[seqfan] Number of squares that can be obscured on Rubiks's cube n × n × n without making any states indistinguishable

[Vladimir Reshetnikov]

Dear SeqFans,

I asked a question at M.SE several days ago:
http://math.stackexchange.com/questions/1694790/obscuring-squares-of-rubiks-cube

What maximal number of squares in Rubik's cube n × n × n can be obscured
without making any states indistinguishable?

Apparently, the answer depends on n and could be made into a (hopefully)
interesting sequence. I'm asking for your help in solving this problem.

--
Thanks
Vladimir Reshetnikov