The game FreeCell is installed on most windows OS’s, at least recently. In the help menu it says “It is believed (although not proven) that every game is winnable.” Has this been proven or disproven? Seems to me that knowledge of the game’s RNG might give some insight as to whether or not the game will generate an “unwinnable” game.
I consider myself to be a very good player, yet I have come across a few games that have presented seemingly impossible game scenarios. An example of which might be, all of the lowest cards being at the top, and the higher ups cascading beneath them. I have never seen such an extreme scenario, but I’ve seen close to it.
So, is the game designed to give some very hard scenarios with all of them concievably winnable, or could it possibly generate an unwinnable one?
I think that people with knowledge of card stats in this situation, or maybe people with knowledge of the game code might know the answer.
Yes, it’s just a silly game, but it presents a potentially provable answer.
Free cell will only deal one of 32,000 games, out of a total number which is significantly larger than that (whatever 52! works out to be). Out of those, only one game is not winnable. This happens to be game number 11982 if you want to try it for yourself.
Oh, I forgot. In Windows XP, the number of deals was increased to a million, and 8 of those are unsolvable. I don’t remember the numbers offhand, though.
I don’t know if it works in XP, but we did it at school. You can actually type in game number -1 and it’s unsolvable… there may be a -2, but I don’t remember…
So the older version with 32 000 deals has about 3.96735978 × 10[sup]-62[/sup]% of all possible deals, and the new one with a million has about 1.23979993 × 10[sup]-60[/sup]% of them. Of course, many of them (I have absolutely no idea how many) are trivially easy, obviously impossible, or essentially identical. For example, swap all the clubs with their corresponding spades, or switch the entire first column with the entire second column.
The Freecell maniacs out there (see links given) have programs to run thru a game to determine if there is a win or not. Not all games have been tested, but the famous ones have been.
So the possible answers are: winnable, not winnable, unknown. The first two can be believed.
Does not make any difference to me. I have actually worked 8166 of the games starting with #1 and going in order and not jumping around nor skipping, just do it until I get it and move to the next. It is a mind thing like playing chess or bridge. See how fast I can do it. or put up my own requirements, like not using all the ‘free cells’ or some such.
so, if running a program to see if a win is possible or not does nada for me. YMMV Some games are kind of tough for me but I am having fun. A game ya know. I have looked at 11982 and I think it won’t fly. No problem, I got 31,999 I can work with.
