SWCR-64: Houd, Fire, Loop M1-T, Equi, Cipo, CraftyNP

[Dr.Wael Deeb]

Hi Wael,

I made different analyzes on a fifth quad core system I have with the Houdini and Fire games.

I compare different of this analyzes with Rybka 2 / 3 / 4.

1. Houdini
Produced many critical moves Rybka plays too. In endgames perhaps a very little bit stronger.

2. Fire
Other engine, too many other moves if I compare with the Rybka 2 / 3 / 4. I collect Rybka 3 / 4 blunders from SWCR games. Fire don't play the most of it. Rybka don't play the two blunders I have from the SWCR Fire 1.31 games. Fire produced more remis games in endgames Rybka 3 / 4 wins.

I don't have fun with Houdini, Fire is after all what I saw OK I think (don't have enough games so far). Clear is, that Houdini 1.03a is around 30-50 ELO stronger as Fire 1.31.

Best
Frank

Hi Frank,
Please check your inbox

As for Houdini it's a little bit stronger in the endgame than Rybka 4....

As I wrote before,FireBird 1.1 is the strongest version of this program and my settings make it more aggressive and yet playing sound moves but not at bullet time controls....

You can compare the default FireBird 1.1 and my settings and choose which one to include in your tournaments....
Cheers,
Dr.D

[Frank Quisinsky]

Hi Wael,

OK, thanks!

Best
Frank

[Frank Quisinsky]

Hi there,

here the results after the second ELO calculation ...

ELO calculation (2)
After 20.253 games (SWCR-64, all games)

Code: Select all

01. IPP Houdini 1.03a x64        2.954 49 46  186 80% 2.719 28% NEW
02. Rybka 4 x64                  2.940 19 19 1209 81% 2.689 28%
--. IPP Fire 1.31 x64            2.905 45 43  191 75% 2.719 35% NEW
--. Rybka 3 x64                  2.905 22 22  840 78% 2.688 28%
03. Stockfish 1.8.0 JA x64       2.900 22 21  849 76% 2.702 31%
--. Stockfish 1.7.1 JA x64       2.896 19 18 1120 76% 2.706 34%
04. Critter 0.80 x64             2.841 21 20  849 69% 2.705 33%
05. Naum 4.2 x64                 2.832 16 16 1449 68% 2.700 35%
--. Critter 0.70 x64             2.807 19 19  880 65% 2.701 38%
06. Shredder 12                  2.800 16 16 1409 64% 2.699 35%
07. Komodo 1.2 JA x64            2.798 17 16 1209 65% 2.694 40%
--. Komodo 1.0 JA x64            2.791 20 20  840 64% 2.693 40%
--. Shredder 12 x64              2.788 18 18 1080 61% 2.707 35%
08. Spark 0.5 x64                2.738 16 16 1209 56% 2.696 39%
09. Thinker 5.4d Inert x64       2.732 15 15 1449 54% 2.703 40%
10. Zappa Mexico II x64          2.723 15 15 1449 53% 2.703 39%
--. Spark 0.4 x64                2.718 19 19  840 53% 2.697 40%
11. Protector 1.3.4 JA x64       2.716 15 15 1409 52% 2.704 37%
12. Fruit 09_07_05 x64           2.712 15 15 1450 51% 2.704 34%
--. Critter 0.60 x64             2.700 20 20  840 50% 2.697 38%
13. Hannibal 1.0a x64            2.689 20 20  849 47% 2.712 36%
14. Sjeng WC-2008 x64            2.686 15 15 1449 48% 2.704 36%
--. Protector 1.3.5 x64          2.684 19 19  840 47% 2.707 39%
15. Junior 11.2 x64              2.683 17 17 1209 48% 2.698 31%
16. Onno 1.2.70 x64              2.675 17 17 1209 47% 2.698 37%
--. Onno 1.1.1 x64               2.671 19 19  840 46% 2.699 40%
--. Junior 11.1a x64             2.648 20 20  840 43% 2.700 32%
17. Loop M1-T x64                2.641 42 43  185 38% 2.732 34% + 9
--. Loop 2007 x64                2.632 15 16 1400 40% 2.706 37%
18. Equinox 0.83 x64             2.614 43 44  186 34% 2.734 30% NEW
--. Twisted Logic 20100131x x64  2.611 18 18 1120 35% 2.716 32%
19. Umko 1.0 x64                 2.607 20 20  850 35% 2.716 36% (ponder not possible)
20. SmarThink 1.20 x64           2.602 15 16 1449 36% 2.707 33%
21. Crafty 23.3 JA x64           2.589 20 20  850 33% 2.717 33%
--. Cipollino 3.25 x64           2.574 44 46  186 28% 2.735 30% NEW
22. BugChess2 1.7 x64            2.563 21 22  800 29% 2.719 33%
23. Scorpio 2.6 JA x64           2.555 18 18 1120 28% 2.718 32%
--. Crafty 23.2 JA x64           2.555 18 18 1120 28% 2.718 30%
24. Chronos 1.99 x64             2.552 18 18 1120 27% 2.719 33% (ponder not possible)
--. Crafty 23.3 JA x64 NP        2.540 42 44  207 24% 2.737 30% (without ponder = -49 ELO)
25. Daydreamer 1.75 JA x64       2.521 19 19 1120 24% 2.720 30%
26. Tornado 3.6.7 x64            2.479 23 24  800 19% 2.723 24%

Frank's Chess Page
http://www.amateurschach.de

Best
Frank

[Sedat Canbaz]

Hello Frank,

Many thanks for the update and for your efforts

Yes...the Crafty test with pondering off/on test is a little bit different than mine and i found your test to be very interesting...
About Crafty with 'ponder off' due to not enough games its early to make any conclusions,
but anyway that is very normal to be weaker approx.20-25 elo,as we know the engine plays at non-equal conditions
(the opponents with 'ponder on' are gaining time on during Crafty's thinking time)
In other words,its something like Crafty (ponder off) is using 3cpu where the opponents are using 4cpu or another example i can give too:
the time management of Crafty (ponder off) is about 40 moves in 8 minutes,where the opponents play at 40 moves in 10 minutes

Best,
Sedat

[gerold]

Thanks Frank.Your tests are greatly appreciated.

Best to you,
Gerold.

[Frank Quisinsky]

Hi Sedat,

yes, think the same ... also what you wrote about 4 CPUs.

Let us wait of the final results.
I wrote it on my actual / aktuell site. For some years I made such a test on two times. The result was -27 ELO for the non ponder version. I think we will get the same results.

Best
Frank

[Frank Quisinsky]

Hi Gerold,

the test from the IPP family engines based on a poll, I started in the German fora CSS. 70% like it to see how strong are the IPP family in SWCR.

I am not a fan of these engines and from Rybka too.

But now let us look how strong are the engines really.

Have fun with SWCR and thanks for your message!

Best
Frank

[Frank Quisinsky]

Hi Sedat,

for the next comming tournament I am thinking about a second Crafty test. In SWCR all engines playing with all the different available 4-pieces TBs. In the past I find out that engines with 5-pieces TBs lost 15-20 ELO in eng-eng tournaments.

So, I can test it again with WB Crafty 23.3 JA x64.
Means after the still running tournament I can add Crafty 23.3 JA x64 5Tbs.

What do you think?
Do you think it isn't necessary to do it again, or interesting to do it again?

Best
Frank

[Sedat Canbaz]

Hi Sedat,

for the next comming tournament I am thinking about a second Crafty test. In SWCR all engines playing with all the different available 4-pieces TBs. In the past I find out that engines with 5-pieces TBs lost 15-20 ELO in eng-eng tournaments.

So, I can test it again with WB Crafty 23.3 JA x64.
Means after the still running tournament I can add Crafty 23.3 JA x64 5Tbs.

What do you think?
Do you think it isn't necessary to do it again, or interesting to do it again?

Best
Frank

Hello Frank,

Endgame Testing-a good idea indeed !

Yes..not often,but sometimes the performance of engines are falling down in case of using the endgames,i am not sure exactly that its a engine ponder bug or due to corrupted tb files

For example,especially i can confirm for Zappa Mexico II (in Playchess 2009 Tournament) that Zappa even on winning positions lost some games in using the endgames,but i am quite sure that the reason was pondering not the Nalimov endgames:
http://sedatchess.110mb.com/index.php?p=1_24

In other words,Zappa has ponder bug in case of pondering on + tablebases and that's why Zappa with ponder off has better results than ponder on:
http://sedatchess.110mb.com/index.php?p=1_66




Regards,
Sedat

[Frank Quisinsky]

Code: Select all

ELO calculation (3)
After 20.886 games (SWCR-64, all games)

01. IPP Houdini 1.03a x64        2.951 38 36  306 79% 2.719 26% NEW
02. Rybka 4 x64                  2.940 19 19 1239 81% 2.690 28%
--. IPP Fire 1.31 x64            2.912 35 33  322 75% 2.721 35% NEW
--. Rybka 3 x64                  2.905 22 22  840 78% 2.687 28%
03. Stockfish 1.8.0 JA x64       2.902 21 21  880 76% 2.702 31%
--. Stockfish 1.7.1 JA x64       2.896 19 18 1120 76% 2.706 34%
04. Critter 0.80 x64             2.841 20 20  881 69% 2.705 33%
05. Naum 4.2 x64                 2.831 16 15 1480 68% 2.700 36%
--. Critter 0.70 x64             2.806 19 19  880 65% 2.701 38%
06. Shredder 12                  2.800 16 15 1440 64% 2.699 35%
07. Komodo 1.2 JA x64            2.798 17 16 1240 64% 2.694 39%
--. Komodo 1.0 JA x64            2.791 20 20  840 64% 2.693 40%
--. Shredder 12 x64              2.787 18 18 1080 61% 2.707 35%
08. Spark 0.5 x64                2.739 16 16 1240 56% 2.696 38%
09. Thinker 5.4d Inert x64       2.732 15 15 1480 54% 2.703 40%
10. Zappa Mexico II x64          2.723 15 15 1480 53% 2.703 39%
--. Spark 0.4 x64                2.717 19 19  840 53% 2.696 40%
11. Protector 1.3.4 JA x64       2.715 15 15 1440 52% 2.704 37%
12. Fruit 09_07_05 x64           2.711 15 15 1480 51% 2.704 34%
--. Critter 0.60 x64             2.700 20 20  840 50% 2.697 38%
13. Hannibal 1.0a x64            2.689 19 19  879 47% 2.712 36%
14. Sjeng WC-2008 x64            2.685 15 15 1480 48% 2.704 36%
--. Protector 1.3.5 x64          2.684 19 19  840 47% 2.706 39%
15. Junior 11.2 x64              2.682 17 17 1240 48% 2.698 31%
16. Onno 1.2.70 x64              2.676 16 16 1240 47% 2.698 37%
--. Onno 1.1.1 x64               2.671 19 19  840 46% 2.699 40%
--. Junior 11.1a x64             2.648 20 20  840 43% 2.700 32%
17. Loop 2007 x64                2.631 15 16 1400 40% 2.706 37%
--. Loop M1-T x64                2.628 33 34  301 36% 2.733 35% - 3 ELO
18. Equinox 0.83 x64             2.615 34 35  300 34% 2.735 29% NEW
--. Twisted Logic 20100131x x64  2.611 18 18 1120 35% 2.716 32%
19. Umko 1.0 x64                 2.606 20 20  881 35% 2.716 36% (ponder not possible)
20. SmarThink 1.20 x64           2.602 15 15 1480 36% 2.707 34%
21. Crafty 23.3 JA x64           2.589 20 20  880 33% 2.716 33%
--. Cipollino 3.25 x64           2.578 34 35  302 29% 2.736 32% NEW
22. BugChess2 1.7 x64            2.563 21 22  800 29% 2.719 33%
23. Scorpio 2.6 JA x64           2.555 18 18 1120 28% 2.718 32%
--. Crafty 23.2 JA x64           2.555 18 18 1120 28% 2.718 30%
24. Chronos 1.99 x64             2.552 18 18 1120 27% 2.718 33% (ponder not possible)
--. Crafty 23.3 JA x64 NP        2.551 34 35  321 26% 2.736 31% (without ponder = -38 ELO)
25. Daydreamer 1.75 JA x64       2.521 19 19 1120 24% 2.719 30%
26. Tornado 3.6.7 x64            2.479 23 24  800 19% 2.723 24%