This is the strange story of a riddle.

In March 2001, on the mailing list of GLIA, Olen Netteburg gave the following riddle (Olen told me later he had seen it in an old - about 1975 - scientific popular magazine) : On a chessboard 5x5, how to put five white queens and three black queens such as none queen can be taken in one move by an opposite colour queen?

I solved it in about 15 minutes, without great difficulty.

Up to now, I gave the problem to 48 chess players (with a time of 20 minutes).

The results are incredible :

- 13 international masters and international grandmasters : only one found the solution.

- 15 experts and masters (including myself) : none of the others found the solution.

- 20 "average chess players" : two found the solution.

Four of the experts asked for more time and solved the problem in about one hour.

This result is very astonishing : One could say that this problem has nothing to do with what happens in a normal chess game. Or, in retroanalyse (in chess, it means - by logical deductions - to find the last (or several lasts) move(s) that have been played), which is a very difficult field, and has nothing to do with a normal chess game, there is a strong correlation between the ability to solve problems and the strength of the player. For the 5+3 problem, the correlation looks very low!

Among non chess players, at Glia level, I think (I did not have a sufficient large sample to assert this) about 50 % would solve the riddle.

This is of course only one item, but the question of the relation between I.Q. and chess strength comes up again.