From MinesweeperWiki
Jump to navigationJump to search

The Dreamboard is a famously easy Intermediate board created by Windows Minesweeper. It resulted in several world records, creation of the Winmine Congress, discovery of Board Cycles, formation of Project Minesweeper Utopia and the development of Clones.


The Dreamboard finished in 10s by Matt McGinley. 3BV=30

The first person known to complete the Dreamboard was Damien Moore (Canada) when he scored 18 on 22 Mar 2000. Gernot Stania (Germany) then set a world record of 15 on 20 May 2000. A week later Lasse Nyholm (Denmark) noticed the two games were identical!

The Guestbook posts from this period are lost, but this was a big surprise. Players had thought boards were completely random. Now it was suggested that the infinite number of boards was created in the same order. Thus every player would get one chance at each board during their lifetime. More examples of duplicate boards were soon found, leading to the discovery and investigation of finite Board Cycles. The story of the Dreamboard is closely linked to developments in Board Cycle research.

Damien posted an article on 1 Oct 2000 called "Random Games?" with several examples of duplicate boards. The main feature was screenshots from different players of what became known as the Dreamboard.

Gernot then set a new world record of 12 seconds on 7 Oct 2000. There was a delay before Damien updated Authoritative Minesweeper then Matt McGinley (USA) posted 16 Nov 2000: "Wow! 12 in intermediate! And Gernot got the SAME EXACT BOARD AGAIN!!!!!!!!!!!!!!! Hmmmmm, something's not right in the game of minesweeper". The new record was accepted, but this meant one person could get the same board multiple times. Over the next few months more players completed the Dreamboard or found other duplicate boards, but it was still believed these boards appeared by chance.

One such player was Matt, who on 27 Feb 2001 posted, "I just got the 12 sec. board everyone got! However the bad thing was I screwed it up and lost it! Im so pissed off now! I was thinking just last night how fast I could probably go if I got that board." Joe Nuss (USA) commented, "With the crazy frequency with which that baord has been showing up, i bet everyone here's had it at least once." Two months later Matt got the Dreamboard again and knocked 3 seconds off his record to score a 14. He posted 28 Apr 2001, "A freakin' 14! It's on the same board that everyone's getting too....the one that I almost killed myself a month ago because I missed. I have the video." His new score was accepted unanimously by the community.

10 Seconds

Emmanuel Brunelliere (France) scored 16 on the Dreamboard 10 Jun 2001. He posted a profile on the Intermediate Hall of Fame website and wrote, "Now I’m wondering if, to beat the inter record ( 12 ) the best thing to do is not “simply” to learn the few boards which are known to allow great scores, learn the places for the most efficient clicks, and just wait to have one of those boards. I must say that I don’t find it exciting to play such a way, but as randomness is no perfect in that game, maybe will it be the only way to do the first sub-10 on inter."

27 Jun 2001 Matt got the Dreamboard again and jumped from 14 to a new world record of 10 seconds. This was despite the Timer Jump and losing time by misflagging in the bottom left corner. Players were unanimous with their congratulations and were eager for their chance at playing the board.

Matt replied, "Thanks guys. The timer jump is the exact same one as in my 14. When I calculated my "real" time using Windows Media Player, I got 9.4 seconds. Despite that, I really dont think that 9 is possible. I flew through that board entirely from my subconscious, and the little misclick wouldn't have gained me a whole second. Im not trying to say that no one will ever beat me, I'd love to see someone get a 9. But Im pretty sure 10 is the fastest you can go on that board. Will somebody find a faster board?"

Emmanuel wrote on 30th June, "When I wrote my bio for Matt's site, 2 or 3 weeks ago, I didn't think that what I said about the way to break the 10-barrier on inter was going to be confirmed so soon."

A discussion followed over the next few days, but it was not about accepting the new record. Instead, players argued whether a 10 made Matt the best Intermediate player or if the volume of Sub20 scores by Damien made him the best player.

Dan Cerveny (USA) posted, "I must disagree that Matt is definitely the fastest intermediate player in the world. Keep in mind that his two fastest games, 14 and now 10, were done on the same single board that is clearly faster than any other board we can get. It is clear from watching the video that Matt has the board mostly memorized. Take the upper right portion of the board, for example, where it is necessary to guess. It is clear that Matt knows exaclty which squares to clear. My point here is that I doubt Matt, or anyone, would be able to get a 10 on that board if they had never seen it before. All that said, Matt should be a proud intermediate world record holder."

Matt answered, "I cannot claim to be the fastest intermediate player. Sure, I have a 10, the world record, but my next best time is a 14, then 16. If you look at people like Damien or Lasse who have nearly three or four times the amount of sub-20's I have, you will see the fastest int. players. Im fine with being ranked third in int. by means of totals. Damien and Lasse deserve to have the 10 more than I do. Dan you're right. The board was all luck, good mine layout, and I did memorize it so that all I had to do was get it and BAM! I had it...But do you blame me? Would you have done anything different?"

Dan replied, "Matt, nope, I don't blame you at all. In fact, right now I am just taking a break from trying to do exactly what you did. [...] do you know how many times you've gotten the dream board? I need to play int more often so I have a better chance to get it." He also noted this particular board was "so much faster than anything else."

David Barry (Australia) warned, "I think that Matt getting a 10 by knowing the board is a turning point for how the best int. players will play - if you need to memorise boards, so be it. To me at least, this is not a turn for the better, but if it generates faster times, then the int. record holders of the future will all memorise the best boards, whether I or Dan or whoever like it or not." He added, "Just to make one point clear from my thoughts in the post below. Matt getting a 10, breaking the record by 2sec, makes him the best int. player in the world. He is just using a tactic that no-one else has before (at least not that effectively)." Later that summer, David did the first serious analysis of Intermediate board cycles and wrote a program to generate the Dreamboard, enabling him to score 6 seconds. He published the results and deleted the program.

Joe Nuss wrote, "matt trained very hard for the ideal condition, and when the ideal condition occured, he performed nearly flawlessly, and thats why he's the best intermediate player in the world. hell, minesweeper's memorization anyway. we've all trained ourselves to react to certain situations, such as patterns, and thats what makes up a whole game anyway. is it really any different if matt just did that on a bigger scale, memorizing the whole board? i think not." Marc Schouten agreed, "Well said, Joe. I couldn't agree more."

Paul Kerry (UK) gave both sides of the argument: "Matt too, has memorized a pattern. It is though, a quite bizarre, extremely rare pattern, involving 40 mines in a 16x16 area, in a board of the same dimensions. That, to me, seems like the same skill all of us use, every day, in every game, to get good scores. We know certain patterns will be present in a game and we take advantage. On the flip side [...] Could the same be said for memorising an entire board that is guarenteed to come up eventually? Minesweeping skill is practically thrown out of the window and it just becomes a mouse test. AFAIK the concept of Minesweeper is about being able to uncover an unknown combination of mines. Learning logical/mathematical patterns is fine. However, abusing bugs is well, for lack of a better word, cheating. I don't want to take from the achievements of the people who have fast scores on a 'dream' board or two. Even if they have a little knowledge beforehand, one still has to make the clicks, avoid making mistakes and so forth, but it seems like a very dodgy area to me."

(Paul, in this post of 29 Jun 2001, became the first person to mention "dream" boards. The following day Roelof Smit (Belgium) used the phrase, "the dream board". Damien first used the exact word "Dreamboard" on 27 Jul 2001.)

Matt defended himself, "When Gernot got a 12, after a 15, dropping three seconds off his already world record time, on the dream board, did anyone doubt his world record holder and fastest int player status? If anything I think that I have proved to be a faster sweeper then Gernot, getting a 14 on my first real go at the board, then dropping 4 seconds and subsequently 2 off the world record my next shot." He mused, "I really wonder how many people are trying to memorize dream board now..."

Lance Votroubek (USA) wrote, "matt is just doing large scale what we all do small scale by memorizing the pattern of the board, but minesweeper is generally supposed to be a game measuring the time that a person uses to reaction to a novel arrangement of patterns. [...] By "randomizing" the structure of the boards, the programmer(s) sought to eliminate/minimize the effects of recognizable boards. By all of us congregating at one minsweeper fansite, and showing all of our best times and boards, we have taken away much of that randomization. People like Matt (whose record I consider as valid as any other) would probably not have realized that there was one particular board that makes 8 seconds on intermediate possible, and therefore would likely not have memorized it from the gameplay, strategy, and videos of others who have happened across it. I have gotten that board once, and without any memorization I got an 18. Most people would thing "hey! sweet time!", and maybe not think twice, but by seeing that at least 6 people have gotten their records off that board, it is within logical reason that someone would think "18 seconds the FIRST time... imagine what I'll do when I get that thing again! What is the most efficient path..."

Owen Fox (Ireland) posted, "I actually had started to work out the most efficient clicks for the "Dream Board" long before Matt got his 10. Unfortunately i've never got the darn board! I even thought that we should exclude that board from the ranking." Paul also admitted to starting games by clicking where the two big openings of the Dreamboards were, but he had quit from boredom.

Dan pointed out that, "When I started this rant and subsequent string of rebuttals a couple days ago, my point was solely to refute what I considered to be a wrong claim (by others, not Matt) that your 10 on the dream board clearly made you the best intermediate player." He added that Damien had told him "his intermediate score list and I now think it is safe to say he is the superior intermediate player. He got more sub-20's in six hours yesterday than Matt has total. The guy has 34 sub-17's, for crying out loud. Wow."

Matt made one last post on 1 Jul 2001: "I just want to clear one thing up. I said something, then somebody interpreted it, then somebody blew it way out of proportion. I did NOT memorize that board. I did not sit down one night and remember all the spaces and patterns. I watched my 14 video so many times....With each viewing, I noticed more slip-up's then previously and grew more and more discontent with my playing. Incidentally, I corrected those mistakes subconsciously and consciously, like everybody does when they watch a video and see a mistake. Going with the denotation of the word, "memorize", I most certainly did not. And lastly, after watching my 10 video, I realize that especially without a timer jump, 9 is possible. But it's someone else's turn to get that board, and maybe a 9."

Players continued to get the Dreamboard, but the community did not see this as a problem. Marc Schouten (Netherlands) scored 16 on 10th Jun, Damien lost it on 27th Jul, Dan scored 13 on 4th Aug, Owen scored 12 on 12th Sep, Stephan Bechtel (Germany) made 17 on 21st Nov, Lasse made 12 on 5th Mar and Dan made 11 on 24 Jun. These examples were just a drop in a bucket that was about to overflow.

Roland Debate

Roland Seibt (Germany) joined the minesweeper community in April 2002 and began sending times to the Active Ranking. In August he set a personal intermediate best of 21 on the famous '6 Circle' board. A few weeks later he made a 17 on the same board. In September he scored 16 then 15 on the Dreamboard. This culminated on 30 Sep 2002, when he announced that he had tied the world record of 10 seconds on Intermediate. Roland explained it was his 6th time playing the Dreamboard and that the board layout was "burned into my brain!"

At first everyone was quick to congratulate him. Steffen Stachna (Germany) said it was 'unbelievable', Dan 'great', Jeroen Courtens (Belgium) 'spectacular', Roelof Smit (Netherlands) 'amazing'. Many other players responded in a similar way. James Lange (Canada) asked for a video to help him memorize the board. Matt, the existing world record holder, thanked Roland for the motivation to keep playing.

Roland responded by posting, "Matt, if you would like to know my strategy, let me know!" He apologised for the lack of video evidence, saying "I just hope you guys believe me until I get the 9 and make a better proof!" A few days later, 5 Oct 2002, he solved the Dreamboard again, this time in 9 seconds and on video.

Most players still saw no problem and continued to congratulate Roland. For example, Steffen posted in the Guestbook, "Great work and a nice video Roland. I think there is room for at least one second. May be i should think about playin for the dreamboard only just like you." James similarly posted, "Good job Roland! How do you get the dreamboard so often?"

Roland replied describing his technique. "I know how to increase the chance to get the dreamboard and I know how to play it and so why not among other boards play for it?! If you knew how to beat a world record, wouldn´t you beat it? [...] I told you before that I have the board memorized! Nobody interjected! [...] Minesweeper, as every other game, is a game with more facetts than speed! I developed a strategy how the record can be beaten and it took me about 3 months of hard improvement to do so! [...] It´s just not the "normal" way! [...] You have to accept that many people tried my way before but didn´t reach it yet. It is a part of the game to find out that some boards come by and by and using this "against" the game is simply strategy! [...] And what should I do in the futuretime then: "Oh, the dreamboard again, but I can´t play it, since I know it!"? To my mind, I would see it as extension to minesweeper."

This post became pivotal in forming public opinion. It was already known boards repeated, but the idea you could manipulate board cycles to play particular boards was alarming. Players still accepted 9 as the new world record, but many started voicing discontent. Matt immediately asked how he played for the dreamboard. Steffen now wrote "Regarding those things Roli said before, i think it is a pity [...] Now playing minesweeper in int is just hoping to get a familiar board [...] and doin everything that one has already constructed in his head [...] This is not my understanding of minesweeper." Dan added "If Roland knows of a way to get the dreamboard frequently and has the board memorized to the point where he knows every necessary click by heart, then as far as I'm concerned the intermediate record is now meaningless." Edu Cros (Spain) wrote "I'd like to congratulate Roland for his job, but I agree that that's not exactly playing minesweeper." Marc Schouten wrote "The way you play, you're eliminating what I feel is the essence of Minesweeper [...] I assume that by bug you mean the fact that a certain board keeps popping up. A bug is an unintended feature. To make use of it is to make use of unintended game mechanics. These two things combined are why I asked whether it was still Minesweeper you were playing." However, he did not reject the score and told Roland "Your 9 is very impressive and shows a lot of skill."

Sorin Manea (Romania) then shocked the community by announcing he had "made a minesweeper like game and played the dreamboard over and over" until scoring a 6 after several attempts. A few days later he made a 5 and offered to upload the video. Sorin stated that he had not yet played the Dreamboard on the original version. (A few months later on 24 Dec 2002, David Barry (Australia) wrote a program to study Board Cycles and used it to also score a 6 on the Dreamboard).

Roland now explained, "Okay, I guess I can´t keep it secret, since everyone wants to know! Playing for the dreamboard: All I thought of was, where the first two clicks would have to be to get the 2 great openings! So everytime it shows up, I recognize it quickly! Perhaps it even increases the chance of getting one?"

Most players still supported Roland. James wrote "I think that Roland's 9 is just as valid as Matt's 10 [...] Personally, the only thing that I do not like about the way Roland plays is that I could not do it as well as he does." Dave McBryan wrote: "I can't help thinking that a lot of people's complaints with Roland's 9 applied to Matt's 10 at the time as well. When he first tried to defend that against sceptics, his main argument was that he had watched his earlier dreamboard video [...] so many times that he knew the board by heart. That was accepted, so I think this should be too. [...] Yes, it does take away from the logical aspect of the game, but it merely continues an onward trend in that direction." He pointed out that players were using similar methods to get 1 second Beginner games. David Barry wrote "The world record for intermediate is now 9, whether we like it or not. There isn't much we can do about this, just as we can't do anything about Matt also memorising the board and scoring a 10. [...] Matt knew precisely where he wanted to click, just as Roland did." He further stated, "Whether you had the intention of memorising the dreamboard or not Matt, you did, and at the same time you worked out the most efficient way of sweeping it, which led to the 10. Roland did the same thing, only explicitly intended to. The result was the same - a memorised board with a memorised way of sweeping it."

Matt now had to defend his own record, "I have to say that yes I was familiar with the board by the time I got the 10, but I had certainly not devised an ideal playing srategy as Roland explicity stated he has. Sure, I knew the locations of the mines well (notably in the upper right corner) and gained some time simply with familiarity but I only played the board from memory of experience, not planned action or drilling of the scenario."

Other players were now put in an awkward spot. The top ranked player, Lasse Nyholm, noted that his record was a 12 on the Dreamboard and suggested highlighting DB scores on the ranking. Marc Schouten, Jon Simonsen, Owen Fox, Gernot Stania, and Gyula Pap also held records of 12 seconds on the Dreamboard. Damien held an 11 on a Dreamboard shift. The only players with Sub13 scores not made on a variant of the Dreamboard were Dan (11 on the 6 Circle Board), Emmanuel Brunelliere (12 on the 6 Circle Board) and Sriram Sridharan with an 11 on the same board David Barry made his 15 with. In addition, Dan had an 11 on the Dreamboard and Matt's next best score of 11 was on the 6 Circle Board!

Roelof, who ran the Active Ranking, made the first big decision: "Roland, do you have a method to get the dreamboard more than others? Because, if you do, I won't accept your 9 for the Active Ranking - unless the community thinks truly different." He listed Roland with an imaginary 15 for Period 21 while the community discussed the problem. He admitted, "Formally, there's nothing wrong with the way he plays, but nevertheless I feel it's not fair. I can't explain why, but after this, I don't see the point of playing intermediate anymore. Something must be done about this problem."

Damien, webmaster of the world ranking, finally wrote on 6 Oct 2002: "i find this whole event disheartening. it was quite a blow when matt got his 10 on the dreamboard and i remember the controversy...i was reluctant to accept his score for the site; the main points against it being it was his [2nd] record on the dreamboard (14 [...] then 10), that he was too new to the game to have earned such a rank, and that he had probably memorized it. the defense is that the board appeared randomly and was not 'played for', a strategy had not been worked out, and the majority of sweepers were in favour of accepting it. it's also to be remembered that the previous world record (by Gernot Stania) was also a personal repeat of the board (he had a 15 then the 12), but at that time no one had known that boards repeated and it was not recognized until later that the boards were identical. [...] the only way to solve something like this is a mixture of lasse's suggestion, identification of the 5 or 6 dream boards [...], a listing of stats for the ranked players or something else. the main issue here is a lack of foresight among us in setting up standards for this community. i particularly respect those involved in the 3bv project, and as soon as it works on all systems i think it should become a standard download and could be included beside the scores in the rankings. i would encourage the creation of an official minesweeper game."

Elmar Zimmermann (Germany) then made a discovery on 7 Oct 2002: "if you take a look at rolands vid, i think it's pretty obvious how he increases his chances of getting the dreamboard. on roland's vid the board before the dreamboard is the same as the one on matt's 10sec vid. this board has a large significant opening in the top left corner. so i think roland's method or at least a possible method is this: you put the cursor in the top left corner and keep hitting the left mouse and the F2 button until ou get that particular opening. pretty dull, but it works! that way you already now that you're getting the dreamboard before you start the board itself." Jon Simonsen had already discovered this in March 2002 while investigating Board Cycles. Emmanuel said he too had already noticed this watching earlier videos. Actually, Matt had noticed this on 4 Jul 2001 and mentioned it in the Guestbook.

Roland proved he had not used this method by providing an unedited video of his record. He added "I didn´t know about any cycles, I just hoped to get it by chance and time. Same what everyone else could have done. I also wonder that no interjection came until to my 6th dreamboard. You could have told me then that you think it´s cheating to get a board by chance more than twice."

The debate continued to evenly split the community. Paul wrote, "Bleh, what a fabulously dull few days here. A style of play designed to take advantage of game bugs is cheating. Only a fool would agree that the repeating board syndrome is not a bug. [...] All bug abuse is cheating." Marc now stated he would never accept the score as official, but declined to call it cheating. Mikael Jonker (Sweden) defended the technique, "Either there are a set number of pre-generated boards that are displayed in a cycle (random or not), or the algorithm that generates the boards is limited to generate quite similar boards and thus increasing the chance of the same boards coming up twice. This would mean that using board rotation to your advantage can not be considered as exploiting a bug since the programmers must have known that boards would cycle. This is simply how the game was designed. Finding out exactly how the boards rotate should be easy enough, unless the cycle is completely random. You could just write a solver which records every game and reports the number of possible boards and in which order they occur. Such a program could also be used to let you skip all boards except the dream board by letting the program stop when the dream board is due. Using a program like this to only play the dream board should definitely be considered as cheating though. If Roland is not using something like this or is doing something to increase the chance of getting the dream board, I don't see any problems with his style of play." Emmanuel wrote "I think that the way Roli played his 9 is in the logical evolution of the intermediate."

In the 9 second video Roland had placed two flags before they could be deduced. Elmar now noticed that Matt had also placed a memorised flag in his 10 second video! (This occurs in the bottom left corner.) Several players pointed out this placed both scores in the same boat.

On 8 Oct 2002, Jon Simonsen (Norway) published an important article showing how he had experimented with Board Cycles. He put his support behind Project Minesweeper Utopia to create a secure minesweeper clone without board cycles. He also suggested creating polls in Minesweeper Addicts and allowing the community to vote on rules. Keith Whitener (USA) suggested forming a committee and James suggested a minesweeper constitution. August DeKock (USA) then made some major proposals: 1) Remove Beginner from rankings, 2) Create an Official Ranking requiring videos seperate from a personal highscore ranking, 3) Document and ban known dream boards, 4) Create a congress to make binding decisions on official scores. This led directly to creation of the Winmine Congress three weeks later.

Later that day James create a poll in Addicts about Dreamboards. The final results were mixed and players voted for multiple options. In the end, 3 rejected Roland's 9s, 2 rejected Roland's 10s, 0 rejected Matt's 10s, 5 rejected any Dreamboard, 4 accepted the Dreamboard if played without UPK, 8 accepted the Dreamboard if played 'normally', and 8 accepted the Dreamboard if no program or cheats were used.

Over the next few days Mikael scored a 14 then was dismayed to recognise it was the Dreamboard, Jon spent several hours marking flags before starting games in an attempt to set a new world record, and Stephan announced he would solve every dream board. Owen pointed out the need for a set of official rules quickly, while other players expressed hope for a new version of minesweeper without board cycles.

When the Active Ranking was released 11 Oct 2002, Dan and Roelof declined to rank Roland with his new scores.

Death of a Dream

The community eventually tired of the topic and stopped discussion, seemingly waiting for a magical solution. Players continued to make records on the Dreamboard but their posts were mostly ignored. The cases piled up as Benny Benjamin (Australia) scored 14 on 3rd Nov, Dan lost it on 8th Nov, Andy Porter (USA) scored 29 on 12th Nov, Elmar lost it on 25th Nov, Brian Cornell (USA) scored 12 on 15th Dec, Ross Berg (USA) lost it on 22nd Dec, Yoni Roll (Israel) lost it on 30th Dec, Case Cantrell (USA) scored 16 on 27th Jan, Dan lost again on 11th Feb, and Grzegorz Gorny (Poland) scored 16 on 27th Feb. This is not a complete list by any means.

James Lange posted an excellent description of the situation on 26 Feb 2003: "Roli got a 9 a few months ago using the "controversial" techique of repeadedly clicking in the same area so he could quickly regognize the DB, and finnish it using his knowledge of the board. There is a poll on the Yahoo site where you can see who thinks the WR is 9, 10, or 11 [...] the counsel was going to decide on it, but it seems as though we have all just decided to wait for the "perfect" minesweeper game (one without cycles, timmer jump, cheats, etc.) so we can start from scratch [...] because of the tension caused by the disagreement over the 9, we have all been a little aprehensive about bringing it up."

In the meantime, several players made use of the lack of clarity on the issue.

Stephan Bechtel scored 11 on the Dreamboard 23 Feb 2003 and posted, "Wow. What a day! Only 45 minutes after my 56 and while watching biathlon, the dreamboard came up surprisingly. Although I noticed after 1 or 2 seconds that it is the DB, I stayed calm and finished in 11 seconds." He added, "I got the DB three times in 19 months and finished in 17 (not knowing it exists), 15 (getting hypernervous as knowing it exists), and 11 (by luck staying calm)." Although public opinion had turned against Roland and Steffen, many of the top players were quick to congratulate Stephan.

Steffen Stachna had asked on 18 Oct 2002 if an official decision had been made and received a negative answer. So he continued to practice his technique. On 2 Nov 2002 he posted a personal best of 13 on the Dreamboard and claimed he had not used memory as an advantage. Yet later the same day he posted a new score of 11, again the Dreamboard. His posts were ignored. However, 10 Mar 2003 he wrote, "After 3 hours of playing all of a sudden the DB was there. I made a 9 on it. I played just like Roland did. May be i´m faster next time." Responding to questions he added, "if you know the distance between the good boards, it is good to press the f2 button for a longer time. I have been playing so much int the last days, in order to find out, how long i have to do that, to come back to the same board on the one hand, and how long i have to do that, to come from one good board to another. I gues, with my knowledge of only 30-50 boards, i can now find any board i can do 14 or better in not more than 15 min." He posted again four days later, "Just got the DB once more. But i "only" did it in 10 sec. This would (if it was allowed) have been 9,10,11,12,13,14,15... all in one period." Finally, he announced 27 March 2003 that he had made 8 seconds on the Dreamboard.

Case Cantrell (USA) also saw this as an opportunity to improve his scores. On 29 Nov 2002 he wrote, "my new goal has been lowering my intermediate score. i've been playing it a lot, and what i've really been hoping for is the board everyone seems to get their records on. Anyone have any idea how often this board cycles through?" His efforts became important later. He score 16 on the Dreamboard 27 Jan 2003. On 2 Feb 2003 he ominously posted "i found during that long stretch of intermediate playing that there are actually several boards that are better than the dreamboard." After Steffen made 8 seconds on 27 Mar 2003 he wrote, "Steffen, why do you bother putting that 8 as your record? I've got a 10 i got on a dreamboard (28-3bv) i had pretty much memorized, but there's no sense in counting it... it's not playing minesweeper. But if you guys are gonna start counting that stuff go ahead and put me down for a 10 intermediate, and i'll have a 7 before the night's out." The next day he posted, "well the 7 is eluding me, but here's a 10 on a NON dreamboard, but still quite done by memorization."

Matthew Ackerman (USA) noticed 20 Mar 2003 that a previous record of his was a shift of the Dreamboard. He posted, "My 12 is a dreamboard, so maybe I should start posting my best as a non dreamboard score? I just got it a week or so ago, I thought I had the board memorized but then when it came I still spent too much time thinking." He later posted on 18 Aug 2003 that he had scored an 8: "My third encounter with the dreamboard, video link is under URL. I haven't posted here in a long time, do people even care about dreamboard scores anymore? Didn't somebody cheat or something? Anyway I am pretty pleased." He went on yo score another 8 and two 9 second games on the Dreamboard. However, by this time Roland was the only person supporting the score as a world record.

These extreme cases further exacerbated the problem. Daniel Lynch posted in despair 10 Mar 2003, "I didn't realise that Intermediate was so dead." After Steffen scored 8, Matt McGinley wrote, "Please stop doing whatever youre doing, I can see int. becoming a joke right before my eyes." Roland was annoyed to be labelled the same as Steffen because he had not calculated the board cycle and Steffen had continued after the community had expressed disapproval. Some players demanded the Active Ranking be halted until a solution was found. Others wanted to remove Beginner and Intermediate scores from the rankings, leading to the idea of a Sayre Board.

Damien summarised things 28 Mar 2003 with this post: "steffen has a simple two-fold purpose: one, to show how ridiculously far memorization can go while discrediting the dreamboard (prompted by Roli's appearance), and secondly, to get the world record if the congress decides db's are legit. matt didn't have the board completely memorized and had no way to practice the board and he did not take advantage of the cycles; however, he was the first to flag the mine 3-in on the bottom corner from memory to save time and he also knew a faster way to clear the middle patch because he had realized his previous inefficeincy from watching his videos. Roli played waiting for the dreamboard and, uncomfortably, he got it very frequently, and through fairly careful planning he took advantage of it when it showed up. Steffen has completely thought out the board and taking full advantage of knowing the cycles. on his vid, he knows it's the db befoe he begins, hesitates as he counts down from the top corner, then immediately takes the one click that everyone else ends up finding last and goes for the openings, then cleans everything on a preplanned routine. so, i see Roli's frustration at being clumped with steffen, but the question is, where did it go too far?"

Case replied, "Some might think that banning "the dreamboard" is the solution, but then people will just do the same thing to other boards. This is what i was trying to show. I admit that in this regard i probably did go overboard. Back when Roli got his 9, i was annoyed and decided to map the intermediate cycle and play boards i'd know so i could get one of those insanely fast intermediate times. Then i got bored of it, because it doesnt really mean anything to get a low score if you didnt actually play the game. By this time i had gotten a 10, and several other times under 12, but decided against submitting them because i didnt think it was right. [...] If the congress decides to accept the 8s and 9s done from memory, you can bet there will be a lot more intentional memorization of intermediate boards. I would hate to see this, because intermediate does not need to be ruined." Steffen added, "I just had to do that 8 because it is so easy for me now to get that board and i just didnt want to share that 9 with roland. [...] Damien wrote a good explanation, though the first purpose dominates."

The Solution

The community decided to wait until a new version of minesweeper could replace the easily manipulated version from Microsoft Windows. The first attempt was Project Minesweeper Utopia. The unrelated release of Global Mines by Stefan Pettersson in October 2002 seemed to achieve this goal but development ceased soon afterwards. Utopia continued until Curtis Bright released Minesweeper X in March 2003, at which point the project saw no need to continue. MSX did not gain immediate popularity and most players still played Winmine until January 2004, when Rodrigo Camargo released the first version of Minesweeper Clone.

The Winmine Congress was divided on how to respond to the crisis. After deliberating for six months they voted in April 2003 to declare official disapproval. Rather than creating rules, the community would remain on the honour system of score submission. The desire for acceptance would be the lone preventative measure, with cheaters taking the risk of rejection from the community. The Congress members, seeing no better solution in sight, were all disheartened by what they saw as the loss of innocence in the community and the demise of meaningful records. Over the next few months, 6 of the 7 members quit and retired permanently. On a more positive note, the solution did have an effect and the majority of players with affected scores voluntarily renounced them.

The Authoritative Minesweeper site was rarely updated at the time but did not accept the Dreamboard records. The most current ranking at the time, the Bestever run by Georgi Kermekchiev, did accept the 9 second games from Steffen and Roland. The Bestever then transferred on 27 Jun 2004 to Planet Minesweeper. Its owner Grégoire Duffez cleaned up the list and removed most Dreamboard and all UPK records from the ranking. Matt Mcginley renounced his 10 and allowed it to be replaced with an 11 on the ranking. With the growing popularity of the Clone, it soon became mandatory to submit videos made on official versions in order to join the world ranking. This requirement continues to the present day.