14th ACCA World Computer Rapid Chess Championships

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

Albert Silver
Posts: 3019
Joined: Wed Mar 08, 2006 9:57 pm
Location: Rio de Janeiro, Brazil

Re: 14th ACCA World Computer Rapid Chess Championships

Post by Albert Silver »

chrisw wrote: Sun Oct 25, 2020 2:40 pm
Joost Buijs wrote: Sun Oct 25, 2020 2:22 pm
chrisw wrote: Sun Oct 25, 2020 12:50 pm On Charles, reply to 2. Above. Mine is not rated on CCRL and won’t be because they don’t rate private engines. It has a rating of about 2750 or so at 5 1 on the nightmare.nl server and it gained maybe 150-200 or so in it’s time there because A) it now has a book, B) I implemented last SMP and C) it got retuned using the A Grant test suites recently published, but mostly, D) those were its first game outings and that spurs programmer to fix and do a bunch of things. There’s still much low hanging fruit.
My engine is not rated at CCRL too, and I don't care because I'm quite capable of testing/rating it myself.

A rating of 2750 on the nightmare.nl server is roughly comparable to 3100 CCRL 40/15, although it is very difficult to compare because most of the top 10 engines on nightmare.nl use very big hardware, while the lower rated engines do not, at least it gives you an idea.
Part of the increase of mine over recent months, I forgot to add, was also a switch from an i7 to an AMD 64 core 3990X and thus the Lazy SMP.

My only intention to apply to play in Charles's tourney would be for fun, same as the nightmare.nl thing, which is indeed fun. I used to be a commercial programmer, now twenty years on, I guess I'm an amateur, but fundamental rule of commercial programmer is security of your source code. It has value and you keep it in secure conditions. I take the same attitude now, there's no reason because I like to enter a tournament for fun purposes, that I want to place myself and code at risk by agreeing, under conditions I can't control, to hand the entire thing over to a committee, where, whatever anybody might claim, security can't be guaranteed, and one is at risk (as has been proven) for several years onwards. If someone wants to object to my box of magic BEFORE the tourney. fine, I can decide whether I want to enter of not, but after entry, or beyond that, I don't want to go signing any virtual agreements with unknown consequences at unknown times into the future. Is not compatible with making a "fun" entry.
Why on earth are you making this big hoopla about sending code and talk about 'committees' in a thread where the only question was what dates seemed doable? I played in the event and the question of originality was resolved (in my case) just via some friendly personal emails exchanged with Charles. I rather imagine that unless special reasons were openly stated, no one else was even questioned. It is run by a massive committee of one: the OP.

In spite of the 'World' in the title, it is a friendly rapid event usually held on ICC where the authors chat and others watch. You should join, it's fun.
"Tactics are the bricks and sticks that make up a game, but positional play is the architectural blueprint."
Joost Buijs
Posts: 1563
Joined: Thu Jul 16, 2009 10:47 am
Location: Almere, The Netherlands

Re: 14th ACCA World Computer Rapid Chess Championships

Post by Joost Buijs »

chrisw wrote: Sun Oct 25, 2020 2:40 pm
Joost Buijs wrote: Sun Oct 25, 2020 2:22 pm
chrisw wrote: Sun Oct 25, 2020 12:50 pm On Charles, reply to 2. Above. Mine is not rated on CCRL and won’t be because they don’t rate private engines. It has a rating of about 2750 or so at 5 1 on the nightmare.nl server and it gained maybe 150-200 or so in it’s time there because A) it now has a book, B) I implemented last SMP and C) it got retuned using the A Grant test suites recently published, but mostly, D) those were its first game outings and that spurs programmer to fix and do a bunch of things. There’s still much low hanging fruit.
My engine is not rated at CCRL too, and I don't care because I'm quite capable of testing/rating it myself.

A rating of 2750 on the nightmare.nl server is roughly comparable to 3100 CCRL 40/15, although it is very difficult to compare because most of the top 10 engines on nightmare.nl use very big hardware, while the lower rated engines do not, at least it gives you an idea.
Part of the increase of mine over recent months, I forgot to add, was also a switch from an i7 to an AMD 64 core 3990X and thus the Lazy SMP.

My only intention to apply to play in Charles's tourney would be for fun, same as the nightmare.nl thing, which is indeed fun. I used to be a commercial programmer, now twenty years on, I guess I'm an amateur, but fundamental rule of commercial programmer is security of your source code. It has value and you keep it in secure conditions. I take the same attitude now, there's no reason because I like to enter a tournament for fun purposes, that I want to place myself and code at risk by agreeing, under conditions I can't control, to hand the entire thing over to a committee, where, whatever anybody might claim, security can't be guaranteed, and one is at risk (as has been proven) for several years onwards. If someone wants to object to my box of magic BEFORE the tourney. fine, I can decide whether I want to enter of not, but after entry, or beyond that, I don't want to go signing any virtual agreements with unknown consequences at unknown times into the future. Is not compatible with making a "fun" entry.
I already wondered why you suddenly got such a high nps. For tournaments I use a 32 core AMD 3970X, the 3990X was a little bit too expensive to my taste.

I used to be a commercial programmer too, besides my job at the Amsterdam univ. I was in the business of writing data-acquisition and analysis packages, probably the reason that I'm not very fond of open-source software.

I don't consider the source of my chess engine very valuable but I keep it private because it became a big mess over the years and I don't feel the urge to clean it up.

Of course it is up to Charles to decide who is eligible to enter his tourney or not.
chrisw
Posts: 4315
Joined: Tue Apr 03, 2012 4:28 pm

Re: 14th ACCA World Computer Rapid Chess Championships

Post by chrisw »

Albert Silver wrote: Sun Oct 25, 2020 4:10 pm
chrisw wrote: Sun Oct 25, 2020 2:40 pm
Joost Buijs wrote: Sun Oct 25, 2020 2:22 pm
chrisw wrote: Sun Oct 25, 2020 12:50 pm On Charles, reply to 2. Above. Mine is not rated on CCRL and won’t be because they don’t rate private engines. It has a rating of about 2750 or so at 5 1 on the nightmare.nl server and it gained maybe 150-200 or so in it’s time there because A) it now has a book, B) I implemented last SMP and C) it got retuned using the A Grant test suites recently published, but mostly, D) those were its first game outings and that spurs programmer to fix and do a bunch of things. There’s still much low hanging fruit.
My engine is not rated at CCRL too, and I don't care because I'm quite capable of testing/rating it myself.

A rating of 2750 on the nightmare.nl server is roughly comparable to 3100 CCRL 40/15, although it is very difficult to compare because most of the top 10 engines on nightmare.nl use very big hardware, while the lower rated engines do not, at least it gives you an idea.
Part of the increase of mine over recent months, I forgot to add, was also a switch from an i7 to an AMD 64 core 3990X and thus the Lazy SMP.

My only intention to apply to play in Charles's tourney would be for fun, same as the nightmare.nl thing, which is indeed fun. I used to be a commercial programmer, now twenty years on, I guess I'm an amateur, but fundamental rule of commercial programmer is security of your source code. It has value and you keep it in secure conditions. I take the same attitude now, there's no reason because I like to enter a tournament for fun purposes, that I want to place myself and code at risk by agreeing, under conditions I can't control, to hand the entire thing over to a committee, where, whatever anybody might claim, security can't be guaranteed, and one is at risk (as has been proven) for several years onwards. If someone wants to object to my box of magic BEFORE the tourney. fine, I can decide whether I want to enter of not, but after entry, or beyond that, I don't want to go signing any virtual agreements with unknown consequences at unknown times into the future. Is not compatible with making a "fun" entry.
Why on earth are you making this big hoopla about sending code and talk about 'committees' in a thread where the only question was what dates seemed doable? I played in the event and the question of originality was resolved (in my case) just via some friendly personal emails exchanged with Charles. I rather imagine that unless special reasons were openly stated, no one else was even questioned. It is run by a massive committee of one: the OP.

In spite of the 'World' in the title, it is a friendly rapid event usually held on ICC where the authors chat and others watch. You should join, it's fun.
I would hope it was fun, I’m sure that’s what the idea is on the first place, and I’ld only apply to enter on the basis it was fun. However, I also read the small print. If you enter you are agreeing on basis complaint etc, to give up your source code for inspection. I called it to a committee for lack of a better term.

It’s obviously highly unlikely my entry would be challenged, but, it’s a finite risk with potential heavy downside. For entering a fun tourney just for fun, there’s absolutely no reason one should (well, that I should) accept any downside risk at all. For me it would be enter, play, enjoy, league table, end, done, all without much expectation of much of a personal result, btw. So, I’m simply asking to know the status beforehand. There’s zero reason my Covid inspired box of magic should have any problems and I’ld like to enter on basis it stays that way, no post start-tourney retroactive complaints. It’s almost certain I’ll never release or distribute it, btw. Again the potential downside to get heavily attacked, for example here, is too great, for effectively zero upside.
chrisw
Posts: 4315
Joined: Tue Apr 03, 2012 4:28 pm

Re: 14th ACCA World Computer Rapid Chess Championships

Post by chrisw »

Joost Buijs wrote: Sun Oct 25, 2020 4:53 pm
chrisw wrote: Sun Oct 25, 2020 2:40 pm
Joost Buijs wrote: Sun Oct 25, 2020 2:22 pm
chrisw wrote: Sun Oct 25, 2020 12:50 pm On Charles, reply to 2. Above. Mine is not rated on CCRL and won’t be because they don’t rate private engines. It has a rating of about 2750 or so at 5 1 on the nightmare.nl server and it gained maybe 150-200 or so in it’s time there because A) it now has a book, B) I implemented last SMP and C) it got retuned using the A Grant test suites recently published, but mostly, D) those were its first game outings and that spurs programmer to fix and do a bunch of things. There’s still much low hanging fruit.
My engine is not rated at CCRL too, and I don't care because I'm quite capable of testing/rating it myself.

A rating of 2750 on the nightmare.nl server is roughly comparable to 3100 CCRL 40/15, although it is very difficult to compare because most of the top 10 engines on nightmare.nl use very big hardware, while the lower rated engines do not, at least it gives you an idea.
Part of the increase of mine over recent months, I forgot to add, was also a switch from an i7 to an AMD 64 core 3990X and thus the Lazy SMP.

My only intention to apply to play in Charles's tourney would be for fun, same as the nightmare.nl thing, which is indeed fun. I used to be a commercial programmer, now twenty years on, I guess I'm an amateur, but fundamental rule of commercial programmer is security of your source code. It has value and you keep it in secure conditions. I take the same attitude now, there's no reason because I like to enter a tournament for fun purposes, that I want to place myself and code at risk by agreeing, under conditions I can't control, to hand the entire thing over to a committee, where, whatever anybody might claim, security can't be guaranteed, and one is at risk (as has been proven) for several years onwards. If someone wants to object to my box of magic BEFORE the tourney. fine, I can decide whether I want to enter of not, but after entry, or beyond that, I don't want to go signing any virtual agreements with unknown consequences at unknown times into the future. Is not compatible with making a "fun" entry.
I already wondered why you suddenly got such a high nps. For tournaments I use a 32 core AMD 3970X, the 3990X was a little bit too expensive to my taste.
i7 without any lazy SMP was what you first saw, I think that was maxing out around 2Mnps. 64 core 3990X, plus SMP, multiply the 2Mnps by 64, almost, I think the last tourney I only used 32 threads, so it was whatever it was. Strange for me because in past engine phase twenty more years ago was really not fast at all compared to everything else.


I used to be a commercial programmer too, besides my job at the Amsterdam univ. I was in the business of writing data-acquisition and analysis packages, probably the reason that I'm not very fond of open-source software.

I don't consider the source of my chess engine very valuable but I keep it private because it became a big mess over the years and I don't feel the urge to clean it up.

Of course it is up to Charles to decide who is eligible to enter his tourney or not.
Yes. I am probably too much trouble.
CRoberson
Posts: 2055
Joined: Mon Mar 13, 2006 2:31 am
Location: North Carolina, USA

Re: 14th ACCA World Computer Rapid Chess Championships

Post by CRoberson »

The voting has completed with 28-29 November being the weekend for the tournament.
Web pages will be up soon.

Charles
CRoberson
Posts: 2055
Joined: Mon Mar 13, 2006 2:31 am
Location: North Carolina, USA

Re: 14th ACCA World Computer Rapid Chess Championships

Post by CRoberson »

The webpages are online and up. Registration is open.
http://aigames.net/ACCA/ACCAWCRCC/2020A ... /home.html

Charles
CRoberson
Posts: 2055
Joined: Mon Mar 13, 2006 2:31 am
Location: North Carolina, USA

Re: 14th ACCA World Computer Rapid Chess Championships

Post by CRoberson »

Entry Nightmare 7.92 by Joost Buijs of The Netherlands.
chrisw
Posts: 4315
Joined: Tue Apr 03, 2012 4:28 pm

Re: 14th ACCA World Computer Rapid Chess Championships

Post by chrisw »

CRoberson wrote: Wed Nov 04, 2020 1:56 am The webpages are online and up. Registration is open.
http://aigames.net/ACCA/ACCAWCRCC/2020A ... /home.html

Charles
Hi Charles,

May I enter with CoronaVirusChess, please.
chrisw
Posts: 4315
Joined: Tue Apr 03, 2012 4:28 pm

Re: 14th ACCA World Computer Rapid Chess Championships

Post by chrisw »

chrisw wrote: Fri Nov 06, 2020 10:04 am
CRoberson wrote: Wed Nov 04, 2020 1:56 am The webpages are online and up. Registration is open.
http://aigames.net/ACCA/ACCAWCRCC/2020A ... /home.html

Charles
Hi Charles,

May I enter with CoronaVirusChess, please.
ah, just discovered the registration URL (it was hidden under the banner header on my iphone)
CRoberson
Posts: 2055
Joined: Mon Mar 13, 2006 2:31 am
Location: North Carolina, USA

Re: 14th ACCA World Computer Rapid Chess Championships

Post by CRoberson »

Entry; Goldbar by Bart Goldhoorn - The Netherlands