mwageringel / everest

A mathematical puzzle game
https://mwageringel.github.io/everest/
GNU General Public License v3.0
70 stars 10 forks source link

Game design criticism (spoilers) #21

Open fohfuu opened 1 year ago

fohfuu commented 1 year ago

There needs to be more help in guiding the player to understand the puzzle.

As a designer, one needs to first introduce a concept, test that the player understands this concept, then challenge them with a harder version of the concept or by combining multiple concepts.

Level 1 is difficult for so many players because the only concept introduced prior to this point is addition. While "X" could be introduced more helpfully, it is introduced, but there is no introduction to the relation between problems. Most of the sums are conventional arithmetic, so patterns won't stand out. The lack of obvious connection between 0 + 0 = 0 and 3 + 1 = 4 fails entirely to teach how the joined answers mechanic works, as adding and subtracting 0 is understood by most people who can do arithmetic, and a right-handed player may miss the design queue entirely due to their hand obscuring the right side of the screen. You may want the player to instead view the entire level's sums before coming to the answer, but the player hasn't been introduced to their ability to scroll the page or skip questions, and may not know this is possible.

It is, in fact, very easy for the player to miss the purpose of the level entirely by brute-forcing individual answers, and still be unable to find any relation between the questions. This means the player does not have to demonstrate they have an understanding of relations between questions, that the sums should not always be solved arithmetically, etc. and doesn't have the skills to proceed.

While it is laudable to attempt a game without a written tutorial, it doesn't mean that there shouldn't be a tutorial at all. As the game is inspired by The Witness, I suggest looking up some analyses of the game's design, which will explain this concept far better than I can. The Game Maker's Toolkit video essay How Jonathon Blow Designs A Puzzle is a good starting point if you haven't seen it before.

I hope this is some helpful feedback.

mwageringel commented 1 year ago

Hi @fohfuu, thanks a lot for your detailed feedback.

Indeed, designing level 1 is quite hard, as it can lead to players quickly giving up if it is too difficult or if the behavior is considered a bug.

For me, the main challenge of level 1 is introducing the player to the meanings of X. As there is not just one meaning, potentially the level could be split in half in order to introduce them one by one. I have chosen to do this in a single level for now to avoid consolidating a single meaning that doesn't show the whole picture.

A secondary goal is to familiarize the player with the joined questions mechanic. Even though understanding this mechanic is not strictly required, it is very useful for proceeding to the next level. I didn't want anyone to get stuck when it is introduced, so I chose simple questions (0+0, 3+1). This allows a player to notice that it exists, to play around with it to understand the mechanic, or to later go back to it in order to explore it in case they get stuck at the end of the level where there are three joined questions. All this is optional though, as many players seem to understand the mechanic directly.

The ability to scroll or to skip questions is something that is difficult to introduce explicitly. It's not really required to know this for proceeding, so I don't think a written tutorial would be needed for this. Instead, I suspect most players will eventually notice that it works, at some point during the game.

As for brute-forcing in level 1, I tried to find a balance between too many and too few questions. If there are too many hints, solving level 1 becomes too obvious and not rewarding – if there are too few and if the answers are brute-forced, it becomes too difficult to understand the rules. The right amount of questions depends a bit on the player. I think there is enough information to solve level 1, but I may need to observe more new players to be sure.

Also, thanks for the link to the video. I've done a bit more research on this topic and encountered many thoughts and ideas that resonate with my own experience after having made this game.