Author: Djordje Vidanovic
Date: 10:10:04 07/18/03
Go up one level in this thread
On July 18, 2003 at 12:55:45, Jonas Bylund wrote:
>On July 18, 2003 at 12:47:14, Djordje Vidanovic wrote:
>
>>On July 18, 2003 at 12:30:57, Will Singleton wrote:
>>
>>>On July 18, 2003 at 11:40:42, Djordje Vidanovic wrote:
>>>
>>>>Hi,
>>>>
>>>>I have some interesting news. A young programmer from my University (the
>>>>University of Nis, Serbia), an electronics engineer, Vladan Vuckovich, has
>>>>created a strong chess program called Geniss Axon. This program had its first
>>>>public tournament against humans the day before yesterday and it won ahead of
>>>>three IMs! It was a rapid chess event and here is one of the games. Look at
>>>>the move Nd4 when Geniss sacrificed its Bishop :)
>>>>
>>>>Vladimir has contacted me saying that he has decided to apply for the Leiden
>>>>tournament if something like that is feasible within the Leiden participation
>>>>rules. Therefore I would like to ask some people from the Leiden organisation
>>>>team to tell me what the participation rules are and in case Geniss Axon could
>>>>be admitted to the tournament would there be someone to operate Geniss Axon for
>>>>Vladan (if that is possible too, because I believe that programmers are usually
>>>>required to operate their own programs). The program is extremely fast (about 2
>>>>million positions per second on my Athlon 2400+) and written in assembler. Just
>>>>like Fritz... It has beaten some commercials several and I saw it beat Crafty as
>>>>well:) Geniss Axon has been in development since 1997, but it has started
>>>>playing friendly games against humans only last year.
>>>>
>>>>I am attaching a game from the day before yesterday when Geniss Axon beat IM
>>>>Miljkovic (the "atc" stands for the attacking version of Geniss Axon, targeting
>>>>human players).
>>>>
>>>>[Event ""]
>>>>[Site "Nis"]
>>>>[Date "16.7.2003 21:31:24"]
>>>>[Round " 65003 sec. per move"]
>>>>[White " Geniss Axon XP , v650_atc, July 16 2003. "]
>>>>[Black "Miroslav Miljkovic"]
>>>>[Result ""]
>>>>[ECO "B33 Sicilian: Sveshnikov variation |e=79| "]
>>>>[PlyCount "77"]
>>>>
>>>>1. e4 c5 2. Ng1-f3 Nb8-c6 3. d4 c:d4 4. Nf3:d4 Ng8-f6 5. Nb1-c3
>>>>e5 6. Nd4-b5 d6 7. Bc1-g5 a6 8. Nb5-a3 b5 9. Bg5:f6 g:f6 10. Nc3-d5
>>>>f5 11. Bf1-d3 Bc8-e6 12. O-O Be6:d5 13. e:d5 Nc6-e7 14. Na3:b5
>>>>e4 15. Nb5-d4 e:d3 16. Rf1-e1 Qd8-a5 17. Qd1:d3 Qa5:d5 18. c4
>>>>Qd5-b7 19. Nd4:f5 O-O-O 20. Nf5:d6+ Rd8:d6 21. Qd3:d6 Qb7-c6
>>>>22. Ra1-d1 Qc6:d6 23. Rd1:d6 Ne7-g6 24. Re1-e8+ Kc7 25. Rd6:a6
>>>>Bf8-g7 26. Re8:h8 Bg7:h8 27. b4 Bh8-d4 28. c5 Ng6-e5 29. Ra6-d6
>>>>Bd4-b2 30. Rd6-d5 Kc6 31. Rd5-d6+ Kc7 32. b5 Bb2-a3 33. Rd6-d5
>>>>f6 34. h4 h5 35. g3 Ba3-b4 36. f4 Ne5-d7 37. c6 Nd7-b6 38. Rd5:h5
>>>>Nb6-c4 39. Rh5-h7+
>>>>
>>>>{ This PGN file was generated by Geniss Axon XP , v650_atc, 16. July
>>>>2003. }
>>>>
>>>>Regards,
>>>>
>>>>Djordje
>>>
>>>Since when is a rapid chess event 18 hours per move? :)
>>>
>>>Mine won't play Nd4. I don't see why Nd4 exd3 Re1 dxc2 (not Qa5) is better than
>>>Bc4.
>>>
>>>Will
>>
>>
>>Perhaps it was a slow rapid chess event :). Seriously, I have no idea why Axon
>>produced that... It was a G/25 min event with 3 IMs and Geniss Axon, a double
>>round robin. About Nd4: my gut feeling tells me that it is a great move, and I
>>guess that Tal would go along with that. In any case it was a good move, I am
>>sure and there are logical grounds for it: opening up the e-file, getting
>>control of the c6 square, weak black pawn at f5, weaving a dangerous net around
>>the black king, etc.
>>
>>Rgds,
>>
>>Djordje
>
>I agree, my gut feeling is too that it is a good move.
>So will this new Geniss Axon be available some time either as a commercial or
>free?
>
>Jonas
From what I've been able to gather so far, it seems that Vladan is planning to
go commercial if he manages to get his SMP version ready soon. He says that it
should be no problem. If this program starts running on a dual Athlon MP, it
should reach about 5 million nodes per second. And it is very good tactically,
I can tell you that. It still lacks TB support and more positional knowledge.
But it has a matesolver built in the program itself so when needing to mate than
what you see is 7-10 million nodes per second. And it mates then. No wonder.
BTW, as it is now, Geniss has a nice, original and functional, GUI. The time
controls are still wobbly, so that is something to work on as well. It even has
some sort of multimedia support. You can hear Vladan's voice announcing moves
(only in Serbian for the moment :( ), move and board sounds and, after getting
out of book, the sounds of an engine revving :). Lots of fun, and a terrific
position analyser too. Altogether about 8 meg of diskspace.
Djordje
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.