Puzzling in 2D vs. First Person

After a burst of productive energy last week, this week started to feel a bit slower. There’s only one puzzle left to do for the whole game, and then after that it’s final boss and ending. I have some ideas for the puzzle on paper, but decided to take some more time to think about it and in the meantime I spent more time this week replaying earlier areas, fixing old bugs I’d never gotten around to, and tweaking more writing.

Once the game’s content is done, I’ll start passing it around to close friends and maybe some people off this blog to get early feedback, squelch more bugs, that sort of thing. I want to give a lot of time for that, and only after I’m totally happy with the game and feel it’s at full functioning power will I put it out for everyone to play. Sorry for being a bit cagey with it–I don’t usually do things so cautiously like this, but then, I also usually come out on release day with a few major bugs!

Anyway–I wanted to get some words out about my experience working on Chiaroscuro in comparison to this game. Once you get into the Spirit Tunnels in Phanta it’s pretty easy to draw comparison to those areas and Chiaroscuro, and that’s not an accident. Level design’s my favorite part of making games and I really love thinking about and putting together puzzles. I approached the level design in both games with a really similar mentality, but from experimentation and watching play testers I picked up some interesting and useful information about building puzzles for 3D space and first-person view as opposed to 2D.

The differences really come down to one key thing; of the conventional types of camera/view set-ups in video games, first person is actually the LEAST informational. That is, it’s a very narrow, limited way to view the game world. Spatial problems that would be really easy to perceive and solve from a 3rd-person view easily become pretty hard when you’re forced to solve them from first person view.

Here’s an example. The above screenshot is of the first light-radius puzzle I made for Phanta. Looking at it from this utilitarian view, the solution practically screams out to you; the light has to go in the middle of the four towers! This is further enforced by the fact that if you click on one of the towers, a textbox comes up saying that the tower wants to be in light. The tools available to me as a designer made building this first introductory situation pretty simple, since it’s easy to see and figure out how to lay things out.

But for Chiaroscuro, players on the first run just didn’t “get” the spatial relationship between puzzle elements in the first level, even though it was essentially the same as in Phanta. They saw each of the flowers as sort of a free-standing entity, with no implied center. So we had to fudge it and put a big, obvious hint in the center of the flowers for players to aim for. The next couple levels gradually made the “center” less obvious, leaving it to players to figure out where the light should go.

This trend affected heavily the arc of the puzzles went for each game, in sort of an interesting way.

Because the layout of the level is so clear in Phanta, puzzles grew naturally as increasingly complex arrangements of light and dark-receiving towers. Here’s the first room from a dungeon late in the game:

Here the context has shifted from light/dark to hot/cold, but the mechanics are essentially the same for this room. The interest here comes from the rather complex way the towers are positioned. There’s only one real solution to this room, but finding it doesn’t require any trial and error for players who have come this far.

But Chiaroscuro, as with many first-person puzzle games that I’ve seen, does a lot more with expanding the space of each level. Having small areas with compact designs feels tight and streamlined in Phanta, but for first-person it’s constrictive. Successive levels focus a lot less on small, complex arrangements and usually have you exploring a large area, looking for clues and necessary pieces to advance. The light-radius mechanic is still the major player, but instead of many light sources with small radii that interact, the puzzles are usually based on dropping the sources from weird places or onto moving platforms or that sort of thing. All of this serves to make levels progressively bigger.

This doesn’t make the levels at all weaker in Chiaroscuro–their focus is just different. Having an open three-dimensional space to explore gave me a lot of fun verticality to play with as a level designer. For players, it’s interesting just to get to walk through these big spaces, and finding an integral puzzle element in the distance or in the corner of their eye feels like a eureka moment all to itself.

It’s interesting, because I found that the expectation was totally opposite in Phanta. If an important puzzle element was “just out of view,” (i.e. on a platform that was off-screen and only visible if the player stood on the far end of a room or platform) many players wouldn’t even notice it or be motivated to search every corner to find it. For the 2D view, players expect everything to be contained in the immediately obvious view.

That’s a pretty basic breakdown of what I learned from making Chiaroscuro. Coming back to Phanta from that gave me a really interesting perspective on my existing work, and inspired me to explore different ideas in 2D space.