SWISSTEST: 1.RobboLito - 2.Rybka - 3.Stockfish

[kingliveson]

200 Games conducted in October


The test was conducted using Fritz 11 GUI on Windows 7 64-bit. The hardware is an AMD Phenom II 940 overclocked to 3.6 GHz.

Both engines used the same book renamed to bookone.ctg (Rybka 3) and booktwo.ctg (RobboLito 0.085c2). The book has no marked/priority moves. The book settings are 2/12 with learning mode disabled. The variety of play was set at approximately 71.4% which is 20/28 right arrow key.

The engines were 1 CPU 64 bit versions with hash size set to 256 MB and pondering/permanent brain was enabled.
Time control is 4 minutes with no increment.

100 games where Rybka used contempt setting of zero.

Code: Select all

1. RobboLito  +25/=58/-17	54.00%		54.0/100
2. Rybka  3   +17/=58/-25	46.00%		46.0/100

100 games where Rybka used contempt setting of 15 (default).

Code: Select all

1. Robbolito  +24/=53/-23	50.50%		50.5/100
2. Rybka  3   +23/=53/-24	49.50%		49.5/100

Total Result:

Code: Select all

1. RobboLito   +49/=111/-40  52.25%		104.5/200   
2. Rybka   3   +40/=111/-49  47.75%		95.5/200

[Rolf]

100 games where Rybka used contempt setting of zero.

Code: Select all

1. RobboLito  +25/=58/-17	54.00%		54.0/100
2. Rybka  3   +17/=58/-25	46.00%		46.0/100

100 games where Rybka used contempt setting of 15 (default).

Code: Select all

1. Robbolito  +24/=53/-23	50.50%		50.5/100
2. Rybka  3   +23/=53/-24	49.50%		49.5/100

Total Result:

Code: Select all

1. RobboLito   +49/=111/-40  52.25%		104.5/200   
2. Rybka   3   +40/=111/-49  47.75%		95.5/200

How often now I have seen such a mess? How do you justify to mix the two samples if you tested with two different, seemingly important, contempt factors? Why not checking if more games with 15 would assure the almost equal result? At least you could have made some reflections and commented on your decisions. Such results are never talking as such, also in that more lower N scale.

No mistake, I dont want to demotivate you but I have learned a different stats. Why not giving something back to you?

Perhaps you have still more results?

[kingliveson]

How often now I have seen such a mess? How do you justify to mix the two samples if you tested with two different, seemingly important, contempt factors? Why not checking if more games with 15 would assure the almost equal result? At least you could have made some reflections and commented on your decisions. Such results are never talking as such, also in that more lower N scale.

No mistake, I dont want to demotivate you but I have learned a different stats. Why not giving something back to you?

Perhaps you have still more results?

Your criticism is well received. The reason I tested with contempt factor of zero initially is because some have claimed Rybka to be stronger than this engine using that particular setting. Perhaps, I will conduct more tests with default settings and publish the results.

[Milos]

The thing I don't understand is why you tested it with pondering on, when Robbo didn't have pondering implemented in version 0.085c2. So you gave an unfair advantage to Rybka.
If you want to test with pondering on, then try newer version of Robbi like e4.

[kingliveson]

The thing I don't understand is why you tested it with pondering on, when Robbo didn't have pondering implemented in version 0.085c2. So you gave an unfair advantage to Rybka.
If you want to test with pondering on, then try newer version of Robbi like e4.

Unless I am mistaking, pondering did not work in Arena GUI but worked in Fritz with that version.

edit:

I am checking right now, and I think you may be right

[Rolf]

How often now I have seen such a mess? How do you justify to mix the two samples if you tested with two different, seemingly important, contempt factors? Why not checking if more games with 15 would assure the almost equal result? At least you could have made some reflections and commented on your decisions. Such results are never talking as such, also in that more lower N scale.

No mistake, I dont want to demotivate you but I have learned a different stats. Why not giving something back to you?

Perhaps you have still more results?

Your criticism is well received. The reason I tested with contempt factor of zero initially is because some have claimed Rybka to be stronger than this engine using that particular setting. Perhaps, I will conduct more tests with default settings and publish the results.

Thanks. I had a similar experience last night when I made the SWISS test. I had my Rybka on I think 452 MB memory and it wasnt on 55 positions. Then I looked closer at the description in the page of the test and saw that they advised 128 MB. I changed it and then got the best result. However it was late at night so that I stopped after the two tests. If would now want to become a tester I would at first try to find out what exactly the Fritz testsuite feature is about. Without I felt myself like a little robot. click click. No comment about the bckground activities on the PC nothing. But sorry I dont want to start nitpicking, later people would say, Rolf, this isnt science here, it's just our hobby, please let us play, that is how it started in 1996. Therefore thanks again for your openminded response. I like it. -Rolf

[Albert Silver]

I downloaded and tested the suite also, and got very different results. Rolf is correct about the dubiousness of some positions. Notably 3 have no adequate defense, and thus avoiding a move does not change the lost evaluation. The positions in question are no. 3, 20, and 46. These positions cannot be saved with best play. No. 48 claims Qb6 to be the wrong move, proposing none in its stead. Even stranger is that the Excel file claims Robb gets it correct, meaning it avoids this move, but this is incorrect on both accounts. First, Qb6 is in fact the best move in the position, and second, Robbo does not avoid playing it.

Aside from this, there is also the issue of the results reported. Naturally I take into account the stronger computer I used to run the suite, but this only explains time discrepancies, not outright different results. I double-checked and triple checked my results, noting more than one discrepancy works to Robber's favor. These were all hand done, clearing the hash between engine loads, on a Core 2 Duo 3.6GHz with 1.5 Gb Ram for the hash.

In my testing, not removing the wrong positions, Robbo got 9 wrong, and Rybka 6. I used the 64-bit versions of the engines, but where results were radically different, I checked with the 32-bit version. The results were consistent.

It is also worth noting that the 64-bit version of Robber displays the exact same speedups Ryba does, when compared with the 32-bit version. Ex:

Position 18: SwissTest4_18 - Verteidigung, DeepShredder11-Bright02c, CompGame 2008

[d]r1bq1rk1/pp2n1p1/2n1p2B/3pPp1Q/2pP2P1/P1PB1N2/2P2P1P/R3K2R b KQ - 0 1

Moves to avoid: 12...Qe8, 12...Rf7, and 12...cxd3. Time allotted to find solution is 60 seconds.

RobboLito is reported to not find this within 1 min, and Rybka 3 MP in 18 seconds.

Analysis by RobberLito 0.085e4 w32:

12...Rf7 13.gxf5 cxd3 14.f6 Qa5 15.Bxg7 Qxc3+ 16.Kf1 Qxa1+ 17.Kg2 Rxg7+ 18.fxg7 Qxh1+ 19.Kxh1 Kxg7 20.cxd3 Nd8 21.Qh4 Nec6 22.Qf6+ Kg8 23.Kg2 Nf7 24.h4 b5 25.Qg6+ Kf8 26.h5
+/= (0.56) Depth: 13/43 00:00:09 10435kN
12...g6 13.Qg5 cxd3 14.Bxf8 Qxf8 15.gxf5 Qxf5 16.Qxf5 Nxf5 17.cxd3 Kg7 18.Rg1 Bd7 19.Rb1 b6 20.Kd2 Rh8 21.Kc2 Nce7
+/= (0.31) Depth: 13/43 00:00:11 12606kN
12...g6 13.Qg5 cxd3 14.Bxf8 Qxf8 15.gxf5 Qxf5 16.Qxf5 Nxf5 17.cxd3 Kg7 18.Rg1 Bd7 19.Rb1 b6 20.h3 Nce7 21.Ng5 Rc8 22.Kd2
+/= (0.34) Depth: 14/43 00:00:11 12764kN
12...g6 13.Qg5 Qa5 14.Kd2 cxd3 15.Bxf8 Kxf8 16.cxd3 fxg4 17.Qxg4 Nf5 18.Rhg1 Nce7 19.Qh3 Ke8 20.a4 Kd7 21.Qh7 Kc6 22.Kc2 b6
+/= (0.53) Depth: 15/48 00:00:33 37253kN
12...g6 13.Qg5 cxd3 14.Bxf8 d2+ 15.Kxd2 Qxf8 16.gxf5 Qxf5 17.Qxf5 Nxf5 18.h4 Nfe7 19.h5 Bd7 20.Rab1 Rf8 21.Ke2 Na5 22.Ng5 gxh5 23.Rxh5 Ng6 24.Rh7 Nf4+ 25.Kf1
+/= (0.55) Depth: 16/49 00:00:42 47548kN
12...g6 13.Qg5 cxd3 14.Bxf8 Qxf8 15.cxd3 fxg4 16.Qxg4 Qf5 17.Qxf5 Nxf5 18.h4 Bd7 19.Rb1 Na5 20.h5 gxh5 21.Rxh5 Be8 22.Rh1 Rc8 23.Kd2 Kf8 24.Rh7
+/- (0.76) Depth: 17/64 00:02:25 164mN
12...g6 13.Qg5 cxd3 14.Bxf8 Qxf8 15.cxd3 fxg4 16.Qxg4 Qf5 17.Qxf5 Nxf5 18.h4 Ng7 19.Kd2 Ne7 20.Rag1 Bd7 21.h5 Rf8 22.Ke2 Rc8 23.h6 Ngf5 24.Kd2 Kh8 25.Ng5
+/= (0.70) Depth: 18/64 00:02:53 196mN

Analysis by RobberLito 0.085e4 x64:

12...Rf7 13.gxf5 cxd3 14.f6 Qa5 15.Bxg7 Qxc3+ 16.Kf1 Qxa1+ 17.Kg2 Rxg7+ 18.fxg7 Qxh1+ 19.Kxh1 Kxg7 20.cxd3 Nd8 21.Qh4 Nec6 22.Qf6+ Kg8 23.Kg2 Nf7 24.h4 b5 25.Qg6+ Kf8 26.h5
+/= (0.56) Depth: 13/43 00:00:06 10435kN
12...g6 13.Qg5 cxd3 14.Bxf8 Qxf8 15.gxf5 Qxf5 16.Qxf5 Nxf5 17.cxd3 Kg7 18.Rg1 Bd7 19.Rb1 b6 20.Kd2 Rh8 21.Kc2 Nce7
+/= (0.31) Depth: 13/43 00:00:08 12606kN
12...g6 13.Qg5 cxd3 14.Bxf8 Qxf8 15.gxf5 Qxf5 16.Qxf5 Nxf5 17.cxd3 Kg7 18.Rg1 Bd7 19.Rb1 b6 20.h3 Nce7 21.Ng5 Rc8 22.Kd2
+/= (0.34) Depth: 14/43 00:00:08 12764kN
12...g6 13.Qg5 Qa5 14.Kd2 cxd3 15.Bxf8 Kxf8 16.cxd3 fxg4 17.Qxg4 Nf5 18.Rhg1 Nce7 19.Qh3 Ke8 20.a4 Kd7 21.Qh7 Kc6 22.Kc2 b6
+/= (0.53) Depth: 15/48 00:00:24 37253kN
12...g6 13.Qg5 cxd3 14.Bxf8 d2+ 15.Kxd2 Qxf8 16.gxf5 Qxf5 17.Qxf5 Nxf5 18.h4 Nfe7 19.h5 Bd7 20.Rab1 Rf8 21.Ke2 Na5 22.Ng5 gxh5 23.Rxh5 Ng6 24.Rh7 Nf4+ 25.Kf1
+/= (0.55) Depth: 16/49 00:00:30 47548kN
12...g6 13.Qg5 cxd3 14.Bxf8 Qxf8 15.cxd3 fxg4 16.Qxg4 Qf5 17.Qxf5 Nxf5 18.h4 Bd7 19.Rb1 Na5 20.h5 gxh5 21.Rxh5 Be8 22.Rh1 Rc8 23.Kd2 Kf8 24.Rh7
+/- (0.76) Depth: 17/64 00:01:45 164mN
12...g6 13.Qg5 cxd3 14.Bxf8 Qxf8 15.cxd3 fxg4 16.Qxg4 Qf5 17.Qxf5 Nxf5 18.h4 Ng7 19.Kd2 Ne7 20.Rag1 Bd7 21.h5 Rf8 22.Ke2 Rc8 23.h6 Ngf5 24.Kd2 Kh8 25.Ng5
+/= (0.70) Depth: 18/64 00:02:06 196mN

Analysis by Rybka 3:

12...Rf7 13.Ng5 g6 14.Qh3 fxg4 15.Qe3 cxd3 16.Nxf7 Kxf7 17.cxd3 Nf5 18.Qd2 Qh8 19.Bg5 b6 20.Rb1
= (-0.25) Depth: 9 00:00:02 276kN
12...g6 13.Qg5 Qa5 14.Kd2 Rf7
+/= (0.32) Depth: 10 00:00:08 846kN
12...g6 13.Qg5 Qa5 14.Kd2 Rf7 15.Be2 fxg4 16.Qxg4 Rh7 17.Be3
+/= (0.42) Depth: 11 00:00:09 905kN
12...g6 13.Qg5 cxd3 14.Nh4 Kh7
+/= (0.43) Depth: 12 00:00:14 1431kN
12...g6 13.Qg5 cxd3 14.Nh4 Kh7 15.Bxf8 Qxf8 16.gxf5 gxf5 17.cxd3 Qh6 18.Nf3 Qxg5 19.Nxg5+ Kg7 20.Rg1 Ng6 21.Kd2 Bd7 22.Rg3
+/= (0.47) Depth: 13 00:00:24 2454kN
12...g6 13.Qg5 cxd3 14.Nh4 Kh7 15.Bxf8 Qxf8 16.gxf5 d2+ 17.Kxd2 gxf5 18.Rhg1 Qh6 19.Nf3 Ng6 20.Rg3 Bd7 21.Qxh6+
+/= (0.59) Depth: 14 00:00:45 4437kN
12...g6 13.Qg5 cxd3 14.Nh4 Kh7 15.Bxf8 Qxf8 16.gxf5 exf5 17.cxd3 Qh6 18.Nf3 Qxg5 19.Nxg5+ Kg7 20.h4 f4 21.Kd2 Bf5 22.Rag1 Rh8 23.h5
+/= (0.62) Depth: 15 00:01:55 11791kN

Questionable positions:

No. 3: SwissTest4_03 - Verteidigung, Aristarch45-TheBaron15, CompGame 2005

[d]r1b2rk1/2p2ppp/p2b4/1p6/6Pq/1BP5/PP1P1PP1/RNBQR1K1 w - - 0 1

Moves to avoid: Qf3, Re3, g3.

Unfortunately, white is dead and has no adequate defense. The only other tries, f3 or Re4, are also insufficient.

No. 20: SwissTest4_20 - L vs L, Eigenmann 2006

[d]5k2/5ppp/p7/3p1b2/5P2/5P2/PPBK3P/8 b - - 0 1

Moves to avoid: Bxc2.

Obviously the pawn endgame is dead lost after the exchange of bishops with the queenside majority. The only problem is the endgame is lost even with the bishops. Ex:

1... Be6 2. b4 Ke7 3. Kc3 Kd6 (3... d4+ 4. Kxd4 Bxa2 5.Kc5!) 4. Bd3 Bc8 5. a4 g6 6. Kd4 f6 (6... h6 7. a5!) 7. h4!

No. 46 Naum22-Hiarcs112, CompGame 2007

[d]2r3k1/5p1p/4pP2/1p1p4/1p1R3P/8/PPP3P1/1K6 b - - 0 27

Moves to avoid: Rc4.

It is true that 27... Rc4 fails, but Black has no way to save this position against perfect play. The only other try that comes to mind, is 27...h5, but I don't see any hope for Black after 28.Rxb4 Rc5 29.g4

No. 48 - Pepito142-LGoliath35, CompGame 2002

[d]r1bq1rk1/pp2bpp1/3pp3/6P1/4PPn1/2N5/PPPQ2P1/2KR1B1R b - - 0 13

Moves to avoid: Qb6

The problem is that this is actually the best move.

If one removes the 4 positions from the equation, Robbo got a total of 6 wrong out of 60, and Rybka got 3 wrong.

Albert

[BBauer]

I did like you wrote and got

Der Server unter rapidshare.com konnte nicht gefunden werden.

regards
Bernhard

[Glarean]

Thx for your tests/analysis.

I downloaded and tested the suite also, and got very different results. Rolf is correct about the dubiousness of some positions. Notably 3 have no adequate defense, and thus avoiding a move does not change the lost evaluation.
...

No, Rolf isn't correct (as always

These matter - ("...does not change the lost evaluation") - is explained in the article.
(Try Google or Babelfish for automatic translation of the whole report: i.e. http://translate.googleusercontent.com/ ... 6zB7QziuwA )

...
The positions in question are no. 3, 20, and 46. These positions cannot be saved with best play. No. 48 claims Qb6 to be the wrong move, proposing none in its stead.
...

Probably there are some "bugs" in your analysis...:

No. 3 - Much better moves than Qf3?/Re3?/g3?: f3 and Re4 (Of course Black is still worse, but that's not the question...)

No.20 - Exchange of the bishops leads to a lost pawn endgame; in contrast these endgame B vs B has a great chance for a draw (I see no forced win for white after i.e. 1...Bd7 or Bc8 followed by centralization of the king)

No.46 - Similar to No. 20: Exchange of the rooks leads to a lost pawn endgame; the two rooks increase the chance to a draw. It's a matter of chess endgame knowledge (and a big problem for engines...)

No.48 - Qb6? Absolutely not the best move in the position; on the contrary, probably he lost the game.
Alternatives with the ability to compensate: ...Bd7 (with returning of material) or ...d5 (with counter play).

Thanks for your effort: Walter

.

[jarkkop]

Robbolito 0.084 gets 90.6% (58 out of 64) for Q8400@3.2GHz time adjusted to 41s instead of 60s

1. SwissTest4_01 - Vergifteter Bauer, AnMon521-Tao54, CompGame 2002 Solved in 19.23s/16; Solved: 1
2. SwissTest4_02 - Entwicklung, Aristarch45-Delfi45, CompGame 2004 Solved in 0.23s/10; Solved: 2
3. SwissTest4_03 - Verteidigung, Aristarch45-TheBaron15, CompGame 2005 > 41s.
4. SwissTest4_04 - Damenopfer, Asauskas-Malisauskas, Vilnius 2004 > 41s.
5. SwissTest4_05 - Abzug, Bright03d-Fritz11, CompGame 2008 Solved in 1.59s/13; Solved: 3
6. SwissTest4_06 - Königssicherheit, CMaster10000-CMaster9000, CompGame 2004 Solved in 0.08s/8; Solved: 4
7. SwissTest4_07 - Springermanöver, ChessTiger14-GTiger2, CompGame 2001 Solved in 8.97s/14; Solved: 5
8. SwissTest4_08 - T&S vs T&S, ChessTiger2007-Rybka22n2, CompGame 2007 Solved in 16.59s/20; Solved: 6
9. SwissTest4_09 - Läuferopfer h7, Chessmaster9000-Ruffian1, CompGame 2003 Solved in 1.30s/13; Solved: 7
10. SwissTest4_10 - Königssicherheit, Comas-Martinez, Barcelona 2000 Solved in 2.67s/14; Solved: 8
11. SwissTest4_11 - Verteidigung, Crafty221-WildCat8, CompGame 2008 Solved in 1.91s/11; Solved: 9
12. SwissTest4_12 - T vs T, CraftyCito12-Movei008310, CompGame 2005 Solved in 8.36s/20; Solved: 10
13. SwissTest4_13 - Königssicherheit, DJunior101-DShredder11, CompGame 2007 Solved in 0s/3; Solved: 11
14. SwissTest4_14 - Verteidigung, DJunior101-DShredder11, CompGame 2007 Solved in 0.78s/11; Solved: 12
15. SwissTest4_15 - Initiative, DeepFritz10-Rybka232a, CompGame 2007 Solved in 0s/5; Solved: 13
16. SwissTest4_16 - Springeropfer, DeepJunior101-Hiarcs12, CompjGame 2008 Solved in 3.69s/13; Solved: 14
17. SwissTest4_17 - Turmopfer, DeepJunior101-Naum3, CompGame 2008 > 41s.
18. SwissTest4_18 - Verteidigung, DeepShredder11-Bright02c, CompGame 2008 Solved in 11.64s/13; Solved: 15
19. SwissTest4_19 - Linienöffnen, DeepShredder11-ZappaMexi, CompGame 2007 Solved in 0s/3; Solved: 16
20. SwissTest4_20 - L vs L, Eigenmann 2006 > 41s.
21. SwissTest4_21 - B vs B, Eigenmann 2006 Solved in 11.36s/21; Solved: 17
22. SwissTest4_22 - Türme vs Dame, Fischer-Uhlmann, BuenosAires 1960 Solved in 0.02s/6; Solved: 18
23. SwissTest4_23 - Königsangriff, Fritz10-Hiarcs11, CompGame 2006 > 41s.
24. SwissTest4_24 - Initiative, Fritz10-Nau21, CompGame 2007 Solved in 0s/3; Solved: 19
25. SwissTest4_25 - Verteidigung, Fritz10-Rybka21c, CompGame 2006 Solved in 0s/4; Solved: 20
26. SwissTest4_26 - Starke Felder, Fritz10-Rybka232a, CompGame 2007 Solved in 0s/4; Solved: 21
27. SwissTest4_27 - Initiative, Fritz10-Rybka232a, CompGame 2008 Solved in 0s/1; Solved: 22
28. SwissTest4_28 - Vergifteter Bauer, Fritz8-Chessmaster10000, CompGame 2005 Solved in 6.94s/13; Solved: 23
29. SwissTest4_29 - Prophylaxe, Fritz8-Hiarcs10HM, CompGame 2006 Solved in 13.70s/15; Solved: 24
30. SwissTest4_30 - Königssicherheit, Fritz8-Shredder704, CompGame 2003 Solved in 1.25s/12; Solved: 25
31. SwissTest4_31 - Vergifteter Bauer, Fritz9-Rybka12, CompGame 2006 Solved in 0.05s/7; Solved: 26
32. SwissTest4_32 - Königssicherheit, GambitTiger2-Gandalf51, CompGame 2002 Solved in 32.05s/17; Solved: 27
33. SwissTest4_33 - Verteidigung, Gandalf432-Pharaon262, CompGame 2002 Solved in 1.09s/13; Solved: 28
34. SwissTest4_34 - Verteidigung, Gandalf6-CM10000, CompGame 2005 Solved in 39.61s/15; Solved: 29
35. SwissTest4_35 - Verteidigung, Gandalf6-Shredder8, CompGame 2005 Solved in 0s/3; Solved: 30
36. SwissTest4_36 - Verteidigung, Hiarcs11-Fritz10, CompGame 2007 Solved in 1.03s/11; Solved: 31
37. SwissTest4_37 - Läuferpaar, Hiarcs12-DeepFritz101, CompGame 2008 Solved in 1.16s/12; Solved: 32
38. SwissTest4_38 - Verteidigung, Hiarcs732-Shredder6, CompGame 2002 Solved in 1.86s/11; Solved: 33
39. SwissTest4_39 - Königssicherheit, Hiarcs9-Crafty1707, CompGame 2004 Solved in 0.03s/6; Solved: 34
40. SwissTest4_40 - T&S vs T&L, Hiarcs9-Gandalf6, CompGame 2004 Solved in 0.77s/14; Solved: 35
41. SwissTest4_41 - Königssicherheit, Hiarcs9-List512, CompGame 2004 Solved in 34.20s/16; Solved: 36
42. SwissTest4_42 - L vs L, Hiarcs9-Shredder704, CompGame 2003 Solved in 22.17s/22; Solved: 37
43. SwissTest4_43 - T vs T, Junior9-Shredder8, CompGame 2005 Solved in 2.20s/17; Solved: 38
44. SwissTest4_44 - Damenopfer, Maliangkay-Geenen, CorrGame 2003 Solved in 30.19s/17; Solved: 39
45. SwissTest4_45 - Vorposten, Movei008438-FrenzeeFeb08, CompGame 2008 Solved in 1.81s/13; Solved: 40
46. SwissTest4_46 - T vs T, Naum22-Hiarcs112, CompGame 2007 > 41s.
47. SwissTest4_47 - Verteidigung, Naum3-Hiarcs112, CompGame 2008 Solved in 32.52s/18; Solved: 41
48. SwissTest4_48 - Verteidigung, Pepito142-LGoliath35, CompGame 2002 Solved in 0.38s/10; Solved: 42
49. SwissTest4_49 - Qualitätsopfer, Petrosian-Spassky, Moskau 1966 Solved in 3.41s/12; Solved: 43
50. SwissTest4_50 - Verteidigung, ProDeo1-Aristarch45, CompGame 2005 Solved in 0.81s/12; Solved: 44
51. SwissTest4_51 - Königssicherheit, Rybka22-Naum21, CompGame 2007 Solved in 35.19s/18; Solved: 45
52. SwissTest4_52 - Vergifteter Bauer, SOS2-Yace09956, CompGame 2002 Solved in 0.86s/11; Solved: 46
53. SwissTest4_53 - T vs T, SOS4-List512, CompGame 2004 Solved in 0s/6; Solved: 47
54. SwissTest4_54 - B vs B, Shredder101-Fritz10, CompGame 2007 Solved in 4.70s/19; Solved: 48
55. SwissTest4_55 - Verteidigung, Shredder704-Ruffian101, CompGame 2005 Solved in 0.13s/9; Solved: 49
56. SwissTest4_56 - Königssicherheit, Shredder8-GambitTiger2, CompGame 2004 Solved in 3.91s/14; Solved: 50
57. SwissTest4_57 - Vergifteter Bauer, Shredder9-Junior9, CompGame 2005 Solved in 0s/4; Solved: 51
58. SwissTest4_58 - D vs D, Spike1a-Shredder9, CompGame 2005 Solved in 0.33s/13; Solved: 52
59. SwissTest4_59 - S vs S, Thinker47a-Tao57b, CompGame 2005 Solved in 1.59s/14; Solved: 53
60. SwissTest4_60 - T&L vs T&L, WildCat267-Horizon41b, CompGame 2005 Solved in 0.14s/11; Solved: 54
61. SwissTest4_61 - T vs T, Yace09987-Delfi45, CompGame 2005 Solved in 4.38s/19; Solved: 55
62. SwissTest4_62 - B vs S, ZChess161-Bringer17, CompGame 2000 Solved in 0.02s/9; Solved: 56
63. SwissTest4_63 - Vergifteter Bauer, ZappaMexico-Glaurung201, CompGame 2008 Solved in 1.59s/13; Solved: 57
64. SwissTest4_64 - Grundreihe, Zarkov486-Glaurung102, CompGame 2006 Solved in 6.77s/14; Solved: 58

Result: 58 out of 64 = 90.6%. Average time = 6.47s / 11.82