AEGT---Crafty.rc

[Heinz van Kempen]

Geschrieben von:/Posted by: Heinz van Kempen at 22 July 2004 03:31:03:

Hi all testers ,
Slobodan is out of our list of testers (my decision). Seems that he anyway never was sure if to join and he did not participate in votings and other stuff. There seemed to be some other doubtful things, so we continue with those we have and I am sure we will get more help when we have started and people will see that it is a good tournament.
I am almost certain that you all know that for running Crafty decently a Crafty.rc file is necessary. If you do not have one or are not certain how to make one, I think Bryan can give you a link where to download it (or I will send you mine).
Bryan might feel free to give as two Crafty.rc files for those who can use 48 and for those who can give 96 MB. Please remember then Bryan that we voted by majority that pawnhash should be included in the total amount of hash allowed and cannot be given additonally.
To Olivier: are we already proceeding concerning this Axon-Benchmark stuff?
Best Regards
Heinz

[Heinz van Kempen]

Geschrieben von:/Posted by: Heinz van Kempen at 22 July 2004 06:09:45:
Als Antwort auf:/In reply to: AEGT---Crafty.rc geschrieben von:/posted by: Heinz van Kempen at 22 July 2004 03:31:03:

Hi Slobodan ,
"other doubtful things" like I wrote before should in no case sound like an accusation. This is just my deficient knowledge of English language, sorry for that.
What I meant are those tournament patterns. I never understood what you meant by those and first thought maybe tournament modes like round robin, swiss system, knockout or anything. In your first and only email you again only repeated 3,5,7,9 tournament patterns.
Was there anything you did not like concerning our project that finally induced you not to take part in the votes and not to participate or did you never intend to participate and I only misunderstood your postings in the discussion under "just a try" etc. as the intention that you might want to join?
Best Regards
Heinz

[Igor Gorelikov]

Geschrieben von:/Posted by: Igor Gorelikov at 22 July 2004 12:35:08:
Als Antwort auf:/In reply to: AEGT---Crafty.rc geschrieben von:/posted by: Heinz van Kempen at 22 July 2004 03:31:03:

Hi all testers ,
Slobodan is out of our list of testers (my decision). Seems that he anyway never was sure if to join and he did not participate in votings and other stuff. There seemed to be some other doubtful things, so we continue with those we have and I am sure we will get more help when we have started and people will see that it is a good tournament.
I am almost certain that you all know that for running Crafty decently a Crafty.rc file is necessary. If you do not have one or are not certain how to make one, I think Bryan can give you a link where to download it (or I will send you mine).
Bryan might feel free to give as two Crafty.rc files for those who can use 48 and for those who can give 96 MB. Please remember then Bryan that we voted by majority that pawnhash should be included in the total amount of hash allowed and cannot be given additonally.
To Olivier: are we already proceeding concerning this Axon-Benchmark stuff?
Best Regards
Heinz

My opinion is that any new tester can be added on to the list any time later if he will be interested (including Slobodan). Then he just should conform to all our previous decisions (votings) made.
Igor

[Heinz van Kempen]

Geschrieben von:/Posted by: Heinz van Kempen at 22 July 2004 13:17:17:
Als Antwort auf:/In reply to: Re: AEGT---Crafty.rc geschrieben von:/posted by: Igor Gorelikov at 22 July 2004 12:35:08:

Hi all testers ,
Slobodan is out of our list of testers (my decision). Seems that he anyway never was sure if to join and he did not participate in votings and other stuff. There seemed to be some other doubtful things, so we continue with those we have and I am sure we will get more help when we have started and people will see that it is a good tournament.
I am almost certain that you all know that for running Crafty decently a Crafty.rc file is necessary. If you do not have one or are not certain how to make one, I think Bryan can give you a link where to download it (or I will send you mine).
Bryan might feel free to give as two Crafty.rc files for those who can use 48 and for those who can give 96 MB. Please remember then Bryan that we voted by majority that pawnhash should be included in the total amount of hash allowed and cannot be given additonally.
To Olivier: are we already proceeding concerning this Axon-Benchmark stuff?
Best Regards
Heinz

My opinion is that any new tester can be added on to the list any time later if he will be interested (including Slobodan). Then he just should conform to all our previous decisions (votings) made.
Igor

Hello Igor,
yes, that is what we want, but anyway when there is no reaction to voting or other questions, I can do nothing and it should then not appear that the tester is proceeding. As I already stated, anyone can also stop as soon as he lost interest.
There are still some votes to come, but I regret that I have to tell you Igor that most are deciding to pick the latest available version exclusively, what will be in case of Gothmog the brandnew Gothmog 1.0 beta10 and also a new beta is allowed if the author propose this to us and can deliver a stable and tested one. For Tao so please all make a short test at least that the beta 5.7b4 is running without problems and other hint concerning Tao. Every game where it participates must be checked, how it finished in order to see if the underpromotion bug still exists. As you all know previous Tao versions simply ignored underpromotion to a knight for example. If it is ignored Tao will lose the game, it will not be replayed.
The votings concerning the other issues are still not decided, only that most think that for Crafty we should take all the same compile. We have really tough discussions concerning List via email, especially here I would like to have the opinion from any tester please.
Best Regards
Heinz

[Fabien Letouzey]

Geschrieben von:/Posted by: Fabien Letouzey at 22 July 2004 13:27:04:
Als Antwort auf:/In reply to: AEGT---engine versions geschrieben von:/posted by: Heinz van Kempen at 22 July 2004 13:17:17:

The votings concerning the other issues are still not decided, only that most think that for Crafty we should take all the same compile. We have really tough discussions concerning List via email, especially here I would like to have the opinion from any tester please.

Heinz,
As an engine author I have avoided any discussions so far.
Now that List is mentionned I should probably take part as PolyGlot is concerned. Since the problem with List is not really technical (writing a work around for *this* bug is easy, since it is well understood), I believe it is in the interest of the tournament that I participate in the discussion.
Of course only about the participation of UCI engines, as the tournament is obviously a WinBoard one. Also I think, not via a forum.
Fabien.

[Igor Gorelikov]

Geschrieben von:/Posted by: Igor Gorelikov at 22 July 2004 13:27:39:
Als Antwort auf:/In reply to: AEGT---engine versions geschrieben von:/posted by: Heinz van Kempen at 22 July 2004 13:17:17:

There are still some votes to come, but I regret that I have to tell you Igor that most are deciding to pick the latest available version exclusively.

I know that the most want to test latest versions (even if they are weaker).
So I have my two-stage proposal which I want to set on voting to all members.
Igor

[Heinz van Kempen]

Geschrieben von:/Posted by: Heinz van Kempen at 22 July 2004 13:45:48:
Als Antwort auf:/In reply to: Re: AEGT---engine versions geschrieben von:/posted by: Fabien Letouzey at 22 July 2004 13:27:04:

The votings concerning the other issues are still not decided, only that most think that for Crafty we should take all the same compile. We have really tough discussions concerning List via email, especially here I would like to have the opinion from any tester please.

Heinz,
As an engine author I have avoided any discussions so far.
Now that List is mentionned I should probably take part as PolyGlot is concerned. Since the problem with List is not really technical (writing a work around for *this* bug is easy, since it is well understood), I believe it is in the interest of the tournament that I participate in the discussion.
Of course only about the participation of UCI engines, as the tournament is obviously a WinBoard one. Also I think, not via a forum.
Fabien.

Hello Fabien,
we already exchanged I think more than 100 emails concerining all the questions.
Thanks for offering your help. I will write an email to you and explain and if the other testers have no objections you can make proposals of course.
Best Regards
Heinz

[Igor Gorelikov]

Geschrieben von:/Posted by: Igor Gorelikov at 22 July 2004 14:31:53:
Als Antwort auf:/In reply to: AEGT---Crafty.rc geschrieben von:/posted by: Heinz van Kempen at 22 July 2004 03:31:03:

Bryan might feel free to give as two Crafty.rc files for those who can use 48 and for those who can give 96 MB.

If you set Crafty to 48:
hash=48M
hashp=1M
that means aclual allocation of 55 MB which is too big (in engine to engine matches) for my 256MB of RAM.
Igor

[Heinz van Kempen]

Geschrieben von:/Posted by: Heinz van Kempen at 22 July 2004 14:35:37:
Als Antwort auf:/In reply to: Crafty.rc geschrieben von:/posted by: Igor Gorelikov at 22 July 2004 14:31:53:

Bryan might feel free to give as two Crafty.rc files for those who can use 48 and for those who can give 96 MB.

If you set Crafty to 48:
hash=48M
hashp=1M
that means aclual allocation of 55 MB which is too big (in engine to engine matches) for my 256MB of RAM.
Igor

Hello Igor,
we already agreed to make an exception for you and we know that you can only give 38 MB max. for each engine.
Best Regards
Heinz

[Bryan Hofmann]

Geschrieben von:/Posted by: Bryan Hofmann at 22 July 2004 14:43:36:
Als Antwort auf:/In reply to: AEGT---Crafty.rc geschrieben von:/posted by: Heinz van Kempen at 22 July 2004 03:31:03:

Hi all testers ,
Slobodan is out of our list of testers (my decision). Seems that he anyway never was sure if to join and he did not participate in votings and other stuff. There seemed to be some other doubtful things, so we continue with those we have and I am sure we will get more help when we have started and people will see that it is a good tournament.
I am almost certain that you all know that for running Crafty decently a Crafty.rc file is necessary. If you do not have one or are not certain how to make one, I think Bryan can give you a link where to download it (or I will send you mine).
Bryan might feel free to give as two Crafty.rc files for those who can use 48 and for those who can give 96 MB. Please remember then Bryan that we voted by majority that pawnhash should be included in the total amount of hash allowed and cannot be given additonally.
To Olivier: are we already proceeding concerning this Axon-Benchmark stuff?
Best Regards
Heinz

I am basing the .rc file from the following rule that was posted.
---Hash for those engines that support the scheme 32...64 should be 64 MB or 128 MB for those who can give that (Igor can only give 38 max.)
...and for those engines who support 24...48...96 it should be 48 MB or 96 MB for those testers who can give it.
From the above it appears to me that the max total memory has been set to 128MB. With this in mind here are 2 recommended .rc file settings for the different memory settings of max 64 & 128 (Note that the comments in () need to be removed.
.rc for max 128
hash = 96
hash = 24
cache = 8
egtb
xboard (This is needed if running under Arena, if under Winboard remove this)
computer
ponder off
exit
.rc for max 64
hash = 48
hash = 12
cache = 8
egtb
xboard (This is needed if running under Arena, if under Winboard remove this)
computer
ponder off
exit
Specifing the EGTB path should be done on the command line and not in the .rc file. For those using Arena the following needs to be added to the command line options in the engine management tbpath=c:. Those using winboard should use the FCP/SCP command line of /FCP "c:crafty tbpath=c:".
On another note, since these matches are long time control a max memory of 38M (hash 24 hashp 12) is no where near enough memory for a program like Crafty to run effectively as the hash becomes saturated quickly.

[Igor Gorelikov]

Geschrieben von:/Posted by: Igor Gorelikov at 22 July 2004 14:54:29:
Als Antwort auf:/In reply to: Re: Crafty.rc geschrieben von:/posted by: Heinz van Kempen at 22 July 2004 14:35:37:

Bryan might feel free to give as two Crafty.rc files for those who can use 48 and for those who can give 96 MB.

If you set Crafty to 48:
hash=48M
hashp=1M
that means aclual allocation of 55 MB which is too big (in engine to engine matches) for my 256MB of RAM.
Igor

Hello Igor,
we already agreed to make an exception for you and we know that you can only give 38 MB max. for each engine.
Best Regards
Heinz

OK, here is my crafty.rc for 36.5Mb allocation:
hash=24M
hashp=11M
egtb
tbpath d:\TBs
resign 6
exit

[Igor Gorelikov]

Geschrieben von:/Posted by: Igor Gorelikov at 22 July 2004 15:55:51:
Als Antwort auf:/In reply to: Re: Crafty.rc geschrieben von:/posted by: Igor Gorelikov at 22 July 2004 14:54:29:

Corrected rc for 38Mb:
hash=24M
hashp=8M
cache=4
egtb
tbpath d:\TBs
resign 6
exit

[Igor Gorelikov]

Geschrieben von:/Posted by: Igor Gorelikov at 22 July 2004 15:57:45:
Als Antwort auf:/In reply to: Re: AEGT---Crafty.rc geschrieben von:/posted by: Bryan Hofmann at 22 July 2004 14:43:36:

.rc for max 128
hash = 96
hash = 24
cache = 8
egtb
xboard (This is needed if running under Arena, if under Winboard remove this)
computer
ponder off
exit
.rc for max 64
hash = 48
hash = 12
cache = 8
egtb
xboard (This is needed if running under Arena, if under Winboard remove this)
computer
ponder off
exit

do you mean hashp instead of the second hash?
Igor

[Bryan Hofmann]

Geschrieben von:/Posted by: Bryan Hofmann at 22 July 2004 16:04:12:
Als Antwort auf:/In reply to: Re: AEGT---Crafty.rc geschrieben von:/posted by: Bryan Hofmann at 22 July 2004 14:43:36:

Hi all testers ,
Slobodan is out of our list of testers (my decision). Seems that he anyway never was sure if to join and he did not participate in votings and other stuff. There seemed to be some other doubtful things, so we continue with those we have and I am sure we will get more help when we have started and people will see that it is a good tournament.
I am almost certain that you all know that for running Crafty decently a Crafty.rc file is necessary. If you do not have one or are not certain how to make one, I think Bryan can give you a link where to download it (or I will send you mine).
Bryan might feel free to give as two Crafty.rc files for those who can use 48 and for those who can give 96 MB. Please remember then Bryan that we voted by majority that pawnhash should be included in the total amount of hash allowed and cannot be given additonally.
To Olivier: are we already proceeding concerning this Axon-Benchmark stuff?
Best Regards
Heinz

I am basing the .rc file from the following rule that was posted.
---Hash for those engines that support the scheme 32...64 should be 64 MB or 128 MB for those who can give that (Igor can only give 38 max.)
...and for those engines who support 24...48...96 it should be 48 MB or 96 MB for those testers who can give it.
From the above it appears to me that the max total memory has been set to 128MB. With this in mind here are 2 recommended .rc file settings for the different memory settings of max 64 & 128 (Note that the comments in () need to be removed.
.rc for max 128
hash = 96
hash = 24
cache = 8
egtb
xboard (This is needed if running under Arena, if under Winboard remove this)
computer
ponder off
exit
.rc for max 64
hash = 48
hash = 12
cache = 8
egtb
xboard (This is needed if running under Arena, if under Winboard remove this)
computer
ponder off
exit
Specifing the EGTB path should be done on the command line and not in the .rc file. For those using Arena the following needs to be added to the command line options in the engine management tbpath=c:. Those using winboard should use the FCP/SCP command line of /FCP "c:crafty tbpath=c:".
On another note, since these matches are long time control a max memory of 38M (hash 24 hashp 12) is no where near enough memory for a program like Crafty to run effectively as the hash becomes saturated quickly.

Rather bad mistake on the above .rc files, I forgot to place a M after the hash settings;
.rc for max 128
hash = 96M
hashp = 24M
cache = 8M
egtb
xboard (This is needed if running under Arena, if under Winboard remove this)
computer
ponder off
exit
.rc for max 64
hash = 48M
hashp = 12M
cache = 8M
egtb
xboard (This is needed if running under Arena, if under Winboard remove this)
computer
ponder off
exit

[Bryan Hofmann]

Geschrieben von:/Posted by: Bryan Hofmann at 22 July 2004 16:05:40:
Als Antwort auf:/In reply to: do you mean hashp instead of the second hash? geschrieben von:/posted by: Igor Gorelikov at 22 July 2004 15:57:45:

.rc for max 128
hash = 96
hash = 24
cache = 8
egtb
xboard (This is needed if running under Arena, if under Winboard remove this)
computer
ponder off
exit
.rc for max 64
hash = 48
hash = 12
cache = 8
egtb
xboard (This is needed if running under Arena, if under Winboard remove this)
computer
ponder off
exit

do you mean hashp instead of the second hash?
Igor

Yes, posted before a full cup of coffee......

[Bryan Hofmann]

Geschrieben von:/Posted by: Bryan Hofmann at 22 July 2004 16:07:12:
Als Antwort auf:/In reply to: correction geschrieben von:/posted by: Igor Gorelikov at 22 July 2004 15:55:51:

Corrected rc for 38Mb:
hash=24M
hashp=8M
cache=4
egtb
tbpath d:\TBs
resign 6
exit

This is not a valid hashp setting, it results in 6M of hashp.

[Igor Gorelikov]

Geschrieben von:/Posted by: Igor Gorelikov at 22 July 2004 16:15:24:
Als Antwort auf:/In reply to: Re: correction geschrieben von:/posted by: Bryan Hofmann at 22 July 2004 16:07:12:

Corrected rc for 38Mb:
hash=24M
hashp=8M
cache=4
egtb
tbpath d:\TBs
resign 6
exit

This is not a valid hashp setting, it results in 6M of hashp.

OK, then
hashp=6M

[Christian Koch]

Geschrieben von:/Posted by: Christian Koch at 22 July 2004 17:05:09:
Als Antwort auf:/In reply to: Re: AEGT---Crafty.rc geschrieben von:/posted by: Bryan Hofmann at 22 July 2004 16:04:12:

Here is my CRAFTY-RC
hash 96M
hashp 24M
cache 8M
ponder off
log on
tbpath=c:\chess\tb5
egtb
The EGTB's work very fine under Arena without putting them in the command-line. Therefore it makes no sense for me.
Try it with this FEN:
8/8/2p1p1k1/2r1p2p/RRP2rP1/7P/4K3/8 w - - 0 44

------------------------------------------------------------------------------
5 piece tablebase files found
13980kb of RAM used for TB indices and decompression tables
Crafty v19.15
White(1): xboard
White(1): protover 2
feature ping=1 setboard=1 san=1 time=1 draw=1
feature sigint=0 sigterm=0 reuse=1 analyze=1
feature myname="Crafty-19.15" name=1
feature playother=1 colors=0
feature variants="normal,nocastle"
feature done=1
White(1): accepted ping
White(1): accepted setboard
White(1): accepted san
White(1): accepted time
White(1): accepted draw
White(1): accepted sigint
White(1): accepted sigterm
White(1): accepted reuse
White(1): accepted analyze
White(1): accepted myname
White(1): accepted name
White(1): accepted playother
White(1): accepted colors
White(1): accepted variants
White(1): accepted done
White(1): new
White(1): random
White(1): level 40 10 0
40 moves/10 minutes primary time control
40 moves/10 minutes secondary time control
White(1): post
White(1): hard
pondering enabled.
White(1): easy
pondering disabled.
White(1): ping 1
pong 1
White(1): name WB Crfaty 19.15
Crafty 19.15 vs WB
White(1): level 40 10 0
40 moves/10 minutes primary time control
40 moves/10 minutes secondary time control
White(1): new
White(1): random
White(1): level 40 10 0
40 moves/10 minutes primary time control
40 moves/10 minutes secondary time control
White(1): post
White(1): hard
pondering enabled.
White(1): easy
pondering disabled.
White(1): ping 2
pong 2
White(1): name Christian Koch
Crafty 19.15 vs Christian
White(1): level 40 10 0
40 moves/10 minutes primary time control
40 moves/10 minutes secondary time control
White(1): new
White(1): random
White(1): level 40 10 0
40 moves/10 minutes primary time control
40 moves/10 minutes secondary time control
White(1): post
White(1): hard
pondering enabled.
White(1): easy
pondering disabled.
White(1): ping 3
pong 3
White(1): name Christian Koch
Crafty 19.15 vs Christian
White(1): new
White(1): random
White(1): level 40 10 0
40 moves/10 minutes primary time control
40 moves/10 minutes secondary time control
White(1): post
White(1): hard
pondering enabled.
White(1): easy
pondering disabled.
White(1): ping 4
pong 4
White(1): name Christian Koch
Crafty 19.15 vs Christian
White(1): force
White(1): setboard 8/8/2p1p1k1/2r1p2p/RRP2rP1/7P/4K3/8 w - - 0 44
+---+---+---+---+---+---+---+---+
8 | | . | | . | | . | | . |
+---+---+---+---+---+---+---+---+
7 | . | | . | | . | | . | |
+---+---+---+---+---+---+---+---+
6 | | . |
| . |
| . || . |
+---+---+---+---+---+---+---+---+
5 | . | || |
| | . |
|
+---+---+---+---+---+---+---+---+
4 |-R-|-R-|-P-| . | ||-P-| . |
+---+---+---+---+---+---+---+---+
3 | . | | . | | . | | . |-P-|
+---+---+---+---+---+---+---+---+
2 | | . | | . |-K-| . | | . |
+---+---+---+---+---+---+---+---+
1 | . | | . | | . | | . | |
+---+---+---+---+---+---+---+---+
a b c d e f g h
White(1): time 120000
time remaining: 20:00 (crafty).
White(1): otim 120000
time remaining: 20:00 (opponent).
White(1): go
clearing hash tables
time surplus 10:00 time limit 22.50 (3:30)
depth time score variation (1)
7 0.12 -2.12 1. gxh5+ Kxh5 2. Rb8 Rcxc4 3. Rxc4
Rxc4 4. Re8 Re4+ 5. Kd3 Rd4+ 6. Ke3
Rd6
7-> 0.14 -2.12 1. gxh5+ Kxh5 2. Rb8 Rcxc4 3. Rxc4
Rxc4 4. Re8 Re4+ 5. Kd3 Rd4+ 6. Ke3
Rd6
8 0.22 -1.94 1. gxh5+ Kxh5 2. Rb8 Rcxc4 3. Rxc4
Rxc4 4. Re8 Re4+ 5. Kd3 Rd4+ 6. Ke3
Rh4 7. Rxe6 Rxh3+
8-> 0.25 -1.94 1. gxh5+ Kxh5 2. Rb8 Rcxc4 3. Rxc4
Rxc4 4. Re8 Re4+ 5. Kd3 Rd4+ 6. Ke3
Rh4 7. Rxe6 Rxh3+
9 0.34 -1.64 1. gxh5+ Kxh5 2. Rb8 Rcxc4 3. Rxc4
Rxc4 4. Re8 Kg5 5. Rxe6 Kf5 6. Rd6
Kf4
9-> 0.39 -1.64 1. gxh5+ Kxh5 2. Rb8 Rcxc4 3. Rxc4
Rxc4 4. Re8 Kg5 5. Rxe6 Kf5 6. Rd6
Kf4
10 0.58 -1.64 1. gxh5+ Kxh5 2. Rb8 Rcxc4 3. Rxc4
Rxc4 4. Re8 Kg6 5. Rxe6+ Kf5 6. Rd6
Kf4 7. Kd3 Rc1
10-> 0.70 -1.64 1. gxh5+ Kxh5 2. Rb8 Rcxc4 3. Rxc4
Rxc4 4. Re8 Kg6 5. Rxe6+ Kf5 6. Rd6
Kf4 7. Kd3 Rc1
11 1.12 -1.69 1. gxh5+ Kxh5 2. Rb8 Rcxc4 3. Rxc4
Rxc4 4. Re8 Kg6 5. Rxe6+ Kf5 6. Rd6
Ke4 7. h4 Rc2+ 8. Kd1 Rc3
11-> 1.38 -1.69 1. gxh5+ Kxh5 2. Rb8 Rcxc4 3. Rxc4
Rxc4 4. Re8 Kg6 5. Rxe6+ Kf5 6. Rd6
Ke4 7. h4 Rc2+ 8. Kd1 Rc3
12 6.61 -1.83 1. gxh5+ Kxh5 2. Ra8 Rcxc4 3. Rxc4
Rxc4 4. Ra5 c5 5. Kd3 Rd4+ 6. Ke3 Rd5
7. Ra6 Kh4 8. Ke4 Kxh3 9. Rxe6
12-> 7.24 -1.83 1. gxh5+ Kxh5 2. Ra8 Rcxc4 3. Rxc4
Rxc4 4. Ra5 c5 5. Kd3 Rd4+ 6. Ke3 Rd5
7. Ra6 Kh4 8. Ke4 Kxh3 9. Rxe6
13 10.40 -1.83 1. gxh5+ Kxh5 2. Ra8 Rcxc4 3. Rxc4
Rxc4 4. Ra5 c5 5. Kd3 Rd4+ 6. Ke3 Rd5
7. Ra6 Kh4 8. Ke4 Kxh3 9. Rxe6
13-> 11.55 -1.83 1. gxh5+ Kxh5 2. Ra8 Rcxc4 3. Rxc4
Rxc4 4. Ra5 c5 5. Kd3 Rd4+ 6. Ke3 Rd5
7. Ra6 Kh4 8. Ke4 Kxh3 9. Rxe6
14 19.94 -1.83 1. gxh5+ Kxh5 2. Ra8 Rcxc4 3. Rxc4
Rxc4 4. Ra5 Kg5 5. Rxe5+ Kf6 6. Rh5
Rd4 7. Rc5 Rd6 8. Ke3 e5 9. Ke4 Rd4+
10. Ke3
time=22.53 cpu=95% mat=-1 n=28921553 fh=91% nps=1.28M
ext-> chk=1578825 cap=85377 pp=10100 1rep=44869 mate=7059
predicted=0 nodes=28921553 evals=13642204 50move=0
endgame tablebase-> probes=485 hits=485
White(1): gxh5+
time used: 22.53
learning position, wtm=1 value=-183
+---+---+---+---+---+---+---+---+
8 | | . | | . | | . | | . |
+---+---+---+---+---+---+---+---+
7 | . | | . | | . | | . | |
+---+---+---+---+---+---+---+---+
6 | | . |
| . |
| . || . |
+---+---+---+---+---+---+---+---+
5 | . | || |
| | . |-P-|
+---+---+---+---+---+---+---+---+
4 |-R-|-R-|-P-| . | || | . |
+---+---+---+---+---+---+---+---+
3 | . | | . | | . | | . |-P-|
+---+---+---+---+---+---+---+---+
2 | | . | | . |-K-| . | | . |
+---+---+---+---+---+---+---+---+
1 | . | | . | | . | | . | |
+---+---+---+---+---+---+---+---+
a b c d e f g h
Black(1): ?
Black(1): force
Black(1): quit
LearnBook() executed

Hi all testers ,
Slobodan is out of our list of testers (my decision). Seems that he anyway never was sure if to join and he did not participate in votings and other stuff. There seemed to be some other doubtful things, so we continue with those we have and I am sure we will get more help when we have started and people will see that it is a good tournament.
I am almost certain that you all know that for running Crafty decently a Crafty.rc file is necessary. If you do not have one or are not certain how to make one, I think Bryan can give you a link where to download it (or I will send you mine).
Bryan might feel free to give as two Crafty.rc files for those who can use 48 and for those who can give 96 MB. Please remember then Bryan that we voted by majority that pawnhash should be included in the total amount of hash allowed and cannot be given additonally.
To Olivier: are we already proceeding concerning this Axon-Benchmark stuff?
Best Regards
Heinz

I am basing the .rc file from the following rule that was posted.
---Hash for those engines that support the scheme 32...64 should be 64 MB or 128 MB for those who can give that (Igor can only give 38 max.)
...and for those engines who support 24...48...96 it should be 48 MB or 96 MB for those testers who can give it.
From the above it appears to me that the max total memory has been set to 128MB. With this in mind here are 2 recommended .rc file settings for the different memory settings of max 64 & 128 (Note that the comments in () need to be removed.
.rc for max 128
hash = 96
hash = 24
cache = 8
egtb
xboard (This is needed if running under Arena, if under Winboard remove this)
computer
ponder off
exit
.rc for max 64
hash = 48
hash = 12
cache = 8
egtb
xboard (This is needed if running under Arena, if under Winboard remove this)
computer
ponder off
exit
Specifing the EGTB path should be done on the command line and not in the .rc file. For those using Arena the following needs to be added to the command line options in the engine management tbpath=c:. Those using winboard should use the FCP/SCP command line of /FCP "c:crafty tbpath=c:".
On another note, since these matches are long time control a max memory of 38M (hash 24 hashp 12) is no where near enough memory for a program like Crafty to run effectively as the hash becomes saturated quickly.

Rather bad mistake on the above .rc files, I forgot to place a M after the hash settings;
.rc for max 128
hash = 96M
hashp = 24M
cache = 8M
egtb
xboard (This is needed if running under Arena, if under Winboard remove this)
computer
ponder off
exit
.rc for max 64
hash = 48M
hashp = 12M
cache = 8M
egtb
xboard (This is needed if running under Arena, if under Winboard remove this)
computer
ponder off
exit

[Igor Gorelikov]

Geschrieben von:/Posted by: Igor Gorelikov at 22 July 2004 17:21:02:
Als Antwort auf:/In reply to: Re: AEGT---Crafty.rc geschrieben von:/posted by: Christian Koch at 22 July 2004 17:05:09:

Here is my CRAFTY-RC
hash 96M
hashp 24M
cache 8M
ponder off
log on
tbpath=c:\chess\tb5
egtb

How many RAM you have on your AMD XP2000+ ?
Igor

[Christian Koch]

Geschrieben von:/Posted by: Christian Koch at 22 July 2004 17:23:32:
Als Antwort auf:/In reply to: Re: AEGT---Crafty.rc geschrieben von:/posted by: Igor Gorelikov at 22 July 2004 17:21:02:

My desktop comp DIED today, perhaps Motherboard or power suplly. So I have to play on my notebook (Centrino 1.6, 1 GB RAM)

Here is my CRAFTY-RC
hash 96M
hashp 24M
cache 8M
ponder off
log on
tbpath=c:\chess\tb5
egtb

How many RAM you have on your AMD XP2000+ ?
Igor