Nice question...the hint made me solve it...First better to find individual probabilities of sum of cubes roled, this one will be,
2 : (1, 1);
3 : (1, 2),(2, 1);
4 : (1, 3),(2, 2),(3, 1);
5 : (1, 4),(2, 3),(3, 2),(4, 1);
6 : (1, 5),(2, 4),(3, 3),(4, 2),(5, 1);
7 : (1, 6),(2, 5),(3, 4),(4, 3),(5, 2),(6, 1);
8 : (2, 6),(3, 5),(4, 4),(5, 3),(6, 2);
9 : (3, 6),(4, 5),(5, 4),(6, 3);
10 : (4, 6),(5, 5),(6, 4);
11 : (5, 6),(6, 5);
12 : (6, 6)
Now direct chances of winning are 7 and 11 and P(Direct Win) = 8/36 and P(Direct loss) = 4/36 (i.e case 2,3 or 12). Now we need to consider cases (4,5,6,8,9,10) where we keep on playing until we get the same event before a 7.
We know P(4) = 3/36 . Suppose 'n' be no of rolls made. So if I get 4 in the first roll next (n-2) rolls should be anything other than roll of 7 until we get 4 again.Let i is any of (4,5,6,8,9,10).
P(Ei,n) = P(i)^2 (30 − Pi)^(n−2) /36^n .Now take summation from n = 2 to $infty$. you will get G.P series kind of sum.Ultimately you will get.
P(i)^2 / 36(6 + P(i)) .Now add the probabilities for each case (i.e (4,5,6,8,9,10)) along with P(Direct win).
Please pardon my writing, not yet well versed inserting symbols in answers.My answer coming as 0.49 almost.