I don't think that you can make any conclusions based on the results you report. Will you also post a smiley based on a single game, if the stronger engine did not win it?eriq wrote:Just a small tourney to test arena, but very surprising as to the results I used little arena book for all.
1: GrapeFruit1beta 4.0/6 ·· =1 0= 11 9.50
2: Spike 1.2 4.0/6 =0 ·· 1= 11 8.50
3: Rybka 2.2 32 bit 3.0/6 1= 0= ·· 10 9.00
4: ProDeo 1.5 1.0/6 00 00 01 ·· 3.00
12 games played / Tournament is finished
Name of the tournament: Arena tournament
Site/ Country: OSIRIS, United States
Level: Blitz 5/0
Hardware: AMD Turion(tm) 64 X2 Mobile Technology TL-60 with 2,944 MB Memory
Operating system: Microsoft Windows Vista Home Edition Service Pack 1 (Build 6001)
If you want to test engines as a "finished product" you should use engines with their own opening books. If you want to understand how strong the engines are by themselves, you need to use a much shorter general book (limited to 7 or 8 moves per side, or at most 12). If you test "Arena 2.0" as a finished product, then reporting engine-engine results is irrelevant?eriq wrote:Yes in fact I watched most of the games and was very disappointed in the
opening book, however this was not a test of engines, so much as, a test of a so-called finished product (Arena 2.0)
Best,
Kirill