As a game I agree that it ultimately doesn't work quite as well as the original minesweeper, but I definitely am glad it was made. It's awesome! Visually the cascading effect of boxes being revealed is done so well and it's fun to see familiar patterns from the 2D version show up here -- I'm sure there are many new patterns in this 3d version that I haven't found yet.
Without giving it any thought really, I wonder what modifications (if any) needed to be made to the flood fill algorithm in this version.
Sometimes it's more about the challenge really. It started as an university assignment. Later I ported it to WebGL.
The original minesweeper rules don't make for a great game in 3D. I added the "sweep mode", where you can remove mines, to make the game more appealing in 3D
This is awesome! I was a pretty serious (classic) minesweeper player and was even ranked pretty highly internationally. Curious: did other people have their intuition translate "incorrectly" at first? I kept on blowing up cells because I hadn't yet fully internalized the changes that occur in 3D (I know the rules are the same, but the "pattern recognition rules" are not quite the same.) It was really interesting having to step back and rework out the implications of things in 3D. In some vague sense I feel like this is a bit like how some mathematical theorems hold in certain number of dimensions but not others (e.g. random walks). Cool game!
Yes, at first I couldn't even beat easy mode because the "obvious" places were unexpectedly mined. Each cell has way more neighbors in 3d, so it becomes hard to use the techniques from classic minesweeper.
I consider myself like an advanced-intermediate player, though my clicking dexterity is not what it once was. Care to share some of your record times? :)
This is a cool concept. The one problem that I had, is that it is difficult to tell which cubes the numbers are paired with. I think that it would be more intuitive if the numbers were printed/etched into each face of the cubes (like dice, but with normal numbers).
This is surprisingly addictive. I'm loving the "bonus" links it sends when you complete a puzzle (although I did get one with near transparent blocks that was murder to attempt).
That's cool. I had the same idea (3d minesweeper with semi-transparent blocks) about 10 years ago, but never came around to actually realize it. Great to see how the gameplay is like.
One possible improvement: Add some keyboard mappings for map rotation. E.g. something like WSAD to rotate the cube and QE,RF to strafe vertically and horizontally. I guess it would need some experimentation to figure out what makes sense.
Assuming the developer is listening... I think I ran into a bug related to resizing the window after starting the game. Clicking selected other blocks than I desired.
Also, I'd like to +1 the request to have a grid remain at least on the numbered cells. A partial transparency of numbered cells might be neat because you could see depth better from overlap of cells.
I kept finding it wanted to clip the right hand edge of the puzzle off the right hand side of the browser window. I tend to use square browsers --- am I weird?
And left-dragging to pan wasn't an entirely happy experience; partly because it kept selecting bits of the HTML on the left, and partly because I felt there was something slightly strange about the way rotation worked. Is it traditional arcball or something else?
...but in general I thought it worked really well. I played sweep mode, and liked it a great deal (I did get an unfair puzzle, but guessed correctly and won it. On, um, easy mode).
Update: has something just changed? I can't delete mines in sweep mode any more --- they just blow up instead.
I spend hours every day refreshing HN looking for the keyword to appear (and secretly, I must admit, also hoping for the next episode In the "PG vs. Seahawks OL" show).
If we start having bots do refreshing and posting for us, then bot will not only be taking our jobs, they'll be taking our hobbies!!
Ditto -- but in the process of making my minesweeper implementation, I got a little sidetracked and ended up making it Conway's Game of Life (sadly with no input, just random seeds/observe):
That guarantee allows for playing with high mine counts, which would ordinarily be impossible way too often. The high mine count boards sometimes contain interesting deductions.
Normally you'd do that by quickly reshuffling the mines if you guess wrong, although he might've gone the route of checking if it's possible to get into a guess-only state during generation and regenerating the board until no such states exist.
Yours displays in a subwindow that's too small for it and has scrollbars. You can click and drag to rotate the cube, but for some reason it doesn't start rotating until after you release the button, which makes rotation very difficult to judge. I spotted a mine, clicked on a "1" next to the mine, and it exploded. If you're not using standard Minesweeper controls, that needs to be clearer. Also, Kongregate made me watch a 30-second ad first.
Check the dismember view I provided (space bar). That open up the cube to give a better spatial understanding of the board.
That is the problem with 3D games, I realized. Trying to visualize is a pain to most people and deviates them from the game. This was my attempt at solving that.
I think in your case, hovering over a cell and highlighting its adjacent numbers might help create better visual understanding of the state.
>Trying to visualize is a pain to most people and deviates them from the game.
But it's a built-in skill of everyone over the age of a few months old!
Isn't the real problem the inability of 3D interfaces to, ahem, interface, with an innate skill 100% of sighted users have? In other words I think the interfaces are wrong, not the users.
I don't think that's the problem. I think the problem is that "3D" versions have us effectively trying to manipulate the internal 3D state of a box - that's something we're terrible at - we usually interact with 3D objects in a sparse state.
>we usually interact with 3D objects in a sparse state.
I don't know that I can agree with this. There's nothing hard about spearing a meatball with your fork, and some spaghetti, even if it's dense 3D space... So it's not about visualization, more about UX/interaction or interface...
57 comments
[ 3.5 ms ] story [ 114 ms ] threadWithout giving it any thought really, I wonder what modifications (if any) needed to be made to the flood fill algorithm in this version.
https://en.wikipedia.org/wiki/Flood_fill
The original minesweeper rules don't make for a great game in 3D. I added the "sweep mode", where you can remove mines, to make the game more appealing in 3D
In fact, a highlight of the 27 cells within the area of a clue on hover would be great as well.
Otherwise, fantastic work.
GitHub Issues (https://github.com/egraether/mine3D/issues)
Feature Requests (https://mine3d.uservoice.com/forums/136434-general)
Now I feel selfish for having the same thoughts as you about the grid and just quitting the game instead of telling OP.
http://www.minesweeper.info/archive/MinesweeperStrategy/mine...
Example: http://egraether.com/mine3d/?grid=30,16,1,99
Currently only at a 20% win rate, which is frustrating.
One possible improvement: Add some keyboard mappings for map rotation. E.g. something like WSAD to rotate the cube and QE,RF to strafe vertically and horizontally. I guess it would need some experimentation to figure out what makes sense.
Assuming the developer is listening... I think I ran into a bug related to resizing the window after starting the game. Clicking selected other blocks than I desired.
Also, I'd like to +1 the request to have a grid remain at least on the numbered cells. A partial transparency of numbered cells might be neat because you could see depth better from overlap of cells.
And left-dragging to pan wasn't an entirely happy experience; partly because it kept selecting bits of the HTML on the left, and partly because I felt there was something slightly strange about the way rotation worked. Is it traditional arcball or something else?
...but in general I thought it worked really well. I played sweep mode, and liked it a great deal (I did get an unfair puzzle, but guessed correctly and won it. On, um, easy mode).
Update: has something just changed? I can't delete mines in sweep mode any more --- they just blow up instead.
Someone says "minesweeper", I post my implementation. It's an established ritual.
Here go: http://www.ronilan.com/bugsweeper/
Few notes:
1. This is dated January 2014. It is a job interview homework.
2. The HN community has detected a bug and indicated that it should be fixed. That was done March 2017.
3. There are currently no known bugs.
4. The goal of the game is to find the unknowns.
5. No puns intended.
6. Code: http://www.ronilan.com/bugsweeper/js/bugsweeper.js
If we start having bots do refreshing and posting for us, then bot will not only be taking our jobs, they'll be taking our hobbies!!
Can't allow that!!!1
http://www.geekwire.com/2016/seahawks-lineman-russell-okung-...
http://game-of-life-manfrin.herokuapp.com/
http://www.chiark.greenend.org.uk/~sgtatham/puzzles/js/mines...
That guarantee allows for playing with high mine counts, which would ordinarily be impossible way too often. The high mine count boards sometimes contain interesting deductions.
https://git.tartarus.org/?p=simon/puzzles.git;a=blob;f=mines...
There is code for solving a grid using deductions in this section:
https://git.tartarus.org/?p=simon/puzzles.git;a=blob;f=mines...
And below that is the code that generates grids and checks that they are solvable without taking risks.
A similar approach is used to analyse the difficulty of Sudoku games here:
http://hodoku.sourceforge.net/en/index.php
with different names for the different types of deduction.
https://github.com/tim-hub/TiminesSweeper-3D-
There are many bugs and bad things in the game I did not fix, because I have no time to do. I have to work and find work to make a living.
Maybe the word I say is not clearly, What I mean is I think my game's mechanic/idea is better.
Certainly, http://egraether.com/mine3d/ this is more stable, more beautiful ui, and more light weight.
Reminds me of a 3D 2048 game I made some years back.
https://kshitij-banerjee.github.io/Cubiks-2048/
Check the dismember view I provided (space bar). That open up the cube to give a better spatial understanding of the board.
That is the problem with 3D games, I realized. Trying to visualize is a pain to most people and deviates them from the game. This was my attempt at solving that.
I think in your case, hovering over a cell and highlighting its adjacent numbers might help create better visual understanding of the state.
But it's a built-in skill of everyone over the age of a few months old!
Isn't the real problem the inability of 3D interfaces to, ahem, interface, with an innate skill 100% of sighted users have? In other words I think the interfaces are wrong, not the users.
I don't know that I can agree with this. There's nothing hard about spearing a meatball with your fork, and some spaghetti, even if it's dense 3D space... So it's not about visualization, more about UX/interaction or interface...