room, whereas I play some community maps that are long rambling adventures, with no moment of player reward for doing something painfully fiddly.
There are a couple of good reasons for this. One is that if you have a funnel, 3 gels, lasers, turrets, hardlight bridges etc in a map all at once it can make mega lag. Another is that having all testing elements available makes it too easy to break the puzzle. It just gets too hard for the mapper to control what the player can do while still making a decent map. And last but certainly not least, it just aint fun to use a bazillion elements together. There is fun in mixing two or three elements together, but using them all together just gives people a headache. May as well play a playground map.
I think in terms of boolean variables. Generally, it makes things easier.
Bumping this thread because I have a similar question to OP. I find it very difficult to come up with a good puzzle that's based around 'AHA moments' rather than having the player work out a ridiculous amount of elements that need to be correctly used(laser puzzles as an example of what's not fun). Does anyone know a good 'systematic' approach to come up with puzzles that are challengine without feeling like work to the player, or does being a good mapper just require creativity?
I tried the 'bottom up' approach from the Valve wiki but it makes it very hard to oversee whether your puzzle will be inventive or frustrating to the player. Also tried making an activity scheme of the steps a player needs to take but it often results in the same problem..
For one, make sure you base your puzzle on an idea, rather than a series of steps you want the player to do. Only limit the basic concept of your puzzle instead of preventing every alternate solution.
A good puzzle would have many ways to solve it, however the way most people would solve it may even be completely different to what you designed it to be. If people still get the 'aha' moment, then brilliant!
Also: Test, test, test! Get people to play your map as a WIP. Do not worry about detailing until you are certain that you are happy with the puzzle. My WIPs usually only have the puzzle elements in it, forget clipping brushes round laser catches and funnels, plop them in front. Be lazy for the sake of science!
Happy designing!
In my opinion multiple solutions is an unwanted defect and is sign of sloppy puzzlemaking. It's like a crossword puzzle where no words intersect, you can just put in anything you like. Takes away the fun. Yes the map should feel like a sandbox, in that sense there are many ways not to solve it, and with multiple things to try out. But there should be only one main solution. I know many people will disagree.
The way I see it if there are 28 fundamental different ways of solving your puzzle, choose one of them and enforce it by making restrictions on your map until only the solution you want remains. Preferably before releasing it or people will get cranky... Yes it requires lots of testing.
Well, I have to agree with that. I believe that the funniest part of a challenge puzzle is to try each of your skills or theories to solve it, until you finally try THE ONE who is the true / valid one. In fact it is very good for a puzzle that you can try several ways, that a map is open enough for allowing you to test your skills... that gives the illusion that you're in front of a not so difficult and solvable problem. When you have tried many of your theories and none of them seems to be the way for solving it, you start to respect the puzzle and the creator... Me at least! Why? because in that moment you learn that the creator has already thought about ALL different ways to solve the puzzle and has put the necessary obstacles to prevent all the ways except the intended one, that one he chose.
Personally, I prefer challenging puzzles, but I still know how to have fun with a good map, even if it is not too hard. That delicate balance that Marise mentioned above is for me the best option, and that's Valve's way! Why? because it sells. This may assure you've got many people playing, enjoying and talking of your map... but it's so commercial. You may want this or you may say (like one friend of mine would say): "you'll hate me later" On the other hand that line is as delicate as hard to get.
"Patent Pending" was weaving calmly onto that delicate line of difficulty... and (apart from the fact that it had also a perfect balance at aesthetics, puzzle elements, atmosphere, lighting techniques, etc.) that made it won the contest. For me, it really is a model to stick to when mapping, not to imitate but to extract the essence (read: Ebola's Patent Pending Design Introspection at: http://www.tomchiu.com/downloads/tom-chiu_patent-pending_design-introspection.pdf)
But, as a follower of recent mappers like jonatan, Nyskrte or Chander, or classics like HMW, Damagepy or Omnicoder, I want challenging puzzles, clever designs, creative solutions or creative ways of using usual puzzle elements.
Now that I'm learning to map, I ask myself many times about that delicate line of my puzzles... are they easy? are they hard enough? The easiest is to decide one of the 2 edges: hardcore - easy... but that balanced line between them, ...The difficulty lies in measuring and determining the difficulty in your map..
Don't like this rule. This is where the mapper can get so creative. Makes a difference between just making a puzzle, and creating art. Plus, it looks better and the player likes the map more.
Correct. I agree with Baca25.
I would say a good puzzle is art but I get your point. It should be the icing on the cake though. A common mistake I see in many maps is they obviously spent a lot of time of making their map pretty, but it lacks in other areas (points 1-9). For me as a player, I don't care about music, fancy hallways or animations. I got enough of that already in the official maps. I just want a clean challenging puzzle to solve.
Also correct. I agree with jonatan.
My point of view is kinda a mixture of both opinions: I've been waiting (learning hammer, testing what I learnt in all tutorials, etc.) until I thought I had a good puzzle/ or puzzles... After that, of course I will give all necessary good look a map MUST have, but as jonatan meant, without good puzzles, why wasting the time doing all that improvement in aestheticals?
My best example would be HMW maps: not just all puzzle elements are used in many different ways and difficulty levels, all of them absolutely funny btw, he even creates new puzzle elements which design is absolutely impressive and creative (and he makes this only with the same given tools we all have!)... And regarding aesthetical elements... well, I can't just say a word: the biggest words I know are awesome or impressive or huge and they are all far away for describing what HMW does...
useful tools and stuff here on TWP 
[spoiler]
[/spoiler]There are a couple of good reasons for this. One is that if you have a funnel, 3 gels, lasers, turrets, hardlight bridges etc in a map all at once it can make mega lag. Another is that having all testing elements available makes it too easy to break the puzzle. It just gets too hard for the mapper to control what the player can do while still making a decent map. And last but certainly not least, it just aint fun to use a bazillion elements together. There is fun in mixing two or three elements together, but using them all together just gives people a headache. May as well play a playground map.
Sorry,
what I meant was that the entire puzzle is contained in the one area, not that a puzzle should utilise every available element (that's just too much for most players).