Do spell-based puzzles inherently take less steps to solve?

I’ve been working on a big game for the last year that is essentially going to be 10 IFComp-sized games stapled together.

The current one I’m working on is a spell-scroll dimension in a dark cave with horrifying creatures in it, and which requires you to die/reset in-game a couple of times.

I work best when I give myself artificial constraints. For an IFComp-size game (which I want this dimension to be), I like to pick the arbitrary goals of lasting around 2 hours for an average player and having a minimal walkthrough of around 200 commands.

I’ve been able to achieve these numbers with my other dimensions (Haunted House with traditional gameplay, murder mystery on a train, wax museum escape room).

But I’ve found that spell scroll gameplay doesn’t lend itself to high command counts. It feels like most of the challenge is in trying to figure out what to do. I looked up walkthroughs and Enchanter, the first spell scroll game I’m familiar with, can be solved in around 200 moves, and that’s a 10+ hour long game!

I don’t want to pad out the puzzles artificially (by adding, like, a long corridor you have to spend five turns walking through). So I wonder, are there ways to make spell scroll puzzles more involved than just ‘select the correct spell and cast it?’

I can think of a few from @kamineko’s recent Spellbreaker Let’s Play, like:

  1. Timing two different spells carefully
  2. Having to revisit an area later on
  3. Recreating a sequence of earlier events.

But I also question whether my desire for symmetry and arbitrary goals is just a bad idea, and if it’s okay to have some areas that are harder but require less actions. On the other hand, the challenge of fitting the arbitrary constraints is fun, a game in itself and one I’m asking for help on.

What are your thoughts?


Nothing is inherent with puzzle design. Conventional, yes,

An obvious example is the Spellbreaker cubes puzzle, where you need to cast a spell three times and juggle a dozen cubes between containers in between. That’s gotta be a lot of moves.

It feels like a symmetry goal that only the author will notice. The player’s experience of a game isn’t measured by the walkthrough command count. It’s a mix of the number of commands needed to experiment with the puzzle and the amount of time spent not typing commands at all. (Taking a shower is often critical puzzle solving time.)


I like puzzles that make me stop and go through my spell list and think about how to solve it. I don’t consider number of moves to be indicative of anything. You can increase number of moves by decreasing implicit actions, but that doesn’t make it a better game, of course.

I might like to see a longer list of spells than I can actually use productively in a game. You could add spells that do humorous things but don’t solve puzzles. In reality, if you had a list of spells, not every one would do something useful for you. But you could throw in some easter eggs or some fun effects for flavor.

I think this is more than OK.


Since you mentioned Enchanter: I think it does a good job of mixing spell-based puzzles with more traditional fare. Sometimes in different proprortions!

So I think there’s a lot of opportunity for changing up the magic-to-solution ratio. For instance:

  • Dreaming about a bedpost, then manipulating it.
  • Summoning the adventurer (magic), making him friendly (magic), asking him to open the door (not magic)
  • The famous map puzzle (not magic)
  • The machine room puzzle: talking to the turtle (magic) speeding up the turtle (magic) asking him to get the scroll (not magic)
  • Entering a dark room to find a glowing painting (not magic)

By “magic,” I mean spellcasting.

So, one bit of advice would be to treat parts of your dimension as a mundane treasure hunt (which those games are at times!). The scrolls are the treasures. As you imply (I agree), knowing what to do with the scroll is not always the “puzzle.”

My Enchanter transcript:
second enchanter transcript.txt (134.3 KB)


Trust your instincts, padding a puzzle with meaningless rooms to traverse is not fun. Figuring out the puzzles is the part that brings the joy. That part does not need to be contained in the game, we have internet and search engines.

For example expecting your players to use a mouse-like animal to scare a cat is illogical and such a puzzle does not lend itself to reasoning and figuring out the solution, but if you convey to the player that the name on the cage of the mouse is mus lupus, that opens up a path for inquiry and enlightenment, and the puzzle becomes more involved(one might need to consult Wikipedia to make the connection if one is not a zoologist and that counts as interaction in my opinion) without actually increasing the commands the player has to type in to progress.

That being said, it is possible to overuse this strategy. Do not force your players to spend more time on Encyclopedia Britannica than in your game, trying out possible scenarios.

In your case with the spells, try to think of multiple plausible uses of your spells to advance the game, and if you want to close off some those paths, add meaningful scenarios and explanations(use humor if you want) to explain why those fail. Even if something fails to solve the puzzle the player will get an enjoyment out of it if they realize they found a way that the author also thought of.

1 Like

Would it be possible to require combinations of spells to get certain puzzles solved, perhaps mixing in non-spell elements? (Even magicians are likely to have some areas their magic can’t influence).


I think the key is, Enchanter magic gives you an assortment of specific tools to work with, and using one of those tools requires only a single action. (Or two actions, since you have to MEMORIZE the spell first.)

This means that, once you’ve figured out the specific spell you need, it’s only one action to use it. The hard part is figuring out what spell on what object will solve your problem.

But, as others have said, that doesn’t necessarily make the problems any faster or easier to solve. It just reduces the move count.


I think not really. For example, Spellbreaker had a lot of going back and forth between rooms, trying out spells until you realise that the answer had been under your nose all along! (Like with the roc puzzle, I had flown using the carpet to the roc’s nest multiple times, but never tried to take the cube, which I could have done immediately.) As long as the spell puzzles are Oh-That-Is-Cool style, you’re doing fine.

1 Like

You could make obtaining the scrolls a puzzle unto themselves: One is written in a dead language and must be translated by finding a book (or a translate spell!), one has been torn in half and its pieces must be reassembled, one requires the use of an alchemical substance you must collect (you know me!), etc.

There’s also the head-fake: Make the puzzle suggestive a particular spell is required (perhaps one they don’t have in their repertoire), but another is an easier solution. A “think outside the box” puzzle.

I think most symmetries will be missed by players, unless they’re somehow organic or integral to the game or story.

I’m not sure there ever was a game where “arbitrary goals” are looked back upon with fondness.


This is all great! I think you’ve all convinced me not to artificially pad out the puzzles.

Instead, I’m just going to make good content and see how long it is. If it feels unbalanced compared to other sections I can always increase it.

I do like many of the suggestions given here, and I think I will incorporate spells that affect each other because that’s cool (like the “reversal” spell in Balances by Graham Nelson).

Now that I’m freed from length considerations I can think of other stuff. This is a horror sections so I can make all the spells horrible, like “to make a creature watch you”, “to cause rot or decay”, etc.


Metamagic! Always fun.


I wanted to answer this to @mathbrush , but the forum software decided it was not a proper sentence:


So I padded it with these surrounding sentences.


I always thought it would be cool to have an accomplished wizard struggle to unearth arcane forgotten magics only to slowly reveal meta elements of the game like allowing undo, and seeing other save files created by alternate universe versions of the wizard.

The wizard has the final horrifying realization that he’s simply a character in a fictional world, at which point it has him hiding in your system files and then using your internet connection to jump elsewhere, which leads to an ARG like breadcrumb trail as you chase this Wizard across the internet as he descends into madness, finally finding yourself trying to talk him down from using his magic to trip off the nuclear silos inside the aging computer systems of Kosvinsky Kamen (Russia’s equivalent of Cheyenne Mountain).

You have to explain to him that he still matters even though he’s fictional and that life has no fixed meaning but only what meaning we give it.

ETA: (Of course inspired by the TNG episode, “Elementary, my dear Data.”)


Reminds me of Dr Moriarty in Star Trek TNG. Always a great storyline.

ETA: haha! Beat me to it!

1 Like

Heh, I edited my post to add that acknowledgement in right before you posted. Good call out!

I thought it’d be a great way to mix formats. The faux websites the Wizard would travel through would lend themselves well to being made with Twine and hosted on their own URLs, whereas the original game would be a parser. Depending on how ambitious you were, you could have a GOG knock off and have the wizard travel through a side-scrolling platform, causing havoc and ruining the plot of the game.


You guys should try Spam Zapper from the last IFComp, it’s actually extremely similar to what you e said and includes weird mystic musings about the nature of existence and is really long.

I have to say thanks for this thread because I’ve been sos tuck on this section but after reading this k decided to just focus on what cool moments/feelings and story I want, and to just let the spells flow from that without worrying about arbitrary constraints. It’s been fun!