Talk:FreeCell

Game creation algorithm
I believe the original creator of Freecell placed the game creation algorithm in the public domain. If someone can track it down and it's not too large (I've seen it before and don't recall it being overly large) perhaps it would be a nice addition to the page. Bonalaw 15:01, 1 Sep 2004 (UTC)

" Don Woods wrote a solver for FreeCell"
This statement in the article is not attributed to anyone or anything. It links to an article that actually lists this article as the source of the statement that Woods wrote a solver! In other words, Wikipedia is verifying something by quoting Wikipedia! I'm just a passerby, so I'm not going to edit it.Geneven (talk) 03:02, 19 April 2009 (UTC)

Jim Horne's Shuffling Algorithm
The shuffling algorithm of MS Freecell depends on the Microsoft C compilers' randomization routines. A complete implementation (with its own random number generator emulating the Microsoft one) capable of generating such boards is available as part of Freecell Solver (as board_gen/pi_make_microsoft_freecell_board.c.

From the revision history comment by dfmclean requesting discussion here justifying my changes. Most specifically he took issue with the game number issue. The Windows shuffle number generator was already discussed in the above paragraph. While a few popular open source implementations have chosen to use the same shuffling algorithm for consistent game numbers, countless minor implementations do not. The original versions of the game, for example, used a completely different random number generator. See the reference in the article to the discussion with the game creator for that. Also during the earlier edit, I had moved the citation along with the relevant portion of about the game numbers. I kept your additional comment about the numbered game generators, but removed the duplicate ref and also removed the erroneous assertion that the game is strikingly similar to Klondike. About the only similarity is that they use a single deck of cards; consider the completely different implementations of movable tableaus, use of cells, allowing any card into empty cascades, and a radically different opening layout. I also removed a confusing statement about how some games are possibly unbeatable (many are provably unbeatable) and how other games are also unbeatable. The article even describes two such unbeatable games using the -1 and -2 game numbers in Windows. Bwagstaff 04:55, 6 October 2007 (UTC)

I agree that the game is nothing like Klondike (even my edit removed that ridiculous assertion). I still think that the summary should have more information about the nature of the game - most especially the fact that very few games are unsolvable but that for a given implementation, many of the unsolvable games are well known. Dfmclean 17:00, 8 October 2007 (UTC)

Number of possible games
Since only the number of cards distinguishes one column from another, the number of possible games is actually 52!/(4!)2, or about 1.4x1065. Hv 15:23, 12 July 2005 (UTC)

Minor variant
I originally learnt this game slightly differently: the cards are dealt 7 to each of the 8 columns, and the remaining 4 cards are dealt into the cells. This makes the game slightly harder; I don't know if there is a separate name for this variation. Hv 15:24, 12 July 2005 (UTC)
 * No, there is not a separate name. The game was originally a different sized board, with 8 cascades of cards becoming standard later on.  This detail has been added to the article. --Bwagstaff 04:58, 6 October 2007 (UTC)

Missing rule or explanation
In the rule section it is not explaned if or which cards can be placed in an empty column. My understanding is that ANY card can be placed in an empty column, but it could also be only Kings?


 * It is indeed ANY card. --Bonalaw 07:35, 20 September 2005 (UTC)


 * For clarity, would it be useful to add in the rules that foundations are built A to K and that a tableau is built down K to A? Also, I feel that the phrase "For games with the standard layout (four open cells and eight cascades) most games are easily solved. The Windows version article contains a section that discusses unsolved games." should be placed in the introduction and not in the rules, as this is not really a rule. Also, the part of that sentence where it says "easily solved" should be changed to "can be solved". Saying it is easily solved is a matter of opinion, as the average FreeCell player may not find each game that easy. AlanDaTalon6789 (talk) 21:28, 5 October 2010 (UTC)

win32s and best of microsoft entertainment pack
a win16 version of Freecell is included in best of microsoft entertainment pack. Does this pre-date the version that came with win32s. (unsigned by Plugwash 13:31, 14 October 2005)
 * The CIX freecell conference was started on 17th May 1995, with the introduction:
 * This conference has been created to mastermind an attempt to prove Microsoft's assertion that all freecell games are solvable.
 * Comparing against the dates in List of Microsoft Windows versions, this suggests it was first introduced no later than Windows NT 3.5, and in particular before Windows NT 3.51 or Windows 95. Hv 15:54, 14 October 2005 (UTC)
 * The Entertainment Pack Freecell About Box: Freecell/by Jim Horne/Copyright 1991 Microsoft Corp. 24.185.31.111 02:29, 15 October 2007 (UTC)

image
The image on the freecell page isn't freecell, its Solitaire! This should be changed asap because it's completely inaccurate.

thers also this image in the commons

The windows image is copyrighted.Since we have an alternative i think that fair use don't aply.I propose to deleat the windows image.--Pixel ;-) 02:21, 13 August 2006 (UTC)


 * No comment on which to use, but we certainly don't need two images that are basically the same. 2005 02:29, 13 August 2006 (UTC)


 * Originaly i replaced the windows image,but someone reverted me.So i simply added the gpl image as a temporery mesure.If you mean that thers two kde images of freecel in the commons, i'm ok with you, that one can go.--Pixel ;-) 02:33, 13 August 2006 (UTC)
 * That image is ugly, wtf!!! Shandristhe azylean 08:32, 15 August 2006 (UTC)
 * Ugliness is irrelevant. Please see the third pillar of Wikipedia, and read the article on free content.  Fair use images may only be used when no free replacement is possible. —Simetrical (talk • contribs) 19:03, 15 August 2006 (UTC)
 * personally i think we should have images of the windows version since that is the original that the article is primerally about. Of course we should have images of free versions as well both to show that there are other vendors and for the benifit or resuers who reject fair use images. Plugwash 00:44, 17 August 2006 (UTC)
 * What information would our readers gain? It's basically identical to the KDE version except it has the Windows-style GUI and a slightly different background. —Simetrical (talk • contribs) 18:03, 17 August 2006 (UTC)
 * i added a new Vista image. but as i was tired. i realised it was normal solitare... oops... but its pretty! --Yoshi navi t 11:10, 21 March 2007 (UTC)
 * i agree we need each OS's editon... so people dont think certian ones are better. as most people are stoopid :-P --Yoshi navi t 11:12, 21 March 2007 (UTC)

Game 517,776
The following comments were moved from the article page
 * Edit: game 517,776 is most certainly beatable


 * If game 517,776 is beatable, as is claimed, the person making the claim should provide proof of the moves that beat it. See the Freecell website for move-coding conventions.  I have tried the game about 10 times, and regard it as unbeatable.

I found, which posted the list of "unsolveable" games in 2005. Several people doubted it and said they would try to beat them, but there was no reported success.

Matchups 03:29, 13 November 2006 (UTC)

The first 'unbeatable' game is also beatable.

Chinese Windows version of FreeCell

 * Chinese Windows version of freecell is called 新接龍 probably someone can make a page for it? --(Blckbird2002 16:44, 8 February 2007 (UTC))
 * 新接龍 (xīnjiēlóng) literally means "new join dragon". Anthony Appleyard (talk) 09:05, 5 May 2013 (UTC)
 * although it doesn’t matter about the name is appears that by your explanation ( correct me if I’m wrong ) the Chinese ‘version’ of free cell is simply the same premise under a different name so it isn’t nessesary 86.22.145.67 (talk) 18:55, 21 October 2022 (UTC)

-1 and -2
-1 and -2 are actually impossible to complete. 217.44.164.30 (talk) 17:22, 19 February 2008 (UTC)
 * These are Easter eggs. The Windows 7 version added games -3 and -4 which are instantly solvable (all cards can be immediately moved to the foundation). - Mike Rosoft (talk) 20:42, 10 July 2018 (UTC)

Removing Sesame Seeds
Consider what "Joel on Software" says about gradually removing sesame seeds from Hamburger buns and look at this diff to the article from my latest edit to the present. As you can see, a lot of material was mercilessly removed and the article now sucks much more than it did then. I don't know whether this is the work of one editor or a gradual change, but it certainly sucks. For example, the "Solvers" contain one dangling paragraph.

As a result, I think what I'll do now, is retrieve both versions, and merge them manually.

Shlomif (talk) 06:21, 7 May 2008 (UTC)

Difficulty section
I have two issues with the section on difficulty:
 * The section is mostly about Minesweeper, not FreeCell.
 * The section isn't clear about the difficulty level of Minesweeper: From my understanding, it isn't the Minesweeper game itself that has been shown to be provably hard — it is the problem of establishing whether or not a given minesweeper state (map, position, whatever it is called) is consistent. I don't think the section does a good job of making the distinction.    SlowJog (talk) 18:17, 27 May 2008 (UTC)

Dude, where's the article?
The revision history contains a lot of excellent and informative versions of this article. But as the current article stands, it's had so much info stripped out that it's next to useless. I realise that the purpose of the Wikipedia editing process is to reduce the amount of information available to the casual reader, but this article seems to be taking it to extremes. -79.71.214.103 (talk) 07:37, 8 July 2008 (UTC)
 * If you're referring to things like strategies and how to play the game that was removed because it's considered inappropriate for an encyclopedia.Caissa&#39;s DeathAngel (talk) 16:58, 8 July 2008 (UTC)

Scoring your game
If you've play Free Cell for awhile you know, when you remove that last card the game is over. I've devised a way to score my game, just for fun, to see how well I've done. First of all, for simplicity sake, all face cards score a 10. Everything scores what number they are. A score of less than 10 is a good game. An average game is less than 21. Here's how it works: When you get to that last play (before the game is over), manually put all the rows (cascades) together that you can. Most of the time there will be four rows and a fifth row that has the winning move. Now manually move all the cards you can into the matching suits area. When you can't move any more cards in without ending the game, you're done. Count the total of the suits section. That's your score. To see if it's really your score, move that last card. If the other cards go to the suits area, you are correct. If they don't, you need to move the card you missed and the new total becomes your score.

Solvers
I've restored the information about the various programs. Maybe it can be cleaned up a bit, but the links are neither spam nor unencylopedia. It's notable information that programs exist which can solve FreeCell, not only from the player's perspective, but from a programming as well as a game theory perspective. And where to find these programs is useful information. As for the link in the references section, a site showing solutions to various hands is exactly the kind of thing external links are for. The solutions would be too much for the article, but they provide more information on the topic that would be inappropriate in the article itself. PaulGS (talk) 05:43, 3 April 2009 (UTC)
 * I removed the inappropriate links. Please do not attempt to re-add them.  External links do not go in the body of an article.  Also, this is an encyclopedia, not a game guide, or a strategy guide.  Discussion of solvers coule be included, but it need to be done in an encyclopedic way that complies with both WP:V and WP:EL. 2005 (talk) 08:59, 3 April 2009 (UTC)
 * WP:EL says that external links do not normally go in the body of an article, although if a particular piece of software is being discussed, I see no reason not to link to it, or at least to put the link in the reference section. The game guide policy does not mean that discussion of game hints or strategy is never permitted, but the solver section wasn't that, anyway. The policy says: "While Wikipedia has descriptions of people, places and things, a Wikipedia article should not read like a how-to style manual of instructions, advice (legal, medical or otherwise) or suggestions, or contain how-tos. This includes tutorials, walk-throughs, instruction manuals, game guides and recipes." A game guide would include specific strategies for FreeCell, such as playing certain cards first or looking for patterns or that sort of thing, which is not what was there - it contained a list of programs and briefly described each. I've revised the section to eliminate the comments about the programs. PaulGS (talk) 02:55, 4 April 2009 (UTC)
 * Perhaps you don't see a reason, but the guideline does. The links are totally inappropriate.  If you want to add content about the encyclopedic notability of solvers in general, with third party reliable sources as referneces, fine.  But linking to somebody's Tripod page is prohibited by the guidelines.  This is an encyclopedia about notable subjects; not a listing of anything that exists. 2005 (talk) 05:26, 4 April 2009 (UTC)
 * So even some basic strategy such as, "Before moving any cards it is often a good idea to look for trouble spots, such as Aces as the bottom card of a tableau or Kings as the top card of a tableau." is not allowed? I was looking to edit this page for strategy as a project for school, but that does not seem likely based on what I have read. AlanDaTalon6789 (talk) 21:22, 5 October 2010 (UTC)

Using real cards

 * How often do people play freecell with real cards, not on a computer? On one holiday in the evenings with nothing else to do I sometimes played freecell with real cards. Anthony Appleyard (talk) 17:12, 6 November 2009 (UTC)
 * A study in 1997 showed that approximately every two hours 53 minutes someone plays freecell with real cards. This is an average of course (it's more often when it's night time in the US, and to a lesser extent in Europe.)  Is this the sort of fact we should add to the article? FreezEll (talk) 17:12, 6 November 2011 (UTC)  — Preceding unsigned comment added by 212.139.227.50 (talk)

Merge from FreeCell (Windows)?
It would make sense, it seems: this article discusses the Windows version, while the Windows version discusses general stuff (e.g. solving) that actually belongs here, so there is a large overlap. On top of that Windows article is tagged for cleanup. GregorB (talk) 16:44, 17 August 2010 (UTC)
 * That would be a good idea, as long as information isn't lost in the merge. 216.80.135.70 (talk) 02:59, 19 August 2010 (UTC)
 * That's right - nothing should be lost (apart from dubious, unsourcable stuff perhaps). I'm volunteering to do it, but right now I'll wait and see if there are any objections. GregorB (talk) 07:52, 19 August 2010 (UTC)
 * Agree it makes sense. If there is anything that relates to Windows version only this can be noted in brackets.  —Preceding unsigned comment added by 87.113.53.98 (talk) 21:09, 29 August 2010 (UTC)
 * I would be more than happy to merge the two articles together. I am taking a technical writing class and we have a project to edit Wiki and make it better. AlanDaTalon6789 (talk) 21:30, 5 October 2010 (UTC)
 * I was about to do it - as soon as I get a little free time - but if you want to do it go ahead. You are a new editor, so I must warn you that merging is (sometimes) not easy, and you might want to consult Help:Merging first. I'll probably join you in the process, don't worry about the end result. In case you need help, ask here or in my talk page. GregorB (talk) 21:40, 5 October 2010 (UTC)
 * OK, I'm starting with the merge now. GregorB (talk) 21:11, 20 December 2010 (UTC)
 * Done. (Phew, that was quicker than I thought it would be...) I'll tweak the article some more this week. GregorB (talk) 21:49, 20 December 2010 (UTC)
 * I agree with that idea. My idea of how to merge them is so long that I put in a new section titled 'Incomplete set of rules[.]' Blackbombchu (talk) 00:12, 2 January 2014 (UTC)


 * Hi. It is a good idea. I advise you to act boldly and do it now. As for the unreferenced info, merge it or not, I might drop by and delete. After all unreferenced info may be challenged or removed. Best regards, Codename Lisa (talk) 00:28, 2 January 2014 (UTC)
 * Are you asking me or GregorB to make the edit? I'm not at all able to merge 2 articles myself and even if I knew how, I would not at all know how to organize the article after merging the 2. All I'm able to do make the edit to the 'Rules' section. If you meant me, it's highly inappropriate to ask me to do the merging myself when I'm not an administrator and so I probably don't have the right to do it myself any more than I have the right to delete an article that was nominated for deletion. Blackbombchu (talk) 00:57, 2 January 2014 (UTC)
 * Hi.


 * If you had paid attention to my message indentation, you'd have realized that I am talking to, who also volunteered the merge in a next message.


 * Administrative permission is not required for merging. Merged pages are not deleted. Only their contents are replaced with:
 * #REDIRECT FreeCell
 * Their history must be kept, per GFDL and CC-BY-SA 3.0 licensing terms.


 * Best regards,
 * Codename Lisa (talk) 01:57, 2 January 2014 (UTC)


 * I'd support the merge for more or less the same reasons that applied back in 2010. However, I'm semi-retired as of late so I'm going to take a pass this time, I'm afraid. Shouldn't be too difficult to do, though. GregorB (talk) 18:16, 21 January 2014 (UTC)

Title in italics
The article title (unlike that of this talk page) is in italics. I'd do a Move, but I see no provision to deitalicise... Rothorpe (talk) 01:38, 21 December 2010 (UTC)

unsolvable example
Here is an example which does not allow any building of the stacks. I do not know what the corresponding game number is, since it was built from scratch. Only two piles are shown because all 6 row columns and all 7 row columns have the same values. The nine 4-card combinations shown offer the most variety of playable cards. This is intended to prove all games are not solvable.

Phyti (talk) 02:15, 21 December 2010 (UTC)phyti

Another unwinnable example, perhaps even easier to analyze. The suits are immaterial. The only available moves are to the free spaces, and any four of them lead to an unplayable position. -- Elphion (talk) 10:47, 9 October 2018 (UTC)

A A  A  A   3  3  3  3 5 5  5  5   7  7  7  7   9  9  9  9   J  J  J  J   K  K  K  K   Q  Q  Q  Q  10 10 10 10   8  8  8  8 6 6  6  6   4  4  4  4   2  2  2  2

Freecell is NP-complete, but...
"This generalized version of the game is NP-complete; it is unlikely that any algorithm more efficient than a brute-force search exists that can find solutions for arbitrary generalized FreeCell configurations."

This is a very simplistic (read: wrong) view of NP-completeness. I suggest replacing it by: "This generalized version of the game is NP-complete; thus, it is unlikely that a fast algorithm exists that can quickly find a solution for arbitrary generalized FreeCell configurations." — Preceding unsigned comment added by 130.225.0.242 (talk) 01:15, 28 October 2011 (UTC)

Complexity
The article states: "The FreeCell game has a constant number of cards. This implies that in constant time, a person or computer could list all of the possible moves from a given start configuration and discover a winning set of moves or, assuming the game cannot be solved, the lack thereof."

Well, it appears finding either a solution or the lack thereof has nothing to do with the number of cards. It is a matter of the number of possible moves until it is proven that no solution exists. If a solution does exist it only requires the number of moves to the first solution. The latter clearly varies and I am quite sure the former does as well. In other words, I don't believe all possible hands have the same total number of possible moves. Therefore this statement would appear to be wrong. The constant number of cards should not imply constant time at all.

Am I missing something? is there a proof? Someone with more inside knowledge on the mathematics of this game might want to check this and correct as necessary. — Preceding unsigned comment added by 76.126.236.18 (talk) 16:59, 28 April 2013 (UTC)


 * The point is that the fixed number of cards means that there is a constant upper bound on the number of moves, regardless of how the cards are laid out. Thus, listing the possible moves also has a constant (albeit large) upper bound.  Determining the answer for a particular game may take less time, but is always bounded by that constant time.  That's what it means to be a constant time algorithm. -- Elphion (talk) 13:19, 24 February 2019 (UTC)

Incomplete set of rules
Wikipedia is supposed to be understandable to everybody. This article isn't supposed to assume that everybody reading it has played FreeCell before with the playing of that game filling in the missing in the missing information about the rules. It is proposed that FreeCell (Windows) be merged into FreeCell I think the Rules section should go as follows, possibly slightly different:

Rules
Construction and layout:
 * One standard 52-card deck is used.
 * There are four open cells and four open foundations. Some alternate rules use between one to ten cells.
 * Cards are dealt into eight cascades, four of which comprise seven cards and four of which comprise six. Some alternate rules will use between four to ten cascades.

Building during play:
 * A tableau is a string of cards going all the way from king to ace going up in ascending order and alternating the colour of the suit.

Moves: A move must be of one of the following forms:
 * Moving any single card from the bottom of a cascade or from a cell to an empty cell, empty cascade or the bottom of a cascade whose bottom card is the number one higher and the opposite colour suit. Moving a card from a foundation is not allowed.
 * Moving any stack of cards from the bottom of a cascade to the bottom of another cascade or into an empty cascade under the following conditions (all moves of this form can be broken down into one card moves):
 * The stack is a piece of a tableau.
 * If it's moved to a nonempty cascade, the number of the card at the top of the moving stack must be one less than the number of the bottom card of the cascade it moves to and the opposite colour suit.
 * The number of spaces that are either empty cells or empty cascades both before and after the move must be at least as large as one less than the number of cards in the stack getting moved.


 * Moving an ace to an empty foundation or moving another type of card into into a foundation that's one higher than the number of and the same suit as the card that was previously placed there. The card that was most recently placed into the foundation is the one that shows there.

Victory:
 * The game is won after all cards are moved into a foundation.

For games with the standard layout (four open cells and eight cascades) most games are easily solved, but it is easy to construct examples that cannot be solved. Various computer versions of Freecell occasionally deal unsolvable games.

This article should have another section near the end of the article that goes as follows:

Windows version
The Windows version is like the original version except for the following: Blackbombchu (talk) 23:36, 1 January 2014 (UTC)
 * It's done in a program on the monitor instead of using real cards.
 * All moves can be undone.
 * Unless it's the beginning of the game, every time there is a move of moving a card to a foundation, that move gets made automatically (automatic moves can also be undone).
 * When ever someone has no legal moves, the program offers them the option of closing the game, restarting the game with the same card layout, or starting a new game with the cards dealt randomly.
 * When ever somebody clicks the closing icon of the program, it offers them two choices of either saving the game or not saving it.


 * Those points are true but are not rules, merely automated continuances of the rule set to streamline gameplay. No need for inclusion. — Huntster (t @ c) 01:16, 2 January 2014 (UTC)
 * Maybe some of the bullet points in this section are worth including and others not. The 'Rules' section is very unclear about what moves are allowed and wht ones aren't. Since you seem to know better than me about which rules are worth including and which ones aren't, I suggest that you edit the 'Rules' section into the way that seems best to you. Blackbombchu (talk) 15:27, 2 January 2014 (UTC)
 * Did you just mean all moves I described except for the ones that move multiple cards at once are real rules? Should I edit the article in almost the same way as I suggested except for stating the rule about moving multiple cards in the 'Windows version' section after I create that section? Blackbombchu (talk) 15:33, 2 January 2014 (UTC)
 * Wow, sorry, I didn't look carefully enough and thought these were two different sections. I'm only referring to the Windows material, since those are merely automated actions which follow existing card rules. I wasn't expressing any opinion on what you wrote in the main section above that. Sorry for the misunderstanding. — Huntster (t @ c) 06:56, 3 January 2014 (UTC)

Solvers Section
I propose removing the text about the specific solvers mentioned in the article as I did here. The section does not have a reliable secondary source. It appears to be original research using primary sources on a couple solvers which do not make verifiable claims of notability to make them suitable for inclusion in this encyclopedia article. ParacusForward (talk) 05:17, 7 January 2014 (UTC)

Offtopic: Under Two Minutes?
One of the "Additional Sources", the "Four Virtues of FreeCell" article in Escapist Magazine, claims that: "Many games last less than two minutes, but those with difficult patches (or "knots") can slow down considerably."

That's just crazy fast; I think that even if the cards are in sorted order such that they can be moved from the piles straight to the foundation, I would not be able to finish in two minutes, using a physical deck of cards.

I've never played a computerized game faster than 6 minutes, and that's with automation of moves to the foundation, as well as automation of moves of card sequences from one column to another. 216.31.219.19 (talk) 00:57, 24 February 2016 (UTC)
 * My best computerized time is 1:23, and I haven't even worked at trying to go fast, so under 2 minutes seems quite plausible. I just tied one, it took 2:01, and I'm rusty.-- S Philbrick (Talk)  17:39, 8 March 2016 (UTC)

Moving cards from foundations? Undoing moves?
Under (1) standard rules and (2) the Windows version, is it allowed: (A) to move a card from a foundation to a tableau (other than undoing an automated move in the Windows version); or (B) after undoing one move, to then undo the previous move, then the move before that and so on? Or can you only undo the last move and no more (again, other than undoing automated moves)? ☺ Coppertwig (talk) 21:45, 12 March 2022 (UTC)
 * Depends on the version you're playing. E.g., Windows 7 version allows multiple undo's back to the beginning layout.  No version I'm aware of allows you to play from a foundation (although you can undo playing to a foundation). -- Elphion (talk) 23:43, 30 November 2022 (UTC)

Negative games
In the older Windows FreeCell you could enter a negative game number (-1) which would produce an unsolvable game. 108.147.192.20 (talk) 20:35, 30 October 2023 (UTC)