World Computer Chess Championship ?

[Guenther]

Ok this arrogant and fossilized organization called ICGA really made me get angry and I looked a bit in their site to know a bit more about them. Apart that it really seems "dead and buried", last interesting content is years ago old, I was really impressed by what they call: "World Computer Chess Championship" , the last edition of this farsa was in 2011 and below the list of participants:

http://www.grappa.univ-lille3.fr/icga/t ... php?id=239

Code: Select all

1	Junior	
2	Hiarcs	
2	Shredder	
4	Pandix	
4	Jonny		
6	The Baron	
7	Booot	
8	Rookie3.4	
9	Woodpusher 1997

Now the first question that comes to my mind is to what "world" they refer in the title of the event ? Mars ? Underwater world ? Surely not chess engines landscape in 2011, this is clear even to the dumbest. So how they have the presumption to call this marketing event, for the only benefit of legacy commercial engines, "World Computer Chess Championship" ? Because of historical reasons ? Sorry, this is not enough IMHO.

Nobody of the authors of the above listed engines (at least the ones I know, there are less known engines I don't know the authors) participates in forums, shares ideas / comments (even not talking about code), contributes in any way (again, I am talking of 2012) to chess engine community.

My opinion of Robet Houdart's Houdini is well known and I have publicly stated many times that I don't second many of his choices, nevertheless I value his contribution to this community much higher than everybody in that list. I have took Houdini example because it is considered the posterchild of "the evil clone" by many, and with this I want to highlight that even the most discussed characters contribute more than that live museum of pretending guys: I am not talking of the engine authors (which I respect all) but of the organization that shameless call itself "computer chess world" !

It seems to me you have gone crazy in the last days. First you defend cloning in one post and now you 'value the contribution' of a known cloner
(BTW what contribution? I have seen nada)higher than long respected members of the chess scene e.g. Richard Pijil, SMK and Mark Uniacke and so on.
YOUR post makes me sick and angry, but I hope you find the right pills next time. Basta!

[marcelk]

Now the first question that comes to my mind is to what "world" they refer in the title of the event ? Mars ? Underwater world ? Surely not chess engines landscape in 2011, this is clear even to the dumbest. So how they have the presumption to call this marketing event, for the only benefit of legacy commercial engines, "World Computer Chess Championship" ? Because of historical reasons ? Sorry, this is not enough IMHO.

What exactly is your problem? You say you are not interested on going to tournaments. And now you think that gives you the right to dump on people that do??? :shock:

I think Marco objects to calling the recent ICGA chess tournament a "world championship", while calling all of its other (and more thriving) tournaments "olympiads" (which is funny because those are yearly events and thus not olympiads). Their chess tournament is not a world championship of course, because for that the computer chess community needs to accept it as such at that is not the case anymore. The real competition in computer chess is at the rating lists, not at the associations.

These "associations" have become relics from pre-Internet days, they serve no real purpose anymore to the computer chess community.

As of my participation last year, lets just say that you have to try everything at least once. It is a nice way to meet with some people, but also somewhat expensive in terms of days off and accommodation.

[Sedat Canbaz]

I think it is absurd to talk of ICGA. It'is simply dead.
This association has done a lot for chess engines... In the '70s and '80s! The trouble is that their leaders have stood firm in those years. They're waving a bogey (called "clone") trying to pretend to be the last barrier against barbarism, but in doing so, they are self-condemned to extinction... Like the dinosaurs. I tell them: "Thanks, you've done so much, you will be forever in our hearts, but now please ... Goodbye and thanks!

The solution is simple, and I turn to the proposal to 'all the authors of goodwill':

It takes a lot to make a new association? I think not. You just get together (virtually. We are no longer in the 80s, now there's internet!), decide the rules (simple and clear) and a few other details then... Simply play! Tournaments, leagues, match ... Everything is possible.
I think it would be important to create certain categories and different types of competition, later you can look a sponsors... I do not think this is very difficult.

Dear Roberto,

Yes...we need really a new organization,otherwise (with the current organization) we will not see the Top Engines of 3300-3400 Elo

Actually its will be great,if the new 'World Championships' to be played in two systems:Swiss and also as Leagues

League system is very interesting and exiting too

For example,two ore more leagues-based Top 16 or Top 20 participants

Super League:
----------------
1 Houdini
2 Critter
3 Ivanhoe
4 Stockfish
5 Rybka
6 Vitruvius
7 Fire
8 Naum
9 Komodo
10 Equinox
11 Spike
12 Fritz
13 Chiron
14 Shredder
15 Protector
16 Loop

Second League:
--------------
1 Junior
2 Hiarcs
3 Sjeng
4 Pandix
5 Spark
6 Zappa
7 Fruit
8 Thinker
9 Tornado
10 Onno
11 MilkoChess
12 Crafty
13 Jonny
14 The Baron
15 Hannibal
16 Gull


Best Regards,
Sedat

[hgm]

I think Marco objects to calling the recent ICGA chess tournament a "world championship", while calling all of its other (and more thriving) tournaments "olympiads" (which is funny because those are yearly events and thus not olympiads). Their chess tournament is not a world championship of course, because for that the computer chess community needs to accept it as such at that is not the case anymore. The real competition in computer chess is at the rating lists, not at the associations.

Well, the computer-chess comunity I know considers it a World Championship... That there is a bunch of vocal clowns around that like to tell others what they should do but are not much good for anything else doesn't alter that.

Rating lists have nothing to do with World Championships. Name me one sport where the World Champion is chosen based on rating lists. Certainly not in soccer. Or cycling. Or ice skating. Or boxing. Or hockey...

[Uri Blass]

I guess that the difference is only something like 150 elo in WCCC conditions and being 150 elo weaker than Carlsen is still a level of almost 2700 that is not a level of a club player.

Uri, there are 15 (!) engines between the world champion (!) and the top: http://www.inwoba.de/index.html

Considering that chess engines are much less than people playing chess (even professional), it is like there would be some hundreds players stronger than the current chess world champion !

I know you like to indulge in "splitting the hair in 4 parts" as we say in Italy, but please try to look at the big picture at least once and try to get the "meaning" of what we are talking about here.

You took a list that is more convenient(one cpu and blitz)

You are going to find less engines between the world champion and the top
in the following list:

http://computerchess.org.uk/ccrl/4040.live/

practically I think that only one derivative of Ivanhoe should be considered
so only one of houdini strelka Ivanhoe) and not supporting more than one cpu is a disadvantage
so the list without book at long time control is like the following:
1)Houdini
2)Stockfish
3)Rybka
4)Critter
5)Komodo
6)Naum
7)Chiron1.1a(the 4 cpu version does not have enough games but it is logical to believe that it is better)
8)Junior

Other engines in the IPON that are above Junior are:
2)Fritz13(support only one cpu bad luck)
3)Deep Shredder12(probably slightly weaker than Junior at long time control 7 elo weaker in the CCRL list)
4)Hannibal1.2(support only one cpu so weaker than Junior)
5)Gull1.2(support only one cpu)
6)Deep Sjeng 2010(only one cpu in the ccrl list)
7)Spike1.4(probably slightly weaker than Junior at long time control
7 elo in the CCRL list)
8)Spark1.0(clearly weaker with 4 cpu based on the ccrl list)
9)Protector1.4(weaker with 4 cpu based on the ccrl list)

[Sedat Canbaz]

Dear Uri,

Ok...but however,just i' like to mention that Fritz is a MP engine since many years, and from my experience i can say that Deep Fritz 13 is stronger than Deep Junior 13

And i see in your estimation:Houdini is approx.150 Elo stronger than Junior
Hmm...maybe you can be right,but only in case of Houdini is using not very strong book and playing on old-dated hardware

I mean,if Junior is playing on faster hardware and using a superior book

On equal conditions, the difference between both engines (Houdini/Junior):
Blitz:250-300 Elo difference
http://www.sedatcanbaz.com/chess/scct-rating/
http://www.clemens-keck.de/base.htm

Slow:200-250 Elo difference
http://www.amateurschach.de/
http://www.husvankempen.de/nunn/40_40%2 ... liste.html

Btw,i expected better performance by latest Junior 13 release (almost no any Elo difference between Junior 12)

Best,
Sedat

[marcelk]

Rating lists have nothing to do with World Championships. Name me one sport where the World Champion is chosen based on rating lists.

Formula 1 is one example. Tennis is another.

[Uri Blass]

Dear Uri,

Ok...but however,just i' like to mention that Fritz is a MP engine since many years, and from my experience i can say that Deep Fritz 13 is stronger than Deep Junior 13

And i see in your estimation:Houdini is approx.150 Elo stronger than Junior
Hmm...maybe you can be right,but only in case of Houdini is using not very strong book and playing on old-dated hardware

I mean,if Junior is playing on faster hardware and using a superior book

On equal conditions, the difference between both engines (Houdini/Junior):
Blitz:250-300 Elo difference
http://www.sedatcanbaz.com/chess/scct-rating/
http://www.clemens-keck.de/base.htm

Slow:200-250 Elo difference
http://www.amateurschach.de/
http://www.husvankempen.de/nunn/40_40%2 ... liste.html

Btw,i expected better performance by latest Junior 13 release (almost no any Elo difference between Junior 12)

Best,
Sedat

I believe good books for both engines reduce the difference in playing strength between engines so I still believe that 150 elo is close to the truth at long time control with books(and not bad book for houdini but good books as possible for both programs).

[rodolfoleoni]

It seems to me you have gone crazy in the last days. First you defend cloning in one post and now you 'value the contribution' of a known cloner
(BTW what contribution? I have seen nada)higher than long respected members of the chess scene e.g. Richard Pijil, SMK and Mark Uniacke and so on.
YOUR post makes me sick and angry, but I hope you find the right pills next time. Basta!

I missed that post, but..... thanks for supporting my friend Richard and all long time chess programmers.

I think this thread from Marco Costalba isn't against programmers, but against International COMPUTER Games Association. Which is not only computer chess. I'm only a CC enthusiast, not an ICGA member, and I've no idea about how many chess programmers are registered members, if they're the majority into ICGA or not. So I don't know if computer chess community is in right to remove Mr. Levy or not. Besides, I've no idea if the majority of CC programmers agrees with Marco's ideas.

I'm sure about one thing: one cannot pretend to be the World Champion without participating the Championship. And he cannot pretend to invalidate an event like that.

[Sedat Canbaz]

Dear Uri,

Ok...but however,just i' like to mention that Fritz is a MP engine since many years, and from my experience i can say that Deep Fritz 13 is stronger than Deep Junior 13

And i see in your estimation:Houdini is approx.150 Elo stronger than Junior
Hmm...maybe you can be right,but only in case of Houdini is using not very strong book and playing on old-dated hardware

I mean,if Junior is playing on faster hardware and using a superior book

On equal conditions, the difference between both engines (Houdini/Junior):
Blitz:250-300 Elo difference
http://www.sedatcanbaz.com/chess/scct-rating/
http://www.clemens-keck.de/base.htm

Slow:200-250 Elo difference
http://www.amateurschach.de/
http://www.husvankempen.de/nunn/40_40%2 ... liste.html

Btw,i expected better performance by latest Junior 13 release (almost no any Elo difference between Junior 12)

Best,
Sedat

I believe good books for both engines reduce the difference in playing strength between engines so I still believe that 150 elo is close to the truth at long time control with books(and not bad book for houdini but good books as possible for both programs).

In my opinion,we can't measure the real Elo strenght of the engines in case of not testing the engines with neutral books

In case of using non-neutral books,the Engines Elo performance is highly effecting from the opening books

Thats why there is possibility to see even 100 Elo difference
Even there is possibility to see 300 Elo difference between both engines
I mean,if we will test Houdini against Junior at slow time control

In other words,X engine which uses a stronger book will gain more Elo...


Greetings,
Sedat