Very strong Igorrit v9!

[Tomcass]

Again Igorrit V9 (not 8!) proves to be the strongest engine today. No more time problems for any of the engines. Ivan (x4) very close to R3 (x8).

Quad i7 975 3,33 Mz. 12 Mz mem Ram. 8 threads for Igor, R3 and DS12. 4 threads for Ivan.

Game in 10 minutes.
Book HS
Gui: Fritz 11
Ponder: Off


2010-01Ivanhoe 8.m4 2010

1 2 3 4
1 Igorrit 0.086v8_x64 ********** 111½½½½½½½ ½½11½½½1½1 11½½01½½1½ 20.0/30
2 Rybka 3 000½½½½½½½ ********** ½11½½½½001 010½½½½101 14.0/30
3 IvanHoe v81 x64 (x4) ½½00½½½0½0 ½00½½½½110 ********** 1½½½½1½1½0 13.5/30
4 Deep Shredder 12 x64 00½½10½½0½ 101½½½½010 0½½½½0½0½1 ********** 12.5/30


Games:

http://www.megaupload.com/?d=IE46DW6N

Regards,

Tom

[yanquis1972]

could you do a test for me and let v9 analyze a position for 10+ minutes? i read reports that it hangs when this is the case.

regardless i very much doubt the strength difference between it & v5fix is significant enough for me to make the change.

supposedly there is a much stronger version of ivanhoe coming very soon, btw.

[Le Fou numerique]

Hi,

No sure

Code: Select all

Dual 2.20Ghz, 2Gb Ram, Windows Vista SP2
5+1 128Mo HS-Book.ctg
                         Score     Rybk Igor Ivan Robb
-------------------------------------------------------
 1: Rybka 3 32-bit      7.5 / 12   XXXX =10= 01=1 ==11
 2: Igorrit 0.086v8     6.0 / 12   =01= XXXX ==== ====
 3: IvanHoe v81         5.5 / 12   10=0 ==== XXXX ====
 4: RobboLito 0.085g3   5.0 / 12   ==00 ==== ==== XXXX
-------------------------------------------------------
24 games: +4 =17 -3

I think I lost most of computer time for this tournament.

Regards,
Patrick

[garybelton]

Tom I recommend using a testset of balanced positions where each side gets to play each position as white for testing engines. Look at this game from your match where black kindly (blindly?) played into a lost game, this is just one example, I didn't look at all the games:

[Event "2010-01Ivanhoe 8.m4"]
[Site "?"]
[Date "2010.01.14"]
[Round "3.3"]
[White "Rybka 3"]
[Black "IvanHoe v81 x64 (x4)"]
[Result "1-0"]
[ECO "B99"]
[Annotator "1.48;1.47"]
[PlyCount "101"]
[EventDate "2010.01.14"]
[Source "LEARNING"]
[TimeControl "600"]

{Intel(R) Core(TM) i7 CPU @ 9750 @ 3.33GHz 3373 MHz W=15.6 plies; 1.711kN/s;
HS-Masterbook.ctg B=17.7 plies; 7.347kN/s; HS-Masterbook.ctg} 1. e4 {0.00/0 0}
c5 {0.00/0 0} 2. Nf3 {0.00/0 0} d6 {0.00/0 0} 3. d4 {0.00/0 0} cxd4 {0.00/0 0}
4. Nxd4 {0.00/0 0} Nf6 {0.00/0 0} 5. Nc3 {0.00/0 0} a6 {0.00/0 0} 6. Bg5 {0.00/
0 0} e6 {0.00/0 0} 7. f4 {0.00/0 0} Qc7 {0.00/0 0} 8. Qf3 {0.00/0 0} Be7 {0.00/
0 0} 9. O-O-O {0.00/0 0} Nbd7 {0.00/0 0} 10. g4 {0.00/0 0} b5 {0.00/0 0} 11.
Bxf6 {0.00/0 0} Nxf6 {0.00/0 0} 12. g5 {0.00/0 0} Nd7 {0.00/0 0} 13. f5 {0.00/
0 0} Nc5 {0.00/0 0} 14. h4 {0.00/0 0} b4 {0.00/0 0} 15. Nce2 {0.00/0 0} e5 {0.
00/0 0} 16. Nb3 {0.00/0 0} Nxe4 {0.00/0 0} 17. Bg2 {0.00/0 0} Bb7 {0.00/0 0}
18. Qe3 {0.00/0 0} d5 {0.00/0 0} 19. Bxe4 {0.00/0 0} dxe4 {0.00/0 0} 20. Ng3 {
0.00/0 0} a5 {0.00/0 0} 21. f6 {0.00/0 0} gxf6 {0.00/0 0} 22. gxf6 {0.00/0 0}
Rc8 {0.00/0 0} 23. Rh2 {0.00/0 0} Bf8 {0.00/0 0} 24. Nxe4 {0.00/0 0} a4 {0.00/
0 0} 25. Nd4 {0.00/0 0} Qb6 {1.47/16 12} 26. Re2 {1.48/17 25} Rd8 {1.85/17 14}
27. Qg5 {1.48/15 0} Rd5 {1.85/16 0} 28. c4 {2.89/15 30 (Cf3)} bxc3 {3.01/15 17
(Rd8)} 29. Nxc3 {3.10/13 0} Bh6 {3.49/18 11} 30. Qxh6 {3.35/17 3} Rxd4 {3.49/
17 0} 31. Qg7 {3.35/15 0} Rxd1+ {3.57/19 11} 32. Kxd1 {3.49/17 9} Bf3 {3.57/18
0} 33. Qxh8+ {3.49/16 0} Kd7 {3.57/21 7} 34. Qxh7 {3.49/18 4} Bxe2+ {3.57/21 8}
35. Nxe2 {3.49/16 1} Ke6 {4.21/23 260} 36. Qc2 {3.57/18 16} Kxf6 {4.33/22 130
(Db5)} 37. Qd2 {3.70/17 6} Ke7 {4.22/20 38} 38. h5 {3.70/15 0} Kf8 {4.22/21 7
(Dc6)} 39. h6 {4.23/17 9} Qf6 {4.95/21 43} 40. Qb4+ {4.40/16 8} Kg8 {4.22/19 0}
41. Qxa4 {4.40/15 0} Qd6+ {4.86/17 3 (Rh7)} 42. Ke1 {5.10/18 18} Kh7 {4.94/18
2 (Dxh6)} 43. Qc2+ {5.28/16 8} Kg8 {4.96/18 2} 44. Qd2 {5.28/14 0 (Df5)} Qg6 {
5.47/15 5} 45. a4 {5.68/16 12} Kh8 {5.69/16 2 (Rh7)} 46. a5 {6.45/15 12} Kh7 {
5.69/16 0} 47. b4 {6.45/13 0} Qb1+ {5.69/15 0} 48. Nc1 {6.83/15 23 (Rf2)} Qe4+
{6.59/14 5} 49. Kd1 {6.83/13 0} Qg6 {6.71/14 3 (Df5)} 50. b5 {7.78/14 10 (Cd3)}
Qg4+ {7.98/14 9 (Dg1+)} 51. Ne2 {8.26/14 4 (De2)} 1-0

(Black could have played 8 ... Nbd7 or 8 ... b5 with much better chances)

[Tomcass]

could you do a test for me and let v9 analyze a position for 10+ minutes? i read reports that it hangs when this is the case.

regardless i very much doubt the strength difference between it & v5fix is significant enough for me to make the change.

supposedly there is a much stronger version of ivanhoe coming very soon, btw.

Yes, with pleasure. Let me know which is this position and I will test it for you.

Any stronger version of IvanHoe will be very welcome.

Regards,

Tom.

[NATIONAL12]

By NATIONAL12 Date 2010-01-15 21:35 i have done 2 tests so far 4+2,when contempt set 0. 8cpu at 3.6

+14=86-14.this took 36 hours.

second test which is still running 40/20. 8 cpu at 3.6

+1=18-1

just in case it gets deleted from Rybka forum.

this is 8cpu Skully at 3.6

[yanquis1972]

i don't trust iggy's scaling at more than 4 CPU. i'm almost positive it wouldn't be close to rybka's.

[NATIONAL12]

i don't trust iggy's scaling at more than 4 CPU. i'm almost positive it wouldn't be close to rybka's.

looking at task manager it runs between 80% and 100%.switching to processes,similar.not bad though R3 is always 100%

[Tomcass]

Tom I recommend using a testset of balanced positions where each side gets to play each position as white for testing engines.

Hi Gary,

Thanks for your comment and your attention to my test. I am totally conscious that my tests are not for scientits purposes. They have a degree of randomness, because of the books I use. But because of this randomness I enjoy following each game very much.

In the long run, if the number of games is high enough, each engine gets its right position. I keep my computers working hours and hours for my own enjoyment and this is only possible when there is enough variation in games.

Regards,

Tom.

[NATIONAL12]

does not seem much point in going on.score now

+1=20-1

its just IMHO R3 playing R3.

i await,like everyone Rybka 4.