General discussion about computer chess...
-
notyetagm
- Posts: 139
- Joined: Thu Jun 10, 2010 6:12 pm
Post
by notyetagm » Thu Jul 15, 2010 6:50 am
Just wanted to give my kudos to IvanHoe developers.
Magnificent achievement.
http://www.chesslogik.com/ivanhoe.htm
robbolito wrote:Code: Select all
Engine Score Iv Ho Ry S-B
1: IvanHoe-B57d_11 65.5/120 ········ 19-10-31 18-16-26 3759.5
2: Houdini 1.02 60.0/120 10-19-31 ········ 21-12-27 3550.5
3: Rybka 4 w32 54.5/120 16-18-26 12-21-27 ········ 3429.5
180 games played / Tournament is finished
Name of the tournament: Ivanhoe B57d
Site/ Country: XP-2G, United States
Level: Blitz 5/0
Hardware: Intel(R) Pentium(R) 4 CPU 2.00GHz 1994 MHz with 1,024 MB Memory
Operating system: Microsoft Windows XP Professional Service Pack 3 (Build 2600)
-----------------Houdini 1.02-----------------
Houdini 1.02 - IvanHoe-B57d_11 : 25.5/60 10-19-31 (==0===011000=0===01=10=0===000001======11010=0==0==0=====11=) 43% -49
Houdini 1.02 - Rybka 4 w32 : 34.5/60 21-12-27 (0===011101==01=1==01110==10======111==10101==00==111=====110) 58% +56
-----------------IvanHoe-B57d_11-----------------
IvanHoe-B57d_11 - Houdini 1.02 : 34.5/60 19-10-31 (==1===100111=1===10=01=1===111110======00101=1==1==1=====00=) 58% +56
IvanHoe-B57d_11 - Rybka 4 w32 : 31.0/60 18-16-26 (0=0=011==0=11=0==011===0==01=0=0111=01=010=11==1100=0===11==) 52% +14
-----------------Rybka 4 w32-----------------
Rybka 4 w32 - Houdini 1.02 : 25.5/60 12-21-27 (1===100010==10=0==10001==01======000==01010==11==000=====001) 43% -49
Rybka 4 w32 - IvanHoe-B57d_11 : 29.0/60 16-18-26 (1=1=100==1=00=1==100===1==10=1=1000=10=101=00==0011=1===00==) 48% -14
-
Uly
- Posts: 838
- Joined: Thu Jun 10, 2010 5:33 am
Post
by Uly » Thu Jul 15, 2010 8:59 am
notyetagm wrote:Magnificent achievement.
It's easy when you base it on something already quite strong. Anyone can now take the ip* source code and make their own beast at home.
I think the praising should be given to other engines that took more effort.
-
notyetagm
- Posts: 139
- Joined: Thu Jun 10, 2010 6:12 pm
Post
by notyetagm » Thu Jul 15, 2010 7:37 pm
Ovyron wrote:notyetagm wrote:Magnificent achievement.
It's easy when you base it on something already quite strong. Anyone can now take the ip* source code and make their own beast at home.
I think the praising should be given to other engines that took more effort.
If it's so easy how come Vas couldn't do it?
-
Martin Thoresen
- Posts: 386
- Joined: Thu Jun 10, 2010 5:27 am
Post
by Martin Thoresen » Thu Jul 15, 2010 8:10 pm
You are basing that on way too few games, the error margins are too big.
Test it with 3-4000 games and if the results stands then it's very likely that it's stronger.
-
Zherkovv
- Posts: 36
- Joined: Thu Jun 10, 2010 10:12 am
Post
by Zherkovv » Fri Jul 16, 2010 8:12 am
notyetagm wrote:Ovyron wrote:notyetagm wrote:Magnificent achievement.
It's easy when you base it on something already quite strong. Anyone can now take the ip* source code and make their own beast at home.
I think the praising should be given to other engines that took more effort.
If it's so easy how come Vas couldn't do it?
That is a very good question!
-
Zherkovv
- Posts: 36
- Joined: Thu Jun 10, 2010 10:12 am
Post
by Zherkovv » Fri Jul 16, 2010 8:16 am
Ovyron wrote:notyetagm wrote:Magnificent achievement.
It's easy when you base it on something already quite strong. Anyone can now take the ip* source code and make their own beast at home.
-
Zherkovv
- Posts: 36
- Joined: Thu Jun 10, 2010 10:12 am
Post
by Zherkovv » Fri Jul 16, 2010 8:33 am
Ovyron wrote:notyetagm wrote:Magnificent achievement.
It's easy when you base it on something already quite strong. Anyone can now take the ip* source code and make their own beast at home.
I think the praising should be given to other engines that took more effort.
Vasik Rajlich has offered _zero_ evidence, just an original vague claim that IppoLit appeared to be derived from Rybka. No code examples or anything. So he has offered exactly zero evidence.
B.t.w. Rybka is certainly questionable in terms of legality, based on the Fruit comparison.
-
notyetagm
- Posts: 139
- Joined: Thu Jun 10, 2010 6:12 pm
Post
by notyetagm » Fri Jul 16, 2010 3:35 pm
Zherkovv wrote:Ovyron wrote:notyetagm wrote:Magnificent achievement.
It's easy when you base it on something already quite strong. Anyone can now take the ip* source code and make their own beast at home.
I think the praising should be given to other engines that took more effort.
Vasik Rajlich has offered _zero_ evidence, just an original vague claim that IppoLit appeared to be derived from Rybka. No code examples or anything. So he has offered exactly zero evidence.
B.t.w. Rybka is certainly questionable in terms of legality, based on the Fruit comparison.
I know that; just try telling that to the Rybka fanboys though.
-
Uly
- Posts: 838
- Joined: Thu Jun 10, 2010 5:33 am
Post
by Uly » Sat Jul 17, 2010 1:15 am
notyetagm wrote:If it's so easy how come Vas couldn't do it?
Maybe for the same reason Mark Uniacke, Stefan Meyer-Kahlen or Frans Morsch "couldn't" do it. They'd rather have an original engine.
Please note I'm not claiming Ippolit is a clone of Rybka, I'm claiming IvanHoe is a clone of Ippolit.