Not logged inRybka Chess Community Forum
Up Topic Rybka Support & Discussion / Rybka Discussion / Parameters Experiment 38: 64 Elo over R4 default
1 2 3 4 5 6 Previous Next  
Parent - By mindbreaker (****) Date 2010-09-13 20:31
Thanks.
Parent - - By Athinira (**) Date 2010-09-13 20:03
The problem with your settings is that they use uneven piece evaluations (eg. one sides pawn is worth more than the other sides).

While this causes no problems in games, because the engine assumes that "White" means it's own pieces even if it is black (meaning that if "White Pawn" is set to be worth more than "Black Pawn", and the engine plays black, then for the sake of piece values, it assumes it plays white), it causes a problem when Rybka is used for analysis. In analysis mode, Rybka always assigns the correct color to the piece evaluations, and this will cause it to misevaluate positions for one side.

If you want to at some point, it would be great if you could experiment with values that use equivalent piece values for both sides. This kind of experiment will be much more useful for evaluation purposes.
Parent - By Fulcrum2000 (****) Date 2010-09-13 20:05
Yes you are right, these asymmetric settings effectively introduce an contempt factor.
Parent - - By mindbreaker (****) Date 2010-09-13 20:28
I heard this claimed about the analysis mode and contempt, but I wish I could hear it from Vas.  I would like to know if it is actually true.  Seems odd if true. Why would behavior so radically shift?
Parent - - By mindbreaker (****) Date 2010-09-13 20:39
Thanks.  I also just checked the engine output with black to move in the initial position instead...it does indeed give a different value.
Parent - - By Uly (Gold) Date 2010-09-16 21:56
Yes, for the black side the piece values need to be reversed for analysis, that is, the user has to create two personalities, one with your values (white) and one with white's values on black's places (black) and use them accordingly.

What you have done with your experiments is finding some contempt that works, the equivalent of Rybka 3's Contempt Play 15, or similar, which was never suggested to be used for analysis. Nevertheless, I'd expect your results to be reproducible by anyone (i.e. that the settings you've found will be stronger than default against engines that are weaker than Rybka 4, but they will be weaker than default against equal or stronger opposition, just like R3 Contempt 15 used to be).
Parent - By Waserman (*) Date 2010-09-18 10:54
I had this idea before you ...
Engines 15/51, 16/61, 18/81, 24/42, 25/52, 26/62 are pairs that change their parameters, white with black.
See results below ...
Parent - - By mindbreaker (****) Date 2010-09-15 20:00
Here is the final table of this cycle:

1    Rybka 4 x64 Exp 31 v1    3348  1323
2    Rybka 4 x64 Exp 37 v1    3343  1319
3    Rybka 4 x64 Exp 30 v1    3342  1323
4    Rybka 4 x64 Exp 38 v1    3341  1318
5    Rybka 4 x64 Exp 34 v1    3340  1322
6    Rybka 4 x64 Exp 24 v2    3339  1331
7    Rybka 4 x64 Exp 26 v2    3336  1329
8    Rybka 4 x64 Exp 32 v1    3336  1322
9    Rybka 4 x64 Exp 36 v1    3337  1319
10    Rybka 4 x64 Exp 33 v1    3334  1322
11    Rybka 4 x64 Exp 28 v1    3332  1327
12    Rybka 4 x64 Exp 25 v2    3331  1330
13    Rybka 4 x64 Exp 11 v1    3332  454
14    Rybka 4 x64 Exp 35 v1    3328  1321
15    Rybka 4 x64 Exp 27 v1    3326  1328
16    Rybka 4 x64 Exp 18 v1    3324  3039
17    Rybka 4 x64 Exp 22 v1    3321  2231
18    Rybka 4 x64 Exp 29 v1    3319  1327
19    Rybka 4 x64 Exp 9 v1    3317  434
20    Rybka 4 x64 Exp 23 v1    3316  2178
21    Rybka 4 x64 Exp 8 v1    3313  1315
22    Rybka 4 x64 Exp 14 Human    3304  900
23    Rybka 4 x64 Exp 7 v1    3302  815
24    Rybka 4 x64 Forum v1    3301  1160
25    Rybka 4 x64 Exp 10 v1    3300  462
26    Rybka 4 x64 Exp 4 v1    3299  900
27    Deep Rybka 4 x64 v1    3298  161
28    Rybka 4 x64 Exp 20 v1    3298  590
29    Rybka 4 x64 Exp 15 v1    3297  1800
30    Rybka 4 x64 Exp 1 v1    3297  916
31    Rybka 4 x64 Exp 3 v2    3296  423
32    Deep Rybka 4 x64    3294  1986
33    Rybka 4 x64 Exp 21 v1    3294  900
34    Rybka 4 x64 Exp 16 v1    3294  557
35    Rybka 4 x64 Exp 19 v1    3290  189
36    Rybka 3 Dynamic    3285  1042
37    Deep Rybka 4 x64 Lasker    3285  479
38    Rybka 4 x64 Exp 13 vC13510    3283  750
39    Rybka 4 x64 Beta 15 v1    3283  426
40    Rybka 4 x64 Exp 2 v1    3283  472
41    Rybka 3    3281  2169
42    Rybka 4 x64 Exp 12 v1    3282  900
43    Rybka 3 Human    3278  1659
44    Rybka 4 x64 Exp 17 v1    3270  355
45    Rybka 4 x64 Exp 6 v1    3258  121
46    Deep Rybka 4 x64 Human    3256  174
47    Stockfish 1.8 JA 64bit    3256  3131
48    Stockfish 1.7.1 JA 64bit 4t    3236  4235
49    Stockfish 1.7 JA 64bit 4t    3223  560
50    Stockfish 1.6.3 JA 64bit    3206  384
51    Critter 0.80 64-bit    3182  1800
52    Deep Fritz 11    3121  4774
53    HIARCS 12 MP    3117  4183
54    Deep Shredder 12    3112  1824
55    Naum 4 4t    3106  4311
56    spark-0.4    3103  1802
57    Critter 0.70 64-bit    3078  2326
58    Zappa Mexico II    3076  2519
59    Deep Shredder 11 UCI    3063  2410
60    Komodo64 1.1 JA    3055  3238
61    Komodo64 1.0 JA    3053  407
62    bright-0.5c    3034  2350
63    Protector 1.3.3 x64 4t    3031  782
64    bright-0.4a    3020  21
65    Protector 1.3.6 x64 4t    2975  900
66    Thinker54AInert-MP64-UCI    2959  1800
67    spark-0.3a    2935  23
68    Deep Fritz 12    2657  385
Parent - - By Gaмßito (****) Date 2010-09-16 01:26
Thanks for your effort and for giving us your best settings.

Regards,
Gaмßito.
Parent - By mindbreaker (****) Date 2010-09-17 03:16
You're welcome.
Parent - - By Bouddha (****) Date 2010-09-17 12:12
Thank you very much for the good job done.

Did you also tested these settings vs default rybka 4 ?

How is it performing ?
Parent - By mindbreaker (****) Date 2010-09-17 19:54
Thanks. No. I am off and running with the next batch of engines Exp 39-47.  It is looking very good.  One even has the preposterous rating of 3419 Elo.  Of course with only 119 games per engine done, it is likely to hand a chunk of that lead back.  But I am still optimistic because as a group they are performing 4.5% higher than the last group.
Parent - - By Lukas Cimiotti (Bronze) Date 2010-09-30 09:29
Thanks for your great tests.
We are using some of your settings in the WCCC.
Parent - - By mindbreaker (****) Date 2010-09-30 10:21
Very cool :)
Parent - - By Sharaf_DG (**) Date 2010-09-30 11:06
Where can i find the parameter settings,
Parent - By mindbreaker (****) Date 2010-09-30 12:02
Everything up to 39 is available in an attachment near the bottom of the first page.
Parent - By mindbreaker (****) Date 2010-09-30 10:52
Here are my latest results: (the new ones are 39-47)

1  Rybka 4 x64 Exp 42 v1    3348  901
2  Rybka 4 x64 Exp 31 v1    3346  1323
3  Rybka 4 x64 Exp 44 v1    3344  901
4  Rybka 4 x64 Exp 37 v1    3341  1319
5  Rybka 4 x64 Exp 41 v1    3342  901
6  Rybka 4 x64 Exp 30 v1    3340  1323
7  Rybka 4 x64 Exp 39 v1    3339  901
8  Rybka 4 x64 Exp 38 v1    3338  1318
9  Rybka 4 x64 Exp 34 v1    3337  1322
10  Rybka 4 x64 Exp 24 v2    3337  1331
11  Rybka 4 x64 Exp 26 v2    3334  1329
12  Rybka 4 x64 Exp 36 v1    3334  1319
13  Rybka 4 x64 Exp 32 v1    3334  1322
14  Rybka 4 x64 Exp 45 v1    3334  901
15  Rybka 4 x64 Exp 33 v1    3332  1322
16  Rybka 4 x64 Exp 28 v1    3330  1327
17  Rybka 4 x64 Exp 47 v1    3331  900
18  Rybka 4 x64 Exp 25 v2    3329  1330
19  Rybka 4 x64 Exp 43 v1    3328  901
20  Rybka 4 x64 Exp 40 v1    3328  901
21  Rybka 4 x64 Exp 46 v1    3327  900
22  Rybka 4 x64 Exp 11 v1    3326  454
23  Rybka 4 x64 Exp 35 v1    3326  1321
24  Rybka 4 x64 Exp 27 v1    3324  1328
25  Rybka 4 x64 Exp 18 v1    3324  3039
26  Rybka 4 x64 Exp 22 v1    3320  2231
27  Rybka 4 x64 Exp 29 v1    3317  1327
28  Rybka 4 x64 Exp 23 v1    3315  2178
29  Rybka 4 x64 Exp 9 v1    3311  434
30  Rybka 4 x64 Exp 8 v1    3311  1315
31  Rybka 4 x64 Exp 14 Human    3304  900
32  Rybka 4 x64 Exp 7 v1    3302  815
33  Rybka 4 x64 Forum v1    3299  1160
34  Rybka 4 x64 Exp 4 v1    3298  900
35  Deep Rybka 4 x64 v1    3298  161
36  Rybka 4 x64 Exp 20 v1    3298  590
37  Rybka 4 x64 Exp 15 v1    3297  1800
38  Rybka 4 x64 Exp 1 v1    3297  916
39  Rybka 4 x64 Exp 3 v2    3296  423
40  Rybka 4 x64 Exp 10 v1    3295  462
41  Deep Rybka 4 x64    3293  1986 Default
42  Rybka 4 x64 Exp 21 v1    3294  900
43  Rybka 4 x64 Exp 16 v1    3294  557
44  Rybka 4 x64 Exp 19 v1    3290  189
45  Rybka 3 Dynamic    3286  1339
46  Deep Rybka 4 x64 Lasker    3285  479
47  Rybka 4 x64 Exp 13 vC13510    3283  750
48  Rybka 4 x64 Beta 15 v1    3283  426
49  Rybka 4 x64 Exp 2 v1    3283  472
50  Rybka 4 x64 Exp 12 v1    3282  900
51  Rybka 3    3279  2482
52  Rybka 3 Human    3274  1956
53  Rybka 4 x64 Exp 17 v1    3270  355
54  Rybka 4 x64 Exp 6 v1    3258  121
55  Deep Rybka 4 x64 Human    3256  174
56  Stockfish 1.8 JA 64bit    3250  4031
57  Stockfish 1.7.1 JA 64bit 4t    3236  4235
58  Stockfish 1.7 JA 64bit 4t    3222  560
59  Stockfish 1.6.3 JA 64bit    3204  384
60  Critter 0.80 64-bit    3180  2700
61  HIARCS 12 MP    3118  5083
62  Deep Fritz 11    3116  5674
63  Deep Shredder 12    3108  2724
64  Naum 4 4t    3105  4311
65  spark-0.4    3102  1802
66  Critter 0.70 64-bit    3078  2326
67  Zappa Mexico II    3074  2519
68  Deep Shredder 11 UCI    3063  2410
69  Komodo64 1.1 JA    3054  3238
70  Komodo64 1.0 JA    3051  407
71  bright-0.5c    3034  2350
72  Protector 1.3.3 x64 4t    3029  782
73  bright-0.4a    3027  21
74  Gull 1.0    3000  900
75  Protector 1.3.6 x64 4t    2974  900
76  Thinker54AInert-MP64-UCI    2957  1800
77  spark-0.3a    2934  23

This run is mildly disappointing given a very strong start but there are still many games to go.  I run them against engines in roughly descending order of those engine's strength, so as a group the latest ones may be better against stronger opponents or better at strong and weak but less good in the middle in comparison with previous groups.  The next week or so should show which it is.
Parent - - By mindbreaker (****) Date 2010-10-05 22:02
I am not quite done with this run but I think the first place is fairly secure.  Here is the updated table and the parameters through 47 are attached.  39-47 were tested this cycle.

1    Rybka 4 x64 Exp 42 v1    3351    1301
2    Rybka 4 x64 Exp 31 v1    3347    1323
3    Rybka 4 x64 Exp 44 v1    3347    1301
4    Rybka 4 x64 Exp 47 v1    3345    1300
5    Rybka 4 x64 Exp 41 v1    3345    1301
6    Rybka 4 x64 Exp 39 v1    3344    1301

7    Rybka 4 x64 Exp 45 v1    3343    1301
8    Rybka 4 x64 Exp 37 v1    3343    1319
9    Rybka 4 x64 Exp 30 v1    3341    1323
10    Rybka 4 x64 Exp 38 v1    3340    1318
11    Rybka 4 x64 Exp 24 v2    3338    1331
12    Rybka 4 x64 Exp 34 v1    3339    1322
13    Rybka 4 x64 Exp 43 v1    3338    1301
14    Rybka 4 x64 Exp 40 v1    3337    1301

15    Rybka 4 x64 Exp 26 v2    3336    1329
16    Rybka 4 x64 Exp 32 v1    3336    1322
17    Rybka 4 x64 Exp 36 v1    3336    1319
18    Rybka 4 x64 Exp 46 v1    3335    1300
19    Rybka 4 x64 Exp 33 v1    3333    1322
20    Rybka 4 x64 Exp 28 v1    3332    1327
21    Rybka 4 x64 Exp 11 v1    3331    453
22    Rybka 4 x64 Exp 25 v2    3331    1330
23    Rybka 4 x64 Exp 35 v1    3328    1321
24    Rybka 4 x64 Exp 27 v1    3325    1328
25    Rybka 4 x64 Exp 18 v1    3325    3039
26    Rybka 4 x64 Exp 22 v1    3321    2231
27    Rybka 4 x64 Exp 29 v1    3319    1327
28    Rybka 4 x64 Exp 9 v1    3317    429
29    Rybka 4 x64 Exp 23 v1    3316    2178
30    Rybka 4 x64 Exp 8 v1    3313    1310
31    Rybka 4 x64 Exp 14 Human    3306    898
32    Rybka 4 x64 Exp 7 v1    3303    815
33    Rybka 4 x64 Forum v1    3301    1159
34    Rybka 4 x64 Exp 10 v1    3300    459
35    Rybka 4 x64 Exp 4 v1    3300    898
36    Deep Rybka 4 x64 v1    3299    161
37    Rybka 4 x64 Exp 20 v1    3299    590
38    Rybka 4 x64 Exp 15 v1    3297    1799
39    Rybka 4 x64 Exp 1 v1    3298    916
40    Rybka 4 x64 Exp 3 v2    3297    422
41    Rybka 4 x64 Exp 16 v1    3295    555
42    Deep Rybka 4 x64    3294    1986 Default
43    Rybka 4 x64 Exp 21 v1    3295    900
44    Rybka 4 x64 Exp 19 v1    3290    189
45    Rybka 3 Dynamic    3289    1339
46    Deep Rybka 4 x64 Lasker    3286    477
47    Rybka 4 x64 Exp 13 vC13510    3284    749
48    Rybka 4 x64 Exp 12 v1    3283    896
49    Rybka 4 x64 Exp 2 v1    3284    471
50    Rybka 4 x64 Beta 15 v1    3284    426
51    Rybka 3    3282    2482
52    Rybka 3 Human    3276    1956
53    Stockfish 1.9 JA 64bit    3275    900
54    Rybka 4 x64 Exp 17 v1    3271    354
55    Rybka 4 x64 Exp 6 v1    3259    121
56    Deep Rybka 4 x64 Human    3257    174
57    Stockfish 1.8 JA 64bit    3255    4067
58    Stockfish 1.7.1 JA 64bit 4t    3237    4225
59    Stockfish 1.7 JA 64bit 4t    3224    560
60    Stockfish 1.6.3 JA 64bit    3206    384
61    Rybka 2.3.2a mp  Mod 8    3192    150
62    Critter 0.80 64-bit    3184    2748
63    HIARCS 12 MP    3121    5133
64    Deep Fritz 11    3119    5720
65    Deep Shredder 12    3114    2774
66    spark-0.4    3109    2750
67    Naum 4 4t    3100    5208
68    Critter 0.70 64-bit    3079    2324
69    Zappa Mexico II    3068    3419
70    Deep Shredder 11 UCI    3064    2410
71    Komodo64 1.1 JA    3055    3233
72    Komodo64 1.0 JA    3053    407
73    bright-0.5c    3035    2350
74    Protector 1.3.3 x64 4t    3031    781
75    bright-0.4a    3030    21
76    Gull 1.0    3008    900
77    Protector 1.3.6 x64 4t    2975    900
78    Thinker54AInert-MP64-UCI    2958    1800
79    spark-0.3a    2935    23
80    Deep Fritz 12    2657    385
Parent - - By Akbarfan (***) Date 2010-10-06 11:39
At the risk of it was already posted, which time control you play? Bullet?  3min per game?
Parent - By mindbreaker (****) Date 2010-10-06 18:04 Edited 2010-10-06 20:58
It is posted but I have not mentioned it in a while.  About half are at 60 moves in a minute followed by 60 moves in a minute followed by a second a move increment.  The more recent half is at 70 moves in a minute followed by 70 moves in a minute followed by a second a move increment.

Fritz 11 interface
Intel(R) Core(TM)2 Quad CPU    Q6600  @ 2.40GHz 3465 MHz
[TimeControl "70/60:70/60:0+1"]
Perfect2010.ctg

Average of 30 sequential games with Rybka 4: Depth: 14.08,  Speed: 316.5kn/s

TB: 3-4-5-man on SD card

Resign/Draw: late

I am running 100 round 900 game gauntlets.
Parent - - By mindbreaker (****) Date 2010-10-06 18:36
I don't like to have too many losses on time so I use "Long Game".  I also think the ratings will more acuratly reflect actual tournament ratings if the timecontrol has simular structure.

Would you believe that there have been 234, 200+ move games.  27 were decisive and none on time.  Hmm, it looks like the Fritz interface is adjudicating every game that would go longer at 255 moves.  Don't like that. There are way too many 255 movers. 24 of 32 games over 249 moves are stopped at 255.  The rest end before move 255.
Parent - - By mindbreaker (****) Date 2010-10-06 18:43
Ah, it was only the database.  It only shows up to 255 moves the games themselves are longer. :)
Parent - By mindbreaker (****) Date 2010-10-06 18:53
A 274-move decisive game.

[Event "Gaunlet att-10"]
[Site "?"]
[Date "2010.06.29"]
[Round "84.7"]
[White "bright-0.5c"]
[Black "Rybka 4 x64 Exp 4 v1"]
[Result "0-1"]
[ECO "C18"]
[Annotator "-0.05;-0.26"]
[PlyCount "548"]
[EventDate "2010.06.28"]
[EventType "simul"]
[Source "Octospasmic"]
[TimeControl "60/60:60/60:0+1"]

{Intel(R) Core(TM)2 Quad CPU    Q6600  @ 2.40GHz 3465 MHz  W=14.7 plies; 7,
525kN/s; Perfect 13.ctg  B=19.8 plies; 248kN/s; 26,032 TBAs; Perfect 13.ctg} 1.
e4 {B 0} e6 {B 0} 2. d4 {B 0} d5 {B 0} 3. Nc3 {B 0} Bb4 {B 0} 4. e5 {B 0} c5 {
B 0} 5. a3 {B 0} Bxc3+ {B 0} 6. bxc3 {B 0} Ne7 {B 0} 7. Qg4 {B 0} O-O {B 0} 8.
Bd3 {B 0} Nbc6 {B 0} 9. Qh5 {B 0} Ng6 {B 0} 10. Nf3 {B 0} Qc7 {B 0} 11. Be3 {
B 0} c4 {B 0} 12. Bxg6 {B 0} fxg6 {B 0} 13. Qg4 {B 0} Qf7 {B 0} 14. Ng5 {B 0}
Qe8 {B 0} 15. h4 {B 0} Ne7 {B 0} 16. O-O {B 0} Nf5 {B 0} 17. Bd2 {B 0} Bd7 {-0.
26/14 4} 18. Rfb1 {-0.05/14 2 (Rab1)} b6 {-0.23/14 3} 19. Rb4 {-0.07/14 2} Ba4
{-0.25/14 3 (Rb8)} 20. Ra2 {-0.06/14 3 (Rc1)} Rb8 {-0.25/14 4} 21. Qh3 {-0.09/
12 1 (Rab2)} h6 {-0.35/13 2 (Bd7)} 22. Nf3 {-0.18/14 1} b5 {-0.38/15 1 (Qd7)}
23. Rb1 {-0.09/13 1 (Rbb2)} Qf7 {-0.37/16 2 (Qc6)} 24. Rab2 {-0.20/14 1} a6 {
-0.37/16 1 (Ne7)} 25. Kh2 {-0.23/14 2 (Rd1)} Ne7 {-0.37/16 1} 26. Be3 {-0.23/
13 1 (Rd1)} Rb7 {-0.37/16 3 (Qf5)} 27. Qg4 {-0.23/14 1 (Re1)} Rd7 {-0.36/14 2
(Qf5)} 28. Ng1 {-0.20/13 1 (Rd1)} Qf5 {-0.34/13 3 (Ra7)} 29. Qxf5 {-0.20/15 1}
Nxf5 {-0.34/15 1} 30. Nf3 {-0.21/15 1} Kf7 {-0.33/16 2 (Kh7)} 31. Kh3 {-0.20/
16 2 (g4)} Ke8 {-0.39/15 1 (Kg8)} 32. Bd2 {-0.20/15 1 (Rf1)} Rdf7 {-0.39/16 1
(Kf7)} 33. Ra2 {-0.14/13 1 (Rf1)} Kd7 {-0.39/16 1 (Rc7)} 34. Rd1 {-0.16/14 2
(Kg4)} Ne7 {-0.39/14 1 (Rc8)} 35. Be3 {-0.13/14 1 (Kg4)} Kc6 {-0.39/15 2 (Nf5)}
36. Rb2 {-0.05/15 2 (Bd2)} Rh8 {-0.36/13 1 (Nf5)} 37. Bd2 {-0.05/16 2 (Kg4)}
Kb7 {-0.38/15 1 (Nf5)} 38. Kg4 {-0.09/14 1 (Kg3)} Rhf8 {-0.38/14 3 (Kb6)} 39.
Rdb1 {-0.08/14 1 (Be3)} Kb6 {-0.38/16 1} 40. Re1 {-0.10/16 1} Kc6 {-0.38/16 1
(Ka5)} 41. Rh1 {-0.10/14 1} Nf5 {-0.38/15 0 (Kb6)} 42. Kh3 {-0.09/14 1 (Ra1)}
Rc7 {-0.38/14 1 (Kb6)} 43. Ra2 {-0.13/14 2 (Kg4)} Kb7 {-0.38/15 2 (Kd7)} 44.
Rf1 {-0.05/14 1 (Kg4)} Rcf7 {-0.38/13 1 (Kc6)} 45. Re1 {-0.06/15 1 (Kg4)} Ne7 {
-0.38/15 1 (Kb6)} 46. Rb2 {-0.04/16 1} Kc6 {-0.37/15 1 (Nf5)} 47. Ra1 {-0.05/
15 2 (Rd1)} Kb6 {-0.37/15 1} 48. Kg3 {-0.05/15 2 (Rd1)} Nc6 {-0.37/15 1 (Nf5+)}
49. Kg4 {-0.04/16 1 (Re1)} Kc7 {-0.37/14 1 (Ne7)} 50. Rh1 {-0.05/14 1 (Re1)}
Kd7 {-0.37/14 1 (Kb6)} 51. Kg3 {-0.05/16 1 (Rd1)} Ne7 {-0.37/16 1 (Kc7)} 52.
Re1 {-0.05/13 1 (Kg4)} Nf5+ {-0.37/15 1} 53. Kg4 {-0.16/14 1} Kc7 {-0.37/15 1
(Ke8)} 54. Rf1 {-0.12/12 2 (Ra2)} Ne7 {-0.37/14 1 (Kd7)} 55. Rc1 {-0.05/13 1
(Kg3)} Kd7 {-0.37/16 1 (Nf5)} 56. Rg1 {-0.08/13 2 (Rh1)} Kd8 {-0.37/16 1 (Ke8)}
57. Rc1 {-0.08/13 1 (Be3)} Kc7 {-0.37/15 1 (Rf5)} 58. Be3 {-0.06/13 1 (Rd1)}
Nf5 {-0.37/15 1} 59. Bd2 {-0.05/14 1} Kc8 {-0.37/14 0 (Ne7)} 60. Rd1 {-0.10/12
1 (Ra2)} Kb7 {-0.37/13 0 (Rc7)} 61. Kh3 {-0.05/15 1 (Ra2)} Rc7 {-0.36/13 2
(Kb6)} 62. Re1 {-0.05/14 1 (Kg4)} Kc8 {-0.30/15 7 (Kb6)} 63. Kh2 {-0.18/13 1
(Bc1)} Rcf7 {-0.36/15 5 (Kd7)} 64. Rf1 {-0.05/15 2 (Kh3)} Kb7 {-0.36/16 3} 65.
Rd1 {-0.12/15 2 (Re1)} Kb6 {-0.35/15 1 (Rc7)} 66. Kh3 {-0.05/16 1} Ne7 {-0.35/
15 2 (Rh8)} 67. Rh1 {-0.05/17 2 (Kg4)} Nc6 {-0.35/15 1 (Nf5)} 68. Kg4 {-0.06/
15 1} Rc8 {-0.35/15 1 (Ka5)} 69. Bf4 {-0.05/14 1 (Be3)} Rh8 {-0.35/15 3} 70.
Bd2 {-0.06/15 1 (Be3)} Kb7 {-0.35/15 1 (Rc8)} 71. Kg3 {-0.06/13 1 (Rf1)} Rhf8 {
-0.35/16 2 (Kb6)} 72. Ra1 {-0.07/14 1 (Kg4)} Ne7 {-0.35/16 2 (Kb6)} 73. Rd1 {
-0.07/13 1 (Kg4)} Rc8 {-0.33/14 1 (Re8)} 74. Rh1 {-0.07/14 1 (Re1)} Ra8 {-0.31/
13 2 (Kb6)} 75. Kh3 {-0.01/14 1 (Rhb1)} Rh8 {-0.24/13 1 (Rc8)} 76. Kg4 {0.00/
15 1 (Ra1)} Nf5 {-0.20/13 1} 77. Rd1 {0.00/15 1 (Ra1)} Rc8 {-0.19/12 1 (Kb6)}
78. Kh3 {0.00/16 1 (Re1)} Rff8 {-0.19/13 1 (Kb6)} 79. Re1 {0.00/17 1} a5 {-0.
19/14 0 (Ne3)} 80. Rd1 {0.20/14 1 (Kg4)} Rf7 {-0.19/14 1 (Rc7)} 81. Rdb1 {0.19/
13 1 (Ra1)} Kc6 {-0.19/14 1 (Rh8)} 82. Rd1 {0.19/13 1 (Kg4)} Rh8 {-0.19/14 1
(Kc7)} 83. Rf1 {0.19/16 1 (Kg4)} Kb6 {-0.19/14 1 (Rhf8)} 84. Ra2 {0.19/14 2
(Kg4)} Ne7 {-0.19/14 1 (Re8)} 85. Rb2 {0.19/14 1} Kc6 {-0.19/15 0 (Nf5)} 86.
Kg4 {0.16/14 1} Rhf8 {-0.19/15 0 (Ra8)} 87. Rc1 {0.16/15 1 (Re1)} Kb6 {-0.19/
14 1 (Rg8)} 88. Rh1 {0.16/15 1 (Re1)} Kb7 {-0.19/15 1 (Kc6)} 89. Ra2 {0.16/15
1 (Re1)} Rf5 {-0.19/14 0 (Kb6)} 90. Rb2 {0.16/16 1} Kb6 {-0.19/14 0 (R5f7)} 91.
Re1 {0.16/15 1 (Rhb1)} Kc6 {-0.19/15 1 (R5f7)} 92. Be3 {0.17/14 1} R5f7 {-0.19/
15 0 (Kb6)} 93. Bd2 {0.17/14 1} Nf5 {-0.19/14 1 (Rg8)} 94. Rg1 {0.17/15 1
(Reb1)} Kb6 {-0.19/13 1 (Ne7)} 95. Ra1 {0.14/14 2 (Rd1)} Kb7 {-0.19/13 1 (Rh8)}
96. Rh1 {0.15/13 1 (Rab1)} Ka6 {-0.19/14 0 (Kb6)} 97. Rhb1 {0.13/13 1 (Rf1)}
Ne7 {-0.19/14 1 (Rg8)} 98. Kg3 {0.17/15 1 (Rd1)} Rc8 {-0.19/13 0 (Rh8)} 99. Kh3
{0.15/16 2 (Kg4)} Kb6 {-0.19/15 0 (Rh8)} 100. Re1 {0.15/16 1 (Kg4)} Rcf8 {-0.
19/14 1 (Ka6)} 101. Kh2 {0.15/16 1 (Kg4)} Nf5 {-0.19/13 0 (Ka6)} 102. Kg1 {0.
15/17 1 (Kh3)} Kc6 {-0.19/13 0 (Ka6)} 103. Rc1 {0.12/15 1 (Kh2)} Ne7 {-0.19/13
0 (Rh8)} 104. Kh2 {0.15/16 1} Kd7 {-0.19/13 0 (Nf5)} 105. Kh3 {0.03/14 3 (Kg3)}
Nc6 {-0.19/14 0 (Ke8)} 106. Kg3 {0.06/11 1 (Kg4)} Kc7 {-0.19/14 0 (Ne7)} 107.
Rd1 {0.16/13 1 (Kg4)} Kb7 {-0.19/14 1 (Ne7)} 108. Rdb1 {0.16/14 2} Kb6 {-0.19/
13 0} 109. Rh1 {0.15/16 1 (Kg4)} Ka7 {-0.19/12 0 (Ne7)} 110. Kg4 {0.15/15 1}
Ka6 {-0.19/13 0 (Ne7)} 111. Be3 {0.16/14 1 (Rd1)} Ne7 {-0.19/14 0} 112. Bd2 {
0.15/14 1 (Rf1)} Kb6 {-0.19/15 1} 113. Rf1 {0.15/16 1} Kc7 {-0.19/14 0 (Rg8)}
114. Re1 {0.16/14 1 (Rh1)} Kb7 {-0.19/14 0} 115. Rf1 {0.17/14 1 (Reb1)} Kc8 {
-0.19/15 1 (Rh8)} 116. Kh3 {0.14/15 1 (Rbb1)} Kc7 {-0.19/15 1 (Kb7)} 117. Ra2 {
0.14/14 1 (Kg4)} Nc6 {-0.19/13 0 (Nc8)} 118. Rb1 {0.15/15 1 (Rb2)} Rd7 {-0.19/
11 0 (Ne7)} 119. Kh2 {0.15/14 0 (Kg4)} Ne7 {-0.19/12 0 (Rdf7)} 120. Rab2 {0.14/
14 1 (Kg3)} Nf5 {-0.19/14 0} 121. Kh3 {0.06/13 1} Rdf7 {-0.19/15 1 (Kd8)} 122.
Ra1 {0.08/12 1 (Kg4)} Kc6 {-0.19/14 1 (Kd8)} 123. Kg4 {0.10/14 1} Kd7 {-0.19/
14 1 (Ne7)} 124. Rf1 {0.06/11 1 (Re1)} Ne7 {-0.17/12 1 (Ke8)} 125. Re1 {0.03/
12 1 (Rfb1)} Kc7 {-0.03/12 1 (Nf5)} 126. Rf1 {0.00/13 1 (Reb1)} Nf5 {0.00/12 0}
127. Ra2 {0.00/14 1 (Rd1)} Kc6 {0.00/13 1 (Kd7)} 128. Rb1 {0.00/14 1 (Rb2)} Rg8
{0.00/14 1 (Kd7)} 129. Rh1 {0.00/15 1 (Rab2)} h5+ {0.00/16 1 (Ne3+)} 130. Kh3 {
0.11/16 1} Nh6 {0.00/16 1} 131. Rb2 {0.11/16 1 (Kg3)} Rh8 {0.00/16 1 (Ng4)}
132. Kg3 {0.10/14 1} Nf5+ {0.00/15 1 (Ng4)} 133. Kh2 {0.10/16 1 (Kh3)} Re8 {0.
00/15 1 (Nh6)} 134. Rf1 {0.23/14 1 (Kh3)} Rc7 {0.00/15 1 (Ref8)} 135. Rfb1 {0.
23/13 1 (Kh3)} Kd7 {0.00/16 1 (Rf7)} 136. Kh3 {0.17/14 1} Rf8 {0.00/16 1 (Ke7)}
137. Re1 {0.22/14 1 (Rg1)} Rcc8 {0.00/16 2 (Nh6)} 138. Re2 {0.18/13 1 (Reb1)}
Rc6 {0.00/15 2 (Nh6)} 139. Bc1 {0.11/13 1 (Re1)} Rb6 {0.00/14 1 (Ke7)} 140. Re1
{0.19/13 1 (Bd2)} Rbb8 {0.00/14 1 (Nh6)} 141. Kh2 {0.17/13 1 (Bd2)} Kc6 {0.00/
15 1 (Rb7)} 142. Bd2 {0.17/13 1 (Kh3)} Rb7 {0.00/15 1 (Rb6)} 143. Kg1 {0.19/14
1 (Kh3)} Re7 {0.00/14 1 (Kd7)} 144. Bf4 {0.21/15 1 (Kh2)} Nh6 {0.00/14 1} 145.
g3 {0.26/15 1} Nf5 {0.07/15 1 (Ng4)} 146. Kg2 {0.26/14 1} Kd7 {0.08/14 1 (Nh6)}
147. Bd2 {0.26/15 3 (Bc1)} Ref7 {0.07/14 1} 148. Bf4 {0.26/15 1 (Rf1)} Re7 {0.
00/16 1} 149. Bd2 {0.24/14 1} Ref7 {0.00/18 1 (Rb8)} 150. Rf1 {0.21/14 1} Rc8 {
0.08/14 1 (Rb8)} 151. Ng5 {0.21/14 1} Rff8 {0.09/15 1 (Re7)} 152. Bc1 {0.26/12
1} Rb8 {0.10/14 1} 153. Rd1 {0.23/13 2 (f3)} Ke7 {0.08/12 1} 154. Bd2 {0.22/13
1 (f3)} Rh8 {0.09/14 1 (Rfe8)} 155. Nh3 {0.22/12 1 (f3)} Kf7 {0.09/14 1} 156.
Bg5 {0.21/14 1 (f3)} Ra8 {0.22/14 1 (Rhe8)} 157. Nf4 {0.21/13 1 (f3)} Ra7 {0.
22/15 1 (Rhf8)} 158. Re1 {0.28/14 1 (Rdb1)} Rc7 {0.22/15 1 (Rb7)} 159. Kh3 {0.
33/15 1 (f3)} Rb7 {0.37/14 2 (Ra7)} 160. f3 {0.44/14 1} Rc7 {0.42/13 0} 161. g4
{0.38/15 1 (Rg1)} Ne7 {0.36/14 0} 162. Bxe7 {0.36/14 1} Rxe7 {0.34/15 0} 163.
Rg1 {0.38/14 1 (Kg3)} Rh6 {0.36/16 1} 164. Kg3 {0.50/14 1} Rh8 {0.36/15 1} 165.
gxh5 {0.53/15 1 (Rg2)} gxh5 {0.37/18 1} 166. Kf2 {0.57/15 1} Rh6 {0.36/17 1
(Ree8)} 167. Ke3 {0.57/15 1} Rb7 {0.37/19 2 (Re8)} 168. Kd2 {0.67/15 1 (Rg2)}
Rb8 {0.37/18 1 (Ke8)} 169. Rbb1 {0.62/16 1} g6 {0.94/16 2 (Rf8)} 170. Rg3 {0.
60/16 1 (Rg2)} Rg8 {1.30/14 1} 171. Rf1 {0.58/15 1 (Rbg1)} Ke7 {1.30/15 1 (Rb8)
} 172. Rb1 {0.58/15 1 (Rfg1)} Kf7 {0.00/21 0} 173. Rc1 {0.56/16 1 (Rbg1)} Rg7 {
1.30/18 1 (Rhh8)} 174. Rg2 {0.55/17 1 (Rh1)} Rh8 {1.30/18 1 (Rg8)} 175. Ke3 {
0.56/16 1 (Rg5)} Rh6 {1.30/18 1} 176. Re1 {0.56/17 1 (Rcg1)} Ke7 {1.30/18 1
(Rg8)} 177. Rf1 {0.55/16 1 (Reg1)} Kf7 {1.30/19 1} 178. Rb1 {0.60/17 1 (Kd2)}
Rg8 {1.30/18 1 (Rh8)} 179. Rbg1 {0.60/16 1 (Ra1)} Rg7 {1.30/19 1} 180. Kd2 {0.
60/16 1} Rg8 {1.30/18 1} 181. Rf2 {0.60/15 1 (Rg5)} Ke7 {1.30/18 1 (Rhh8)} 182.
Rg5 {0.60/15 1} Kf7 {1.30/19 1} 183. Re2 {0.60/16 1 (Rfg2)} Rh7 {1.30/18 1
(Rg7)} 184. Rf2 {0.59/15 1 (Reg2)} Rh6 {0.00/22 0 (Rhh8)} 185. Re2 {0.59/15 1}
Rh7 {0.00/22 1 (Rg7)} 186. Re1 {0.66/16 1 (Reg2)} Rgg7 {1.30/18 1 (Rh6)} 187.
Reg1 {0.59/14 1} Rh6 {1.30/19 0} 188. Rf1 {0.60/15 1 (R1g2)} Ke7 {1.30/19 1
(Rhh7)} 189. Rh1 {0.58/14 1 (Rf2)} Kd7 {1.30/18 2} 190. Rg2 {0.60/16 1 (Rh2)}
Ke7 {1.30/15 1} 191. Rb1 {0.59/15 1 (Rhg1)} Kf7 {1.30/18 1 (Rg8)} 192. Rf1 {0.
56/15 1 (Re1)} Rgh7 {1.30/19 1 (Rg8)} 193. Kc1 {0.57/15 1 (Rfg1)} Rg7 {1.30/18
1 (Rh8)} 194. Rg5 {0.55/15 1 (Rfg1)} Rg8 {1.30/15 1 (Rh8)} 195. Rfg1 {0.56/15
1 (Rd1)} b4 {0.90/15 2 (Rg7)} 196. cxb4 {1.31/15 1} axb4 {0.90/15 0} 197. axb4
{1.35/16 1} c3 {0.90/16 0} 198. Kd1 {1.33/15 1} Rb8 {0.92/16 1 (Bb5)} 199. Rxg6
{1.22/15 1 (Nd3)} Rxg6 {0.70/14 0} 200. Rxg6 {1.26/16 1} Rxb4 {0.70/15 0} 201.
Rf6+ {1.34/15 1} Kg8 {0.75/15 0} 202. Nxe6 {1.52/14 1} Rb2 {0.75/16 1} 203. Ke2
{1.51/13 1 (Rf8+)} Rxc2+ {0.32/13 1 (Bb5+)} 204. Ke3 {1.49/13 1} Rh2 {0.00/15
2 (Bb5)} 205. Rf8+ {1.57/14 1} Kh7 {0.00/0 0} 206. Rf7+ {1.26/14 1} Kh8 {0.00/
15 0} 207. Rf8+ {1.23/14 1 (Rc7)} Kh7 {0.00/22 0} 208. Rf7+ {1.23/15 1} Kh8 {
0.00/21 1} 209. Rc7 {0.96/15 2} c2 {0.00/15 0} 210. Nf4 {0.95/15 1} Kg8 {0.00/
16 2} 211. e6 {0.52/14 1 (Nxd5)} Rh1 {0.00/17 2} 212. Ne2 {0.30/14 1} Kf8 {0.
00/17 1 (Rxh4)} 213. f4 {0.24/13 1} Ke8 {-0.12/17 5 (Rxh4)} 214. f5 {0.34/13 1
(Rc5)} Rf1 {-0.27/12 0 (Rxh4)} 215. Rc3 {0.00/13 1 (e7)} Rxf5 {-0.66/18 2} 216.
Nf4 {0.00/16 1} Ke7 {-0.78/18 3 (Kd8)} 217. Rc5 {-0.18/15 1} Kd6 {-0.69/16 1}
218. Rc3 {-0.79/15 2} Rf8 {-0.69/16 1} 219. Rc5 {-1.14/14 1} Bb3 {-0.72/15 0}
220. e7 {-0.91/14 1} Re8 {-0.72/16 1} 221. Ng6 {-0.57/15 1} Kd7 {-0.72/17 1
(Rg8)} 222. Kd2 {-0.63/15 1} Rg8 {-0.72/17 0} 223. e8=Q+ {-0.43/15 1} Rxe8 {-0.
72/18 0} 224. Rc3 {-0.92/15 2} Ba4 {-0.72/18 1} 225. Ne5+ {-0.54/15 1} Ke6 {-0.
72/18 2} 226. Ra3 {-0.51/15 1} Rc8 {-0.58/18 1} 227. Kc1 {-0.52/15 1} Be8 {-0.
58/18 1 (Bb5)} 228. Rf3 {-0.43/16 3} Bb5 {-0.58/16 1} 229. Rb3 {-0.43/14 1} Be2
{-0.65/15 1} 230. Ng6 {-0.33/14 1} Bd1 {-0.65/17 1} 231. Nf4+ {-0.25/14 1} Kf5
{-0.65/16 1} 232. Nxd5 {-0.41/15 1} Kg4 {-0.69/13 0} 233. Ne3+ {-0.33/15 1}
Kxh4 {-0.69/14 1} 234. d5 {-0.26/15 1} Bg4 {-0.55/14 1 (Rd8)} 235. d6 {-0.16/
13 1} Rd8 {-0.57/15 2} 236. Ng2+ {-0.16/13 1 (Rd3)} Kg5 {-1.04/8 0} 237. Rb5+ {
-0.17/15 1} Kf6 {-1.04/13 1 (Kh6)} 238. Kxc2 {-0.17/14 1} Rxd6 {-1.00/14 1}
239. Kc3 {-0.18/13 1 (Ne3)} Rd8 {-1.05/12 1 (Ra6)} 240. Ra5 {-0.17/12 1 (Rc5)}
Re8 {-0.95/13 1 (Bf3)} 241. Kd3 {-0.17/12 1} Be2+ {-0.89/13 1} 242. Kd2 {-0.16/
13 1 (Kd4)} Bf3 {-0.70/14 1 (Bg4)} 243. Nh4 {-0.18/14 1 (Ne3)} Re2+ {-1.29/14
1 (Bg4)} 244. Kd3 {-0.17/16 1} Bg4 {-0.73/13 1} 245. Nf5 {-0.17/15 1 (Rf5+)}
Re5 {-6.17/11 0 (Rh2)} 246. Rxe5 {-0.22/17 1 (Ra8)} Kxe5 {-#32/29 0} 247. Nh4 {
-0.23/19 1} Kf4 {-#31/59 0} 248. Ng6+ {-0.22/20 1} Kf3 {-#30/59 0 (Kg3)} 249.
Kd4 {-0.23/21 1 (Ne5+)} Be6 {-#29/59 0 (Bc8)} 250. Ke5 {-0.24/21 1} Bd7 {-#28/
59 0} 251. Nh4+ {-0.31/21 1} Kg4 {-#27/59 0} 252. Ng2 {-0.31/22 1} Bc6 {-#26/
59 0} 253. Ne3+ {-0.32/21 1} Kg5 {-#25/59 0} 254. Nd1 {-0.34/21 1} h4 {-#24/59
0 (Kg4)} 255. Nf2 {-6.98/21 1} Bg2 {-#23/59 0} 256. Kd4 {-6.96/21 1} Kf4 {-#22/
59 0} 257. Kd3 {-6.98/20 1} Kf3 {-#21/59 0 (Bf1+)} 258. Ne4 {-7.10/21 1} Bh1 {
-#20/59 0 (Bf1+)} 259. Kd4 {-10.05/21 1} Kf4 {-#19/59 0} 260. Nf2 {-10.26/21 1}
Bb7 {-#18/59 0 (Ba8)} 261. Nh3+ {-10.28/19 1} Kg3 {-#17/59 0} 262. Ng1 {-10.37/
21 1} Bf3 {-#15/59 0} 263. Ke5 {-10.67/21 1} Kg2 {-#14/59 0} 264. Nxf3 {-10.71/
17 1} Kxf3 {-#11/59 0} 265. Kf5 {-#10/20 1} h3 {-#10/59 0} 266. Kg5 {-#9/17 0}
h2 {-#9/59 0} 267. Kf6 {-#8/15 0} h1=Q {-#8/59 0 (Ke4)} 268. Ke7 {-#7/13 0} Qd1
{-#7/59 0 (Ke4)} 269. Kf7 {-#6/11 0} Kf4 {-#6/59 0 (Qd6)} 270. Ke8 {-#5/9 0}
Kg5 {-#5/59 0 (Qd5)} 271. Ke7 {-#4/7 0} Kf5 {-#4/59 0} 272. Kf7 {-#3/6 0} Qd7+
{-#3/59 0} 273. Kg8 {-#2/3 0} Kf6 {-#2/59 0} 274. Kh8 {-#1/1 0} Qg7# {-#1/59 0}
0-1
Parent - By mindbreaker (****) Date 2010-10-06 19:02
Another decisive 255+ move game.

[Event "Gaunlet att 2-28"]
[Site "?"]
[Date "2010.09.05"]
[Round "50.5"]
[White "Rybka 4 x64 Exp 34 v1"]
[Black "not permitted"]
[Result "0-1"]
[ECO "A07"]
[Annotator "0.37;0.14"]
[PlyCount "514"]
[EventDate "2010.09.04"]
[EventType "simul"]
[Source "Octospasmic"]
[TimeControl "70/60:70/60:0+1"]

{Intel(R) Core(TM)2 Quad CPU    Q6600  @ 2.40GHz 3465 MHz  W=14.3 plies; 256kN/
s; 2,408 TBAs; Perfect2010.ctg  B=17.5 plies; 5,760kN/s; Perfect2010.ctg} 1.
Nf3 {B 0} Nf6 {B 0} 2. g3 {B 0} d5 {B 0} 3. Bg2 {B 0} c6 {B 0} 4. O-O {B 0} Bg4
{B 0} 5. d3 {B 0} Nbd7 {B 0} 6. Nbd2 {B 0} e5 {B 0} 7. e4 {B 0} dxe4 {B 0} 8.
dxe4 {B 0} Bc5 {B 0} 9. h3 {0.37/13 3} Bh5 {0.14/16 1} 10. Qe2 {0.46/13 4} O-O
{0.16/16 1} 11. Rd1 {0.49/12 2} Qc7 {0.13/15 2 (Qe7)} 12. g4 {0.54/13 3} Bg6 {
0.13/13 0} 13. Nh4 {0.35/12 1} Rfd8 {0.02/14 1} 14. Nxg6 {0.36/12 1} hxg6 {0.
02/12 0} 15. Nf3 {0.37/12 1} Nf8 {0.05/16 1 (Be7)} 16. Bd2 {0.37/13 1} Ne6 {0.
05/16 0} 17. Ng5 {0.36/13 1} Qe7 {0.06/15 1 (Nxg5)} 18. Nxe6 {0.44/12 1} Qxe6 {
0.07/16 1} 19. a4 {0.35/12 1} Ne8 {0.06/16 2} 20. b4 {0.34/11 1} Be7 {0.01/16 1
} 21. Bf1 {0.36/11 1 (Bc3)} a6 {0.00/16 1 (Rd7)} 22. c4 {0.49/13 2 (Bc3)} a5 {
0.08/16 2 (c5)} 23. bxa5 {0.57/13 1} Bc5 {0.09/17 1} 24. Rab1 {0.49/13 3 (Be3)}
Qe7 {-0.01/16 1} 25. Bc3 {0.56/14 1} Nd6 {0.00/16 0 (Bd4)} 26. Bg2 {0.66/12 1
(Rb2)} g5 {-0.01/17 1} 27. Rb2 {0.61/12 1} Qe6 {0.00/16 0 (Ne8)} 28. Bf1 {0.61/
12 1} Re8 {0.00/17 2 (Qf6)} 29. Qd3 {0.51/12 2 (Qf3)} Qf6 {-0.01/16 0 (Rad8)}
30. Kg2 {0.36/13 4 (Bg2)} Re6 {-0.12/16 1 (Qf4)} 31. f3 {0.48/11 1 (Kg1)} Re7 {
-0.06/16 1 (Rd8)} 32. Rbd2 {0.52/13 2} Qe6 {-0.04/16 1 (Rd8)} 33. Be2 {0.54/12
1} Ree8 {0.00/17 5 (Rd7)} 34. Rb1 {0.69/13 1 (Bf1)} Qe7 {0.00/17 1} 35. Qc2 {
0.69/13 1 (Rbd1)} Rad8 {0.00/16 3} 36. Rdd1 {0.71/12 0} Rc8 {0.18/16 1 (Ra8)}
37. Qc1 {0.71/12 1 (Bd2)} Rc7 {0.19/16 2 (Ra8)} 38. Qd2 {0.72/12 1} f6 {0.22/
15 0} 39. Qb2 {0.73/11 1 (Qa2)} Ra8 {0.07/15 1} 40. Qa2 {0.72/10 0 (Rh1)} Nf7 {
-0.36/14 1 (Rd7)} 41. Qd2 {0.45/11 1} Nh8 {-0.35/15 1 (Rd8)} 42. a6 {0.25/10 0}
Rxa6 {-0.35/13 0} 43. Qd8+ {0.17/11 0} Kh7 {-0.31/19 1} 44. Qxe7 {0.11/12 0}
Rxe7 {-0.34/20 1} 45. a5 {0.07/12 0} Ng6 {-0.42/19 0} 46. Kg3 {0.04/12 0} Nf4 {
-0.42/18 0 (Bd4)} 47. Bf1 {0.11/15 1} Ne6 {-0.48/17 0 (Bd4)} 48. Be1 {0.05/13
0 (Bd2)} Bd4 {-0.45/16 0 (Kg6)} 49. Bb4 {0.00/14 1 (Rdc1)} Rd7 {-0.50/17 1
(Bc5)} 50. c5 {0.00/14 1 (Be1)} Ra8 {-0.50/17 1} 51. Rd2 {0.00/14 0 (Bc4)} Kg6
{-0.51/18 1} 52. Rc1 {0.00/13 0 (Bc4)} Rdd8 {-0.59/17 1} 53. Rcd1 {0.00/14 0
(Rdc2)} Rh8 {-0.59/17 1} 54. Rc2 {0.00/13 0} Ra7 {-0.59/17 0 (Rhd8)} 55. Rc4 {
0.00/14 1 (Rb1)} Kf7 {-0.61/18 1 (Nf4)} 56. Rc2 {0.00/14 0} Ke7 {-0.58/18 1
(Kg6)} 57. Bc4 {0.00/13 1} Nf4 {-0.58/17 1 (Raa8)} 58. Rh2 {0.00/15 1 (Bf1)}
Rh4 {-0.60/15 1 (Raa8)} 59. Bf1 {0.00/14 1 (Rb1)} g6 {-0.57/17 3 (Kf7)} 60. Rc2
{0.13/12 1 (Ra2)} Ra8 {-0.61/17 2 (Rh8)} 61. Rb1 {0.14/12 1 (Bd2)} Kd7 {-0.58/
16 1 (Ne6)} 62. Rh2 {0.04/12 1} Rhh8 {-0.57/16 1 (Bg1)} 63. Rd2 {0.13/12 0} Ke7
{-0.57/16 0} 64. Rc2 {0.13/12 0 (Ra2)} Ra7 {-0.57/17 1} 65. Rb3 {0.12/12 0
(Bd2)} Kd7 {-0.57/17 0 (Rd8)} 66. Ra2 {0.12/12 0 (Bd2)} Ne6 {-0.57/16 1 (Ke7)}
67. Rd3 {0.12/12 0 (Rc2)} Ke8 {-0.57/16 1 (Ke7)} 68. Rc2 {0.13/13 0} Nf4 {-0.
57/19 1} 69. Rd1 {0.13/13 0 (Rb3)} Rh7 {-0.57/8 0 (Ke7)} 70. Rb1 {0.12/12 0
(Bd2)} Kd7 {-0.58/8 0 (Ne6)} 71. Rd2 {0.02/15 5} Kc7 {-0.55/18 5 (Ne6)} 72. Ra2
{0.02/17 3 (Rc2)} Ne6 {-0.57/18 1} 73. Rc2 {0.02/14 1} Rh8 {-0.57/18 2 (Kd7)}
74. Rb3 {0.00/15 2} Kd7 {-0.57/17 0 (Nf4)} 75. Ra3 {0.00/16 2 (Rd2)} Nf4 {-0.
56/18 1} 76. Rd2 {0.00/15 1} Ke7 {-0.56/17 0 (Kc7)} 77. Rh2 {0.00/16 2 (Rda2)}
Rh7 {-0.57/16 2 (Ne6)} 78. Raa2 {0.00/15 3 (Rha2)} Rh6 {-0.57/16 1 (Ne6)} 79.
Rhd2 {0.00/15 3 (Ra3)} Ra8 {-0.57/15 1 (Rh7)} 80. Rh2 {0.00/15 3 (Rd1)} Kd7 {
-0.66/14 1 (Ne6)} 81. Ra3 {0.01/15 2} Rh4 {-0.60/16 1 (Rh7)} 82. Rb3 {0.00/15 1
} Bg1 {-0.60/17 3} 83. Rh1 {0.00/14 0} Bd4 {-0.60/15 0} 84. Rb1 {0.00/14 1
(Rh2)} Rb8 {-0.56/15 1 (Rh7)} 85. Rh2 {0.00/14 1 (Ba3)} Rh7 {-0.55/17 1} 86.
Ba3 {0.00/14 1} Kc7 {-0.50/17 1} 87. Bb4 {0.00/15 1} Rhh8 {-0.49/17 2 (Kd7)}
88. Ra2 {0.09/14 1} Ne6 {-0.48/18 1} 89. Rc2 {0.06/13 1} Kd7 {-0.48/17 1} 90.
Ba3 {0.04/14 1 (Rcc1)} Kc8 {-0.48/16 1 (Kc7)} 91. Rb6 {0.04/14 1} Kc7 {-0.47/
16 1} 92. Rb3 {0.00/14 1} Ra8 {-0.47/17 1 (Rh7)} 93. Bb4 {0.01/13 0} Kd7 {-0.
47/17 0 (Ra7)} 94. Rc4 {0.00/13 1 (Rc1)} Rh4 {-0.48/15 1 (Kc7)} 95. Rb1 {0.00/
14 1} Rh7 {-0.47/16 1} 96. Rc2 {0.00/14 0 (Rcc1)} Nf4 {-0.47/17 1} 97. Rb3 {0.
00/13 0 (Bc3)} Rah8 {-0.48/15 1} 98. Rh2 {0.00/12 0} Ne6 {-0.48/16 1} 99. Rc2 {
0.00/14 1 (Bc4)} Ke8 {-0.45/15 1 (Nf4)} 100. Rb1 {0.00/13 1 (Rd2)} Bg1 {-0.52/
15 1 (Nf4)} 101. Bg2 {0.00/11 0} Be3 {-0.45/18 1 (Bd4)} 102. Re2 {0.00/14 1}
Bd4 {-0.45/17 0} 103. Bf1 {0.00/14 1} Kd7 {-0.43/16 1} 104. Rc2 {0.00/14 0} Kc8
{-0.44/15 1 (Nf4)} 105. a6 {0.00/13 1 (Ba3)} Nf4 {-0.44/17 1 (bxa6)} 106. Rh2 {
0.00/13 1} bxa6 {-0.44/17 0} 107. Ba5 {0.00/13 1} Ra7 {-0.40/18 1} 108. Rc2 {
0.00/13 0 (Bb6)} Kd7 {-0.50/17 1} 109. Ra2 {-0.01/12 1 (Rb4)} Ne6 {-0.54/15 1}
110. Bb6 {-0.10/11 0 (Rd1)} Raa8 {-0.63/16 0} 111. Ra5 {-0.10/12 0} Nf4 {-0.55/
17 1} 112. Ra3 {-0.05/11 0 (Ra2)} Rh4 {-0.88/17 1 (Rh6)} 113. Ra2 {-0.09/7 0
(Bg2)} Rah8 {-0.95/15 1 (Ke6)} 114. Rh2 {-0.26/8 0} Bc3 {-0.86/16 1} 115. Rc1 {
-0.39/11 0} Bb4 {-0.92/16 0} 116. Rd1+ {-0.39/12 0} Ke7 {-0.92/16 0} 117. Ra1 {
-0.40/12 0} R4h7 {-0.98/16 0} 118. Bc7 {-0.40/12 1} a5 {-0.92/15 1 (Bxc5)} 119.
Bxa5 {-0.36/11 0} Bxc5 {-0.92/14 0} 120. Rc1 {-0.37/12 1 (Bd2)} Kd6 {-1.04/15 0
} 121. Rd2+ {-0.39/12 0} Bd4 {-1.04/13 0} 122. Bb4+ {-0.54/12 1} Kd7 {-1.15/17
1} 123. Rh2 {-0.54/12 0} Rb8 {-1.15/17 0} 124. Be1 {-0.82/12 1} Kd6 {-1.21/16 0
} 125. Rhc2 {-0.95/12 1} c5 {-1.22/16 1} 126. Rd1 {-0.99/12 0 (Bd2)} Rbh8 {-1.
21/17 2 (Rhb7)} 127. Rh2 {-0.74/10 0} Ra8 {-1.21/18 1 (Rb7)} 128. Rb2 {-0.82/
12 1 (Rc2)} Kc6 {-1.24/17 1} 129. Rc2 {-0.87/12 0} Rah8 {-1.39/17 0 (Rha7)}
130. Rh2 {-0.87/10 0} Ra7 {-1.35/17 0} 131. Rc2 {-0.96/12 0} Rha8 {-1.35/17 0}
132. Bd2 {-1.03/13 0} Ra2 {-1.56/17 0} 133. Rdc1 {-0.98/12 0} R8a4 {-1.60/18 0}
134. Kh2 {-0.98/13 0} Rxc2 {-1.63/18 1 (R2a3)} 135. Rxc2 {-1.16/9 0} Ra3 {-1.
58/20 1 (Ra1)} 136. Bxf4 {-1.43/15 1} Rxf3 {-1.58/22 4} 137. Bxe5 {-1.43/15 0}
Bxe5+ {-1.63/21 0} 138. Kg1 {-1.40/17 1 (Kg2)} Re3 {-1.58/19 0 (Kd6)} 139. Ra2
{-1.40/14 1} Rc3 {-1.60/21 2 (Bd4)} 140. Ra6+ {-1.32/11 0} Kb7 {-1.60/8 0 (Kd7)
} 141. Ra4 {-1.40/15 1} Bd4+ {-1.60/19 0} 142. Kg2 {-1.38/17 1} Rc2+ {-1.44/19
1} 143. Kg3 {-1.40/17 1 (Kh1)} Kc6 {-1.68/19 1} 144. Rc4 {-1.38/16 1} Ra2 {-1.
63/19 1 (Rb2)} 145. Rc1 {-1.40/17 1} Ra3+ {-1.63/19 0 (Be5+)} 146. Kg2 {-1.40/
18 1} Re3 {-1.63/20 1 (Be5)} 147. Bc4 {-1.48/14 1 (Kh2)} Rxe4 {-1.72/17 0} 148.
Rf1 {-1.49/16 1} Re3 {-1.70/19 1 (Re7)} 149. Rf3 {-1.49/17 1} Re1 {-1.68/18 0}
150. Rf1 {-1.54/17 1} Re4 {-1.70/20 1 (Re7)} 151. Bd3 {-1.50/14 1} Re7 {-1.70/
19 2 (Re3)} 152. Rb1 {-1.50/15 1} Ra7 {-1.78/18 1} 153. Re1 {-1.56/14 1 (Bxg6)}
Ra2+ {-1.85/19 1} 154. Re2 {-1.65/15 1} Ra3 {-1.85/19 1} 155. Bxg6 {-1.66/16 1
(Rd2)} c4 {-1.85/17 1} 156. Bf7 {-1.66/15 1 (Rc2)} Rc3 {-2.03/18 1 (Kc5)} 157.
Re6+ {-1.92/13 2 (Kf1)} Kc5 {-2.32/19 1 (Kb5)} 158. Re1 {-2.09/15 1} Be5 {-2.
33/19 1} 159. Rh1 {-2.15/16 1} Rb3 {-2.36/18 1 (Rc2+)} 160. Kf2 {-2.15/13 1
(h4)} Rb2+ {-2.44/19 1} 161. Kf3 {-2.15/15 1} c3 {-2.47/20 1} 162. Bg6 {-2.30/
15 1} Rd2 {-2.47/19 1} 163. Rc1 {-2.30/15 1 (Re1)} Rh2 {-2.59/17 0 (Kd4)} 164.
Re1 {-2.33/15 1 (Rc2)} Rxh3+ {-2.60/19 1} 165. Ke2 {-2.33/16 1} Bg3 {-2.60/19 1
} 166. Rd1 {-2.33/16 3 (Rc1)} Bf4 {-2.61/18 0} 167. Rd8 {-2.34/16 1 (Rd7)} Kb4
{-2.61/19 1 (Rh4)} 168. Rd7 {-2.33/16 1} Rg3 {-2.61/19 1 (Rh2+)} 169. Bf5 {-2.
33/17 1} Rg2+ {-2.61/19 0} 170. Kd1 {-2.33/18 0} Kc5 {-2.61/20 1 (Kc4)} 171.
Rd8 {-2.33/18 1} Kc4 {-2.61/20 1} 172. Rd7 {-2.33/18 1} Be5 {-2.61/20 1} 173.
Kc1 {-2.33/19 2 (Be6+)} Kc5 {-2.61/19 1 (Rh2)} 174. Rd8 {-2.33/18 1 (Kd1)} Kb4
{-2.61/20 0 (Bg3)} 175. Rd7 {-2.33/19 1} Kc4 {-2.61/21 1 (Rh2)} 176. Be6+ {-2.
33/17 1} Kc5 {-2.61/20 1} 177. Bf5 {-2.33/17 0} Bf4+ {-2.61/20 0 (Bd6)} 178.
Kd1 {-2.33/18 0} Rf2 {-2.61/21 1 (Rh2)} 179. Rd8 {-2.33/19 1 (Be4)} Kc4 {-2.61/
22 2} 180. Rd7 {-2.33/19 2} Rg2 {-2.61/21 1} 181. Be6+ {-2.33/18 1} Kb5 {-2.61/
19 0} 182. Bf5 {-2.33/19 1} Rh2 {-2.61/21 1 (Be5)} 183. Be4 {-2.33/18 2 (Bc2)}
Kc4 {-2.61/21 1} 184. Bf5 {-2.34/17 1} Kb4 {-2.61/21 1 (Kc5)} 185. Be4 {-2.33/
19 1 (Bc2)} Kc5 {-2.61/20 1} 186. Rd5+ {-2.34/19 1 (Bc2)} Kc4 {-2.61/21 1} 187.
Rd7 {-2.34/19 0} Kb4 {-2.61/21 1} 188. Bf5 {-2.34/20 2 (Rb7+)} Rf2 {-2.61/21 1
(Kc5)} 189. Be4 {-2.34/17 1 (Bc2)} Kc5 {-2.61/21 2 (Be5)} 190. Rd5+ {-2.34/18
1 (Bc2)} Kc4 {-2.15/21 3} 191. Rd7 {-2.34/18 0} Bg3 {-2.42/21 2 (Bb8)} 192. Kc1
{-2.34/17 1 (Bf5)} Be5 {-2.42/17 0 (Bb8)} 193. Kd1 {-2.34/19 1 (Bf5)} Kb4 {-2.
42/21 2} 194. Kc1 {-2.34/19 1 (Rb7+)} Kb5 {-2.42/20 1 (Rh2)} 195. Bf5 {-2.34/
18 1} Kc5 {-2.42/20 1 (Kc4)} 196. Rd8 {-2.34/18 1 (Kd1)} Kb4 {-2.42/20 1 (Rh2)}
197. Rd7 {-2.34/18 1} Kc4 {-2.42/20 1 (Bf4+)} 198. Be6+ {-2.34/19 2 (Kd1)} Kc5
{-2.14/20 1} 199. Bf5 {-2.34/20 1} Bf4+ {-1.96/19 1} 200. Kd1 {-2.34/20 0} Be5
{-1.96/19 1 (Rh2)} 201. Kc1 {0.00/25 0 (Rd8)} Bf4+ {-2.42/18 0} 202. Kd1 {0.00/
31 0} Bd6 {-2.15/17 1 (Rh2)} 203. Kc1 {-2.34/19 3 (Rd8)} Ra2 {-2.00/20 1 (Rh2)}
204. Bc2 {-2.34/18 1 (Rd8)} Ra1+ {-2.07/16 1 (Bf4+)} 205. Bb1 {-2.34/6 0} c2 {
-2.05/18 1 (Bf4+)} 206. Kxc2 {-2.03/10 0} Ra4 {-2.05/17 0} 207. Kd3 {-2.10/15
2 (Kd2)} Rxg4 {-2.22/15 1} 208. Ra7 {-2.03/15 2 (Ke3)} Rg1 {-2.26/18 1} 209.
Bc2 {-2.08/15 1} g4 {-2.26/17 1} 210. Ra5+ {-2.07/15 1} Kb6 {-2.19/20 1} 211.
Rf5 {-2.19/15 1} Be5 {-2.19/19 1} 212. Ke2 {-2.16/16 1 (Rf2)} Ra1 {-2.29/18 1}
213. Be4 {-2.15/16 1 (Rf2)} Ra2+ {-2.36/20 1} 214. Ke3 {-2.21/16 2} Ra3+ {-2.
36/20 1} 215. Ke2 {-2.21/17 1} Kc5 {-2.36/20 1 (g3)} 216. Rf1 {-2.30/14 1} Kd4
{-2.36/19 1} 217. Bb1 {-2.30/14 0} Re3+ {-2.32/19 1 (g3)} 218. Kf2 {-2.27/14 0}
g3+ {-2.31/20 2 (Rh3)} 219. Kg2 {-2.05/16 1} Re2+ {-2.28/20 1} 220. Kf3 {-2.14/
16 0} Rd2 {-2.25/21 1 (Rb2)} 221. Bf5 {-2.14/19 2} Bc7 {-2.25/20 1 (Kd5)} 222.
Bc8 {-2.14/17 2 (Bg4)} Bb8 {-2.25/19 1} 223. Be6 {-2.14/17 2 (Bh3)} Bd6 {-2.24/
18 1} 224. Bc8 {-2.14/16 1 (Bd7)} Bc7 {-2.19/19 1 (Kd5)} 225. Be6 {-2.14/18 1
(Bd7)} Bb8 {-2.24/19 1} 226. Bf5 {-2.14/19 1 (Bc8)} Be5 {-2.25/20 1} 227. Bg6 {
-2.14/18 1 (Bh3)} Kd5 {-2.25/19 1} 228. Bf5 {-2.29/17 1} Kd6 {-2.25/19 0} 229.
Bh3 {-2.31/16 1 (Bg4)} Kc5 {-2.25/18 0 (Ke7)} 230. Bf5 {-2.16/15 1 (Bc8)} Rd4 {
-2.23/17 1 (Kd5)} 231. Bh3 {-2.17/15 1 (Kg2)} Rd3+ {-2.25/18 0} 232. Ke2 {-2.
18/16 1 (Ke4)} Rb3 {-2.25/20 2 (Ra3)} 233. Rd1 {-2.18/15 0} Bf4 {-2.25/19 1}
234. Ra1 {-2.22/16 1} Rb2+ {-2.25/18 0} 235. Kf3 {-2.19/17 1} Be5 {-2.25/19 1}
236. Rd1 {-2.19/16 1 (Rf1)} Rf2+ {-2.36/20 2} 237. Ke4 {-2.18/16 1} Rh2 {-2.64/
20 1} 238. Bg4 {-2.18/16 1 (Rc1+)} Rh4 {-2.48/20 1 (Rb2)} 239. Kf5 {-2.36/16 1
(Rc1+)} Kc4 {-2.73/19 1} 240. Be2+ {-2.69/12 0} Kc3 {-2.94/18 0} 241. Rg1 {-2.
79/15 1} Rf4+ {-3.19/17 1} 242. Kg6 {-2.86/17 1} Rf2 {-3.57/18 1} 243. Bb5 {-2.
98/16 2 (Ba6)} Kd2 {-3.53/16 0 (Kd4)} 244. Bc6 {-2.98/15 1} Ke3 {-3.53/14 0}
245. Kh5 {-2.98/13 1} f5 {-4.18/16 1 (Bd6)} 246. Re1+ {-3.26/13 1} Re2 {-4.18/
15 0} 247. Rf1 {-4.64/11 1} Rh2+ {-3.36/16 1} 248. Kg6 {-5.79/12 1 (Kg5)} f4 {
-4.22/18 2} 249. Kf5 {-5.92/10 1} Bc7 {-4.22/17 0 (g2)} 250. Rf3+ {-4.15/12 1
(Rc1)} Ke2 {-5.55/16 1} 251. Rc3 {-5.12/11 1} Ba5 {-5.55/14 0} 252. Bf3+ {-5.
34/11 1} Kf2 {-5.75/17 0} 253. Rd3 {-5.34/11 0} Bd2 {-5.75/15 0 (Rh4)} 254. Ke4
{-5.77/11 1} Be3 {-5.89/17 0} 255. Rb3 {-6.59/12 2} g2 {-5.89/15 0} 256. Bxg2 {
-6.59/13 1} Rxg2 {-6.23/18 1} 257. Kd3 {-6.59/9 0 (Rc3)} f3 {-8.24/16 1} 0-1
Parent - By mindbreaker (****) Date 2010-10-06 19:09
There is a tie for longest game at 290 moves.  Anyone actually want to see the 290-move draws?
Parent - - By Omayra Date 2010-11-05 16:47
nbvvcxfrty
Parent - By Akbarfan (***) Date 2010-11-05 18:23
Wow, for your first post, very good :lol::lol::lol:
Parent - - By mindbreaker (****) Date 2010-10-05 22:18
Lukas, I am curious which Experiments were used in the WCCC and in which games?
Parent - By Lukas Cimiotti (Bronze) Date 2010-10-06 03:21
In the cluster many things are different. Timing has  nothing in common with R4's timing. So what we looked at were your adjustments of pieces values.The most important conclusion we drew was that raising the own queen's value too much is harmful - before we thought this would help avoid draws against weaker opponents. So we used +1 cp for white pawns, +2 cp for all white minor pieces, + 4 cp for white rooks and + 6 cp for the white queen. In the last game all settings were 0.
Parent - - By Sharaf_DG (**) Date 2010-10-09 14:28
Hi
if rybka is black, do i use the white settings, and vice versa. Please assist
Parent - By Uly (Gold) Date 2010-10-09 14:35 Edited 2010-10-10 20:45
For analysis, yes, you'd switch the white and black parameters around.

Please note that if your opponent is Rybka this will just decrease performance.

[EDIT - Used the be the other way around...]
Parent - By mindbreaker (****) Date 2010-10-10 04:37
For games you don't need to adjust.  "White" is whatever side the engine is playing.  Only in analysis is "White" actually white and "Black" black.
Parent - - By M ANSARI (*****) Date 2010-11-21 07:55
Do you have more details on your test conditions?  Personally I only tried modifying the time control parameters and found that at fast time controls that are sudden death, R4 is extremely sensitive to those parameters.  At longer time controls or with increment, things improve quite a bit and TC parameters are not so critical.  The confusing part is that R4 seems to do much better as cores increase and at 8 cores it is pretty much  good at everything and anything and will do well at all TC's.  There is no doubt that there is at least 50 ELO to be gained from R4 by changing some parameters, but after testing things for quite a while I have come to a conclusion that to modify R4 to play well on all platforms, you would need to setup some sort of table and then a subroutine would identify your hardware and choose a setting from the table.  A lot of this has to do with R4 or Rybka in general, not having a fast response between moves ... I would guess that when it has to play instantly it has maybe 3x (or more) lag in response time ... so if engine X has a .1ms response to come up with a move then R4 would need .3ms.  While this may not seem like a lot, it can be very detrimental if Rybka falls behind on time.  Still not sure why Rybka has higher lag than most engines, but the only common denominator I can think of is that Rybka is a process based engine while all others are threaded engines ... maybe that has something to do with it.

What I think is important to note is that Rybka 4 as a chess engine giving chess evaluation output is pretty well optimized (something that cannot be said about TC parameters).  I think it will be very difficult to gain something by adjusting piece parameters except against much weaker opponents due to increased contempt.  But increased contempt usually backfires against stronger opponents.
Parent - By mindbreaker (****) Date 2010-11-21 19:52
Fritz 11 interface
Windows XP 64-bit Pro
Intel(R) Core(TM)2 Quad CPU    Q6600  @ 2.40GHz 3465 MHz water cooled with 55 gallon drum and copper tubing...my own bizarre setup. Current temps at 100% 44C 44C 38C 44C.  I lapped (sanded and polished) the CPU.
4GB DDR2, 5 Hard drives, 2, 24" monitors both @ 1920x1200, video card, audio card, etc.
[TimeControl "70/60:70/60:0+1"]
Book: Perfect2010.ctg
TB: 3-4-5-man on SD card using card reader
Resign/Draw: late
32mb hash (small is better at short time controls)
I am running mostly 100 round 900 game gauntlets (9 modified Deep Rybka 4 Experiments at a time).
Opponents:(Rybka 3/Rybka 3 Dynamic/Rybka 3 Human)<--together 150-200 games, Stockfish 1.9 100 games, Critter 0.80 100 games, Hiarcs 12 100 games, Deep Fritz 11 100 games, Deep Shredder 12 100 games, Spark .4 100 games, Naum 4 100 games, Zappa Mexico II 100 games, Komodo 1.1 100 games, Protector 1.3.3 100 games, and Gull 1.0 100 games.  The last two or three may very or may not be included if the group is not doing well or it is fairly clear who the leaders are.  And I have also updated versions of Stockfish and Critter.  I was running two unmentionable engines also 100 each (the two strongest on my machine).

CPU could go faster if it had not been so abused by 3 years of heavy overclocking and 24/7 100% activity :)  Part of the time I was using peltiers with the water cooling system which left frost and condensation at least twice:eek: If the machine was not running at least 25% continuously it would fall to -4C or colder.  If the engines crashed and I was not there bad things could happen.  Even booting up was dangerous.  Glad I took them out and don't have that headache anymore.
Parent - - By kosmicnut2 (*) Date 2011-01-26 19:14
Just to let you know, your parameter juggling is the primary reason
I visit this forum.  I find the dynamics most fascinating and I hope
you continue these experiments, which not only help to understand
chess engines, they should ultimately be of valuable benefit for
OTB chess.

Looking forward to new discoveries regards,
Ronaldo
Parent - By mindbreaker (****) Date 2011-01-28 06:09
Thanks, Ronaldo.
- - By Waserman (*) Date 2010-09-18 06:11 Edited 2010-09-19 03:50
Results after 200 games :
===============================================================================
Time   : 1m + 1s
Hash   : 128 MB
Ponder : off (4cpu + 4cpu)
Book   : SoecialBook2Ply.ctg
Tour   : RoundRobin
used : 1,2,3,4,5,6 TBS
GUI : Fritz 12
================================================================================
Rybka4 ParametersTest-1  2010

                                                               1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7
1   Deep Rybka 4 SSE42 x64x37  3200   +41  *   1 ½ 1 ½ ½   ½ 0 ½ ½ 0     ½ ½ 1   ½   1 ½       ½   9.5/17  72.50
2   Deep Rybka 4 SSE42 x64x34  3200     0    * 1 0 0   1 0 ½ 0 0 0 1   ½ ½ ½ 1     ½ 1   1   ½ ½   9.5/19  67.00
3   Deep Rybka 4 SSE42 x64x30  3200   +19  0 0 *   ½     0 0     1 0 1 ½ 1   0 ½ 1 1 1   ½ 1 ½     9.5/18  64.75
4   Deep Rybka 4 SSE42 x64x16  3200  +102  ½ 1   * 1   ½ 1     ½   ½   1 ½     1     ½   0 ½ ½     9.0/14  69.25
5   Deep Rybka 4 SSE42 x64x31  3200     0  0 1 ½ 0 *     ½ ½ ½       0 1 ½ 1 0   1 0 ½ 1 0 1       9.0/18  66.00
6   Deep Rybka 4 SSE42 x64x61  3200   +75  ½         * 1   ½ ½ ½ ½   ½ ½ ½   ½     1   ½ ½ 1       8.5/14  61.75
7   Deep Rybka 4 SSE42 x64x06  3200  +110  ½ 0   ½   0 * ½         1 1 ½       1 ½     1 1   1     8.5/13  59.25
8   Deep Rybka 4 SSE42 x64x25  3200   +81    1 1 0 ½   ½ *   1 ½ 1 1     ½ 0               1 0     8.0/13  63.75
9   Deep Rybka 4 SSE42 x64x81  3200   +81  ½ ½ 1   ½ ½     *   ½ ½     ½     ½ 1   ½   1   ½       8.0/13  62.25
10  Deep Rybka 4 SSE42 x64x20  3200     0  1 1     ½ ½   0   *   0   1 ½ 1 0   0 ½ 1   ½   0 ½     8.0/16  62.00
11  Deep Rybka 4 SSE42 x64mp   3200   +49  ½ 1   ½   ½   ½ ½   *     0 ½         0     ½ ½ 1 1 1   8.0/14  55.00
12  Deep Rybka 4 SSE42 x64x62  3200   +24  ½ 1 0     ½   0 ½ 1   *   1 ½ 1     0   ½ ½       ½     7.5/14  57.75
13  Deep Rybka 4 SSE42 x64x18  3200   +53  1 0 1 ½     0 0         *     ½       ½ 1 ½     1 1 ½   7.5/13  53.25
14  Deep Rybka 4 SSE42 x64x38  3200     0      0   1 ½ 0     0 1 0   * 0     1 ½     ½ ½ ½   1 1   7.5/15  50.00
15  Deep Rybka 4 SSE42 x64x42  3200   -23    ½ ½ 0 0 ½ ½   ½ ½ ½ ½   1 *   ½ ½     ½           ½   7.0/15  53.75
16  Deep Rybka 4 SSE42 x64x32  3200   -78  ½ ½ 0 ½ ½ ½   ½   0   0 ½     * 1   ½ 0 0 ½ ½   ½ ½     7.0/18  52.75
17  Deep Rybka 4 SSE42 x64x17  3200   +26  ½ ½     0     1   1         ½ 0 * ½ 0 1 1   ½ ½         7.0/13  52.50
18  Deep Rybka 4 x64mp            3200   -23  0 0 1   1 ½     ½         0 ½   ½ * ½ 0 ½   ½ ½ 1       7.0/15  52.25
19  Deep Rybka 4 SSE42 x64x28  3200   +26      ½ 0     0   0 1   1   ½   ½ 1 ½ *   1 ½         ½   7.0/13  50.25
20  Deep Rybka 4 SSE42 x64x26  3200     0  ½   0   0   ½     ½ 1   ½     1 0 1   *   ½     0 ½ 1   7.0/14  48.50
21  Deep Rybka 4 SSE42 x64x33  3200   -61    ½ 0   1 0     ½ 0   ½ 0   ½ 1 0 ½ 0   * 0     ½ 1 1   7.0/17  47.25
22  Deep Rybka 4 SSE42 x64x19  3200   -24  0 0 0 ½ ½             ½ ½ ½   ½     ½ ½ 1 *       ½ 1   6.5/14  44.25
23  Deep Rybka 4 SSE42 x64x52  3200   -24  ½       0 ½ 0   0 ½ ½     ½   ½ ½ ½         * 1 1   ½   6.5/14  44.25
24  Deep Rybka 4 SSE42 x64x51  3200   -29    0 ½ 1 1 ½ 0       ½     ½     ½ ½         0 * ½       5.5/12  44.50
25  Deep Rybka 4 SSE42 x64x36  3200  -112      0 ½ 0 0   0 ½ 1 0   0     ½   0   1 ½   0 ½ *   1   5.5/16  37.25
26  Deep Rybka 4 SSE42 x64x24  3200  -120    ½ ½ ½     0 1   ½ 0 ½ 0 0   ½       ½ 0 ½       * 0   5.0/15
27  Deep Rybka 4 SSE42 x64x15  3200  -140  ½ ½                 0   ½ 0 ½       ½ 0 0 0 ½   0 1 *   4.0/13

Average elo: 3200 <=> Category: 38
gm = 0.00 m = 0.00
Parent - By ThudanBlunder (**) Date 2010-09-18 07:16
I would turn TBs off.
- - By ghost (**) Date 2010-09-29 21:03
Are there any new on this?

Has anyone confirmed the strength of one of these settings?
Parent - - By Athinira (**) Date 2010-10-07 08:16
I'm testing Experiment 31 (which is the strongest one where the settings have been posted, although I'm using different TC settings). So far, it's kicking the crap out of Stockfish 1.9.1, with 49-19 after 68 games.
Parent - - By Gaмßito (****) Date 2010-10-07 14:53
I am testing ''Exp 42'' (this is the best and not Exp 31) against R4 default, but as expected, results are negative for Exp 42.

I wonder what is the best setting against Rybka default.

Regards,
Gaмßito.
Parent - - By Athinira (**) Date 2010-10-07 17:50
Me too. So far, I'm using Queen cp 7 for both sides, which yields about +11 elo against default Rybka 4, but i don't have the hardware nor time to mass test different Rybka 4 settings.

Also: Where did you get the settings for Exp 42? The link posted on page one doesn't contain the settings for that one.
Parent - By Gaмßito (****) Date 2010-10-08 02:21

> Where did you get the settings for Exp 42?


Mindbreaker gave a new file in the post above. Now it says: ''Attachment: MindbreakerR4Experiments2.htm (106.1k)''

It's interesting that you used ''Queen cp 7'' and you got +11. Thanks, I will try to test it. But, in how many games it is based that?

Regards,
Gaмßito.
Parent - - By mindbreaker (****) Date 2010-10-10 08:54
I was not going for a configuration that would out-perform default head to head, but because of the interest, I decided to pause the gauntlets and run them head to head under similar conditions as I have been running the engines.  Something went wrong with the computer the first run...a memory leak or something...it was saying "low memory" and being nonresponsive though still playing when I got up.  The results of that run were:

1m/60+1m/60+(0m+1s)  0

                                
1   Rybka 4 x64 Exp 42 v1  3353  +87/=213/-82 50.65%  193.5/382
2   Deep Rybka 4 x64       3294  +82/=213/-87 49.35%  188.5/382

It was doing better than that when I went to bed.

The second run (after restart) current standings:

1m/60+1m/60+(0m+1s)  0

                                
1   Rybka 4 x64 Exp 42 v1  3353  +96/=184/-76 52.81%  188.0/356
2   Deep Rybka 4 x64       3294  +76/=184/-96 47.19%  168.0/356

I will let it run overnight and hopefully the same glitch will not recur.

I highly doubt it is the best setting against default but it does not look as bad as has been reported, at least at this point.

The openings are set to reverse colors.  Book is Perfect2010.  Time is 1m/60+1m/60+0m+1s

Would someone try a long match (500+ games) with the same time control and book with Exp 42 vs Default using the 64-bit versions on a reasonably fast machine using 4 threads?
Parent - - By mindbreaker (****) Date 2010-10-11 03:27
Ok, all the games included.  Here are the results:

1m/60+1m/60+(0m+1s)  0

                                
1   Rybka 4 x64 Exp 42 v1  3353  +261/-236/=555 51.19%  538.5/1052
2   Deep Rybka 4 x64       3294  +236/-261/=555 48.81%  513.5/1052

I think that amounts to 9 Elo head to head.

Not exactly awe inspiring, but it did win.  Most of the difference is probably attributable to the timings. 

Back to the gauntlets.
Parent - - By Banned for Life (Gold) Date 2010-10-11 03:31
Most of the difference is probably attributable to the timings. 

Could be more than all of the difference is due to the timing, and the changed parameter values are actually a negative. Mixing testing of timing with testing of parameter changes leads to indeterminate results.
Parent - - By mindbreaker (****) Date 2010-10-11 04:09
Sure it could--I am not seeing it though.  It was only designed to kill other engines not other versions of Rybka 4.  I was addressing the claims that it was weaker than default as it is.  I think that has been tested.

I am not seeing the timings as contributing more than the difference.  It is basically tested already. The timings alone (Forum) is on the ratings charts I have generated:

34    Rybka 4 x64 Forum v1    3302    1159
41    Deep Rybka 4 x64    3294    1986

As you can see, there is a difference of 8 Elo.  Which is virtually the same as the 9 Elo difference.  If Rybka 4 is not giving anything up to play with the Exp 42 settings against Default and it stomps other engines better, what's the harm?
Up Topic Rybka Support & Discussion / Rybka Discussion / Parameters Experiment 38: 64 Elo over R4 default
1 2 3 4 5 6 Previous Next  

Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill