Tornado 4.88 running for the IPON

[Engin]

ok many thx too

[IWB]

I am sorry, but I have to interrupt the run of Tornado 4.88! Not because of the engine, this is running fine, biut because of my "engine server". I used it parallel for something else and it crashed completly ... which means that I can continue earliest on Friday late (CET)!

Sorry for that, but I already have some kind of backup in mind which will prevent this happening again!

Bye
Ingo

[IWB]

Tornado 4.88 continous:

http://www.inwoba.de

Sorry for the delay
Ingo

[IWB]

Complete run of Tornado 4.88 vs 24 opponents finished with a nice 55 Elo increase!

Have a look at http://www.inwoba.de

Bye
Ingo

[lucasart]

Complete run of Tornado 4.88 vs 24 opponents finished with a nice 55 Elo increase!

Have a look at http://www.inwoba.de

Bye
Ingo

I don't really understand how your top 25 list work. For example Fruit 05/11/03 is stronger than Crafty 23.4, but was removed... ?
Also you could test Scorpio 2.7, the extensive testing done by CCRL suggests that it would find a place in the top 25.

[IWB]

I don't really understand how your top 25 list work. For example Fruit 05/11/03 is stronger than Crafty 23.4, but was removed... ?

I try to leave the best engine of a family in that list. In your example Strelka 2.0 is the much better "Fruit" ...
But do not stress that too much. I try to make that with some commen sense as it is sometimes impossible to have 100% certianty. If you check closer you will find that Toga 1.4 is now 1 Elo better than Strelka and is Fruit for sure as well ... but I check even closer and see that Strelka perfomed better vs the same opponents (The same goes for ... as I said, a bit of "opinion" is in that for sure!
The main purpose of that list is to have engines wich are most different. For example if someone want to analyse a position it makes no sense to check that position with 20 different Littos ...

Also you could test Scorpio 2.7, the extensive testing done by CCRL suggests that it would find a place in the top 25.

Thx for the tip. According to the CCRL 40/4 it would be in my list, comparing with the 40/40 it would be out ... I will check that.

Bye
Ingo

[Uri Blass]

Complete run of Tornado 4.88 vs 24 opponents finished with a nice 55 Elo increase!

Have a look at http://www.inwoba.de

Bye
Ingo

I don't really understand how your top 25 list work. For example Fruit 05/11/03 is stronger than Crafty 23.4, but was removed... ?
Also you could test Scorpio 2.7, the extensive testing done by CCRL suggests that it would find a place in the top 25.

For the case of Fruit there is rule one engine per author and the IPON list include strelka in the top 25 list as fruit derivative.

Note that my opinion is that it is better to include Toga II 1.4 beta5c BB and not strelka as fruit derivative and the main reason is that toga is clearly more similiar to fruit relative to strelka(most of the code of strelka is clearly different than fruit and it is not the case for toga).

[lucasart]

I try to leave the best engine of a family in that list. In your example Strelka 2.0 is the much better "Fruit" ...

Strelka is not a "new version" of Fruit. It is a different engine, written from scratch by someone else. It is true nonetheless, that Strelka did a lot of "almost copy paste" from Fruit (I went through the code of Strelka 2.0 and can attest it).
As for Toga it is indeed an official derivative of Fruit, so I guess it would make sense to put Toga instead of Fruit.
But Strelka would be a different engine. Whether you decide to censor Strelka or not is your call, but it's not a new version of Fruit. If you follow that logic, then Rybka should be kept and Toga/Strelka/Fruit removed, no ?
Also, if you consider that all engines that took ideas and code from Fruit are the same "family", and you only keep the best of the family, then your list would not have 25 elements but only ONE. Everyone found great ideas in Fruit... And Fruit didn't invent the wheel from scratch either. That's how open source evolves: it's a give and take game that ultimately results in the field of computer chess improving drastically. Closed sources engines are a "take only" situation, so in a world where all software is closed source, there would be no sustainable progress. I wish more people could understand that...
Anyway, it's your list, so you can do what you like, of course

[IWB]

Note that my opinion is that it is better to include Toga II 1.4 beta5c BB and not strelka as fruit derivative and the main reason is that toga is clearly more similiar to fruit relative to strelka(most of the code of strelka is clearly different than fruit and it is not the case for toga).

That is for sure a point, just Strelka is the better Fruit (when judging with the same opponents) ...
And then there are even other engines which might considered as "Fruity"

Bye
Ingo

[lucasart]

Note that my opinion is that it is better to include Toga II 1.4 beta5c BB and not strelka as fruit derivative and the main reason is that toga is clearly more similiar to fruit relative to strelka(most of the code of strelka is clearly different than fruit and it is not the case for toga).

That is for sure a point, just Strelka is the better Fruit (when judging with the same opponents) ...
And then there are even other engines which might considered as "Fruity"

Bye
Ingo

Anyway, the reason why I pointed Fruit 05/11/03 is because it's the latest version of Fabien Letouzey. So it's the only genuine one of the Fruit family, and as such it deserves its place more than the Fruit derivatives or Fruit plagiarized engines. And when you compare it to Toga, you'll see that after all these years of butchering the Fruit code, and the who-hah that's been going on about Toga, they never came up with that much improvement over the Fruit codebase.