Page 1 of 1

EGTB probing bug(Aristarch,Hagrid,Adam,JanWillem etc.)?

PostPosted: 22 Jun 2005, 08:18
by George Lyapko
After finishing my Rapid tournament and tinkering little bit with EPD2WB I decided to start again testing engines on MES testsuite. Engines which have EGTB support will be tested 2 times (with and without Nalimov's help).
One of the first positions (MES.0002) which is simple KPKP position gave some interesting results. Some engines with EGTB support can't solve it!
Here is the position:

8/8/8/2p5/8/8/1kP5/3K4 w - - bm c4;

and here is the list of such engines:
SOS9911, Aristarch450, Jonny275, Hagrid0756, EnginMax511c, Adam29, Ant2005k, JanWillem113, Freyr1067, FireFly131, KKFChess266, Tytan515, Madeleine02, Mustang292.

Of course I have all 4-men EGTB. Other engines with EGTB support find the solution easily.

Best regards,
George

Re: EGTB probing bug(Aristarch,Hagrid,Adam,JanWillem etc.)?

PostPosted: 22 Jun 2005, 18:38
by Heiner Marxen
George Lyapko wrote:Here is the position:

8/8/8/2p5/8/8/1kP5/3K4 w - - bm c4;



My Nalimov tables tell me that this is a draw. More exactly:
Code: Select all
EGTB value = 1 = draw
    c3       --> [ 31] #14
    c4     ! --> [  1] draw
    Ke1      --> [ 31] #14
    Kd2      --> [ 33] #15
    Ke2      --> [ 31] #14

Is that correct for you?

Cheers, Heiner