SSDF Rating List 2009-01-16

[Thoralf Karlsson]

THE SSDF RATING LIST 2009-01-16 118048 games played by 311 computers
Rating + - Games Won Oppo
------ --- --- ----- --- ----
1 Deep Rybka 3 x64 2GB Q6600 2,4 GHz 3227 36 -31 673 86% 2924
2 Deep Fritz 11 2GB Q6600 2,4 GHz 3085 31 -30 577 69% 2941
3 Deep Rybka 3 256MB Athlon 1200 MHz 3066 51 -50 190 54% 3034
4 Zappa Mexico II x64 2GB Q6600 2,4 GHz 3064 33 -32 486 65% 2954
5 Naum 3.1 x64 2GB Q6600 2,4 GHz 3046 38 -37 358 57% 2995
6 Deep Shredder 11 2GB Q6600 2,4 GHz 3043 31 -29 564 64% 2941
7 Deep Hiarcs 12 x64 2GB Q6600 2,4 GHz 3033 27 -26 716 66% 2920
8 Hiarcs 11.2 MP 2GB Q6600 2,4 GHz 3008 30 -30 546 56% 2964
9 Deep Junior 10.1 2GB Q6600 2,4 GHz 2981 31 -31 510 52% 2970
10 Rybka 2.3.1 Arena 256MB Athlon 1200 MHz 2925 24 -24 842 56% 2884
11 Fritz 11 256MB Athlon 1200 MHz 2919 33 -33 453 58% 2865
12 Deep Shredder 11 256MB Athlon 1200 2913 34 -34 410 49% 2920
13 Deep Fritz 8 2GB Q6600 2,4 GHz 2909 28 -28 629 41% 2977
14 Rybka 1.2 256MB Athlon 1200 MHz 2902 25 -24 893 71% 2747
15 Shredder 8 MP 2GB Q6600 2,4 GHz 2899 33 -34 441 43% 2948
16 Hiarcs 11.1 256MB Athlon 1200 MHz 2875 24 -24 836 53% 2854
17 Junior 10.1 256MB Athlon 1200 MHz 2857 22 -22 1006 54% 2831
18 Fritz 10 256MB Athlon 1200 MHz 2848 37 -35 416 69% 2704
19 Junior 10 256MB Athlon 1200 MHz 2841 23 -22 1002 65% 2728
20 Zap!Chess Zanzibar 256MB Athlon 1200 MH 2840 23 -22 956 53% 2817
21 Deep Junior 8 2GB Q6600 2,4 GHz 2835 37 -39 343 41% 2898
22 Hiarcs 10 HypMod 256MB Athlon 1200 MHz 2831 20 -19 1416 69% 2695
23 Fruit 2.2.1 256MB Athlon 1200 MHz 2828 19 -19 1385 62% 2745
24 Shredder 10 UCI 256MB Athlon 1200 MHz 2824 20 -20 1246 58% 2767
25 Spike 1.2 256MB Athlon 1200 MHz 2812 26 -26 714 57% 2761
25 Fritz 9.0 256MB Athlon 1200 MHz 2812 18 -18 1515 56% 2768
27 Shredder 9.0 UCI 256MB Athlon 1200 MHz 2809 20 -20 1280 66% 2696
28 Junior 9.0 256MB Athlon 1200 MHz 2781 25 -24 868 63% 2684
29 Rybka 1.0 beta 128MB K6-2 450 MHz 2774 64 -69 115 38% 2858
30 Chess Tiger 2007 256MB Athlon 1200 MHz 2766 27 -27 662 49% 2774
31 Spike 1.1 256MB Athlon 1200 MHz 2758 31 -30 537 57% 2705
32 Zap!Chess 256MB Athlon 1200 MHz 2731 30 -29 562 53% 2707
33 Gandalf 6.0 256MB Athlon 1200 MHz 2728 24 -24 835 56% 2686
34 Chess Tiger 2004 256MB Athlon 1200 MHz 2721 21 -21 1088 52% 2704
35 Chessmaster 9000 256MB Athlon 1200 MHz 2714 36 -35 385 56% 2674
36 Pro Deo 1.1 256MB Athlon 1200 MHz 2708 24 -23 876 57% 2654
37 Deep Sjeng 1.5a 256MB Athlon 1200 MHz 2668 31 -31 493 52% 2657
38 Revelation Rybka 2.2 XScale 500 MHz 2636 60 -56 154 64% 2532
39 Ruffian 2.0.0 256MB Athlon 1200 MHz 2626 49 -49 205 46% 2656
40 Gromit 3.11.9 256MB Athlon 1200 MHz 2604 44 -46 246 43% 2656
41 Yace Paderborn 256MB Athlon 1200 MHz 2602 35 -35 389 48% 2616
42 Crafty 19.17 256MB Athlon 1200 MHz 2525 41 -44 304 30% 2676
43 Pocket Fritz 2 XScale 400 MHz 2508 52 -50 191 60% 2437
44 Resurrection Rybka 2.2 StrongARM 203 MH 2499 50 -50 196 53% 2479
45 Resurrection Fruit '05 StrongARM 203 MH 2396 67 -63 120 60% 2321
46 Hiarcs 9.5a/9.6 Palm TungstenE OMAP 126 2395 35 -35 404 45% 2431
47 CEBoard Crafty 2004 HP RX4240 400 MHz 2372 64 -64 122 51% 2366
48 R30 v. 2.5 2273 41 -38 343 69% 2134
49 Chess Genius 1.4 SX1 OMAP 310 120 MHz 2150 50 -48 210 60% 2080
50 Chess Tiger 14.9 Palm m515 16MB 42MHz 2101 69 -74 100 39% 2180



2 Deep Fritz 11 2GB Q6600 2,4 GHz, 3085
DRybka3 Q6600 14-26 DRybka3 A1200 20-20 Naum3.1 Q6600 20-20
DShre11 Q6600 26-14 DHiarcs Q6600 17-7 Hiar112 Q6600 31.5-20.5
DJun101 Q6600 18.5-7.5 Rybk231 A1200 4.5-2.5 Fritz11 A1200 28-12
DShre11 A1200 29.5-10.5 DFritz8 Q6600 31.5-8.5 Hiar111 A1200 13-4
Juni101 A1200 33.5-6.5 Shredd8 A1200 5-1 DpFrit8 A1200 30.5-9.5
Junior8 A1200 39.5-3.5 Hiarcs9 A1200 39-3

3 Deep Rybka 3 256MB Athlon 1200 MHz, 3066
DFrit11 Q6600 20-20 Zap!Mx2 Q6600 20-20 DShre11 Q6600 20-20
DHiarcs Q6600 21.5-18.5 Fritz11 A1200 22-8

5 Naum 3.1 x64 2GB Q6600 2,4 GHz, 3046
DRybka3 Q6600 12.5-27.5 DFrit11 Q6600 20-20 Zap!Mx2 Q6600 19-21
DHiarcs Q6600 19-21 Hiar112 Q6600 22-18 DJun101 Q6600 21-19
DFritz8 Q6600 30-10 Shr8 MP Q6600 28-12 Hiarcs9 A1200 33.5-4.5


For the first time SSDF has tested a chess program made by
Alexandar Naumov. When we began to test Naum 3.1 on Q6600
2,4 GHz it was his latest program version. After 358 tourna-
ment games it´s rating is 3046, which makes it the fourth
strongest program on the SSDF Rating List.

Frans Morschs Deep Fritz 11 running on Q6600 2,4 GHz takes
the second place on the list With 3085! Compared to Fritz 11
on Athlon 1200 MHz, the rating increase is 166 points.

Deep Rybka 3 has now also been tested on Athlon 1200 MHz,
which resulted in a rating of 3066. As expected Vas Rajlichs
program is strongest also on this slower single processor.
On this hardware Deep Rybka 3 has 141 points more than
Rybka 2.3.1 A1200, and 161 points less than on Q6600 2,4 GHz.

Thoralf Karlsson

[pichy]

PostPost subject: SSDF Rating List 2009-01-16 Posted: Fri Jan 16, 2009 11:24 pm Reply to topic Reply with quote Report Post
THE SSDF RATING LIST 2009-01-16 118048 games played by 311 computers
Rating + - Games Won Oppo
------ --- --- ----- --- ----
1 Deep Rybka 3 x64 2GB Q6600 2,4 GHz 3227 36 -31 673 86% 2924
2 Deep Fritz 11 2GB Q6600 2,4 GHz 3085 31 -30 577 69% 2941
3 Deep Rybka 3 256MB Athlon 1200 MHz 3066 51 -50 190 54% 3034
4 Zappa Mexico II x64 2GB Q6600 2,4 GHz 3064 33 -32 486 65% 2954
5 Naum 3.1 x64 2GB Q6600 2,4 GHz 3046 38 -37 358 57% 2995


If that is Naum 3.1 then Naum 4 should be almost touching Rybka 3 tail

[icander]

Code: Select all

THE SSDF RATING LIST 2009-01-16    118048 games played by  311 computers
                                           Rating   +     -  Games   Won  Oppo
                                           ------  ---   --- -----   ---  ----
   1 Deep Rybka 3 x64  2GB Q6600 2,4 GHz     3227   36   -31   673   86%  2924
   2 Deep Fritz 11  2GB Q6600 2,4 GHz        3085   31   -30   577   69%  2941
   3 Deep Rybka 3  256MB Athlon 1200 MHz     3066   51   -50   190   54%  3034
   4 Zappa Mexico II x64  2GB Q6600 2,4 GHz  3064   33   -32   486   65%  2954
   5 Naum 3.1 x64  2GB Q6600 2,4 GHz         3046   38   -37   358   57%  2995
   6 Deep Shredder 11  2GB  Q6600  2,4 GHz   3043   31   -29   564   64%  2941
   7 Deep Hiarcs 12 x64  2GB Q6600 2,4 GHz   3033   27   -26   716   66%  2920
   8 Hiarcs 11.2 MP  2GB  Q6600  2,4 GHz     3008   30   -30   546   56%  2964
   9 Deep Junior 10.1  2GB  Q6600  2,4 GHz   2981   31   -31   510   52%  2970
  10 Rybka 2.3.1 Arena 256MB Athlon 1200 MHz 2925   24   -24   842   56%  2884
  11 Fritz 11  256MB Athlon 1200 MHz         2919   33   -33   453   58%  2865
  12 Deep Shredder 11  256MB  Athlon 1200    2913   34   -34   410   49%  2920
  13 Deep Fritz 8 2GB  Q6600  2,4 GHz        2909   28   -28   629   41%  2977
  14 Rybka 1.2  256MB Athlon 1200 MHz        2902   25   -24   893   71%  2747
  15 Shredder 8 MP 2GB  Q6600  2,4 GHz       2899   33   -34   441   43%  2948
  16 Hiarcs 11.1  256MB  Athlon 1200 MHz     2875   24   -24   836   53%  2854
  17 Junior 10.1  256MB Athlon 1200 MHz      2857   22   -22  1006   54%  2831
  18 Fritz 10  256MB Athlon 1200 MHz         2848   37   -35   416   69%  2704
  19 Junior 10  256MB Athlon 1200 MHz        2841   23   -22  1002   65%  2728
  20 Zap!Chess Zanzibar 256MB Athlon 1200 MH 2840   23   -22   956   53%  2817
  21 Deep Junior 8  2GB Q6600 2,4 GHz        2835   37   -39   343   41%  2898
  22 Hiarcs 10 HypMod 256MB Athlon 1200 MHz  2831   20   -19  1416   69%  2695
  23 Fruit 2.2.1  256MB Athlon 1200 MHz      2828   19   -19  1385   62%  2745
  24 Shredder 10 UCI  256MB Athlon 1200 MHz  2824   20   -20  1246   58%  2767
  25 Spike 1.2  256MB Athlon 1200 MHz        2812   26   -26   714   57%  2761
  25 Fritz 9.0  256MB Athlon 1200 MHz        2812   18   -18  1515   56%  2768
  27 Shredder 9.0 UCI  256MB Athlon 1200 MHz 2809   20   -20  1280   66%  2696
  28 Junior 9.0  256MB Athlon 1200 MHz       2781   25   -24   868   63%  2684
  29 Rybka 1.0 beta 128MB K6-2 450 MHz       2774   64   -69   115   38%  2858
  30 Chess Tiger 2007  256MB Athlon 1200 MHz 2766   27   -27   662   49%  2774
  31 Spike 1.1  256MB Athlon 1200 MHz        2758   31   -30   537   57%  2705
  32 Zap!Chess  256MB Athlon 1200 MHz        2731   30   -29   562   53%  2707
  33 Gandalf 6.0  256MB  Athlon 1200 MHz     2728   24   -24   835   56%  2686
  34 Chess Tiger 2004  256MB Athlon 1200 MHz 2721   21   -21  1088   52%  2704
  35 Chessmaster 9000  256MB Athlon 1200 MHz 2714   36   -35   385   56%  2674
  36 Pro Deo 1.1  256MB Athlon 1200 MHz      2708   24   -23   876   57%  2654
  37 Deep Sjeng 1.5a  256MB Athlon 1200 MHz  2668   31   -31   493   52%  2657
  38 Revelation Rybka 2.2  XScale 500 MHz    2636   60   -56   154   64%  2532
  39 Ruffian 2.0.0  256MB  Athlon 1200 MHz   2626   49   -49   205   46%  2656
  40 Gromit 3.11.9  256MB Athlon 1200 MHz    2604   44   -46   246   43%  2656
  41 Yace Paderborn  256MB Athlon 1200 MHz   2602   35   -35   389   48%  2616
  42 Crafty 19.17  256MB Athlon 1200 MHz     2525   41   -44   304   30%  2676
  43 Pocket Fritz 2 XScale 400 MHz           2508   52   -50   191   60%  2437
  44 Resurrection Rybka 2.2 StrongARM 203 MH 2499   50   -50   196   53%  2479
  45 Resurrection Fruit '05 StrongARM 203 MH 2396   67   -63   120   60%  2321
  46 Hiarcs 9.5a/9.6 Palm TungstenE OMAP 126 2395   35   -35   404   45%  2431
  47 CEBoard Crafty 2004 HP RX4240  400 MHz  2372   64   -64   122   51%  2366
  48 R30 v. 2.5                              2273   41   -38   343   69%  2134
  49 Chess Genius 1.4 SX1  OMAP 310 120 MHz  2150   50   -48   210   60%  2080
  50 Chess Tiger 14.9 Palm m515 16MB 42MHz   2101   69   -74   100   39%  2180

[Yarget]

Thanks a lot for the new SSDF Ratinglist. I know that it is "old news" but once again the strength of the Junior book is confirmed by the latest SSDF list. If you for example take a look at the rating difference between Deep Fritz 11 and Deep Junior 10.1 it is 104 ratingpoints. Compare this to:

CEGT 120/40 Quad list: 3031 - 2842 = 191 ratingpoints difference
CCRL 40/40: 3094 - 2921 (Deep Junior 10) = 184 ratingpoints difference

Or take the difference between Junior and Naum 3/3.1. At SSDF the difference is 65 points in favour of Naum. Compare this to:

CEGT 120/40 Quad list: 3009 (Naum 3) - 2842 = 167 ratingpoints difference
CCRL 40/40: 3075 - 2921 = 154 ratingpoints difference

The junior book is indeed optimized for the special playing style of Junior and increases the strength of the program by at least 80 ratingpoints (compared to standard/neutral books).

I wonder when Junior 11 appear.

Regards
Per

[Uri Blass]

Thanks a lot for the new SSDF Ratinglist. I know that it is "old news" but once again the strength of the Junior book is confirmed by the latest SSDF list. If you for example take a look at the rating difference between Deep Fritz 11 and Deep Junior 10.1 it is 104 ratingpoints. Compare this to:

CEGT 120/40 Quad list: 3031 - 2842 = 191 ratingpoints difference
CCRL 40/40: 3094 - 2921 (Deep Junior 10) = 184 ratingpoints difference

Or take the difference between Junior and Naum 3/3.1. At SSDF the difference is 65 points in favour of Naum. Compare this to:

CEGT 120/40 Quad list: 3009 (Naum 3) - 2842 = 167 ratingpoints difference
CCRL 40/40: 3075 - 2921 = 154 ratingpoints difference

The junior book is indeed optimized for the special playing style of Junior and increases the strength of the program by at least 80 ratingpoints (compared to standard/neutral books).

I wonder when Junior 11 appear.

Regards
Per

A book can reduce the difference in playing strength so I believe that part of the advantage of Junior in rating in ssdf is thanks to the fact that is has a weaker rating.

Here is a comparison

Deep Rybka 3 x64 2GB Q6600 2,4 GHz 3227 SSDF 3225 CCRL
Deep Fritz 11 2GB Q6600 2,4 GHz 3085 SSDF 3094 CCRL
Zappa Mexico II x64 2GB Q6600 2,4 GHz 3064 SSDF 3074 CCRL
Naum 3.1 x64 2GB Q6600 2,4 GHz 3046 SSDF 3075 CCRL


Deep Shredder 11 2GB Q6600 2,4 GHz 3043 SSDF 3018 CCRL
Deep Hiarcs 12 x64 2GB Q6600 2,4 GHz 3033 SSDF 3002 CCRL
Hiarcs 11.2 MP 3008 SSDF 2968 CCRL
Deep Junior 10.1 2GB Q6600 2,4 GHz 2981 SSDF 2910 CCRL

You can see that the weaker programs have higher SSDF rating when the stronger programs except rybka have higher CCRL rating.

Uri

[Yarget]

Thanks for your answer Uri. You´re right, there seems to be a general coherence between playingstrength and the benefit of using own book (the weaker the engine is, the more it benefits from using its own book and vice versa). I admit that I wasn't aware of this coherence. Still, I believe that the book of Junior is extremely optimized. As your results show, "weaker" engines like Hiarcs and Shredder benefit around 25-40 ratingpoints while Junior earns more than 70 ratingpoints when using its own book. Quite a difference I think.

I think it's fair to conclude that the high rating of Junior in SSDF is a combination of being a "weak" engine and a highly optimized book.

I wonder if there are engines that benefit even more than Junior when using its own book? Quite frankly I don't think so but I could be wrong.

Regards
Per

[Laskos]

Thanks for your answer Uri. You´re right, there seems to be a general coherence between playingstrength and the benefit of using own book (the weaker the engine is, the more it benefits from using its own book and vice versa). I admit that I wasn't aware of this coherence. Still, I believe that the book of Junior is extremely optimized. As your results show, "weaker" engines like Hiarcs and Shredder benefit around 25-40 ratingpoints while Junior earns more than 70 ratingpoints when using its own book. Quite a difference I think.

I think it's fair to conclude that the high rating of Junior in SSDF is a combination of being a "weak" engine and a highly optimized book.

I wonder if there are engines that benefit even more than Junior when using its own book? Quite frankly I don't think so but I could be wrong.

Regards
Per

Very good observation about Junior's book. I observed that from the time I bought Junior 8. Junior seems very sensitive to books, at least when playing with weaker programs. It has a very peculiar play, and plays much better with its own book. I saw also on your CSS (I don't know if you are still active there) rating list that Junior has some high preferences for some openings. The success percentages were oscillating wildly. Anyway, Junior is a good alternative, as it is Fritz, to Rybka for searching some wild tactics.

Regards,
Kai

[Vinvin]

I think it's fair to conclude that the high rating of Junior in SSDF is a combination of being a "weak" engine and a highly optimized book.
...
Regards
Per

I don't think Junior is weak ( http://ticc.uvt.nl/icga/news/wccc/ Junior won the WCCC in 2002, 2004 and 2006 !), I think it's very strong in some positions (with initiative/attack/bishops pair/opened/many check involved) and less strong in other ... but the word "weak" for a multi-worldchampion shock me ...

My best,
Vincent

[icander]

You can download the short list, the long list and the comments at my site:
http://home.interact.se/~w100107/welcome.htm

There you also find our testgames, and I've updated my testsuite with programs on Q6600.

[Uri Blass]

I think it's fair to conclude that the high rating of Junior in SSDF is a combination of being a "weak" engine and a highly optimized book.
...
Regards
Per

I don't think Junior is weak ( http://ticc.uvt.nl/icga/news/wccc/ Junior won the WCCC in 2002, 2004 and 2006 !), I think it's very strong in some positions (with initiative/attack/bishops pair/opened/many check involved) and less strong in other ... but the word "weak" for a multi-worldchampion shock me ...

My best,
Vincent

weak is relative term and Junior is certainly weak relative to 2008-2009 software.

It may be strong in specific positions but the facts show that many programs have higher ssdf rating and higher ccrl rating.

Uri