Preparations for Purdue FOSCEC

Discussion of computer chess matches and engine tournaments.

Moderator: Ras

Gusev
Posts: 1476
Joined: Mon Jan 28, 2013 2:51 pm

Preparations for Purdue FOSCEC

Post by Gusev »

The Internet told me,
El FOSCEC, c'est le Fonds de Solidarité de Développement pour l'Emploi et la Protection des Chômeurs.
:)

Well, our Purdue FOSCEC is a student team project in my CNIT 242 System Administration: a Free, Open-Source Chess Engine Contest. 10 engines have been configured in Deep Fritz 14, and a short TC test tournament has began. (The test tournament is not being broadcast.) The current list of competitors is as follows:

1. Stockfish 220414 (standard local compile)
2. DON 1.0b (factory compile: the standard local compile started on another machine, but failed to run on the server)
3. Gull 3 (factory compile: the standard local compile started, but crashed in Deep Fritz 14)
4. Firenzina 2.4.2 (standard local compile)
5. Ivanhoe 1945b (standard local compile; a minor modification of 1945a)
6. Protector 1.6.0 (standard local compile)
7. Senpai 1.0 (standard local compile)
8. Texel 1.03 (factory compile: cannot make a standard local compile of Texel's C++11)
9. Toga II 3.0 (standard local compile)
10. MinkoChess 1.3 (factory compile)

Has a compile of ComStockCC ever been released?.. Here we have two descendants of Glaurung (Stockfish and DON), two descendants of Ippolit (Firenzina and Ivanhoe), two descendants of Fruit (Toga and Minko).
mcostalba
Posts: 2684
Joined: Sat Jun 14, 2008 9:17 pm

Re: Preparations for Purdue FOSCEC

Post by mcostalba »

Gusev wrote:The Internet told me,
El FOSCEC, c'est le Fonds de Solidarité de Développement pour l'Emploi et la Protection des Chômeurs.
:)

Well, our Purdue FOSCEC is a student team project in my CNIT 242 System Administration: a Free, Open-Source Chess Engine Contest. 10 engines have been configured in Deep Fritz 14, and a short TC test tournament has began. (The test tournament is not being broadcast.) The current list of competitors is as follows:

1. Stockfish 220414 (standard local compile)
2. DON 1.0b (factory compile: the standard local compile started on another machine, but failed to run on the server)
3. Gull 3 (factory compile: the standard local compile started, but crashed in Deep Fritz 14)
4. Firenzina 2.4.2 (standard local compile)
5. Ivanhoe 1945b (standard local compile; a minor modification of 1945a)
6. Protector 1.6.0 (standard local compile)
7. Senpai 1.0 (standard local compile)
8. Texel 1.03 (factory compile: cannot make a standard local compile of Texel's C++11)
9. Toga II 3.0 (standard local compile)
10. MinkoChess 1.3 (factory compile)

Has a compile of ComStockCC ever been released?.. Here we have two descendants of Glaurung (Stockfish and DON), two descendants of Ippolit (Firenzina and Ivanhoe), two descendants of Fruit (Toga and Minko).
Are the student supposed to learn something out of running a tournament of existing engines?

P.S: DON is a directly fork of SF, moreover the author quickly merges in any new sf development so it is 100% equivalent to latest dev version of SF modulo code formatting and bugs.
Vinvin
Posts: 5304
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

Re: Preparations for Purdue FOSCEC

Post by Vinvin »

mcostalba wrote:
Gusev wrote:The Internet told me,
El FOSCEC, c'est le Fonds de Solidarité de Développement pour l'Emploi et la Protection des Chômeurs.
:)

Well, our Purdue FOSCEC is a student team project in my CNIT 242 System Administration: a Free, Open-Source Chess Engine Contest. 10 engines have been configured in Deep Fritz 14, and a short TC test tournament has began. (The test tournament is not being broadcast.) The current list of competitors is as follows:

1. Stockfish 220414 (standard local compile)
2. DON 1.0b (factory compile: the standard local compile started on another machine, but failed to run on the server)
3. Gull 3 (factory compile: the standard local compile started, but crashed in Deep Fritz 14)
4. Firenzina 2.4.2 (standard local compile)
5. Ivanhoe 1945b (standard local compile; a minor modification of 1945a)
6. Protector 1.6.0 (standard local compile)
7. Senpai 1.0 (standard local compile)
8. Texel 1.03 (factory compile: cannot make a standard local compile of Texel's C++11)
9. Toga II 3.0 (standard local compile)
10. MinkoChess 1.3 (factory compile)

Has a compile of ComStockCC ever been released?.. Here we have two descendants of Glaurung (Stockfish and DON), two descendants of Ippolit (Firenzina and Ivanhoe), two descendants of Fruit (Toga and Minko).
Are the student supposed to learn something out of running a tournament of existing engines?

P.S: DON is a directly fork of SF, moreover the author quickly merges in any new sf development so it is 100% equivalent to latest dev version of SF modulo code formatting and bugs.
May be replace it by the latest GNUchess (based on fruit) or Fruit-reloaded ( see here : http://www.talkchess.com/forum/viewtopic.php?t=51638 )
Gusev
Posts: 1476
Joined: Mon Jan 28, 2013 2:51 pm

Re: Preparations for Purdue FOSCEC

Post by Gusev »

Are the student supposed to learn something out of running a tournament of existing engines?

P.S: DON is a directly fork of SF, moreover the author quickly merges in any new sf development so it is 100% equivalent to latest dev version of SF modulo code formatting and bugs.
The students will learn how to set up two communicating servers for a web broadcast. This is a system administration class, not a programming class. They began by installing Windows Server 2008 R2 from scratch and running all updates.

I saw four patches at http://abrok.eu/stockfish/ that were either "from DON", or "inspired by DON". Besides, we found out that DON is not as easy to compile as Stockfish (bugs?). I thought Amitis was a direct fork. Or are Amitis and DON the same project?
User avatar
velmarin
Posts: 1600
Joined: Mon Feb 21, 2011 9:48 am

Re: Preparations for Purdue FOSCEC

Post by velmarin »

Gusev wrote:
I saw four patches at http://abrok.eu/stockfish/ that were either "from DON", or "inspired by DON". Besides, we found out that DON is not as easy to compile as Stockfish (bugs?). I thought Amitis was a direct fork. Or are Amitis and DON the same project?
Amitis closed was a clone and not a code change.

Don it now has 907 commints and seven branches,
one with Syzygy bases.
It has more than 80 files.

I did not compile because the author ends not put Stockfish recognition, where I was born.

Also if Marco said to be a copy, do not argue.

The author provides the project in Visual C + + and compiles perfect.
Gusev
Posts: 1476
Joined: Mon Jan 28, 2013 2:51 pm

Re: Preparations for Purdue FOSCEC

Post by Gusev »

The preliminary short TC test seems to indicate that something is wrong with our home compile of Protector 1.6.0 :?
test08 Columbus, IN 2014

1 2 3 4 5 6 7 8 9 0
1 Stockfish 220414 64 ****** ½0½110 11111½ 111111 111½11 111111 111111 111111 111111 111111 50.0/54
2 DON 1.0b x64 ½1½001 ****** 1110½½ ½11½11 0½1011 1½1110 11½111 111111 111111 111111 43.5/54
3 Gull 3 x64 00000½ 0001½½ ****** 0½0½½0 ½½1111 1½11½1 111111 111½1½ 111111 111111 37.0/54
4 Ivanhoe version 1945b 000000 ½00½00 1½1½½1 ****** 1½½1½1 111½01 101011 111½½1 111111 111111 35.5/54
5 Firenzina 2.4.2 xTreme x64 000½00 1½0100 ½½0000 0½½0½0 ****** 101111 1½½111 111½11 111111 111111 33.0/54
6 Senpai 1.0 000000 0½0001 0½00½0 000½10 010000 ****** 1½100½ 111101 111110 111111 24.0/54
7 Toga II 3.0 000000 00½000 000000 010100 0½½000 0½011½ ****** 1½0½01 11111½ 111111 21.0/54
8 MinkoChess 1.3 x64 000000 000000 000½0½ 000½½0 000½00 000010 0½1½10 ****** 111111 111111 18.5/54
9 Texel 1.03 64-bit 000000 000000 000000 000000 000000 000001 00000½ 000000 ****** 111110 6.5/54
10 Protector 1.6.0 x64 000000 000000 000000 000000 000000 000000 000000 000000 000001 ****** 1.0/54
Gusev
Posts: 1476
Joined: Mon Jan 28, 2013 2:51 pm

Re: Preparations for Purdue FOSCEC

Post by Gusev »

I have started a new short TC preliminary test tournament, this time with a factory compile of Protector 1.6.0.

Jose, have you compiled ComStockCC, by any chance?
User avatar
velmarin
Posts: 1600
Joined: Mon Feb 21, 2011 9:48 am

Re: Preparations for Purdue FOSCEC

Post by velmarin »

Gusev wrote:I have started a new short TC preliminary test tournament, this time with a factory compile of Protector 1.6.0.

Jose, have you compiled ComStockCC, by any chance?
ComStock3, yes,

ComStock 4, It gave me problems.

ComStockCC, which is "CC"?

Ah, OK is ComStock DD,, is OK.
Gusev
Posts: 1476
Joined: Mon Jan 28, 2013 2:51 pm

Re: Preparations for Purdue FOSCEC

Post by Gusev »

Where can I find ComStock DD, then? The tarball at the Ippolit site provides code (but no compiles) for some ComStockCC. I had difficulty compiling from that code.
User avatar
velmarin
Posts: 1600
Joined: Mon Feb 21, 2011 9:48 am

Re: Preparations for Purdue FOSCEC

Post by velmarin »

Gusev wrote:Where can I find ComStock DD, then? The tarball at the Ippolit site provides code (but no compiles) for some ComStockCC. I had difficulty compiling from that code.
Between ComStock 3 and ComStockCC, there was a ComStockDD (how Stocfish DD)

ComStockCC I had not seen.

I've already arranged quickly, you have to look in case.
Have source code, proyect Visual C++, libraries Robbo and an executable for testing.
https://www.dropbox.com/s/ki4lnh6nd7dqjyz/Comstock.7z

Recompiling and tester.
Maybe put the last code Stockfish, perhaps it is compatible?

Greetings, Jose.