Not logged inRybka Chess Community Forum
Up Topic Rybka Support & Discussion / Rybka Discussion / Rybka 2.3.2a unable to solve a simple problem
- - By JhorAVi (***) Date 2007-08-02 19:53
1.. Rxb2! is an obvious win for back. Even none masters can spot it.

Shredder 10 and Deep Junior 10 solved it in around 1 minute!

But Rybka 2.3.2a didn't see it even after 1hour until I gave up.

Maybe I just wrongly configured Rybka in my PIV 2.6mhz 1Gig RAM.    Can you try it in your system??

8/pR4pk/1b6/2p5/N1p5/8/PP1r2PP/6K1 b - - 0 2


SOLUTION 1..Rxb2!
Parent - - By herakleitos (***) Date 2007-08-02 20:00 Edited 2007-08-02 20:08
Hiarcs finds it in about a half minute (3200+ sp). Fritz, Rybka, Shredder, and Toga do not find it within 5 minutes (I didn't wait around longer).

Check. WinFinder spots it in 20 seconds. And Naum 2.0 (!) finds it in 30 seconds. Interesting.
Parent - - By Tester (*) Date 2007-08-03 01:48 Edited 2007-08-03 02:36
Some results on my computer with a similar CPU:
Rybka 2.3.2a - 90 seconds
Ktulu 8 - 60 seconds
Zap Zanzibar - 60 seconds
Shredder 10.1 - 40 seconds
Junior 10 - 40 seconds
Naum 2.0 - 30 seconds
Alaric 7.07 - 20 seconds
Glaurung 2e5 - 15 seconds
Yace Paderborn - 10 seconds
Rybka Winfinder - 4 seconds
Chess Tiger 2007 - 4 seconds
Spike 1.2 - 2 seconds!!
Parent - - By JhorAVi (***) Date 2007-08-03 02:26
How about Rybka 2.3.2a on your system? Did it find the solution too? How long?

thanx..
Parent - - By Tester (*) Date 2007-08-03 02:30 Edited 2007-08-03 02:35
Hmm, yes Rybka 2.3.2a found it in 90 seconds, guess I was just delusional the first time around. :)
Parent - - By BigBen (****) Date 2007-08-03 07:36
Hi,
    Useing chessbase 9 Rybka 2.3.2a sees it in under a minute and Hiarcs  11.2 in 4 seconds!!

Regards
Parent - - By Tester (*) Date 2007-08-03 18:11
With what hardware (CPU)?
Parent - By BigBen (****) Date 2007-08-04 07:47
Hi,
    Intel quad at 4Ghz on the P35 chipset

Regards
Parent - By hebroots (**) Date 2007-08-06 16:41
Looks like 2 issues; one issue is, did the engine "find" the solution?, and the other is "how long did it take?".

Looks like the first point had to do with Rybka not finding the solution in a reasonable amount of time.  Chessmaster 9000 did find that 1....Rxb2 was the turning point for Black.  It took 8 min 40 sec, but being connected to the Internet at the same time slows it down a bit.  Some settings on Chessmaster could have been tweaked to maybe get a faster solution, but I am satisfied that it at least got the answer on a vintage 2002 program (lol).
Parent - By George Tsavdaris (****) Date 2007-08-04 00:10

>Fritz, Rybka, Shredder, and Toga


What versions of these and in what hardware?

My Fritz 8.00.23 finds this in 36 seconds in a AMD 3800+ 2.2 GHz.
Also Shredder 10 finds this in 37 seconds.
Also Toga II 1.3x4 finds this in 3:57 minutes

Rybka default has indeed problems with this.....
Parent - - By sylvain.renard.27 (**) Date 2007-08-03 22:46
Hello, how much memory did you use for hashtables? With 1 Gig of RAM, 512 Mb should be OK.
I got this result :

CPU0: GenuineIntel x86 Family 6 Model 14 Stepping 8 1828 MHz
CPU1: GenuineIntel x86 Family 6 Model 14 Stepping 8 1828 MHz
Engine: Rybka 2.3.2a 32-bit (512 MB)
by Vasik Rajlich

UCI_AnalyseMode set to true
  5.00   0:00   +0.38   1...c3 2.bxc3 (840) 26
  5.00   0:00   +0.24   1...Bd8 2.h3 (2.224) 35
  6.00   0:00   +0.10   1...Bd8 2.h3 Bf6 (3.476) 45
  7.00   0:00   +0.07   1...Bd8 2.g3 Bf6 3.h4 (7.667) 55
  8.00   0:00   +0.02   1...Bd8 2.g3 Bf6 3.h3 Kg6 (14.944) 60
  9.00   0:00   -0.04   1...Bd8 2.g3 Bf6 3.h3 Kg6 4.Kf1 (24.220) 65
10.00   0:00   -0.02   1...Bd8 2.g3 Bf6 3.h3 Kg6 4.Kf1 Kf5 (36.972) 71
11.00   0:00   -0.06   1...Bd8 2.g3 Bf6 3.h3 Kg6 4.Kf1 Kf5 5.Rb5 (63.446) 75
12.00   0:02   -0.14   1...Bd8 2.Kf1 Bf6 3.h4 Kg6 4.g3 Be5 5.g4 Rh2 (171.557) 75
13.01   0:03   -0.19   1...Bd8 2.Kf1 Bf6 3.h3 Kg6 4.Kg1 Be5 5.Kf1 Bd4 6.g4 Rf2+ 7.Ke1 (271.683) 78
14.01   0:04   -0.19   1...Bd8 2.Kf1 Bf6 3.h3 Kg6 4.Kg1 Be5 5.Kf1 Bd4 6.g4 Rf2+ 7.Ke1 Rh2 (387.352) 81
15.01   0:08   -0.16   1...Bd8 2.Kf1 Bf6 3.Nxc5 Rxb2 4.Rxb2 Bxb2 5.Ke2 Kg6 6.Ne4 Be5 7.Nd2 Bxh2 8.Nxc4 (659.370) 81
16.01   0:11   -0.11   1...Bd8 2.Kf1 Bf6 3.Nxc5 Rxb2 4.Rxb2 Bxb2 5.Ke2 Kg6 6.Ne4 Be5 7.Nd2 Bxh2 8.Nxc4 (943.574) 81
17.01   0:17   -0.12   1...Bd8 2.Kf1 Bf6 3.Nxc5 Rxb2 4.Rxb2 Bxb2 5.Ke2 Kg6 6.Ne4 Be5 7.Nd2 Bxh2 8.Nxc4 (1.416.643) 83
18.01   0:26   -0.12   1...Bd8 2.Kf1 Bf6 3.Nxc5 Rxb2 4.Rxb2 Bxb2 5.Ke2 Kg6 6.Ne4 Be5 7.Nd2 Bxh2 8.Nxc4 (2.158.694) 83
19.01   0:48   -0.04   1...Bd8 2.Kf1 Bf6 3.Nxc5 Rxb2 4.Rxb2 Bxb2 5.Ke2 Kg6 6.Nb7 Bc3 7.Nd6 Be5 8.Nxc4 (3.731.796) 79
20.01   1:24   -0.03   1...Bd8 2.Kf1 Bf6 3.Nxc5 Rxb2 4.Rxb2 Bxb2 5.Ke2 Kg6 6.Nb7 Bc3 7.Nd6 Be5 8.Nxc4 (6.243.738) 75  TB:2
20.04   2:22   -2.54   1...Rxb2 2.Nxb2 c3 3.Rxb6 c4 4.Rb4 a5 5.Nd3 cxd3 6.Rh4+ Kg8 (11.456.865) 82  TB:10
20.01   5:52   -2.71   1...Rxb2 2.Re7 Rxa2 3.Nc3 Ra3 4.Nd5 c3 5.Re2 Ra1+ 6.Kf2 Rd1 7.Nf4 c4+ 8.Kf3 (31.225.860) 90  TB:17
21.01   6:07   -2.77   1...Rxb2 2.Re7 Rxa2 3.Nc3 Ra3 4.Nd5 c3 5.Re2 Ra1+ 6.Kf2 Rd1 7.Nf4 c4+ 8.Kf3 (32.597.841) 90  TB:18

I have sereral 6-men tablebases, they are useful to find 1...Rxb2 faster. It took 4 min 32 s to find 1...Rxb2 without my 6-men tablebases.
One hour without finding 1..Rxb2 seems strange to me.
  Best regards,
    Sylvain
 
Parent - - By JhorAVi (***) Date 2007-08-04 02:24
Thanx man! It's probably because I forgot to install any tablebase :)

But it's interesting that Shredder & Junior 10 solved it without tablebase...
Parent - By sylvain.renard.27 (**) Date 2007-08-04 13:56
In fact, with my computer and no tablebase access, Rybka 2.3.2a found the solution in 4 min 32 s.
But it is clearly a position where Rybka is not brilliant. Even my program Capture is faster than Rybka!
  Best regards,
     Sylvain
Parent - - By diskamyl (**) Date 2007-08-05 21:06
can you specifically name the masters that can't find this?
Parent - By Felix Kling (Gold) Date 2007-08-06 09:55
He said that also players which are not masters can find this. And that's true since this combination is really well known :)
Up Topic Rybka Support & Discussion / Rybka Discussion / Rybka 2.3.2a unable to solve a simple problem

Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill