CCRL 40/40, 40/4 and FRC lists updated (16th February 2019)

[Rebel]

Let's continue the discussion.

I am not arguing for myself, I retired from the competition in 2004, not interested any longer. What I am interested in is fair competition and I want Stockfish, Komodo, Houdini and all the rest to have tools to fight back against the learning engines, LZ maybe the first but certainly will not be the last.

I posted this previously.

Imagine you have a PGN with games of your engine and one (and likely many) game(s) went as follows:

Move 1-10 : pre-defined
Move 11 : score +0.20
Move 15 : score +0.25
Move 20 : score -0.10
Move 25 : score -0.80
Move 30 : score -3.xx

Something got wrong between move 16-25 and so you put a marker on those moves in a learn file till it plays something else and hopefully better.

What would be against it?

AZ and LZ work that way too only via a different path.

And yes, the markers are recognized by hashing.

[konsolas]

Let's continue the discussion.

I am not arguing for myself, I retired from the competition in 2004, not interested any longer. What I am interested in is fair competition and I want Stockfish, Komodo, Houdini and all the rest to have tools to fight back against the learning engines, LZ maybe the first but certainly will not be the last.

I posted this previously.

Imagine you have a PGN with games of your engine and one (and likely many) game(s) went as follows:

Move 1-10 : pre-defined
Move 11 : score +0.20
Move 15 : score +0.25
Move 20 : score -0.10
Move 25 : score -0.80
Move 30 : score -3.xx

Something got wrong between move 16-25 and so you put a marker on those moves in a learn file till it plays something else and hopefully better.

What would be against it?

AZ and LZ work that way too only via a different path.

And yes, the markers are recognized by hashing.

It looks like CCRL have already updated their testing conditions to allow this:

Code: Select all

Position learning: Off for all engines. If learning files exist they must be set to read-only. Any learning during the games played in the rating process is not permitted.

So your suggestion is allowed, provided that you create the learn file beforehand and it does not change during testing.

[hgm]

No. And there is no reason why should be allowed to get away with this.

Ed regards you as a stalker. He doesn't know what impels you to do it, but you do it - on him. He quit these forums some months ago, precisely because of your stalking, as you well know. Ed returned to posting recently. He is a mild-mannered guy and he posts quiet technical, often helpful things. He posted here, and you came again with the old, old stuff that you do, that he regards as rude. It is rude. He has put you on block, as you also know.

I'm not so mild mannered, as you know. You drove him off the forum once. And now you are doing it again. If you don't get the message to stop stalking the guy, stop being rude to him with absolutely no provocation, who is going to stop you? Ed has made his limits very clear, but you continue to breach them. It's not useful complaining to moderators, you are a moderator. So I'm going to stop you. Okay? Just pack it in. Stop with the gratuitous insults, or take them to CTF or something, but when you are moderator on a forum, you need to behave, you've had enough signals that you're being a problem. Stop. Goodbye and have a nice day.

There are other moderators beside me, you can complain to those, and if I cross a line, I am sure they will notify me. I am not stalking Ed specifically, I am just allergic to bullshit. Consider me the avenger of evil. If Ed or anyone else posts here that I know to be a gross falsehood, I will allways withspeak that poster. I won't make an exception for Ed just because he is Ed. My comments are always on the subject matter, that is not stalking. If someone repeats the same falsehoods again and again, he will naturally get the same rebuttal again and again. You seem to make a big deal out of that Ed might have put me on his ignore list for the forum, but if you had payed attention to his posts, you would know that basically everyone is already on the ignore list of his mind, as he never addresses the concerns of anyone who withspeaks him, and instead just goes on repeating the same stuff, as if that would tire out his detractors.

If anyone has been stalking here lately it is you, with your off-topic ad-hominem you injected in many threads as soon as a post of mine appeared there. But stalk me as much as you want, most of the time it just makes you look silly. If you want to jump on a sinking ship, like you are doing here, be my guest and sink with it...

[hgm]

I am not arguing for myself, I retired from the competition in 2004, not interested any longer. What I am interested in is fair competition and I want Stockfish, Komodo, Houdini and all the rest to have tools to fight back against the learning engines, LZ maybe the first but certainly will not be the last.

They can 'fight back' by improving their evaluation. That is fair competition, because it is all what AZ/LZ do. How this is done is not relevant. Just because AZ/LZ have a more efficient way to tune their evaluation is no reason to allow others to cheat. I never saw proposals to grant liberties to other engines to fight back agains Stockfish, because the latter was so much advantaged by employing fishtest. It would not be in the interest of anyone if they were allowed to fight back by cheap tricks unrelated to engine quality, such as opening books or their equivalent (i.e. tables of positions).

[Rebel]

Let's continue the discussion.

I am not arguing for myself, I retired from the competition in 2004, not interested any longer. What I am interested in is fair competition and I want Stockfish, Komodo, Houdini and all the rest to have tools to fight back against the learning engines, LZ maybe the first but certainly will not be the last.

I posted this previously.

Imagine you have a PGN with games of your engine and one (and likely many) game(s) went as follows:

Move 1-10 : pre-defined
Move 11 : score +0.20
Move 15 : score +0.25
Move 20 : score -0.10
Move 25 : score -0.80
Move 30 : score -3.xx

Something got wrong between move 16-25 and so you put a marker on those moves in a learn file till it plays something else and hopefully better.

What would be against it?

AZ and LZ work that way too only via a different path.

And yes, the markers are recognized by hashing.

It looks like CCRL have already updated their testing conditions to allow this:

Code: Select all

Position learning: Off for all engines. If learning files exist they must be set to read-only. Any learning during the games played in the rating process is not permitted.

So your suggestion is allowed, provided that you create the learn file beforehand and it does not change during testing.

I would say so too but hashing is considered as position learning.

[Chessqueen]

The latest CCRL Rating Lists and Statistics are available for viewing from the following links:
http://computerchess.org.uk/ccrl/4040/ (40/40)
http://www.computerchess.org.uk/ccrl/404/ (40/4)
http://www.computerchess.org.uk/ccrl/404FRC/ (FRC 40/4)

Please note that the three lists are often updated separately to each other. The FRC list is only updated when a new engine or engine version is being/has been tested.
Also, please note that the 40/40 and 40/4 lists are constructed from totally different databases and not comparable to each other.

Thanks to all of our currently active testers (Graham, Ray, Tirsa, Gabor, Sergio, Andreas and Brent), plus to Chessdom.com for providing an online server to assist with resources.

40/40 testing this week that I'm aware of will include (with live broadcast port for TLCV noted where applicable):

When will LCZero be included? I believe that the latest version of LCZero could easy end up in 2nd or 3rd place on your CCRL Rating list.

[Modern Times]

When will LCZero be included? I believe that the latest version of LCZero could easy end up in 2nd or 3rd place on your CCRL Rating list.

Which version is that, and where to get it from ?