Certain simulation gets in an very slow loop

[michiguel]

{ --> 1 When this file "simulated_0092.pgn" is executed like this ordo -p simulated_0092.pgn -w30 -u3 -D -s100 it hangs in the first simulation

--> 2 if it is executed like this ordo -p simulated_0092.pgn -W -D

It gets stuck for a several seconds after phase 2 Deviation increases suddenly too much:

phase iteration deviation resolution 0 6 6.099335035 13.00241 1 1 34.018978935 11.76903 2 117 0.000001841 0.00000 3 1 10.744243516 0.00000 4 21 0.000000051 0.00000 done

}

[White "1"] [Black "4"] [Result "1/2-1/2"] 1/2-1/2

[White "1"] [Black "2"] [Result "1/2-1/2"] 1/2-1/2

[White "1"] [Black "3"] [Result "1/2-1/2"] 1/2-1/2

[White "4"] [Black "1"] [Result "1-0"] 1-0

[White "4"] [Black "2"] [Result "1-0"] 1-0

[White "4"] [Black "3"] [Result "1-0"] 1-0

[White "2"] [Black "1"] [Result "1-0"] 1-0

[White "2"] [Black "4"] [Result "0-1"] 0-1

[White "2"] [Black "3"] [Result "0-1"] 0-1

[White "3"] [Black "1"] [Result "1-0"] 1-0

[White "3"] [Black "4"] [Result "1/2-1/2"] 1/2-1/2

[White "3"] [Black "2"] [Result "1-0"] 1-0

[michiguel]

Fixed by v0.9.9.5-wa3-cln1-wdl6 1) formula for probabilities of W, D, L lack accuracy and returned a slightly negative number rather than zero 2) deltas, when search to adjust white advantage did not expand fast enough