I have recently went through an amazingly unusual sequence of events regarding minesweeper. Actually, much of the unusualness comes from the accumulated sum of these events happening, so I guess it actually is more properly describable as a series of events. Anyway, I have recently joined Minefield, a “‘Massively’ Multiplayer Online” (‘Massively’ in further quotes because given the load the server could actually hold, not actually massively) “Minesweeper” (not actually minesweeper, because here score is associated with marking mines rather than winning associated with finding numbers (one could argue that this is what the word “minesweeper” originally intended)) and one day decided to try to score 1000000 points in one day.

At around 475000, though, something happened and Minefield banned me for using a bot, even though I have not used a bot and am not a bot. I suspect this occurrence was related to lag registering my clicks in a weird pattern, but whatever the reason, this happened. It was especially annoying because the only method of communication they provided for contact was Twitter, and I did not have a Twitter. Because I didn’t particularly want a Twitter, I looked for the developer via other means and tried to contact as such, but the developer didn’t respond. I finally decided to, *sigh*, make a Twitter account. By the way, you should never make a Twitter account. The account making process is ridiculously long and annoying, but I eventually got my Twitter, @lnitak.

I tweeted the Minefield page and was pleased that they swiftly dealt with the issue and allowed me to continue playing Minefield, but just 15 minutes later, the game detected me as a bot again, and so I sent a second tweet over with a suggestion that perhaps bot detection should be less sensitive. (I also had to edit my suspicions the first time about lag being what made the game think I was a bot, as this second time there was no lag and I was just playing a bit fast.) The second tweet never made it over, because Twitter decided to ban me while that was happening. That, of course, meant that I had to look over to the Twitter rules again and see what exactly I might have violated, and I cannot find anything in their terms that I got close to violating, as far as I understand, so I have no idea how, as one of my friends inform me, I have managed to get banned from Twitter faster than terrorist-linked Twitter accounts get banned. (This is sadly reflective on my Twitter handle: Alnitak, ζ Orionis, is a Type-O star and thus one with a particularly short lifetime, like that of my Twitter account named after it. I may have chosen a different handle if not so many have already been claimed. (Firefox spell checker evidently wants me to write “Genital” instead of “Alnitak.” I don’t even know what to say.))

Twitter’s ban appeal process is ridiculously long and complicated, even compared to its account-making process, and I seriously do not want to go through it, so I’m just going to not attempt being able to continue playing Minefield, especially as it has many bugs beside its ridiculous bot detector. (Also, seriously, who puts Twitter as the only means of contact?) Actually, over-sensitive bot detection seems to be very popular in minesweeper games, because on that day I got detected as a bot in two different minesweeper games, although the other just kicked me so that I could reasonably still continue playing.

But anyway, what I came to talk about today was minesweeper and me, and why I have came to consider minesweeper an integral part of my life and why I still continue to intensely enjoy it.

There once was a time when I liked chess. I once really liked chess, during my second and third years of playing chess. But then, my dad decided that he will take over my chess learning experience, and buy me chess books without consulting me about which books I feel I would optimally benefit from, and buy chess software without consulting me either, and make me regularly study chess. A whole lot. And specifically had me study what he wanted me to study. At certain times, maybe if he didn’t make me do it, I would have done that amount of studying, but it does not feel as achieving when my father is moderating my study, specifically as derived from the implicit “you’re this achieved because of me” and the occasional actual verbal utterance of that statement. Anyway, due to this and several other things, chess is permanently ruined for me. At least math was not permanently ruined for me, but it was considerably less enjoyable since starting 10th grade my parents started yelling at me to become better at math and was generally more enjoyable when I got into secretly doing the math that I hid under my SAT book when my parents forced me to study for the SAT. (Also fortunately, due to neither of my parents having advanced science education, I could get away with telling them I’m studying for the SAT II Chemistry when I’m actually on a Wikipedia page finding out why Beryllium-8 is so curiously unstable. (Incidentally, Beryllium-8 being that unstable is actually very fascinating; you should read about it too.)) Music has strangely not yet been damaged that much by parental tyranny, although that may partly be because they didn’t yell that much about it (given that their amounts of music experience and knowledge are 0 and ε, they don’t quite have the right to yell at me to do things with music, but they still occasionally do (this argument also works for other subjects that they direct demands for accelerated improvement in that they also know nothing about)), but definitely I actually have more motivation to keep music close to me.

But minesweeper is something my parents never told me to do more of. In fact, it has always been something my parents have been specifically not okay with: they were one of those parents that exclusively believed that computer and video games were strictly wastes of time. (By the way, my video game collection at home is exclusively DDR, something I could pass by them as “exercise.”)

All that I have achieved in minesweeper, I can fully attribute to work, pattern analyzing, and practice on minesweeper squeezed in during times my parents were not looking. Even with having to deal with the issues of working on something not approved by parents, for example having to immediately quit a game upon parental entrance to the room I was playing in, minesweeper progression was much more emotionally worth it, and I actually felt truly proud of my progress along the way to my current beginner+intermediate+expert summed best time of 66 seconds, at peak 120th in the world. Even though I haven’t improved on this for more than a year, I still find it very much worth it to continue playing minesweeper, because when I’m playing it I remind myself that I have proven that my parents are wrong in believing that my achievements are due to their push. It makes me feel worth where I am given very few opportunities to display so, and is where I can assure myself that I can be free to explore and do well being free to explore.

Obsessive Symmetry and Staircases Modulo Four

I like symmetry. I really do. (I also, at certain points, really like asymmetry.) I sometimes get obsessive over symmetry. But because I found out from the internet that I wasn’t the only person who frequently went out of my own way to put the same number of steps in each sidewalk segment (I also sometimes like to make sure if one of my feet steps on a crack that I get my other foot to step on a commensurate crack), I wondered if maybe I wasn’t the only person who obsessed over symmetry on a staircase, and even if I was, I might as well share it.

I like to ascend staircases two staircase steps per one step by my feet. I don’t know why; one step always feels so small going up. The thing, though, is that two steps also feels slightly too large going up, so I often feel that I wish to give my two legs the same experience. Thus, I have a strong preference for staircases with a 0 (mod 4) number of steps, so that I can ascend it with exclusively double steps and with the same in each leg. When staircases do not have 0 (mod 4) steps, I usually prefer the single step earlier; I frequently feel like I want to end on a double step; I don’t know why: I guess a subconscious affirmation of accomplishment? Anyway, the problem with that is that I have to calculate ahead of time where my steps near the end go. If you have ever watched me ascending or descending a staircase, you might notice that I do strange things, most notably (1) stopping for an extended while before engaging on an ascent or descent (I’m deciding which steps to step on) and (2) making huge steps with one leg after an ascent or descent (I realized I made some sort of mistake in my trajectory and am making up for the leg I ripped off by giving it actions that stretch it as taking a double step does). (I should also note that I much more frequently descend in single steps, since double steps are actually not as fast and more possibly dangerous on descent.) This property of my staircase traversing explained, interesting results time.

Back at my home in California, the staircase between the floors had one initial step, then a τ/8 turn, then two steps, then a τ/4 turn, then a 2 (mod 4) (I don’t specifically remember what exact amount) number of steps. That was really nice, because after the bottom step, which was physically in a special location anyway, I could just take double steps the rest of the way, as there was a 0 (mod 4) total of steps after the bottom, and the long flight was close enough to the set of two for me to be able to conveniently make the turn over one step.

Desmond is different. The two staircases between the first and second floors are (a) 9 steps, then τ/2 turn, then 9 steps, and (b) 11 steps, then τ/4 turn, then 2 steps, then τ/4 turn, then 5 steps. One-stepping over τ/2 turns are almost never good ideas. For the (a) staircase, the clear solution is 2 sets of double double-steps for each sub-flight, but I’ve actually wavered among a variety of solutions for the (b) staircase. I still use a variety of methods to ascend that staircase, but I think the method I lean the most toward is (where a and b is one of the two permutations of left leg and right leg) a double (2/11)-b double (4/11)-a double (6/11)-b double (8/11)-a double (10/11)-b single (11/11)-a double (2/2)-b double(2/5)-a single(3/5)-b double (5/5), with four double-steps for each leg.

At W20, I frequently ascend from the second floor to the fourth floor on the inner staircase. Second floor-to-third floor is 10 steps, then τ/2 turn, then 9 steps, and third floor-to-fourth floor is 9 steps, then τ/2 turn, then 10 steps. At first, this was quite an issue on apparent wasted singles, but after I got to know the layout, 5 doubles-4 doubles plus single-4 doubles plus single-5 doubles (the second time with the other leg represented more to balance the original imbalance from one of the leg in the first 5 doubles) has made me perfectly happy. Fourth floor-to-fifth floor, though, is 11 steps, then τ/2 turn, then 13 steps, so I actually use two singles on the way up the lower subflight there, and then 3 sets of double double-steps for the upper subflight.

A few weeks ago, I realized that when ascending, I liked to ascend for the second floor to the fourth floor in the inner staircase, but when descending, I liked to descend from the fourth floor to the third floor, then go to the main staircase, and descend from the third floor to the second floor there. A while later, I affirmed it was indeed the result of me liking completion in double symmetry that can be achieved from second to fourth, but that because I have a much smaller desire for double-steps on the way down, upon descent I instead defer to my desire to take the main staircase, from which I can see more of what interesting things at the moment may be going on in the student center.

Minesweeper Statistics Pre-College

(just decided to log here some of my minesweeper statistics before I go to college)

Top 9 Beginner times
0.88 on 26 Mar 2011
1.21 on 21 May 2011
1.24 on 06 Jun 2011
1.27 on 09 Aug 2012
1.28 on 05 Mar 2012
1.30 on 12 Jun 2012
1.31 on 16 Sep 2011
1.41 on 02 Mar 2012
1.42 on 10 Dec 2011
Beginner time World Rank=89th, USA Rank=12th, California Rank=2nd

Top 9 Intermediate times
13.29 on 11 Dec 2011
14.45 on 18 Jul 2012
14.93 on 16 Dec 2011
14.93 on 20 Dec 2011
15.04 on 22 Mar 2012
15.06 on 27 Nov 2011
15.07 on 06 Mar 2012
15.22 on 16 May 2012
15.27 on 05 May 2012
Intermediate time World Rank=126th, USA Rank=21st, California Rank=2nd

Top 9 Expert times
51.82 on 26 May 2012
52.54 on 28 May 2012
54.75 on 08 Mar 2012
54.81 on 09 Mar 2012
55.13 on 13 Aug 2012
56.19 on 27 Feb 2012
56.31 on 24 Mar 2012
56.35 on 28 Dec 2011
56.82 on 18 Aug 2012
Expert time World Rank=134th, USA Rank=18th, California Rank=3rd

Top 9 Beginner 3BV/s
6.54 on 08 Apr 2012
6.45 on 27 Apr 2012
5.92 on 13 Jun 2012
5.87 on 03 Apr 2012
5.85 on 23 Nov 2011
5.77 on 23 Jul 2012
5.74 on 16 Dec 2011
5.69 on 10 Dec 2011
5.67 on 01 Apr 2012
Beginner 3BV/s World Rank=19th, USA Rank=3rd

Top 9 Intermediate 3BV/s
3.95 on 19 Jan 2012
3.84 on 11 Jul 2012
3.84 on 21 Nov 2011
3.81 on 27 Mar 2012
3.80 on 27 Dec 2011
3.79 on 15 Mar 2012
3.78 on 13 Dec 2011
3.77 on 22 Dec 2011
3.75 on 26 May 2012
Intermediate 3BV/s World Rank=51st, USA Rank=8th

Top 9 Expert 3BV/s
3.35 on 07 Aug 2012
3.31 on 17 May 2012
3.14 on 27 Apr 2012
3.13 on 08 May 2012
3.06 on 12 Jan 2012
3.02 on 16 Aug 2012
3.02 on 11 May 2012
3.01 on 26 Feb 2012
3.00 on 03 Aug 2012
Expert 3BV/s World Rank=41st, USA Rank=9th

NF Records
Beginner time: 0.88 (Flag record: 1.30)
Beginner 3BV/s: 4.61
Intermediate time: 18.34
Intermediate 3BV/s: 2.92
Expert time: 78.10
Expert 3BV/s: 2.20

IOE Records
Beginner: 1.43
Intermediate: 1.25
Expert: 1.02
NF Beginner: 1.00
NF Intermediate: 0.85
NF Expert: 0.79

The Rules of Minesweeper

I occasionally get asked for some advice as to how to get better at minesweeper.

Oftentimes, one way to get better that is often underused is to better know the rules. Really. And minesweeper is no exception. Many people are not aware of certain specifics of the rules of minesweeper, and what they therefore imply they can do. Thus, here I am making a post about the rules of minesweeper. If you wish to get better, you might want to take a closer look at statements bolded and what they imply.

(In the following rules, a “square” indicates any of the units in the game of minesweeper that contain either a number, a blank (“0”), or a mine.)

The object of the game of minesweeper is to locate all the mines as quickly as possible. Specifically, the game of minesweeper is won at the point when all squares that are not mines are clicked open.

The game of minesweeper is lost when a square that the player clicks open contains a mine.

The game of minesweeper begins upon the opening of the first square the player clicks open, which starts the timer.

The grid of mines for a board is pre-generated before the start of the game. The first square clicked never contains a mine. If it did contain a mine in the board pre-generation, the mine is moved to the upper-left hand corner of the board, and if that was a mine, the next square over to the right, and so on.

There are two ways to click open a square: left-clicking the square and chording.

To click open a square by left-clicking, click the left mouse button. The square the cursor is over at the point of release of the left mouse button is the square that is opened.

When a square is successfully opened without containing a mine, it shows a number. The number indicates the number of mines that exist in the eight squares touching the square the number was in.

If the number would have been a 0, the number 0 is not shown, and all squares touching that square are opened as well.

When a square is right-clicked, a flag appears over the square. Right-clicking is intended for marking a square as a mine. Right-click again to remove the flag. (Note: In the original windows minesweeper, the second click moves to the ?, and only the third click removes the mark. The ? is obsolete in the professional minesweeper world, and by default nearly all professional minesweeper players turn off the ? option if it exists at all.) Each time a square is flagged, the number of “mines left” display is decremented.

(This process is known as chording.) If the number in a square is equal to the number of squares touching that square that are flagged, double clicking on the number opens up all remaining squares around the number. The game is lost if there is any inaccuracy in the flagging around the number. Specifically, the left mouse button must be released over the square in which the number is while the right mouse button is depressed.

The number of mines in a board is pre-determined. A Beginner board has 10 mines on an 8-by-8 board, an Intermediate board has 40 mines on a 16-by-16 board, and an Expert board has 99 mines on a 16-by-30 board. (Note: In Windows XP and later, Beginner is on a 9-by-9 board. The 9-by-9 board is not used in the official minesweeper community for records.)

(Note: Windows Minesweeper is not accepted officially for minesweeper records, because it is too easy to cheat with. The accepted versions are Clone, Arbiter, X, and VS, which can be downloaded at http://minesweeper.info/.)

(Note: In addition, one might add the rule that in official versions, the 3BV of a Beginner board is at least 2 (and 3 in some), the 3BV of an Intermediate board is at least 30, and the 3BV of an Expert board is at least 100. If you wish to learn more about things like 3BV, visit this page.)

(Note: Vista and later minesweepers in Windows are so distant from the original that they might as well not be considered minesweeper.)

This should actually be a fair speed-up in minesweeper skills, if considered thoroughly.

Minesweeper World Records: A Chart

One could find pretty much all the rankings available at Authoritative Minesweeper, but there’s many pages to go through, so I decided to try to make one chart containing probably the most important records so they can be viewed at-a-glance. Note that the records are likely to still get broken quite a few times. Additionally, please realize that I did not try to make this chart aesthetically pleasing. Hopefully there’s enough numbers to get over that.

Minesweeper Highests and Top 3s per State and Country

The following map shows the highest Beginner, Intermediate, and Expert minesweeper times per state (for all times recorded to greater precision levels than the integer, times are rounded up)(hopefully, you will be able to quickly deduce which time is which):

You should not be disappointed at California yet, because it’s actually second when you instead look at the sum of the top three individuals by Beginner/Intermediate/Expert sum (thus, the numbers in the following are the sum of 9 times):

Here is the same thing, with countries:

Minesweeper: All Mine Densities for 2×2, 3×3, 4×4, 5×5

So I decided to try to see how many different combinations of square side length and number of mines I could do. I managed all possibilities for the number of mines for square boards 2×2 up to 5×5 (the 5×5 at high mine densities is extremely guess-diabolical). The images are below. If you download Minesweeper X, you can watch videos here. (Note: “all mine densities” refers to between 1 and n-2, where n is number of squares, mines inclusive, because all other cases are trivial i.e. impossible to lose).