Hi, i know this is a jam game that's almost a year old at this point, but i was looking though past jam entries for inspiration and it piqued my mathematical interest as to why it's so challenging. The platforming is really solid, but i felt under very tight timing pressure regardless of using the slow or normal fuse.
The problem stems from the values chosen for the die. It's a six sided die with 2,4,6,8,0,-1. Assuming it's a fair die (i'm not going to test if the unreal physics engine has any bias), the "expected value" or the average amount from a roll is 3.17. Because your fuse timer is always ticking down, you must roll the dice at least once every 3 seconds or you will be losing time regardless of a fast or slow fuse. From a bit of testing, it takes about 3 seconds for the die to land after throwing, so you barely have any time for platforming.
Second, we have to look at the worst case scenario (the player loses due to bad RNG), and the likelihood for it. Consider the following: when a roll takes 3 seconds, any value rolled below a 3 means you have a net time loss. This means that rolling 2, 0, or -1 actually results in a net time of -1s, -3s or -4s. On a macro scale, this means that there's a 3/6 or 1/2 chance that you lose time with each dice roll; the expected value of those three outcomes is -2.67s. That means with two bad rolls (25% chance), you typically lose 5.33s and with three bad rolls (12.5%), you typically lose 8s. In the absolute worst case scenario (2.78%), you can lose 8s with just two rolls. This is on top of the already very tight timing for "optimal" rolls. Assuming the player is already on low fuse, this can lead to considerable frustration, and once again, the fuse timer does not affect this outcome.
So what is the solution? Instead of offering the fuse timer as a difficulty option, offer different dies. Consider making a die with a expected value of at least 6 (giving the player 3 seconds between throws), and only 1/6 chance to lose time per roll (reducing the cost of bad RNG exponentially). This might appear to make the game super easy, but consider that the average player is probably trying to get a handle on the platforming first. You can keep the current die as a "hard" difficulty, for after the player has finished the game and is comfortable with the levels and platforming.
I hope this isn't too harsh of a critique, and keep on making awesome games :)
← Return to game
Comments
Log in with itch.io to leave a comment.
can you upload the models for sfm/or some thing?
Hi, i know this is a jam game that's almost a year old at this point, but i was looking though past jam entries for inspiration and it piqued my mathematical interest as to why it's so challenging. The platforming is really solid, but i felt under very tight timing pressure regardless of using the slow or normal fuse.
The problem stems from the values chosen for the die. It's a six sided die with 2,4,6,8,0,-1. Assuming it's a fair die (i'm not going to test if the unreal physics engine has any bias), the "expected value" or the average amount from a roll is 3.17. Because your fuse timer is always ticking down, you must roll the dice at least once every 3 seconds or you will be losing time regardless of a fast or slow fuse. From a bit of testing, it takes about 3 seconds for the die to land after throwing, so you barely have any time for platforming.
Second, we have to look at the worst case scenario (the player loses due to bad RNG), and the likelihood for it. Consider the following: when a roll takes 3 seconds, any value rolled below a 3 means you have a net time loss. This means that rolling 2, 0, or -1 actually results in a net time of -1s, -3s or -4s. On a macro scale, this means that there's a 3/6 or 1/2 chance that you lose time with each dice roll; the expected value of those three outcomes is -2.67s. That means with two bad rolls (25% chance), you typically lose 5.33s and with three bad rolls (12.5%), you typically lose 8s. In the absolute worst case scenario (2.78%), you can lose 8s with just two rolls. This is on top of the already very tight timing for "optimal" rolls. Assuming the player is already on low fuse, this can lead to considerable frustration, and once again, the fuse timer does not affect this outcome.
So what is the solution? Instead of offering the fuse timer as a difficulty option, offer different dies. Consider making a die with a expected value of at least 6 (giving the player 3 seconds between throws), and only 1/6 chance to lose time per roll (reducing the cost of bad RNG exponentially). This might appear to make the game super easy, but consider that the average player is probably trying to get a handle on the platforming first. You can keep the current die as a "hard" difficulty, for after the player has finished the game and is comfortable with the levels and platforming.
I hope this isn't too harsh of a critique, and keep on making awesome games :)
hey how are you so damn good at making satisfying 3D platforming mechanics
Time to tas this game lol