Devlog #2 - I can finally talk about my puzzles

Hey everyone! Sorry this update is really late from the previous one. We were really busy with the demo (please play it here!), and then a ton of stuff in life afterwards :(

But Glowkeeper is not dead! And there’s a bunch of new exciting stuff! Unfortunately we can’t talk about the new puzzles cause of spoilers, but here’s a teaser for a new area :o

Anyways, the topic of today shall be the puzzles. This will be spoiling much of the demo so please please go and play it first if you haven’t :)

A process of discovery

One of the things I commonly hear from puzzle designers is how designing a puzzle itself is like a puzzle, and I very much understand this now after spending hours staring at the level editor. What makes a puzzle a puzzle is that you are discovering what the puzzle is about.

In this sense, designing the puzzles and mechanics of a game is similar - when I came up with the idea of fusing the match-3 mechanic with platforming, I really had no idea what I was getting into. I just played around by plopping stuff down onto the level editor and experimented, discovering the interesting interactions that came out of it. At that time, the only elements in the game were fire, wind, earth, ice and void (and fire and wind didn’t have magic powers like they do now).

So what did I discover? For me the interesting interactions were:

• Jump height was a very important limitation to the game - changing the jump height to 1 or 3 blocks basically changed the gameplay entirely

• There are 2 major obstacles: the ones where you have to build a platform to jump up somewhere, and the ones where you have to tear down a wall

• Positioning is important too, sometimes you must position yourself nicely before matching things (e.g. narrow gate, elevator, hollowed pillars).

Eventually, I decided to create a short 20-ish single-screen levels for a prototype, and I got some of my friends and people from the thinky-puzzle-games discord server to playtest them.

I’m gonna talk about how some of these old prototypes evolved into the current state of puzzles

Jumping

Playing back my old prototype, I can’t believe how slow the player character moved - and how short the jump distance is:

When making puzzles, I realized that jump height is a much more meaningful limitation than jump distance. Jump height is very clear whether you can or cannot make it, while distance is like you don’t know if you’re just not platforming hard enough. So I decided to make the jump distance of the player much further to make it clear that distance shouldn’t be a concern when puzzling.

This difference between jump height and jump distance inspired the puzzle Floating Earth!

The bottom ledge is kind of a trap as you can’t jump high enough, cementing the fact that you can only jump 2 spaces. Then hopefully you figure out that you can step back to the above ledge to jump up. You can see it for yourself in action when ambie played this puzzle (by the way go watch the VOD, it’s fun!).

Behind You

This was the first iteration of conveying the idea of positioning - you gotta jump to the right side first before jumping to the left. Unfortunately most people didn’t really get the memo and instinctively, not consciously, moved themselves over to the right ledge.

I guess to get an idea across, you gotta create some friction so that the player can consciously overcome the challenge. And so this became the “infamous” Hollowed Pillars.

Unexpected Surprises

Looking back, this is a pretty silly level - it’s just supposed to introduce you to the fact that ice is match-able as well. But well, I never expected this reaction from bucketfish, who helped playtest the original prototype too:

And this kind of gave me inspiration for “re-contextualizing” existing mechanics. I really love it when puzzle games do things where they take existing mechanics and give new dimensions to them later on in the game.

So I decided that I shall introduce earth and ice relatively early on in the game, but only show the full potential later on. In the original prototype, the fact that you can link earth is revealed to you as early as the 2nd puzzle. Now, I decided to reserve it until about 7 puzzles later in Crumbly Overhang.

I must say this puzzle is even more infamous, with people staring at it for up to 20 minutes until they realize that they can match the earth too. But I think the payoff really worth it when they finally get it.

Playing my prototype that is almost 3 years old now was quite nostalgic to be honest. I’m quite proud of how the levels have evolved from then. When I watched people play the newer version of Glowkeeper, it always makes me happy when there’s reactions like “I need to jump 7 gaps this is not possible!” and then going “WAIT I CAN MATCH THE EARTH” just a few minutes later.

I guess my main takeaway from all of this is that ultimately the puzzle and mechanic design is there to serve the player’s experience, and keeping in mind what kind of experience I want them to feel through these puzzles is what helped me refine the design of the puzzles to the current state.

Anyways, that’s all for today’s somewhat long reflection on puzzle design. See yall in the next one!

~oolimry