I hope you will run the single long tournament of 2000 games rather than 10 tournaments of 200 games. Because the results will then be more realistic and easy-to-follow.
As I said from the beginning I'm testing these chess engines. As has mounted another PC, you will devote exclusively to play a tournament with many more rounds, not just one (165 games) but with about 10 rounds per tournament, 1650 games. But I'm only on this project and I can devote my PC's only this. I suggested a collaboration to develop more realistic test, although I processor, ram, motherboard prepared but lacking the other components. And we all know that the crisis can not do much more.
Thanks for your words.
Abril first day of officially will begin league UCI chess engines, with participants appearing on the list. The book used for all engines will be generic, open learning strongbook2010.ctg, permanent brain off, 128 hash, 16Mb cache size egtbs = 3,4,5,6 (Total Nalimov I) the configuration of the chess engine is by default . He played a total of 2,000 items per week, and will be posted on my website. I hope that these tests can serve you to improve your opening books for the game mode blitz, 3+0
after playing several test between chess engines, and changing in some test participants in this experimental duel in which not all chess engines have played the same items and therefore the test should only be taken as anecdotal, Rybka is which has added more points in the tests performed. However, given that most free chess engines UCi monopolize the top positions, I see no logical reason to buy a commercial product, when there is a free one with a similar playing strength. At least as shown by the test in blitz. Another major adventure would be to test a long time controls, perhaps there may be a greater difference between the engines and free trade. While experience tells me that there will be such a difference to justify spending $ 100.