Featured Match: Same IvanHoe Version- Different Compile!!

[geots]

Last results were from IvanHoe 46f compiled by Peter Pan. Today, we have the results from IvanHoe 46f compiled by KLO.

Of course the results will not tell you that one of the guys is a better compiler than the other. It only gives you the results from one released version. For me, the value in the comparison would be that I would compare the 2 compiles, and probably download and use the one that was strongest in the matches ag. Rybka 4.1. Remembering that most compiles of the same engine are generally going to give similar results.

Only deviation here from the norm is that Peter Pan is sometimes
a little more of a risk taker. So his compiles may vary more (+ or -) than KLOs, who is more of a tried and true compiler. But keep in mind that Peter Pan and KLO are both very good, and you cannot go wrong with either.

Yesterday Peter Pan's results- today, KLO's results:


12/10/2011

1CPU/32bit
128MB hash
Bases=NONE (Rybka 4.1 used Nalimov)
Ponder_Learning=OFF
Perfect 12.32 book w/12move limit
40/3 Repeating
Match=50 games


IvanHoe 9.46f........0 +14/-14/=22 50.00% 25.0/50
Rybka 4.1..............0 +14/-14/=22 50.00% 25.0/50



This was even all the way, tho if IvanHoe had squeezed out a win in game 50, it would have been a 2 point difference- and much closer to Peter Pan's results.

PGNs available as always at my email address.

Would love to stay and talk, but my new dual core 64 bit system came in last night- and I gotta set it up and network it.


Best,

george.