TogaII 1.4.1SE breaks 3000 elo barrier!

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

Edward German

Re: BTW...

Post by Edward German »

Thank You Denis!

I can not write here long sentences. Sorry.
(My english is bad)

With my setting I played on playchess (live with kibitzers) with only 2 cores 2.66 GHz against fast Quad's, but Toga played very well!

(With Toga 1.4.1SE default I played about 10 games, but lost 90%)

My final score with this Setting was last +2 = 2 -1.

I think it was very good, and good preferences for this Setting.

I like such live games on the Server! :)

Best!
User avatar
Denis P. Mendoza
Posts: 415
Joined: Fri Dec 15, 2006 9:46 pm
Location: Philippines

Re: BTW...

Post by Denis P. Mendoza »

Your welcome Eduard! But sorry to disappoint you on the KingSafetyMargin Issue. Maybe Thomas is the right person to answer why he chose to terminate using this option in the new versions. The versions are still on beta stage so the code hasn'r been cleaned yet, even 1.4.1SE as well.
User avatar
Denis P. Mendoza
Posts: 415
Joined: Fri Dec 15, 2006 9:46 pm
Location: Philippines

Re: BTW...

Post by Denis P. Mendoza »

Eduard quote:
PS: Is it possible to compile my Setting as "µ"-Setting by Eduard Nemeth?
Hey Eduard, have you checked your PM? You can get there your personalized "toy" with all the UCI-options who can tweak as promised. This is the original code with additional options. This is just a follow-up if you received it!
Jouni
Posts: 3730
Joined: Wed Mar 08, 2006 8:15 pm
Full name: Jouni Uski

Re: BTW...

Post by Jouni »

After 200 blitz games I got EXACTLY same rating for 1.4 beta and 1.4.1SE.
But I quess 1.4.1 is better with longer time control? Note there can be a bug in Toga, see Rybka forum:

http://rybkaforum.net/cgi-bin/rybkaforu ... id=3229#fp

Jouni
Edward German

Re: BTW...

Post by Edward German »

Jouni wrote:After 200 blitz games I got EXACTLY same rating for 1.4 beta and 1.4.1SE.
But I quess 1.4.1 is better with longer time control? Note there can be a bug in Toga, see Rybka forum:

http://rybkaforum.net/cgi-bin/rybkaforu ... id=3229#fp

Jouni
I cannot see here a BUG on my Core 2 Duo 2.66 GHz, and with my own Settings (e2) under the Shredder 10.1 GUI:

[ENGINE]
Name=Toga II 1.4.1SE T2e2
Author=Thomas Gaksch and Fabien Letouzey
Filename=C:\Programme\ShredderChess\Shredder 10 UCI\Engines\TogaIIv1.4.1SE\toga141se-2cpu.exe
[OPTIONS]
Verification Reduction=4
Toga Lazy Eval Margin=225
Toga King Safety Margin=1300



[d]5r1k/3r1p1p/p1p2R2/1q2p3/p3P1Q1/2P4P/1P4P1/5R1K b - -

Engine: Toga II 1.4.1SE T2e2 (256 MB)
von Thomas Gaksch and Fabien Letouzey

1/04 0:00 +4.19 1...Dxb2 2.Dxd7 (46)

1/08 0:00 +3.10 1...a3 2.Dxd7 axb2 (66)

1/09 0:00 +0.33 1...Dd3 (90)

1/09 0:00 -0.35 1...Td2 (191)

2/15 0:00 -0.86 1...Td2 2.T1f2 Dxb2 3.Txd2 Dxd2
4.Txc6 (1.063)

3/15 0:00 -0.81 1...Td2 2.Tb1 Tg8 (1.638)

4/23 0:00 -0.86 1...Td2 2.T1f2 Dxb2 3.Txd2 Dxd2
4.Txc6 (8.092)

5/23 0:00 -0.87 1...Td2 2.T1f2 Dxb2 3.Txd2 Dxd2
4.Txc6 De1+ 5.Kh2 Tg8 (10.792)

6/23 0:00 -0.27 1...Td2 2.Dh5 Txb2 3.Dh6 Dc5 4.Txc6 De7
5.Txa6 (20.360)

7/23 0:00 +0.11 1...Td2 2.Dg5 Txb2 3.Dh6 Dc5 4.Tg6 Tg8
5.Txf7 (34.922)

7/20 0:00 0.00 1...Td2 2.Dg5 Txb2 3.Dh6 Kg8 4.Dg5+ Kh8 (26.832)

8/30 0:00 +0.07 1...Td2 2.Dg5 Td7 3.Dh6 Tdd8 4.Tb1 (57.402)

9/30 0:00 +0.54 1...Td2 2.Dg5 Txb2 3.Dh6 Dc5 4.Txc6 De7
5.Tc7 Dd8 6.Tfxf7 Txf7 7.Txf7 (107.759)

10/43 0:00 +0.56 1...Td2 2.Dh4 a3 3.Dh6 Tdd8 4.bxa3 Dc4
5.Dg5 Dxe4 6.Txf7 Txf7 7.Txf7 (362.006)

10/42 0:00 +0.51 1...Tb7 2.Dh5 Dxb2 3.Dxe5 Tb5 4.De7 Tg8
5.T6f2 Dxc3 6.Txf7 (420.071)

11/42 0:00 +0.57 1...Tb7 2.Dh5 Dxb2 3.Dxe5 Tb5 4.De7 Tg8
5.T6f2 Dxc3 6.Txf7 Th5 7.h4 a3 8.Td1 (570.641)

11/43 0:00 +0.52 1...Td2 2.Dh4 Dc4 3.Dh6 Tdd8 4.Kh2 a3
5.bxa3 Tb8 6.Tb1 (709.565)

12/43 0:00 +0.65 1...Td2 2.Dh4 Dc4 3.Dh6 Tdd8 4.Te1 Dc5
5.Kh2 De7 6.Txc6 Tg8 7.Df6+ Dxf6
8.Txf6 (1.012.757)

13/43 0:02 +1.62 1...Td2 2.Df5 De2 3.Tg1 Td1 4.Txd1 Dxd1+
5.Kh2 Dd2 6.Dxe5 Kg8 7.c4 Td8 8.Txc6 (2.300.822) 2200

13/42 0:02 +1.61 1...Td2 2.Df5 De2 3.Tg1 Td1 4.Txd1 Dxd1+
5.Kh2 Dd2 6.Dxe5 Kg8 7.c4 Tc8 8.De7 (2.194.880) 2200

13/42 0:02 +0.73 1...Tb7 2.Dh5 Dxb2 3.Dxe5 Tb5 4.Dd4 Kg8
5.T6f2 Db3 6.Tf3 Tg5 7.Tf6 (2.761.066) 2200

14/43 0:04 +0.91 1...Db7 2.Df5 Tg8 3.g4 Dc7 4.Txf7 Txf7
5.Dxf7 Dxf7 6.Txf7 Tb8 7.Tf2 Kg7
8.Kg2 Td8 9.Kg3 (5.001.186) 2171

14/43 0:05 +0.86 1...Tb7 2.Dh5 a3 3.bxa3 Tg8 4.Kh2 Db2
5.Df3 Tc7 6.Td6 f5 7.exf5 Tcg7 (6.070.799) 2168

15/47 0:07 +0.93 1...Db7 2.Df5 Tg8 3.g4 Dc7 4.Txf7 Txf7
5.Dxf7 Dxf7 6.Txf7 Tb8 7.Tf2 Kg7
8.Kg2 Td8 9.c4 Kg6 (8.494.679) 2171

15/47 0:09 +0.89 1...Dd3 2.Df5 Tg8 3.Tg1 Tg6 4.Dxe5 Kg8
5.De8+ Kg7 6.Tf2 Dd6 7.Te1 De6 8.Db8 Dd6
9.Dc8 (10.844.910) 2171

16/51 0:19 +0.77 1...Dd3 2.Dh4 Dc4 3.T1f5 Tg8 4.Th5 Tg7
5.Txe5 Td8 6.Tf2 Tdg8 7.Kh2 a3
8.bxa3 Dxc3 9.Te7 Tf8 10.Tf6 (20.989.866) 2181

17/51 0:26 +0.84 1...Dd3 2.Dh4 Dc4 3.T1f5 Tg8 4.Th5 Tg7
5.Txe5 Td8 6.Tf2 Tdg8 7.Kh2 a3
8.bxa3 Dxc3 9.Te7 Dxa3 10.Texf7 Dd6+
11.Kh1 Txf7 12.Txf7 (28.750.283) 2190

18/53 0:43 +0.99 1...Dd3 2.Df5 Tg8 3.Tf2 Db1+ 4.Kh2 Td1
5.g4 Th1+ 6.Kg3 De1 7.Txf7 De3+
8.Df3 Dxf3+ 9.T2xf3 Tb1 10.Tf2 Tb8
11.Tf8+ Txf8 12.Txf8+ Kg7 13.Tf2 Td1
14.g5 (48.202.215) 2208

18/65 1:06 +0.80 1...Db7 2.Df5 Tg8 3.g4 Dc7 4.Txf7 Txf7
5.Dxf7 Dxf7 6.Txf7 Tb8 7.Tf2 Td8
8.Kg2 Kg7 9.h4 Kg6 10.Kf1 c5 11.g5 (73.085.395) 2204

18/61 1:06 +0.67 1...Db7 2.Df5 Tg8 3.g4 Dc7 4.Txf7 Txf7
5.Dxf7 Dxf7 6.Txf7 Tb8 7.Tf2 Tb3
8.Kg2 Kg7 9.Td2 a3 10.bxa3 Txc3
11.Td7+ Kf6 12.a4 h6 (72.569.687) 2204

19/61 1:45 +0.95 1...Db7 2.Dh4 Tg8 3.T6f5 Tg7 4.Df6 Db8
5.Tg5 Df8 6.Dxe5 Td2 7.c4 Kg8 8.h4 h6
9.Txg7+ Dxg7 10.De8+ Df8 11.Dxc6 Txb2
12.Dxa4 De7 13.Tf4 (115.883.545) 2210

20/65 2:50 +0.96 1...Db7 2.Dh4 Tg8 3.T6f5 Tg7 4.Df6 Db8
5.Tg5 Df8 6.Dxe5 Td2 7.c4 Kg8
8.Txg7+ Dxg7 9.De8+ Df8 10.Dxc6 Txb2
11.Dxa4 De7 12.Dc6 De6 13.Da8+ Kg7 (188.492.977) 2211

No Qxb2.
User avatar
Eelco de Groot
Posts: 4681
Joined: Sun Mar 12, 2006 2:40 am
Full name:   Eelco de Groot

Re: BTW...

Post by Eelco de Groot »

Denis P. Mendoza wrote:Eduard quote:
PS: Is it possible to compile my Setting as "µ"-Setting by Eduard Nemeth?
Hey Eduard, have you checked your PM? You can get there your personalized "toy" with all the UCI-options who can tweak as promised. This is the original code with additional options. This is just a follow-up if you received it!
I think Eduard is still reading the forum in thread view so he may have missed your reply Denis. Curious what you did with the Toga King Safety Option? :P I do remember that we had a discussion about the same issue quite a way back with Toga 1.2.1a I believe but I can't remember details or what the conclusion was. I vaguely remember I tried to repeat games with the same settings and openings but there were often some -inexplainable?- differences from the same starting positions. It may have been just timing effects or some bugs still, so that makes it a bit more difficult to check the truth. But I'm not buying that Toga King Safety is helping much in games on Playchess when there is no difference in Analysis mode. In Toga 1.2.1a the option still worked maybe the discussion was about 1.3.4

Denis I was hoping maybe you would yelp with some compilation work soon? I thought maybe it would be nice to compile the Beta 3 and see if the multithreading really works still, I might have broken it with some changes and I can't check it on my computer. But the code is not ready yet...

Regards, Eelco
User avatar
Denis P. Mendoza
Posts: 415
Joined: Fri Dec 15, 2006 9:46 pm
Location: Philippines

Re: BTW...

Post by Denis P. Mendoza »

Eelco, you maybe right as there's still no response from Eduard. But the one I did for him does not change the TKSM - disabled. It's still the original code with only about 14 additional UCI-options for Razoring depth, Razoring Margin, more material weight options, etc..as he likes to tweak parameters. Maybe I'll test adding more eval UCI-options like in Toga-per of Jim, for variety and testing purposes only.

Just send me something to do, if you want a compile for your mods. Even Master Jim could do a better compile for you :wink: It's what we are here for :) .
ThatsIt
Posts: 992
Joined: Thu Mar 09, 2006 2:11 pm

Re: BTW...

Post by ThatsIt »

I like such live games on the Server! :)

You should not spend too much time on the cb-server..
User avatar
Eelco de Groot
Posts: 4681
Joined: Sun Mar 12, 2006 2:40 am
Full name:   Eelco de Groot

Re: BTW...

Post by Eelco de Groot »

Denis P. Mendoza wrote:Eelco, you maybe right as there's still no response from Eduard. But the one I did for him does not change the TKSM - disabled. It's still the original code with only about 14 additional UCI-options for Razoring depth, Razoring Margin, more material weight options, etc..as he likes to tweak parameters. Maybe I'll test adding more eval UCI-options like in Toga-per of Jim, for variety and testing purposes only.

Just send me something to do, if you want a compile for your mods. Even Master Jim could do a better compile for you :wink: It's what we are here for :) .
Thanks Denis! I don't have your e-mail, I think I will just place the sources for the Beta 3 in the original thread in the CCRL forum, it would be great if you could make some compiles also for multithreaded versions of it! I will add the GPL license, maybe you could add it in a zip file so the exes can be distributed as a GPL package together? I have not tested it much but I can't do tests on a Dual or higher anyway. It is still a Beta but I think some of the modifications at least should work! Not that I made a complete new version, it is still a lot of trial and error, and especially the latter :)

Regards, Eelco
User avatar
Denis P. Mendoza
Posts: 415
Joined: Fri Dec 15, 2006 9:46 pm
Location: Philippines

Re: BTW...

Post by Denis P. Mendoza »

Eelco, I just edited my profile for ease. Now, you can see my e-mail and site. Your request was also done as well. You can visit your post at http://kirr.homeunix.org/chess/discussi ... 92&p=30403. Multithreading still works when I tested it. Thanks.

computer.chess@hotmail.com