Winboard Forum

by **Will Singleton** » 14 Aug 2004, 09:26

Geschrieben von:/Posted by: Will Singleton at 14 August 2004 10:26:25:

The Averno gauntlet, unlike the Queen 2.44 result, shows a stiking difference between the classes.

Averno 0.70 vs. Bishop class
MONARCH, 2004.08.13 - 2004.08.14
                          Score       1    2 
----------------------------------------------
01: Averno 0.70         15.5 / 24    9w+  9b=
02: Spike 0.6            2.0 /  2    1b+  1w+
03: Djinn 0.870          1.5 /  2    1b+  1w=
04: Bruja 1.7            1.0 /  2    1b-  1w+
05: Alarm 0.93.1         1.0 /  2    1b-  1w+
06: Trace 1.29           1.0 /  2    1b-  1w+
07: Booot 4.03           1.0 /  2    1b+  1w-
08: DanChess 1.06 DC     0.5 /  2    1b-  1w=
09: Hermann 1.1.8        0.5 /  2    1b-  1w=
10: Cerebro 1.27         0.0 /  2    1b-  1w-
11: Snitch 0.5.5         0.0 /  2    1b-  1w-
12: BlackBishop 0.9.7i   0.0 /  2    1b-  1w-
13: BigLion 2.23i        0.0 /  2    1b-  1w-
----------------------------------------------
24 games: +13 =3 -8

Averno 0.70 vs. Rook class
MONARCH, 2004.08.11 - 2004.08.13
                             Score       1    2 
-------------------------------------------------
01: Averno 0.70             6.5 / 22    4w-  4b-
02: Arasan 7.4              2.0 /  2    1b+  1w+
03: The Baron 1.4.0 b 2     2.0 /  2    1b+  1w+
04: Terra 3.3 beta11        2.0 /  2    1b+  1w+
05: Naum 1.2                1.5 /  2    1b=  1w+
06: Pepito 1.59             1.5 /  2    1b+  1w=
07: Knightdreamer 3.3       1.5 /  2    1b+  1w=
08: Postmodernist 1010a     1.0 /  2    1b=  1w=
09: Frenzee 159             1.0 /  2    1b+  1w-
10: Pharaon 2.62            1.0 /  2    1b=  1w=
11: Comet B.68              1.0 /  2    1b=  1w=
12: The Crazy Bishop 0052   1.0 /  2    1b+  1w-
-------------------------------------------------
22 games: +4 =9 -9

The Averno gauntlet was played on an FX-51 2.2ghz, 40/25, ponder off, 4-man egtb.
Rook class excludes Amateur, those 2 games to be added later.
Pgn available at Heinz site soon.

by **Heinz van Kempen** » 14 Aug 2004, 09:35

Geschrieben von:/Posted by: Heinz van Kempen at 14 August 2004 10:35:29:
Als Antwort auf:/In reply to: AEGT: rook/bishop vs. Averno 0.70 geschrieben von:/posted by: Will Singleton at 14 August 2004 10:26:25:

The Averno gauntlet, unlike the Queen 2.44 result, shows a stiking difference between the classes.

The Averno gauntlet was played on an FX-51 2.2ghz, 40/25, ponder off, 4-man egtb.
Rook class excludes Amateur, those 2 games to be added later.
Pgn available at Heinz site soon.

>Averno 0.70 vs. Bishop class
>MONARCH, 2004.08.13 - 2004.08.14
>                          Score       1    2 
>----------------------------------------------
>01: Averno 0.70         15.5 / 24    9w+  9b=
>02: Spike 0.6            2.0 /  2    1b+  1w+
>03: Djinn 0.870          1.5 /  2    1b+  1w=
>04: Bruja 1.7            1.0 /  2    1b-  1w+
>05: Alarm 0.93.1         1.0 /  2    1b-  1w+
>06: Trace 1.29           1.0 /  2    1b-  1w+
>07: Booot 4.03           1.0 /  2    1b+  1w-
>08: DanChess 1.06 DC     0.5 /  2    1b-  1w=
>09: Hermann 1.1.8        0.5 /  2    1b-  1w=
>10: Cerebro 1.27         0.0 /  2    1b-  1w-
>11: Snitch 0.5.5         0.0 /  2    1b-  1w-
>12: BlackBishop 0.9.7i   0.0 /  2    1b-  1w-
>13: BigLion 2.23i        0.0 /  2    1b-  1w-
>----------------------------------------------
>24 games: +13 =3 -8
>
>Averno 0.70 vs. Rook class
>MONARCH, 2004.08.11 - 2004.08.13
>                             Score       1    2 
>-------------------------------------------------
>01: Averno 0.70             6.5 / 22    4w-  4b-
>02: Arasan 7.4              2.0 /  2    1b+  1w+
>03: The Baron 1.4.0 b 2     2.0 /  2    1b+  1w+
>04: Terra 3.3 beta11        2.0 /  2    1b+  1w+
>05: Naum 1.2                1.5 /  2    1b=  1w+
>06: Pepito 1.59             1.5 /  2    1b+  1w=
>07: Knightdreamer 3.3       1.5 /  2    1b+  1w=
>08: Postmodernist 1010a     1.0 /  2    1b=  1w=
>09: Frenzee 159             1.0 /  2    1b+  1w-
>10: Pharaon 2.62            1.0 /  2    1b=  1w=
>11: Comet B.68              1.0 /  2    1b=  1w=
>12: The Crazy Bishop 0052   1.0 /  2    1b+  1w-
>-------------------------------------------------
>22 games: +4 =9 -9
>
Hi Will ,
thanks for another interesting gauntlet. Games including those from Amateur will be available for download early evening. Want to watch a bit the Olympic games before.
Best Regards
Heinz

by **Daniel Shawul** » 14 Aug 2004, 09:56

Geschrieben von:/Posted by: Daniel Shawul at 14 August 2004 10:56:17:
Als Antwort auf:/In reply to: AEGT: rook/bishop vs. Averno 0.70 geschrieben von:/posted by: Will Singleton at 14 August 2004 10:26:25:

The Averno gauntlet, unlike the Queen 2.44 result, shows a stiking difference between the classes.

The Averno gauntlet was played on an FX-51 2.2ghz, 40/25, ponder off, 4-man egtb.
Rook class excludes Amateur, those 2 games to be added later.
Pgn available at Heinz site soon.

>Averno 0.70 vs. Bishop class
>MONARCH, 2004.08.13 - 2004.08.14
>                          Score       1    2 
>----------------------------------------------
>01: Averno 0.70         15.5 / 24    9w+  9b=
>02: Spike 0.6            2.0 /  2    1b+  1w+
>03: Djinn 0.870          1.5 /  2    1b+  1w=
>04: Bruja 1.7            1.0 /  2    1b-  1w+
>05: Alarm 0.93.1         1.0 /  2    1b-  1w+
>06: Trace 1.29           1.0 /  2    1b-  1w+
>07: Booot 4.03           1.0 /  2    1b+  1w-
>08: DanChess 1.06 DC     0.5 /  2    1b-  1w=
>09: Hermann 1.1.8        0.5 /  2    1b-  1w=
>10: Cerebro 1.27         0.0 /  2    1b-  1w-
>11: Snitch 0.5.5         0.0 /  2    1b-  1w-
>12: BlackBishop 0.9.7i   0.0 /  2    1b-  1w-
>13: BigLion 2.23i        0.0 /  2    1b-  1w-
>----------------------------------------------
>24 games: +13 =3 -8
>
>Averno 0.70 vs. Rook class
>MONARCH, 2004.08.11 - 2004.08.13
>                             Score       1    2 
>-------------------------------------------------
>01: Averno 0.70             6.5 / 22    4w-  4b-
>02: Arasan 7.4              2.0 /  2    1b+  1w+
>03: The Baron 1.4.0 b 2     2.0 /  2    1b+  1w+
>04: Terra 3.3 beta11        2.0 /  2    1b+  1w+
>05: Naum 1.2                1.5 /  2    1b=  1w+
>06: Pepito 1.59             1.5 /  2    1b+  1w=
>07: Knightdreamer 3.3       1.5 /  2    1b+  1w=
>08: Postmodernist 1010a     1.0 /  2    1b=  1w=
>09: Frenzee 159             1.0 /  2    1b+  1w-
>10: Pharaon 2.62            1.0 /  2    1b=  1w=
>11: Comet B.68              1.0 /  2    1b=  1w=
>12: The Crazy Bishop 0052   1.0 /  2    1b+  1w-
>-------------------------------------------------
>22 games: +4 =9 -9
>
Is Averno 0.7a much different than 0.7.
The 0.7a playing in WBEC seems much weaker than 0.7.
14.5 pts difference between DanChess and Averno 0.7a there.
despite the result,I don't think there is as such a sticking difference b/n rook and bishop classes. I am sure some of the top engines in bishop classes
did convincingly beat those in rook class in other tournaments.
best
daniel

by **Heinz van Kempen** » 14 Aug 2004, 14:00

Geschrieben von:/Posted by: Heinz van Kempen at 14 August 2004 15:00:18:
Als Antwort auf:/In reply to: AEGT: rook/bishop vs. Averno 0.70 geschrieben von:/posted by: Will Singleton at 14 August 2004 10:26:25:

The Averno gauntlet, unlike the Queen 2.44 result, shows a stiking difference between the classes.

The Averno gauntlet was played on an FX-51 2.2ghz, 40/25, ponder off, 4-man egtb.
Rook class excludes Amateur, those 2 games to be added later.
Pgn available at Heinz site soon.

>Averno 0.70 vs. Bishop class
>MONARCH, 2004.08.13 - 2004.08.14
>                          Score       1    2 
>----------------------------------------------
>01: Averno 0.70         15.5 / 24    9w+  9b=
>02: Spike 0.6            2.0 /  2    1b+  1w+
>03: Djinn 0.870          1.5 /  2    1b+  1w=
>04: Bruja 1.7            1.0 /  2    1b-  1w+
>05: Alarm 0.93.1         1.0 /  2    1b-  1w+
>06: Trace 1.29           1.0 /  2    1b-  1w+
>07: Booot 4.03           1.0 /  2    1b+  1w-
>08: DanChess 1.06 DC     0.5 /  2    1b-  1w=
>09: Hermann 1.1.8        0.5 /  2    1b-  1w=
>10: Cerebro 1.27         0.0 /  2    1b-  1w-
>11: Snitch 0.5.5         0.0 /  2    1b-  1w-
>12: BlackBishop 0.9.7i   0.0 /  2    1b-  1w-
>13: BigLion 2.23i        0.0 /  2    1b-  1w-
>----------------------------------------------
>24 games: +13 =3 -8
>
>Averno 0.70 vs. Rook class
>MONARCH, 2004.08.11 - 2004.08.13
>                             Score       1    2 
>-------------------------------------------------
>01: Averno 0.70             6.5 / 22    4w-  4b-
>02: Arasan 7.4              2.0 /  2    1b+  1w+
>03: The Baron 1.4.0 b 2     2.0 /  2    1b+  1w+
>04: Terra 3.3 beta11        2.0 /  2    1b+  1w+
>05: Naum 1.2                1.5 /  2    1b=  1w+
>06: Pepito 1.59             1.5 /  2    1b+  1w=
>07: Knightdreamer 3.3       1.5 /  2    1b+  1w=
>08: Postmodernist 1010a     1.0 /  2    1b=  1w=
>09: Frenzee 159             1.0 /  2    1b+  1w-
>10: Pharaon 2.62            1.0 /  2    1b=  1w=
>11: Comet B.68              1.0 /  2    1b=  1w=
>12: The Crazy Bishop 0052   1.0 /  2    1b+  1w-
>-------------------------------------------------
>22 games: +4 =9 -9
>
Hi all ,
results overview and downloads for this gauntlet are also available now.
http://www.husvankempen.de/nunn/
I have a question concerning older engines in AEGT tournaments. Volker Pittlik already mentioned that he would like it more to test those that are currently under development and I would support this concept and think that we do not need to have those engines in that are below the level of Pharaon and not updated for a long time and so we do not have to include them in the gauntlets for qualify. An exception could be made if one of those authors tells us, that he is still working on his engine and will soon come up with a new version.
In my list there are the following stronger ones not updated for more than two years:
Bringer
Leila
Sjeng
ESC
Phalanx
RESP
Exchess
InMiChess
Ant
and some more on weaker level
and already close to two years without update:
Chezzz
Lord King
Madeleine
CyberPagno
and some others
Opinions from my fellow testers and from the readers or authors concerned are of course welcome.
Best Regards
Heinz

by **Olivier Deville** » 14 Aug 2004, 17:07

Geschrieben von:/Posted by: Olivier Deville at 14 August 2004 18:07:29:
Als Antwort auf:/In reply to: uploaded..older engines in AEGT? geschrieben von:/posted by: Heinz van Kempen at 14 August 2004 15:00:18:

The Averno gauntlet, unlike the Queen 2.44 result, shows a stiking difference between the classes.

The Averno gauntlet was played on an FX-51 2.2ghz, 40/25, ponder off, 4-man egtb.
Rook class excludes Amateur, those 2 games to be added later.
Pgn available at Heinz site soon.
Hi all ,
results overview and downloads for this gauntlet are also available now.
http://www.husvankempen.de/nunn/
I have a question concerning older engines in AEGT tournaments. Volker Pittlik already mentioned that he would like it more to test those that are currently under development and I would support this concept and think that we do not need to have those engines in that are below the level of Pharaon and not updated for a long time and so we do not have to include them in the gauntlets for qualify. An exception could be made if one of those authors tells us, that he is still working on his engine and will soon come up with a new version.
In my list there are the following stronger ones not updated for more than two years:
Bringer
Leila
Sjeng
ESC
Phalanx
RESP
Exchess
InMiChess
Ant
and some more on weaker level
and already close to two years without update:
Chezzz
Lord King
Madeleine
CyberPagno
and some others
Opinions from my fellow testers and from the readers or authors concerned are of course welcome.
Best Regards
Heinz

>>Averno 0.70 vs. Bishop class
>>MONARCH, 2004.08.13 - 2004.08.14
>>                          Score       1    2 
>>----------------------------------------------
>>01: Averno 0.70         15.5 / 24    9w+  9b=
>>02: Spike 0.6            2.0 /  2    1b+  1w+
>>03: Djinn 0.870          1.5 /  2    1b+  1w=
>>04: Bruja 1.7            1.0 /  2    1b-  1w+
>>05: Alarm 0.93.1         1.0 /  2    1b-  1w+
>>06: Trace 1.29           1.0 /  2    1b-  1w+
>>07: Booot 4.03           1.0 /  2    1b+  1w-
>>08: DanChess 1.06 DC     0.5 /  2    1b-  1w=
>>09: Hermann 1.1.8        0.5 /  2    1b-  1w=
>>10: Cerebro 1.27         0.0 /  2    1b-  1w-
>>11: Snitch 0.5.5         0.0 /  2    1b-  1w-
>>12: BlackBishop 0.9.7i   0.0 /  2    1b-  1w-
>>13: BigLion 2.23i        0.0 /  2    1b-  1w-
>>----------------------------------------------
>>24 games: +13 =3 -8
>>
>>Averno 0.70 vs. Rook class
>>MONARCH, 2004.08.11 - 2004.08.13
>>                             Score       1    2 
>>-------------------------------------------------
>>01: Averno 0.70             6.5 / 22    4w-  4b-
>>02: Arasan 7.4              2.0 /  2    1b+  1w+
>>03: The Baron 1.4.0 b 2     2.0 /  2    1b+  1w+
>>04: Terra 3.3 beta11        2.0 /  2    1b+  1w+
>>05: Naum 1.2                1.5 /  2    1b=  1w+
>>06: Pepito 1.59             1.5 /  2    1b+  1w=
>>07: Knightdreamer 3.3       1.5 /  2    1b+  1w=
>>08: Postmodernist 1010a     1.0 /  2    1b=  1w=
>>09: Frenzee 159             1.0 /  2    1b+  1w-
>>10: Pharaon 2.62            1.0 /  2    1b=  1w=
>>11: Comet B.68              1.0 /  2    1b=  1w=
>>12: The Crazy Bishop 0052   1.0 /  2    1b+  1w-
>>-------------------------------------------------
>>22 games: +4 =9 -9
>>
Hi Heinz

Agreed, too many engines. We have no room for the old ones, unless the author says he is interested.
Olivier

ChessWar + AEGT

by **Dan Honeycutt** » 14 Aug 2004, 18:11

Geschrieben von:/Posted by: Dan Honeycutt at 14 August 2004 19:11:38:
Als Antwort auf:/In reply to: Re: uploaded..older engines in AEGT? geschrieben von:/posted by: Olivier Deville at 14 August 2004 18:07:29:

Hi Heinz
Agreed, too many engines. We have no room for the old ones, unless the author says he is interested.
Olivier

Hello Olivier:
In general I agree with you. Problem is everyone has favorites (Pepito in my case) that they would like to see play even though there is no active development. Nothing is ever easy.
Dan H.

by **Heinz van Kempen** » 14 Aug 2004, 18:22

Geschrieben von:/Posted by: Heinz van Kempen at 14 August 2004 19:22:37:
Als Antwort auf:/In reply to: Re: uploaded..older engines in AEGT? geschrieben von:/posted by: Dan Honeycutt at 14 August 2004 19:11:38:

Hi Heinz
Agreed, too many engines. We have no room for the old ones, unless the author says he is interested.
Olivier
Hello Olivier:
In general I agree with you. Problem is everyone has favorites (Pepito in my case) that they would like to see play even though there is no active development. Nothing is ever easy.
Dan H.

Hi Dan,
although I answer for Olivier now....sorry. Pepito although roughly 18 months old is above the level of Pharaon and will of course play.
Best Regards
Heinz

by **Uri Blass** » 14 Aug 2004, 18:35

Geschrieben von:/Posted by: Uri Blass at 14 August 2004 19:35:44:
Als Antwort auf:/In reply to: Re: uploaded..older engines in AEGT? geschrieben von:/posted by: Heinz van Kempen at 14 August 2004 19:22:37:

Hi Heinz
Agreed, too many engines. We have no room for the old ones, unless the author says he is interested.
Olivier
Hello Olivier:
In general I agree with you. Problem is everyone has favorites (Pepito in my case) that they would like to see play even though there is no active development. Nothing is ever easy.
Dan H.
Hi Dan,
although I answer for Olivier now....sorry. Pepito although roughly 18 months old is above the level of Pharaon and will of course play.
Best Regards
Heinz

I do not know how did you get that Pepito is above the level of pharaon.
From Leo's rating list:
12 9 Pharaon 2.62 2562 50 520
26 0 Pepito 1.59 2502 12 504
Uri

by **Heinz van Kempen** » 14 Aug 2004, 18:42

Geschrieben von:/Posted by: Heinz van Kempen at 14 August 2004 19:42:47:
Als Antwort auf:/In reply to: Re: uploaded..older engines in AEGT? geschrieben von:/posted by: Uri Blass at 14 August 2004 19:35:44:

Hi Heinz
Agreed, too many engines. We have no room for the old ones, unless the author says he is interested.
Olivier
Hello Olivier:
In general I agree with you. Problem is everyone has favorites (Pepito in my case) that they would like to see play even though there is no active development. Nothing is ever easy.
Dan H.
Hi Dan,
although I answer for Olivier now....sorry. Pepito although roughly 18 months old is above the level of Pharaon and will of course play.
Best Regards
Heinz

I do not know how did you get that Pepito is above the level of pharaon.
From Leo's rating list:
12 9 Pharaon 2.62 2562 50 520
26 0 Pepito 1.59 2502 12 504
Uri

Hello Uri,
okay, both are pretty close. You will find differences in all rating lists, although Leo´s is without doubt one of the best. In most of my tournaments Pepito was a bit better. But maybe after thousand games more...who knows exactly?
Best Regards
Heinz

by **Dan Honeycutt** » 14 Aug 2004, 18:53

Geschrieben von:/Posted by: Dan Honeycutt at 14 August 2004 19:53:25:
Als Antwort auf:/In reply to: Re: uploaded..older engines in AEGT? geschrieben von:/posted by: Heinz van Kempen at 14 August 2004 19:22:37:

Hi Heinz
Agreed, too many engines. We have no room for the old ones, unless the author says he is interested.
Olivier
Hello Olivier:
In general I agree with you. Problem is everyone has favorites (Pepito in my case) that they would like to see play even though there is no active development. Nothing is ever easy.
Dan H.
Hi Dan,
although I answer for Olivier now....sorry. Pepito although roughly 18 months old is above the level of Pharaon and will of course play.
Best Regards
Heinz

Hi Heinz:
IIRC you define "old" as > 2 years. That takes care of my favorite for the next 6 months (during which time I can hope Carlos del Cacho will have rekindled interest in chess programming) but the problem will still exist for someone else's favorite.
Dan H.

by **Heinz van Kempen** » 14 Aug 2004, 19:12

Geschrieben von:/Posted by: Heinz van Kempen at 14 August 2004 20:12:35:
Als Antwort auf:/In reply to: Re: uploaded..older engines in AEGT? geschrieben von:/posted by: Dan Honeycutt at 14 August 2004 19:53:25:

Hi Heinz
Agreed, too many engines. We have no room for the old ones, unless the author says he is interested.
Olivier
Hello Olivier:
In general I agree with you. Problem is everyone has favorites (Pepito in my case) that they would like to see play even though there is no active development. Nothing is ever easy.
Dan H.
Hi Dan,
although I answer for Olivier now....sorry. Pepito although roughly 18 months old is above the level of Pharaon and will of course play.
Best Regards
Heinz
Hi Heinz:
IIRC you define "old" as > 2 years. That takes care of my favorite for the next 6 months (during which time I can hope Carlos del Cacho will have rekindled interest in chess programming) but the problem will still exist for someone else's favorite.
Dan H.

Hi Dan,
you are right, but we have more than 200 amateur engines and each month more.
I would love to have them all in, but therefore we would need 30 or 40 testers more and many testers only want to play with the strongest.
For those who stopped programming...they will not need games anymore to improve something at this time, although they will still have their fans. So I think we give a chance to those who are now working a lot on their engines and for example prefer NagaSkaki, Muse, Eeyore etc. to Ant, Exchess, Requiem.
Anyway I might be wrong and there is some author of an "old" engine still working on it but not releasing, because he is still not satisfied. In this case it is sufficient just to write to us and we will include the old engine hoping that some day there will be an update.
One thing is for sure, those who want to be included should show a certain interest. Personally I like to test more for those knowing that they look at least at some games.
Best Regards
Heinz

by **Dan Honeycutt** » 14 Aug 2004, 19:34

Geschrieben von:/Posted by: Dan Honeycutt at 14 August 2004 20:34:52:
Als Antwort auf:/In reply to: Re: uploaded..older engines in AEGT? geschrieben von:/posted by: Heinz van Kempen at 14 August 2004 19:42:47:

Hi Heinz
Agreed, too many engines. We have no room for the old ones, unless the author says he is interested.
Olivier
Hello Olivier:
In general I agree with you. Problem is everyone has favorites (Pepito in my case) that they would like to see play even though there is no active development. Nothing is ever easy.
Dan H.
Hi Dan,
although I answer for Olivier now....sorry. Pepito although roughly 18 months old is above the level of Pharaon and will of course play.
Best Regards
Heinz

I do not know how did you get that Pepito is above the level of pharaon.
From Leo's rating list:
12 9 Pharaon 2.62 2562 50 520
26 0 Pepito 1.59 2502 12 504
Uri
Hello Uri,
okay, both are pretty close. You will find differences in all rating lists, although Leo´s is without doubt one of the best. In most of my tournaments Pepito was a bit better. But maybe after thousand games more...who knows exactly?
Best Regards
Heinz

Hi Heinz:
For selfish reasons I like your answer with regard to Pepito, but Uri's post does point out a possible bone of contention if you set the criteria to be "above the level of pharaon" when that is not well defined/subject to this or that rating list/statistical uncertainty etc. How about letting AEGT decide - any engine that maintains a position in King (or King/Queen) Class is exempt from the update rule. If it gets relegated to Queen (or Rook) class and has not been updated then it is dropped until such time as a new update is available. You then have the young bucks fighting to knock the dinosaurs out of the ring.
Dan H.

by **Dan Honeycutt** » 14 Aug 2004, 19:51

Geschrieben von:/Posted by: Dan Honeycutt at 14 August 2004 20:51:27:
Als Antwort auf:/In reply to: Re: uploaded..older engines in AEGT? geschrieben von:/posted by: Heinz van Kempen at 14 August 2004 20:12:35:

Hi Heinz
Agreed, too many engines. We have no room for the old ones, unless the author says he is interested.
Olivier
Hello Olivier:
In general I agree with you. Problem is everyone has favorites (Pepito in my case) that they would like to see play even though there is no active development. Nothing is ever easy.
Dan H.
Hi Dan,
although I answer for Olivier now....sorry. Pepito although roughly 18 months old is above the level of Pharaon and will of course play.
Best Regards
Heinz
Hi Heinz:
IIRC you define "old" as > 2 years. That takes care of my favorite for the next 6 months (during which time I can hope Carlos del Cacho will have rekindled interest in chess programming) but the problem will still exist for someone else's favorite.
Dan H.
Hi Dan,
you are right, but we have more than 200 amateur engines and each month more.
I would love to have them all in, but therefore we would need 30 or 40 testers more and many testers only want to play with the strongest.
For those who stopped programming...they will not need games anymore to improve something at this time, although they will still have their fans. So I think we give a chance to those who are now working a lot on their engines and for example prefer NagaSkaki, Muse, Eeyore etc. to Ant, Exchess, Requiem.
Anyway I might be wrong and there is some author of an "old" engine still working on it but not releasing, because he is still not satisfied. In this case it is sufficient just to write to us and we will include the old engine hoping that some day there will be an update.
One thing is for sure, those who want to be included should show a certain interest. Personally I like to test more for those knowing that they look at least at some games.
Best Regards
Heinz

Hi Heinz:
OK, I agree with you and Olivier. Personal "favorites" should not factor in the decision. Engines should have to play their way into the tournament. And I agree (especially being the author of one) that new/updated engines should be given a chance to show their stuff over old engines that are no longer being worked on.
Dan H.

by **Heinz van Kempen** » 14 Aug 2004, 20:01

Geschrieben von:/Posted by: Heinz van Kempen at 14 August 2004 21:01:32:
Als Antwort auf:/In reply to: Re: uploaded..older engines in AEGT? geschrieben von:/posted by: Dan Honeycutt at 14 August 2004 20:34:52:

Hi Heinz
Agreed, too many engines. We have no room for the old ones, unless the author says he is interested.
Olivier
Hello Olivier:
In general I agree with you. Problem is everyone has favorites (Pepito in my case) that they would like to see play even though there is no active development. Nothing is ever easy.
Dan H.
Hi Dan,
although I answer for Olivier now....sorry. Pepito although roughly 18 months old is above the level of Pharaon and will of course play.
Best Regards
Heinz

I do not know how did you get that Pepito is above the level of pharaon.
From Leo's rating list:
12 9 Pharaon 2.62 2562 50 520
26 0 Pepito 1.59 2502 12 504
Uri
Hello Uri,
okay, both are pretty close. You will find differences in all rating lists, although Leo´s is without doubt one of the best. In most of my tournaments Pepito was a bit better. But maybe after thousand games more...who knows exactly?
Best Regards
Heinz
Hi Heinz:
For selfish reasons I like your answer with regard to Pepito, but Uri's post does point out a possible bone of contention if you set the criteria to be "above the level of pharaon" when that is not well defined/subject to this or that rating list/statistical uncertainty etc. How about letting AEGT decide - any engine that maintains a position in King (or King/Queen) Class is exempt from the update rule. If it gets relegated to Queen (or Rook) class and has not been updated then it is dropped until such time as a new update is available. You then have the young bucks fighting to knock the dinosaurs out of the ring.
Dan H.

Hi Dan,
you can be sure that everything is given for voting. I do not decide alone and there is some voting underway and I am waiting for decisions from some testers still because of holiday time.
Those in King Class and Queen Class will be safe anyway I think, but one of the votings is, if we include those "hybrid" engines like Ktulu, Patriot and soon SmarThink and Fafis. And with still some votes to come it seems likely that they will excluded from AEGT 2 despite of being very strong (maybe we will keep Ruffian for comparison).
Another voting seems to go that way that List will be excluded from AEGT 2, mainly because of this one move prior to mate bug, as it gives also the problem to us how to rate this games (or if to rate them at all) for not giving disadvantage to engines that do resign.
And finally we are voting if we have the same number or more or less classes with next AEGT. Maybe because we do not need to run gauntlets then it might be possible to run a Knight and a Pawn class additionally. Not all want this, but no one wants less classes, so maybe the four or five testers wanting more classes will run them before the main classes start and the others can recover a bit and run their own tournaments in the meantime.
Most engines from this AEGT will drop one class because there are still many "in-between" engines we did not vote for, because we could not include all, but they are strong enough to push some others down after the gauntlets. This will be SOS, Pro Deo (against the trend changing from commercial to free engine, thanks to Ed for this great decision :-)

, Little Goliath, LambChop, Abrok, Nejmet, Betsy, Patzer, Chispa, Averno, Queen, etc..
To give the running tournaments a certain value I will propose to my fellow testers that the first two one from the classes now should be guaranteed promotion and at least ranks 3, 4, 5 and 6 do not have to go down, independent of rating gained by those who qualify by gauntlets.
I repeat, all will be done by votings, only that I want to give some rest at this time, because we already had a lot of discussions and even some hard times with testers being not very cooperative for team work and who did not survive the "storming" (like Volker calls it and I learnt from him that team building has several steps: forming, storming, norming and performing).
Best Regards
Heinz

by **Johan Havegheer** » 14 Aug 2004, 21:11

Geschrieben von:/Posted by: Johan Havegheer at 14 August 2004 22:11:19:
Als Antwort auf:/In reply to: uploaded..older engines in AEGT? geschrieben von:/posted by: Heinz van Kempen at 14 August 2004 15:00:18:

I have a question concerning older engines in AEGT tournaments. Volker Pittlik already mentioned that he would like it more to test those that are currently under development and I would support this concept and think that we do not need to have those engines in that are below the level of Pharaon and not updated for a long time and so we do not have to include them in the gauntlets for qualify. An exception could be made if one of those authors tells us, that he is still working on his engine and will soon come up with a new version.
In my list there are the following stronger ones not updated for more than two years:
Bringer : development stopped ??
Leila : A very promising engine a few years ago
Sjeng : commercial as Deep Sjeng so still in development !!
ESC
Phalanx : development stopped ??
RESP
Exchess
InMiChess
Ant : Still in development, but private at the moment. New version not much stronger than Ant 6.06
and some more on weaker level
and already close to two years without update:
Chezzz
Lord King
Madeleine
CyberPagno
and some others
Opinions from my fellow testers and from the readers or authors concerned are of course welcome.
Best Regards
Heinz

Best
Johan

by **Heinz van Kempen** » 14 Aug 2004, 21:24

Geschrieben von:/Posted by: Heinz van Kempen at 14 August 2004 22:24:42:
Als Antwort auf:/In reply to: Re: uploaded..older engines in AEGT? geschrieben von:/posted by: Johan Havegheer at 14 August 2004 22:11:19:

Bringer : development stopped ??
Leila : A very promising engine a few years ago
Sjeng : commercial as Deep Sjeng so still in development !!
ESC
Phalanx : development stopped ??
RESP
Exchess
InMiChess
Ant : Still in development, but private at the moment. New version not much stronger than Ant 6.06
and some more on weaker level
and already close to two years without update:
Chezzz
Lord King
Madeleine
CyberPagno
and some others
Opinions from my fellow testers and from the readers or authors concerned are of course welcome.
Best Regards
Heinz

Best
Johan

Hi Johan,
thanks for information concerning Ant. I also read that ESC is still under development, maybe a new version will be released some day.
Like you I regret a lot the "loss" of Leila, Bringer and Phalanx (the latter for aggressive playing style). Sjeng could be out because of majority vote that we do not take in engines where a commercial version is on the market or will soon be out. All who still can vote would now have to vote in favour of the "hybrids" to have them in, but many testers voted that they want to keep Ruffian for comparison (being by far the strongest of this "hybrid" group, compared to Ktulu 4.2, Patriot 0.172 light, but maybe not to the commercial Ktulu 5.1, that for an amateur tournament is not of any interest).
Please remember to send games to Olivier when you have a new bunch. Results and games from you and Olivier are the only ones we have at this moment for King Class, because the others will give their games later all at once I suppose.
Best Regards
Heinz

by **Jose Carlos** » 16 Aug 2004, 00:50

Geschrieben von:/Posted by: Jose Carlos at 16 August 2004 01:50:57:
Als Antwort auf:/In reply to: Re: AEGT: rook/bishop vs. Averno 0.70 geschrieben von:/posted by: Daniel Shawul at 14 August 2004 10:56:17:

The Averno gauntlet, unlike the Queen 2.44 result, shows a stiking difference between the classes.

The Averno gauntlet was played on an FX-51 2.2ghz, 40/25, ponder off, 4-man egtb.
Rook class excludes Amateur, those 2 games to be added later.
Pgn available at Heinz site soon.
Is Averno 0.7a much different than 0.7.
The 0.7a playing in WBEC seems much weaker than 0.7.
14.5 pts difference between DanChess and Averno 0.7a there.
despite the result,I don't think there is as such a sticking difference b/n rook and bishop classes. I am sure some of the top engines in bishop classes
did convincingly beat those in rook class in other tournaments.
best
daniel

>>Averno 0.70 vs. Bishop class
>>MONARCH, 2004.08.13 - 2004.08.14
>>                          Score       1    2 
>>----------------------------------------------
>>01: Averno 0.70         15.5 / 24    9w+  9b=
>>02: Spike 0.6            2.0 /  2    1b+  1w+
>>03: Djinn 0.870          1.5 /  2    1b+  1w=
>>04: Bruja 1.7            1.0 /  2    1b-  1w+
>>05: Alarm 0.93.1         1.0 /  2    1b-  1w+
>>06: Trace 1.29           1.0 /  2    1b-  1w+
>>07: Booot 4.03           1.0 /  2    1b+  1w-
>>08: DanChess 1.06 DC     0.5 /  2    1b-  1w=
>>09: Hermann 1.1.8        0.5 /  2    1b-  1w=
>>10: Cerebro 1.27         0.0 /  2    1b-  1w-
>>11: Snitch 0.5.5         0.0 /  2    1b-  1w-
>>12: BlackBishop 0.9.7i   0.0 /  2    1b-  1w-
>>13: BigLion 2.23i        0.0 /  2    1b-  1w-
>>----------------------------------------------
>>24 games: +13 =3 -8
>>
>>Averno 0.70 vs. Rook class
>>MONARCH, 2004.08.11 - 2004.08.13
>>                             Score       1    2 
>>-------------------------------------------------
>>01: Averno 0.70             6.5 / 22    4w-  4b-
>>02: Arasan 7.4              2.0 /  2    1b+  1w+
>>03: The Baron 1.4.0 b 2     2.0 /  2    1b+  1w+
>>04: Terra 3.3 beta11        2.0 /  2    1b+  1w+
>>05: Naum 1.2                1.5 /  2    1b=  1w+
>>06: Pepito 1.59             1.5 /  2    1b+  1w=
>>07: Knightdreamer 3.3       1.5 /  2    1b+  1w=
>>08: Postmodernist 1010a     1.0 /  2    1b=  1w=
>>09: Frenzee 159             1.0 /  2    1b+  1w-
>>10: Pharaon 2.62            1.0 /  2    1b=  1w=
>>11: Comet B.68              1.0 /  2    1b=  1w=
>>12: The Crazy Bishop 0052   1.0 /  2    1b+  1w-
>>-------------------------------------------------
>>22 games: +4 =9 -9
>>
Hi, Averno 0.70a and Averno 0.70 CCT6 is the same engine. I sent 0.70a to some people before CCT6 and then, seeing it was stable, I released it as 0.70 CCT6. I possibly made some microscopic change in compiler optimization, but nothing important at all. They're the same.
The difference in results might come from several sources. Opening book, ponder on/off, hash settings, time control, personality file... or simply random noise.
José C.

by **Daniel Shawul** » 16 Aug 2004, 07:52

Geschrieben von:/Posted by: Daniel Shawul at 16 August 2004 08:52:26:
Als Antwort auf:/In reply to: Re: AEGT: rook/bishop vs. Averno 0.70 geschrieben von:/posted by: Jose Carlos at 16 August 2004 01:50:57:

The Averno gauntlet, unlike the Queen 2.44 result, shows a stiking difference between the classes.

The Averno gauntlet was played on an FX-51 2.2ghz, 40/25, ponder off, 4-man egtb.
Rook class excludes Amateur, those 2 games to be added later.
Pgn available at Heinz site soon.
Is Averno 0.7a much different than 0.7.
The 0.7a playing in WBEC seems much weaker than 0.7.
14.5 pts difference between DanChess and Averno 0.7a there.
despite the result,I don't think there is as such a sticking difference b/n rook and bishop classes. I am sure some of the top engines in bishop classes
did convincingly beat those in rook class in other tournaments.
best
daniel
Hi, Averno 0.70a and Averno 0.70 CCT6 is the same engine. I sent 0.70a to some people before CCT6 and then, seeing it was stable, I released it as 0.70 CCT6. I possibly made some microscopic change in compiler optimization, but nothing important at all. They're the same.
The difference in results might come from several sources. Opening book, ponder on/off, hash settings, time control, personality file... or simply random noise.
José C.

>>>Averno 0.70 vs. Bishop class
>>>MONARCH, 2004.08.13 - 2004.08.14
>>>                          Score       1    2 
>>>----------------------------------------------
>>>01: Averno 0.70         15.5 / 24    9w+  9b=
>>>02: Spike 0.6            2.0 /  2    1b+  1w+
>>>03: Djinn 0.870          1.5 /  2    1b+  1w=
>>>04: Bruja 1.7            1.0 /  2    1b-  1w+
>>>05: Alarm 0.93.1         1.0 /  2    1b-  1w+
>>>06: Trace 1.29           1.0 /  2    1b-  1w+
>>>07: Booot 4.03           1.0 /  2    1b+  1w-
>>>08: DanChess 1.06 DC     0.5 /  2    1b-  1w=
>>>09: Hermann 1.1.8        0.5 /  2    1b-  1w=
>>>10: Cerebro 1.27         0.0 /  2    1b-  1w-
>>>11: Snitch 0.5.5         0.0 /  2    1b-  1w-
>>>12: BlackBishop 0.9.7i   0.0 /  2    1b-  1w-
>>>13: BigLion 2.23i        0.0 /  2    1b-  1w-
>>>----------------------------------------------
>>>24 games: +13 =3 -8
>>>
>>>Averno 0.70 vs. Rook class
>>>MONARCH, 2004.08.11 - 2004.08.13
>>>                             Score       1    2 
>>>-------------------------------------------------
>>>01: Averno 0.70             6.5 / 22    4w-  4b-
>>>02: Arasan 7.4              2.0 /  2    1b+  1w+
>>>03: The Baron 1.4.0 b 2     2.0 /  2    1b+  1w+
>>>04: Terra 3.3 beta11        2.0 /  2    1b+  1w+
>>>05: Naum 1.2                1.5 /  2    1b=  1w+
>>>06: Pepito 1.59             1.5 /  2    1b+  1w=
>>>07: Knightdreamer 3.3       1.5 /  2    1b+  1w=
>>>08: Postmodernist 1010a     1.0 /  2    1b=  1w=
>>>09: Frenzee 159             1.0 /  2    1b+  1w-
>>>10: Pharaon 2.62            1.0 /  2    1b=  1w=
>>>11: Comet B.68              1.0 /  2    1b=  1w=
>>>12: The Crazy Bishop 0052   1.0 /  2    1b+  1w-
>>>-------------------------------------------------
>>>22 games: +4 =9 -9
>>>
That's right. my objection is to the phrase "sticking difference b/n the classes".
daniel

Winboard Forum

Who is online