Hello everyone!
For the time being there are no further strong MP engines that I would like to add to my current testproject (The Positional and Gambit Ratinglist, look here for details: http://64.68.157.89/forum/viewtopic.php?t=18891&start=0 ). I would like to test the two latest entries (Naum 3 and Toga MP) under different conditions and consequently I have started a strong tournament a couple of days ago with these participants:
Rybka 2.3.2a mp 32 bit
Naum 3
Deep Junior 10.1
Deep Fritz 10
Hiarcs 11.1 MP
Deep Shredder 11
Zap!Chess Zanzibar
Toga II 1.4 beta5c
As you see quite a strong field and I'm looking forward to see how the new "guys" (Naum and Toga) will perform. I expect an interesting tournament in which Rybka (as always) is the clear favourite. Here are the tournamentconditions:
Windows XP Pro 32 bit
(Deep) Fritz 10 GUI
AMD Athlon 64 X2 4200 (each engine is using 2 CPU's)
512 MB Hashtables for each engine
3-4-5 Tablebases (32 MB cache)
Pondern OFF
Timecontrol: 120/40 + 60/20 + 30 minutes for the rest of the game
Books: each engine playing with its own book, Naum with the "tiny book" and Toga with "Performance.bin" book by Marc Lacrosse
Double-rounded tournament meaning that each engine will play every engine with both white and black (all together 14 games for each engine)
At this moment 4 rounds are played and here comes the results so far:
Round 1:
Naum - Fritz 1-0
Toga - Rybka 1-0 (!)
Junior - Shredder 0-1
Hiarcs - Zappa ½-½
No doubt, the upset of the tournament so far was that Toga defeated Rybka! You can see the game below.
Round 2:
Fritz - Zappa ½-½
Shredder - Hiarcs 1-0
Rybka - Junior 1-0
Naum - Toga ½-½
Round 3:
Toga - Fritz ½-½
Junior - Naum 0-1
Hiarcs - Rybka ½-½
Zappa - Shredder ½-½
Round 4:
Fritz - Shredder ½-½
Rybka - Zappa 1-0
Naum - Hiarcs 0-1
Toga - Junior ½-½
These results result in the following current stand:
1. Deep Shredder 11 3 points
2-4. Naum 3, Rybka 2.3.2a mp & Toga II 1.4 beta5c each 2½ points
5. Hiarcs 11.1 MP 2 points
6-7. Deep Fritz 10 & Zap!Chess Zanzibar each 1½ points
8. Deep Junior 10.1 ½ points
Shredder has made a strong start but also Naum and Toga are doing very well. After the defeat in round 1 against Toga Rybka has started to score points and is now close to Shredder. I feel sorry for Junior which is one of my favourite engines. Playing with its own book Junior is a very strong and dangerous engine but so far without luck in this tournament.
Here are the round 5 pairings:
Junior - Fritz
Hiarcs - Toga
Zappa - Naum
Shredder - Rybka (a true topgame)
Regards
Per
As promised the defeat of Rybka in round 1:
[Date "2008.02.21"]
[Round "1.1"]
[White "Toga II 1.4 beta5c"]
[Black "Rybka 2.3.2a mp 32-bit"]
[Result "1-0"]
[ECO "D76"]
[Annotator "0.18;0.12"]
[PlyCount "117"]
[EventDate "2008.02.21"]
[EventType "tourn"]
[Source "Elbæk"]
[TimeControl "40/7200:20/3600:1800"]
{AMD Athlon(tm) 64 X2 Dual Core Processor 4200+ 2202 MHz W=22.1 ply; 1.781kN/
s B=22.6 ply; 129kN/s; 11.909 TBAs; RybkaII.ctg} 1. d4 {
[%eval 1,1] [%emt 0:00:00]} Nf6 {[%eval 0,0] [%emt 0:00:00]} 2. c4 {
[%eval 1,1] [%emt 0:00:00]} g6 {[%eval 0,0] [%emt 0:00:00]} 3. g3 {
[%eval 1,1] [%emt 0:00:00]} Bg7 {[%eval 0,0] [%emt 0:00:00]} 4. Nf3 {
[%eval 1,1] [%emt 0:00:00]} O-O {[%eval 0,0] [%emt 0:00:00]} 5. Bg2 {
[%eval 1,1] [%emt 0:00:00]} d5 {[%eval 0,0] [%emt 0:00:00]} 6. cxd5 {
[%eval 18,19] [%emt 0:03:48]} Nxd5 {[%eval 0,0] [%emt 0:00:00]} 7. O-O {
[%eval 10,19] [%emt 0:02:51]} Nb6 {[%eval 0,0] [%emt 0:00:00]} 8. Nc3 {
[%eval 1,1] [%emt 0:00:00]} Nc6 {[%eval 0,0] [%emt 0:00:00]} 9. d5 {
[%eval 1,1] [%emt 0:00:00]} Na5 {[%eval 0,0] [%emt 0:00:00]} 10. Qc2 {
[%eval 1,1] [%emt 0:00:00]} c6 {[%eval 0,0] [%emt 0:00:00]} 11. dxc6 {
[%eval 1,1] [%emt 0:00:00]} Nxc6 {[%eval 0,0] [%emt 0:00:00]} 12. Rd1 {
[%eval 1,1] [%emt 0:00:00]} Qe8 {[%eval 0,0] [%emt 0:00:00]} 13. Be3 {
[%eval 1,1] [%emt 0:00:00]} Bf5 {[%eval 0,0] [%emt 0:00:00]} 14. Qc1 {
[%eval 12,19] [%emt 0:03:40]} Rd8 {[%eval 12,17] [%emt 0:03:27]} 15. Bh6 {
[%eval 6,19] [%emt 0:04:00]} Rxd1+ {[%eval 5,16] [%emt 0:00:54]} 16. Nxd1 {
[%eval 3,18] [%emt 0:01:02]} e5 {[%eval 5,17] [%emt 0:03:50]} 17. Bxg7 {
[%eval 6,19] [%emt 0:02:59]} Kxg7 {[%eval 5,15] [%emt 0:00:00]} 18. Qc5 {
(Ne3) [%eval 7,19] [%emt 0:03:31]} f6 {(Bg4) [%eval 5,18] [%emt 0:05:11]} 19.
Nc3 {[%eval 1,18] [%emt 0:03:03]} Rf7 {(Qf7) [%eval 5,17] [%emt 0:04:37]} 20.
Nd2 {(Nh4) [%eval 22,18] [%emt 0:02:38]} Qe6 {
(Rd7) [%eval 14,18] [%emt 0:04:56]} 21. Nb3 {[%eval 18,19] [%emt 0:04:55]} Rd7
{[%eval 9,17] [%emt 0:01:06]} 22. Bxc6 {(e3) [%eval 23,19] [%emt 0:03:09]} bxc6
{(Qxc6) [%eval 20,19] [%emt 0:04:09]} 23. Rc1 {[%eval 47,19] [%emt 0:02:46]}
Nc4 {[%eval 21,19] [%emt 0:03:51]} 24. Nd1 {[%eval 51,19] [%emt 0:04:58]} a6 {
(g5) [%eval 31,19] [%emt 0:06:12]} 25. Ne3 {[%eval 56,19] [%emt 0:02:46]} Nxe3
{[%eval 29,19] [%emt 0:01:22]} 26. Qxe3 {[%eval 51,20] [%emt 0:04:41]} Qd5 {
[%eval 34,19] [%emt 0:05:11]} 27. f3 {[%eval 45,20] [%emt 0:04:38]} h5 {
[%eval 37,19] [%emt 0:04:03]} 28. Kf2 {[%eval 62,20] [%emt 0:05:29]} Qb5 {
(a5) [%eval 32,19] [%emt 0:01:40]} 29. h4 {(a4) [%eval 72,20] [%emt 0:04:03]}
Rd5 {[%eval 45,20] [%emt 0:04:07]} 30. Qa7+ {[%eval 63,21] [%emt 0:04:46]} Rd7
{[%eval 46,21] [%emt 0:01:12]} 31. Qc5 {[%eval 66,21] [%emt 0:02:56]} Qa4 {
[%eval 53,22] [%emt 0:05:23]} 32. Qxc6 {(Qa5) [%eval 85,21] [%emt 0:05:00]}
Qxa2 {[%eval 62,19] [%emt 0:02:00]} 33. Nc5 {[%eval 72,21] [%emt 0:05:02]} Re7
{(Rf7) [%eval 70,20] [%emt 0:07:23]} 34. Qb6 {
(Qd6) [%eval 89,20] [%emt 0:05:04]} a5 {(Qd5) [%eval 64,20] [%emt 0:06:09]} 35.
Qd8 {(Nb7) [%eval 109,20] [%emt 0:04:55]} Rf7 {[%eval 81,22] [%emt 0:05:02]}
36. b3 {[%eval 123,21] [%emt 0:03:48]} Rf8 {[%eval 81,22] [%emt 0:06:45]} 37.
Qd1 {[%eval 115,21] [%emt 0:04:01]} e4 {[%eval 81,22] [%emt 0:07:01]} 38. fxe4
{(Nxe4) [%eval 123,20] [%emt 0:04:02]} Bg4 {[%eval 74,21] [%emt 0:05:27]} 39.
Ra1 {(Rc2) [%eval 125,21] [%emt 0:05:21]} Rd8 {[%eval 77,22] [%emt 0:03:01]}
40. Rxa2 {[%eval 131,22] [%emt 0:09:38]} Rxd1 {[%eval 71,20] [%emt 0:00:00]}
41. Nd3 {(Rxa5) [%eval 129,23] [%emt 0:02:53]} Rb1 {
[%eval 93,23] [%emt 0:07:07]} 42. Ra3 {[%eval 133,23] [%emt 0:03:17]} Kg8 {
(Bc8) [%eval 94,23] [%emt 0:07:53]} 43. Nf4 {(e5) [%eval 164,22] [%emt 0:02:13]
} g5 {[%eval 105,22] [%emt 0:01:49]} 44. hxg5 {[%eval 166,22] [%emt 0:02:09]}
fxg5 {[%eval 107,23] [%emt 0:01:12]} 45. Nd3 {[%eval 181,22] [%emt 0:02:58]} h4
{(Kg7) [%eval 113,22] [%emt 0:01:04]} 46. Rxa5 {[%eval 186,21] [%emt 0:03:35]}
h3 {[%eval 115,23] [%emt 0:06:00]} 47. Rxg5+ {[%eval 192,23] [%emt 0:02:57]}
Kf7 {[%eval 116,25] [%emt 0:03:29]} 48. Rxg4 {[%eval 201,22] [%emt 0:02:35]} h2
{[%eval 116,25] [%emt 0:01:01]} 49. Rh4 {[%eval 207,23] [%emt 0:01:55]} h1=Q {
[%eval 139,25] [%emt 0:02:10]} 50. Rxh1 {[%eval 207,23] [%emt 0:02:52]} Rxh1 {
[%eval 140,27] [%emt 0:01:00]} 51. b4 {[%eval 210,22] [%emt 0:02:44]} Ke6 {
[%eval 140,27] [%emt 0:01:34]} 52. g4 {[%eval 222,22] [%emt 0:03:04]} Rh7 {
(Rh8) [%eval 141,27] [%emt 0:04:40]} 53. b5 {[%eval 229,24] [%emt 0:03:15]} Rg7
{(Rh1) [%eval 189,26] [%emt 0:03:46]} 54. Kf3 {[%eval 279,23] [%emt 0:02:20]}
Rg8 {[%eval 228,27] [%emt 0:03:36]} 55. e5 {(b6) [%eval 354,23] [%emt 0:03:14]}
Rb8 {[%eval 428,28] [%emt 0:06:35]} 56. Ke4 {[%eval 502,25] [%emt 0:02:42]} Kd7
{(Rxb5) [%eval 428,27] [%emt 0:05:07]} 57. g5 {[%eval 561,25] [%emt 0:03:05]}
Rxb5 {[%eval 509,28] [%emt 0:03:58]} 58. Kf5 {
(Nf4) [%eval 595,27] [%emt 0:03:02]} Ke7 {(Rb8) [%eval 508,24] [%emt 0:03:05]}
59. Nf4 {(g6) [%eval 969,24] [%emt 0:03:04]} 1-0
Strong Naum 3 tournament with long TC (120/40)
Moderator: Ras
-
Heinz Van Kempen
Re: Strong Naum 3 tournament with long TC (120/40)
Hi Per 
,
thanks for first results from this interesting tournament. How long until Rybka will take a clear lead?
By the way I had some trouble to detect your posting due to long threads here and because flat view is not defaulted like in Rybka Forum for example (I usually read from time to time without logging on). In view of some people posting very frequently flat view default would be helpful.
,thanks for first results from this interesting tournament. How long until Rybka will take a clear lead?
By the way I had some trouble to detect your posting due to long threads here and because flat view is not defaulted like in Rybka Forum for example (I usually read from time to time without logging on). In view of some people posting very frequently flat view default would be helpful.
-
Yarget
Re: Strong Naum 3 tournament with long TC (120/40)
Hello Heinz!
Thanks for your interest in my current tournament. Needless to say, Rybka is (as always) the favourite to win but with "only" 14 games per engine I guess everything is possible. If it was 28, 42 or more games per engine then the outcome would be 99,9% sure (Rybka would win) but now with only 14 games per engine almost everything is possible (which makes it more exciting to watch).
Round 5 is just completed and here are the results:
Junior - Fritz 0-1 (Another defeat for Junior)
Hiarcs - Toga 1-0 (First defeat for Toga!)
Zappa - Naum ½-½ (Fourth draw for Zappa in 5 games)
Shredder - Rybka ½-½ (Exciting topgame that ended with a draw)
The current rank after round 5:
1. Deep Shredder 11 3½ points
2-4. Naum 3, Rybka 2.3.2a mp & Hiarcs 11.1 MP each 3 points
5-6. Deep Fritz 10 & Toga II 1.4 beta5c each 2½ points
7. Zap!Chess Zanzibar 2 points
8. Deep Junior 10.1 ½ points
So Shredder defended the lead but it's very close in the top. Hiarcs moved further ahead while Toga fell back to a 50% score. With only 1½ points from number 1 to number 7 it's still a very open tournament. Only Junior seems to have lost the chance to win the tournament.
The round 6 pairings:
Fritz - Rybka (So far no black wins for Rybka, perhaps the first one will come here?)
Naum - Shredder (IMO the game of the round. Will Shredder still be the only undefeated engine after this game?)
Toga - Zappa (I wouldn't be surprised if this will end with a draw)
Junior - Hiarcs (Will Hiarcs win its 3rd game in a row?)
By the way Heinz, I saw your first Naum 3 results on 40/120 yesterday. This is indeed a very strong performance by Naum 3. If I read your results correct then Naum 3 is 105 ELO stronger than Naum 2.2! I know it is still early days in your Naum 3 tests but still...... Perhaps I was right when I (and also you) predicted that Naum 3 would have some chances to reach +100 ELO improvement under your testconditions:
http://64.68.157.89/forum/viewtopic.php ... 44&t=19708
I am looking forward to your next results as well as your marathon match
Best regards
Per
Thanks for your interest in my current tournament. Needless to say, Rybka is (as always) the favourite to win but with "only" 14 games per engine I guess everything is possible. If it was 28, 42 or more games per engine then the outcome would be 99,9% sure (Rybka would win) but now with only 14 games per engine almost everything is possible (which makes it more exciting to watch).
Round 5 is just completed and here are the results:
Junior - Fritz 0-1 (Another defeat for Junior)
Hiarcs - Toga 1-0 (First defeat for Toga!)
Zappa - Naum ½-½ (Fourth draw for Zappa in 5 games)
Shredder - Rybka ½-½ (Exciting topgame that ended with a draw)
The current rank after round 5:
1. Deep Shredder 11 3½ points
2-4. Naum 3, Rybka 2.3.2a mp & Hiarcs 11.1 MP each 3 points
5-6. Deep Fritz 10 & Toga II 1.4 beta5c each 2½ points
7. Zap!Chess Zanzibar 2 points
8. Deep Junior 10.1 ½ points
So Shredder defended the lead but it's very close in the top. Hiarcs moved further ahead while Toga fell back to a 50% score. With only 1½ points from number 1 to number 7 it's still a very open tournament. Only Junior seems to have lost the chance to win the tournament.
The round 6 pairings:
Fritz - Rybka (So far no black wins for Rybka, perhaps the first one will come here?)
Naum - Shredder (IMO the game of the round. Will Shredder still be the only undefeated engine after this game?)
Toga - Zappa (I wouldn't be surprised if this will end with a draw)
Junior - Hiarcs (Will Hiarcs win its 3rd game in a row?)
By the way Heinz, I saw your first Naum 3 results on 40/120 yesterday. This is indeed a very strong performance by Naum 3. If I read your results correct then Naum 3 is 105 ELO stronger than Naum 2.2! I know it is still early days in your Naum 3 tests but still...... Perhaps I was right when I (and also you) predicted that Naum 3 would have some chances to reach +100 ELO improvement under your testconditions:
http://64.68.157.89/forum/viewtopic.php ... 44&t=19708
I am looking forward to your next results as well as your marathon match
Best regards
Per
-
Heinz Van Kempen
Re: Strong Naum 3 tournament with long TC (120/40)
Hi Per ,
I just saw that I can change to flat view without being logged in. Looks much better.
Yes, with a more limited amount of rounds we have more suspense.
Can we download the games somewhere? Would be better than only the results.
Regarding Naum so far it continues with slightly more than +100 ELO against the current opponents Hiarcs 11.2, DF 10.1 and Loop M1P. Convincing big lead against all currently. But one catastrophical result against the "weakest" in the list would be sufficient to drag it down.
,I just saw that I can change to flat view without being logged in. Looks much better.
Yes, with a more limited amount of rounds we have more suspense.
Can we download the games somewhere? Would be better than only the results.
Regarding Naum so far it continues with slightly more than +100 ELO against the current opponents Hiarcs 11.2, DF 10.1 and Loop M1P. Convincing big lead against all currently. But one catastrophical result against the "weakest" in the list would be sufficient to drag it down.
-
Uri Blass
- Posts: 10898
- Joined: Thu Mar 09, 2006 12:37 am
- Location: Tel-Aviv Israel
Re: Strong Naum 3 tournament with long TC (120/40)
I see nothing close to it based on the ccrl resultsHeinz Van Kempen wrote:Hi Per
I just saw that I can change to flat view without being logged in. Looks much better.
Yes, with a more limited amount of rounds we have more suspense.
Can we download the games somewhere? Would be better than only the results.
Regarding Naum so far it continues with slightly more than +100 ELO against the current opponents Hiarcs 11.2, DF 10.1 and Loop M1P. Convincing big lead against all currently. But one catastrophical result against the "weakest" in the list would be sufficient to drag it down.
Naum seems to be relatively stronger in blitz based on CCRL results
40/40
Naum 3 64-bit 4CPU 3047 +34 −34 57.6% −52.1 44.3% 271 (+37 elo)
Naum 2.2 64-bit 4CPU 3010 +17 −17 56.5% −41.7 46.2% 1030
Naum 3 64-bit 2CPU 3047 +32 −31 63.4% −88.7 41.0% 322 (+69 elo)
Naum 2.2 64-bit 2CPU 2978 +23 −23 56.9% −43.0 49.6% 568
Naum 3 64-bit 2952 +45 −45 46.4% +21.6 52.2% 138 (+51 elo)
Naum 2.2 64-bit 2901 +26 −26 50.3% −2.8 43.0% 472
Naum 3 32-bit 2973 +41 −41 58.7% −56.2 49.1% 173(+79 elo)
Naum 2.2 32-bit 2894 +17 −17 54.6% −28.4 46.8% 1122
40/4
Naum 3 64-bit 4CPU 3077 +31 −30 61.7% −81.3 34.9% 364 (+81 elo)
Naum 2.2 64-bit 4CPU 2996 +12 −12 55.5% −37.9 39.3% 2274
Naum 3 64-bit 2959 +37 −37 62.3% −85.0 33.6% 244 (+82 elo)
Naum 2.2 64-bit 2877 +15 −15 44.9% +32.9 39.1% 1449
-
Heinz Van Kempen
Re: Strong Naum 3 tournament with long TC (120/40)
Hi Uri ,
you know that I am cautious, but not always pessimistic.
When we take a look at the rating progression list we had for the single and dual list with long time controls, we can indeed see a lot of changes between 200 and 500 games.
http://www.husvankempen.de/nunn/Replay/ ... ession.htm
Moreover later this year and with more engine versions we will have also more games for each one in the quad list.
But it seems obvious that Naum does better with this conditions. I usually compare here with our own CEGT rating lists 40/20 where we currently have +76 ELO and not with other rating lists.
,you know that I am cautious, but not always pessimistic.
When we take a look at the rating progression list we had for the single and dual list with long time controls, we can indeed see a lot of changes between 200 and 500 games.
http://www.husvankempen.de/nunn/Replay/ ... ession.htm
Moreover later this year and with more engine versions we will have also more games for each one in the quad list.
But it seems obvious that Naum does better with this conditions. I usually compare here with our own CEGT rating lists 40/20 where we currently have +76 ELO and not with other rating lists.
-
Yarget
Re: Strong Naum 3 tournament with long TC (120/40)
Speaking of performances by Naum 3 I am now able to compare the performance of Naum 3 in my 2 Ratinglists (The positional and Gambit Ratinglist) with the referencelist (CEGT 40/4 in which Naum 3 made its entry recently, thanks to CEGT for your great testwork ). Allow me once more to present the two latest ratinglists (last engine tested was Naum 3):
The Positional Ratinglist:
And here comes the Gambit Ratinglist:
Perhaps it's worth explaining how I compare my lists and the CEGT 40/4 ratinglist. It goes like this for the Positional Ratinglist (the method is ofcourse the same for the Gambit Ratinglist):
I compare Naum 3 (like I do with every new engine entering the list) with all engines in my ratinglist starting with Rybka 2.3.2a mp. The referencelist (CEGT) shows a ratingdifference between Rybka and Naum of 50 ratingpoints (2981 - 2931), in my Positional list the difference is 72 ratingpoints meaning that Naum is performing 22 ratingpoints less than expected. Compared with Toga it is also -22, Shredder +10 and so on. In the end you add all the +'s and -'s and you get +162 and -144 meaning that Naum 3 has 18 ratingpoints "too much" compared to the the referencelist. This number must be divided with the number of opponents: 18/11 = 1,64 ratingpoints. In other words: Naum 3 has made a performance that is 1,64 ratingpoints higher than expected which ofcourse is very close to the referencelist. This is a remarkable improvement compared to Naum 2.2 (minus 36,33 ratingpoints for the positional ratinglist compared to CEGT 40/4).
The performance of Naum 3 in the Gambit Ratinglist is even better: plus 37,64 ratingpoints compared to CEGT 40/4 (Naum 2.2 was also doing very fine in the Gambitgames: +23,67 ratingpoints compared to CEGT 40/4).
Generally the performance of Naum 3 in my tests are close to that of CEGT 40/4. In the Positional Ratinglist it's almost equal while Naum 3 is performing better than expected (around 38 ratingpoints) in the Gambit Ratinglist.
Regards
Per
). Allow me once more to present the two latest ratinglists (last engine tested was Naum 3):The Positional Ratinglist:
Code: Select all
Program Elo + - Games Score Av.Op. Draws
1 Rybka 2.3.2a mp 32-bit : 2928 40 39 220 69.1 % 2788 31.8 %
2 Naum 3 : 2856 38 38 220 58.6 % 2795 33.6 %
3 Toga II 1.4 beta5c : 2848 37 37 220 57.5 % 2796 35.0 %
4 Deep Shredder 11 UCI : 2841 39 39 220 56.4 % 2796 30.0 %
5 Deep Fritz 10 : 2832 39 39 220 55.0 % 2797 30.0 %
6 SpikeMP 1.2 Turin : 2788 38 38 220 48.2 % 2801 32.7 %
7 Zap!Chess Zanzibar : 2787 37 37 220 48.0 % 2801 36.8 %
8 LoopMP 11A.32 : 2783 37 37 220 47.3 % 2802 34.5 %
9 Deep Junior 10.1 : 2781 42 42 220 47.0 % 2802 18.6 %
10 HIARCS 11.1 MP UCI : 2767 37 37 220 44.8 % 2803 35.0 %
11 bright-0.2c : 2700 40 41 220 34.8 % 2809 27.7 %
12 Glaurung 2.0.1 : 2690 40 40 220 33.4 % 2810 28.6 % And here comes the Gambit Ratinglist:
Code: Select all
Program Elo + - Games Score Av.Op. Draws
1 Rybka 2.3.2a mp 32-bit : 2958 43 42 220 73.0 % 2786 25.9 %
2 Naum 3 : 2889 41 41 220 63.6 % 2792 25.5 %
3 Toga II 1.4 beta5c : 2860 39 39 220 59.3 % 2794 30.5 %
4 Deep Shredder 11 UCI : 2848 40 40 220 57.5 % 2796 25.9 %
5 HIARCS 11.1 MP UCI : 2825 40 39 220 53.9 % 2798 26.8 %
6 Deep Fritz 10 : 2822 41 41 220 53.4 % 2798 21.4 %
7 LoopMP 11A.32 : 2810 39 39 220 51.6 % 2799 28.6 %
8 Zap!Chess Zanzibar : 2778 39 39 220 46.6 % 2802 27.7 %
9 Glaurung 2.0.1 : 2746 40 40 220 41.6 % 2805 25.0 %
10 bright-0.2c : 2701 42 43 220 35.0 % 2809 20.0 %
11 Deep Junior 10.1 : 2687 44 45 220 33.0 % 2810 16.8 %
12 SpikeMP 1.2 Turin : 2677 42 43 220 31.6 % 2811 23.2 % Perhaps it's worth explaining how I compare my lists and the CEGT 40/4 ratinglist. It goes like this for the Positional Ratinglist (the method is ofcourse the same for the Gambit Ratinglist):
I compare Naum 3 (like I do with every new engine entering the list) with all engines in my ratinglist starting with Rybka 2.3.2a mp. The referencelist (CEGT) shows a ratingdifference between Rybka and Naum of 50 ratingpoints (2981 - 2931), in my Positional list the difference is 72 ratingpoints meaning that Naum is performing 22 ratingpoints less than expected. Compared with Toga it is also -22, Shredder +10 and so on. In the end you add all the +'s and -'s and you get +162 and -144 meaning that Naum 3 has 18 ratingpoints "too much" compared to the the referencelist. This number must be divided with the number of opponents: 18/11 = 1,64 ratingpoints. In other words: Naum 3 has made a performance that is 1,64 ratingpoints higher than expected which ofcourse is very close to the referencelist. This is a remarkable improvement compared to Naum 2.2 (minus 36,33 ratingpoints for the positional ratinglist compared to CEGT 40/4).
The performance of Naum 3 in the Gambit Ratinglist is even better: plus 37,64 ratingpoints compared to CEGT 40/4 (Naum 2.2 was also doing very fine in the Gambitgames: +23,67 ratingpoints compared to CEGT 40/4).
Generally the performance of Naum 3 in my tests are close to that of CEGT 40/4. In the Positional Ratinglist it's almost equal while Naum 3 is performing better than expected (around 38 ratingpoints) in the Gambit Ratinglist.
Regards
Per
-
Yarget
Re: Strong Naum 3 tournament with long TC (120/40)
Round 6 is now completed and here are the results:
Deep Fritz 10 - Rybka 2.3.2a mp 0-1 (A typical Rybka-win, an almost invisible advantage that slowly accumulated, great game)
Naum 3 - Deep Shredder 11 ½-½ (An interesting game but the game never got out of balance, Shredder is still the only undefeated engine in the tournament)
Toga II 1.4 beta5c - Zap!Chess Zanzibar 0-1 (Finally Zappa won a game in this tournament, for Toga it was the second defeat in a row)
Deep Junior 10.1 - Hiarcs 11.1 MP 1-0 (First win for Junior in this tournament and a strong one in only 37 moves)
And here the updated rankings after round 6:
1-2. Deep Shredder 11 & Rybka 2.3.2a mp each 4 points
3. Naum 3 3½ points
4-5 Hiarcs 11.1 MP & Zap!Chess Zanzibar each 3 points
6-7. Deep Fritz 10 & Toga II 1.4 beta5c each 2½ points
8. Deep Junior 10.1 1½ points
It is interesting to see that my favourites for rank 1,2 and 3 (Rybka, Shredder and Naum) are now in Top 3. Having said that it should be emphasized that it is still a very tight tournament with only 1½ points between number 1 and 7. Junior decreased the gap to the field but I wonder if Junior will escape from rank 8.
Here follows the pairings for round 7 after which each engine has played each opponent one time (I play 14 rounds all together):
Hiarcs 11.1 MP - Deep Fritz 10 (both engines need a win to get closer to the top)
Zap!Chess Zanzibar - Deep Junior 10.1 (If Zappa can win the second game in a row then it will move close to the top)
Deep Shredder 11 - Toga II 1.4 beta5c (an important game for both, Shredder needs at least a draw to maintain the topposition while Toga has to score again after 2 defeats in a row)
Rybka 2.3.2a mp - Naum 3 (IMO the game of the round, Naum needs at least a draw to stay close to the top while Rybka with a win can start to dominate the tournament)
If anyone is interested in the games please send me a PM and I'll send you the games.
Regards
Per
Deep Fritz 10 - Rybka 2.3.2a mp 0-1 (A typical Rybka-win, an almost invisible advantage that slowly accumulated, great game)
Naum 3 - Deep Shredder 11 ½-½ (An interesting game but the game never got out of balance, Shredder is still the only undefeated engine in the tournament)
Toga II 1.4 beta5c - Zap!Chess Zanzibar 0-1 (Finally Zappa won a game in this tournament, for Toga it was the second defeat in a row)
Deep Junior 10.1 - Hiarcs 11.1 MP 1-0 (First win for Junior in this tournament and a strong one in only 37 moves)
And here the updated rankings after round 6:
1-2. Deep Shredder 11 & Rybka 2.3.2a mp each 4 points
3. Naum 3 3½ points
4-5 Hiarcs 11.1 MP & Zap!Chess Zanzibar each 3 points
6-7. Deep Fritz 10 & Toga II 1.4 beta5c each 2½ points
8. Deep Junior 10.1 1½ points
It is interesting to see that my favourites for rank 1,2 and 3 (Rybka, Shredder and Naum) are now in Top 3. Having said that it should be emphasized that it is still a very tight tournament with only 1½ points between number 1 and 7. Junior decreased the gap to the field but I wonder if Junior will escape from rank 8.
Here follows the pairings for round 7 after which each engine has played each opponent one time (I play 14 rounds all together):
Hiarcs 11.1 MP - Deep Fritz 10 (both engines need a win to get closer to the top)
Zap!Chess Zanzibar - Deep Junior 10.1 (If Zappa can win the second game in a row then it will move close to the top)
Deep Shredder 11 - Toga II 1.4 beta5c (an important game for both, Shredder needs at least a draw to maintain the topposition while Toga has to score again after 2 defeats in a row)
Rybka 2.3.2a mp - Naum 3 (IMO the game of the round, Naum needs at least a draw to stay close to the top while Rybka with a win can start to dominate the tournament)
If anyone is interested in the games please send me a PM and I'll send you the games.
Regards
Per
-
Heinz Van Kempen
Re: Strong Naum 3 tournament with long TC (120/40)
Hi PerIf anyone is interested in the games please send me a PM and I'll send you the games.
Regards
Per
,please send your games as soon as your tournament is finished to jaque64 ad arcor dot.de (replace ad and dot like usual).
I have started a collection of games with tournament time control and better on dual core and more for my private use and want to run some stats on them (which will not be public).
Also other testers can send me such game if the have them. But only those who are sure that they always gave correct number of threads, what is critical in case of Naum for example. Errors here can happen easily when starting something new and not checking the task manager frequently.
No doubts of course that you will do this correctly Per.
-
Yarget
Re: Strong Naum 3 tournament with long TC (120/40)
Unfortunately my old PC seems to be dead 
When I returned to my home wednesday night I was confused because the PC was shut down (It should have been running with the current 40/120 tournament as it have been doing non-stop since game 1). It's impossible to turn on the computer, a good friend visited me tonight and he believes (I agree) that the motherboard is dead/broken.
Needless to say, this is a huge disappointment but I admit that I was afraid that it would happend sooner or later. I bought this motherboard (socket 939, Asus A8V) several years ago with the AMD San Diego 4000+ CPU and continued with this motherboard when I bought the X2 4200+ CPU. While running the old PEJ-Ratinglist, The CSS SMP Ratinglist, my current tests and a lot of private tests never published my PC has been running for days and weeks without breaks. Perhaps I'm now paying the price for this exhausting use of my PC but as a tester you want to produce results as quick as possible. I wonder if other testers also let their computers run 24/7 for weeks or do you give it a break every day??
My girlfriend has borrowed me an old laptop and fortunately I have internetconnection on this one so that I can write these words. I won't buy a new socket 939 motherboard (if you can buy them at all in 2008??) but try and save some money and buy a Quadcore PC. Until then I will continue to participate in this Forum but ofcourse without reporting testresults. I already miss it.....
Regards
Per
When I returned to my home wednesday night I was confused because the PC was shut down (It should have been running with the current 40/120 tournament as it have been doing non-stop since game 1). It's impossible to turn on the computer, a good friend visited me tonight and he believes (I agree) that the motherboard is dead/broken.Needless to say, this is a huge disappointment but I admit that I was afraid that it would happend sooner or later. I bought this motherboard (socket 939, Asus A8V) several years ago with the AMD San Diego 4000+ CPU and continued with this motherboard when I bought the X2 4200+ CPU. While running the old PEJ-Ratinglist, The CSS SMP Ratinglist, my current tests and a lot of private tests never published my PC has been running for days and weeks without breaks. Perhaps I'm now paying the price for this exhausting use of my PC but as a tester you want to produce results as quick as possible. I wonder if other testers also let their computers run 24/7 for weeks or do you give it a break every day??
My girlfriend has borrowed me an old laptop and fortunately I have internetconnection on this one so that I can write these words. I won't buy a new socket 939 motherboard (if you can buy them at all in 2008??) but try and save some money and buy a Quadcore PC. Until then I will continue to participate in this Forum but ofcourse without reporting testresults. I already miss it.....
Regards
Per