Soberango 0.00.0 realized.

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

Moderators: hgm, Rebel, chrisw

User avatar
Luis Babboni
Posts: 464
Joined: Sat Feb 28, 2015 4:37 pm
Location: Argentina

Re: Soberango 0.00.0 realized.

Post by Luis Babboni »

Norbert Raimund Leisner wrote: ...
Perhaps you can contact respectively send to him debug-files (about games running Soberango 0.00.3 on WinBoard /ScidvsPC)
...
OK! Thanks! I´ll do, but actually I´m still madly involved in just playing development.
User avatar
Sylwy
Posts: 4465
Joined: Fri Apr 21, 2006 4:19 pm
Location: IASI - the historical capital of MOLDOVA
Full name: SilvianR

Re: Soberango 0.00.0 realized.

Post by Sylwy »

Luis Babboni wrote:Thanks for your time Norbert and Sylwy!
Sylwy wrote: ...
The engine has around 410 Elo CCRL 4/40 !
...
1.-How do you know it?

2.-What is 4/40? 4 minutes for 40 moves? ¿And after move 40?
1.-Well, a long test versus Pos 1.20 engine (used like marker);
2.-I used in my test TC=5'/game. The total time per game is very similar to TC=4/40 moves of CCRL.

THANK YOU for the new version !
User avatar
Luis Babboni
Posts: 464
Joined: Sat Feb 28, 2015 4:37 pm
Location: Argentina

Re: Soberango 0.00.0 realized.

Post by Luis Babboni »

5 minutes per game?
Is what I use too as only mode for the moment.
You gives Pos 1.2 the ELO that appears here:
http://www.computerchess.org.uk/ccrl/404/
and see the final ELO for Soberango after some number of games?
Wich ELO do you choice for Soberango at the start?
User avatar
Sylwy
Posts: 4465
Joined: Fri Apr 21, 2006 4:19 pm
Location: IASI - the historical capital of MOLDOVA
Full name: SilvianR

Re: Soberango 0.00.0 realized.

Post by Sylwy »

Luis Babboni wrote:5 minutes per game?
Is what I use too as only mode for the moment.
You gives Pos 1.2 the ELO that appears here:
http://www.computerchess.org.uk/ccrl/404/
and see the final ELO for Soberango after some number of games?
Yes. Pos is a weak but stable chess engine. I tested it a lot ! Well, statistically the rating after a single match is not correct.......but the result is not too far. Maybe CCRL will test it ! :lol:
User avatar
Luis Babboni
Posts: 464
Joined: Sat Feb 28, 2015 4:37 pm
Location: Argentina

Re: Soberango 0.00.0 realized.

Post by Luis Babboni »

I read in CCRL that actually they just test engines at least as good as the worst in this list:
http://www.computerchess.org.uk/ccrl/4040/
That is Ziggurat with ELO 1790..... too far from Soberango! :D

I just ended 20 5/0 matchs against Pos and the score was 19-0-1 (+676 ELO for Soberango 0.01.1 for starting in 288, the same as POS)

I use this list of CCRL:
http://www.computerchess.org.uk/ccrl/404/
Where Pos is the last and trying to play one by one from bottom to top 10 5/0 matches each and do not found all and somes do not works but nº492 Kace with ELO 975 stop 0.01.1 clearly and NSV being the stop for 0.00.3, now ended all in tie cause neither it neither Soberango knows 3rd repetition rule and none use random but with whites NSV was clearly better and slighty better with blacks.
I tried Microchess and the problem is that cause Soberango 0.01.1 lost the random ability cause the implementation of the Alpha-Beta, it plays allways the same game against some other engines that do not have random neither. Microchess is one of them and they play just 2 differents games on wich have whites and who have blacks. Both won by Soberango.

My next target, and I´ll put myself right now on that is to gave back some random to Soberango!
:twisted:
User avatar
Sylwy
Posts: 4465
Joined: Fri Apr 21, 2006 4:19 pm
Location: IASI - the historical capital of MOLDOVA
Full name: SilvianR

Re: Soberango 0.00.0 realized.

Post by Sylwy »

Here is my result after 90 games: +60-29=1 +123 Elo for Soberango 0.00.0 !

Image
User avatar
Luis Babboni
Posts: 464
Joined: Sat Feb 28, 2015 4:37 pm
Location: Argentina

Re: Soberango 0.00.0 realized.

Post by Luis Babboni »

Thanks!

Which version of Pos you have? It seems is better than wich I use or depends on the speed of the hardware, Soberango do not depends on the speed hardware for the moment if it have enough time to complete the game.
I have a Phenom 955 at 3.2 MHz. :?

Wait!! I use 0.01.1, not 0.00.0.
Anyway with 0.00.0 I had 9 0 1.

Nice to see the flag and the logo in Arena, just after saw your image I went to Arena I learn how to do it! :D
User avatar
Luis Babboni
Posts: 464
Joined: Sat Feb 28, 2015 4:37 pm
Location: Argentina

Re: Soberango 0.00.0 realized.

Post by Luis Babboni »

LOL!! Soberango did its first tournament.... and lost all its matches!! :D
http://chessengines.blogspot.com.ar/201 ... chess.html
Thanks Dydolec for include it!! I hope new versions will have better performances in the future! :oops:
User avatar
Luis Babboni
Posts: 464
Joined: Sat Feb 28, 2015 4:37 pm
Location: Argentina

Re: Soberango 0.00.0 realized.

Post by Luis Babboni »

Version 0.01.3 incorporates some of the random that lose when change from minimax to alphabeta:
http://www.luisbtenis.com.ar/Soberango/Soberango.html
Plays better than 0.01.1 and 0.01.2
supersharp77
Posts: 1242
Joined: Sat Jul 05, 2014 7:54 am
Location: Southwest USA

Re: Soberango 0.00.0 "tested"

Post by supersharp77 »

Luis Babboni wrote:Well, 1st step completed.
Starting from zero knowledge, after 5 months (3 trying to understand what´s the matter included failed attempt to learn C++ and 2 in the making and tests), the very first version of Soberango see the sunlight.
It´s played some games against myself (1500 in chess.com) and in 5/0 blitz is a good opponent. Playing 10 games against the weakest engine I could found: Safrad, won 1, tie 3 and loses 6. :roll:

Site: http://www.luisbtenis.com.ar/Soberango/Soberango.html

Comments:
-Version: 0.00.0 (Very basic winboard engine, just works and for the moment just checked successfully in Arena 3.5 and Arena 3.5.1).
-Options: there isn´t, just plays matchs from the initial position without ponder.
-Searching algorithm: just Mini-Max, no alphabeta.
-Search depth: fixed in 4 plies.
-No time management. In a Phenom 955 3.2MHz, could play Blitz from minimum time 1:30/0 losing by time very rarely.
-Evaluation: just material difference.
-Not detection of 50 moves rules neither 3rd repetition.
-Not show think.
-Known bug: if lose by time, could not start a new game without restart.
-32 bits engine.
-8x8 matrix board representation.
-Wrote in FreeBasic .

Thanks to the forum for the help you give me no matter how noob my questions use to be! :D
Engine a bit buggy...tends to hang and not close after games ended..(use task mgr to close) works using only wb2uci adapter in Fritz and Lokasoft...(xboard and winboard gives no moves in lokasoft)...not quite a random mover..shows some ability (tactics) reasonable strength for a new engine..(needs some work)....rating based on some swiss games
1100-1300?!?! rough guess...AR :) :wink:

Soberango pgn games

http://s000.tinyupload.com/?file_id=023 ... 5582231550