10 Nov
2003
10 Nov
'03
4:08 p.m.
At 04:25 PM 11/10/2003, Dan Hoey wrote:
By the way, 16!/2 milliseconds is "only" about 331.5 years, and I'd guess the time taken to examine a position is probably more like a few dozen microseconds than a millisecond. But it's still cpu-years.
I think it might take longer than that per position. In the reference for the 4x4 case, http://citeseer.nj.nec.com/cache/papers/cs/4003/http:zSzzSzwww.cs.ucla.eduzS... they found the optimal solution of 10 random starting configurations, and each of them generated over 8 billion nodes. Three of the 10 generated over a trillion nodes. Of course, there are probably ways to cut that down, and someone could come up with an insight into the problem.