On 6/1/07, Kevin Leeds <[EMAIL PROTECTED]> wrote: > Some possibly interesting info: > > WORD COUNT > > Beyond length 2 and 3, > there are more words with R,U in them than with N,U in them. > > From Zyzzyva searches > > Length, RU* anagrams, NU* anagrams: > 3, 16, 18 > 4, 136, 105 > 5, 471, 336 > 6, 1230, 955 > 7, 2519, 2051 > 8, 3683, 3209 > Total, 24378, 23851 > (total is length 2-15) > > However this doesn't consider the average score of the words, or > probabilities - only counts them
Encouraged by Russell Honeybun, I've resurrected my old code from my 2001 experiment with global analysis and this time at the least I am committed to turning out a useful analysis tool even if I don't get as far as building it into a player. (The tool would enumerate all the next possible plays and give exact figures as to the value of the rackleave in the context of the potential next play) I mention that because last time we discussed the subject (on wgp) we spent some time working out the math for handling probabilities of partial rack draws and actually implemented some code to calculate them. The code is still online and you are welcome to take it if you think you can usefully incorporate it into Zyzzyva (or Quackle). It's here: http://www.gtoal.com/wordgames/scrabble_solver/fixedprob.c.html I think it can be speeded up for repeated use by making the nCr function into a memo function at the cost of a 100*7 array. (Or even just a const array) (For a text version drop the .html from the url) It's been some time since I looked at this but I think that this is the same code: http://www.gtoal.com/wordgames/scrabble_solver/global_analysis/probfn.c.html - a quick diff suggests that the only changes are checking for an invalid set of tiles in the bag, plus some testing code. Graham
