chiark / gitweb /
Having played Keen a bit following the clue-generation fix in r9165,
authorSimon Tatham <anakin@pobox.com>
Sat, 7 May 2011 13:22:17 +0000 (13:22 +0000)
committerSimon Tatham <anakin@pobox.com>
Sat, 7 May 2011 13:22:17 +0000 (13:22 +0000)
commitbd539ae8707422d82955a4838585e17fd01c6975
treed336821f45f620c4c1d89b7005a082677618b583
parent4a172274f2abde1c2ad23ba46b183e1d767fb902
Having played Keen a bit following the clue-generation fix in r9165,
I've decided that the extremely low density of one-option
multiplication clues is not a universally good idea after all: it
seems to me to make puzzles _quantitatively_ harder, even if Keen's
difficulty-level system can't see any difference in the set of modes
of reasoning required at least once to solve the grid.

So I've readjusted the clue selection, so that multiplicative clues
with only one workable pair of factors are restored to 'good' status
at Normal difficulty level and below, and only considered less-than-
fully-desirable at Hard and above. I think that's a reasonable
compromise.

[originally from svn r9170]
[r9165 == e7b2a9dd8d8915802fb69ce2242b1f913b7f3172]
keen.c