A lot of recent sudoku news has focused on the
reported proof of 17 being the minimal number of clues needed to specify a unique classic sudoku solution. I haven't had time to look too deeply into the proof yet but at first read it seems to be one of the better pieces of sudoku scholarship to come out since the full counting of solutions done by
(
Read more... )
Comments 9
Cheers
Rob
Reply
Reply
Reply
As I remember udosuk managed to do a 10-cager following your rules.
I do remember that he did a single cager with constraints - although I cannot recall which constraints he used.
HATMAN
Reply
I forgot to add:
good puzzle; and
congratulations on your victory.
HATMAN
Reply
some of you may know me as 999_springs on the new player's forums. i'm Zapmeister on reddit, if you want proof i'll state my username in a comment there. i'd just like to point out
i win
http://redd.it/10az6m
Reply
some of you may know me as 999_springs on the new player's forums. i'm Zapmeister on reddit, if you want proof i'll state my username in a comment there. i'd just like to point out
i win
ok your blog seems to be flagging links up as spam, which is totally unneccessary, but here's the url:
http and then redd.it and then 10az6m
Reply
x........
xxxx.....
xxxxxxx..
xx.......
xxxxx....
xxxxxxxx.
xxx......
xxxxxx...
xxxxxxxxx
x adds up to 165
one cage ;) :P
if you don't allow repeats you can split the cage into 9 in the obvious way and discard one of them to get an 8 cage killer, but that's boring
Reply
Your novel cage design reminds me of the same kind of property I exploited here. So a neat example of using the canonical grid with min (or max) sums.
Reply
Leave a comment