Chess System Tal II & Engine Style

[AGove]

Where can we get the patches for Chess System Tal? Thanks!

[mclane]

you have to differenciate between Patch for CSTAL DOS
and patch for CSTAL windows.

The patch for DOS is the PARIS version.

the patch for WIN makes the version from 2.0 to 2.03.

----------
here some data about the patch for the DOS version.
if somebody needs the patch, send me an email.

README - This file

PARISEXE.ZIP - Contains the new CHESS.EXE used in Paris World Championship
- with a few more improvements (larger hash tables, better
- tree searching); and a fix for the draw 'bug' found in
- the Virtual-CSTal Paris WMCCC game
-
- Unzip it using PKUNZIP PARISEXE.ZIP
-
- To install: save the old CHESS.EXE for safekeeping, and copy
- this new CHESS.EXE into your CHESSTAL directory.

CHESS.STY - Was the STYLE used by operator Thorsten Czub in the Paris
- tournament. You can either: rename it to PARIS.STY, copy
- it into the STYLES directory, and select it via the menu;
- or, copy it directly into the DATAFILE directory where it
- will become the default STYLE.

DATAFILE.ZIP - Contains several test suites, including the new COVAX test
-
- Unzip it using PKUNZIP DATAFILE.ZIP; then copy the .POS
- files into the DATAFILE directory

PARIS97.PGN - All the games of the Paris World Championship in PGN format


Good luck,

Chris Whittington

***********************************************************************************************************
This download is for owners of Chess System Tal CD only.
The contents can be added to the CHESSTAL directory on your harddrive.

Tests have revealed a more than usual discrepancy between CSTal playing
white and CSTal playing black. We've created two versions with differing
characteristics to try and overcome this problem.

Testers for computer-computer games should certainly use these versions.
If you are playing human versus Chess System Tal, there is no reason why
you shouldn't just use only the CSTWHITE. It is more speculative than
CSTBLACK; and probably, therefore, leads to more interesting games.
Version number is c289_c


Contents:

README - This file

CSTWHITE.EXE - Is the latest version (December 1997), has a few bug fixes
- and improvements. Use this version for CSTal playing as
- white. Or use it if you have created a special gambit
- opening book. This is the better version if you can
- manage to create a high proportion of wild opening
- positions.

CSTBLACK.EXE - This is a slightly less speculative (and therefore more
- materialistic) version. We found that high speculative
- values in the evaluation help if the program can find
- the sorts of positions that tend to arise from wilder
- openings. Since creating wilder openings is more difficult
- with the black pieces (especially for cptr-cptr games), we
- recommend using this version for computer-computer test
- games where CSTal is playing the black pieces.

USAGE: - To install either version: save the old CHESS.EXE for
- safekeeping, and copy the two CST*.EXE versions into
- your CHESSTAL directory. Rename the one you wish to
- currently use as CHESS.EXE

CHESS.STY - Is the recommended STYLE
- You can either: rename it to NEW.STY, copy
- it into the STYLES directory, and select it via the menu;
- or, copy it directly into the DATAFILE directory where it
- will become the default STYLE.

DATAFILE.ZIP - Contains several test suites, including the new COVAX test
-
- Unzip it using PKUNZIP DATAFILE.ZIP; then copy the .POS
- files into the DATAFILE directory

PARIS97.PGN - All the games of the Paris World Championship in PGN format


SCREEN INFO - On the F5 screen, inner eye, we've added two new informations,
- htal, and ctal. These are a measure of king attack potential
- for each side. ctal is the computers measure, and htal the
- opponent's (human's). A higher value indicates greater king
- attack possibilities.

Good luck,

Chris Whittington

[ArmyBridge]

Hi Thorsten :
Since this topic was started my interest has grown to see CST games, I´ve been looking for.. but unfortunately not found many games, I have SSDF big database and DieCOMPlas update and I get only 200 games where I can get more games?? do you have a database with more games if you have would you share it with me thanks
Regards

[Uri Blass]

thank you.

here a few examples how CSTAL (DOS) played 1995

[Event "Revange for Paderborn 95"]
[Site "?"]
[Date "????.??.??"]
[Round "?"]
[White "XXXX2"]
[Black "CSystem Tal"]
[Result "0-1"]

1. e4 b6 2. d4 Bb7 3. Bd3 e6 4. Ne2 c5 5. c3 d5 6. exd5 Qxd5 7. Nf4 Qd7 8. dxc5
Bxc5 { oh - this is of course a position CSTal likes very much!} 9. O-O Nf6 10. Re1 O-O 11. c4 e5 12. Nh3 Rd8 13. Be2 Qc8 {oops - white gets into trouble....} 14. Nd2 Bxg2 { ! Ha ! played with a score of 3.63 !! and from the first second of the search-process !! Standard Tal-move ....you could see the flickering in Martins' eyes when I told him Tals evaluation-....} 15.
Kxg2 e4 16. Qc2 Qf5 {as discussion started if this sac was accurate. Evaluation of XXXX2 said: no this sac was stupid....} 17. Nf1 Nc6 18. Ng3 Qc8 19. Nxe4 { chess system tal says: this move was bad, no we can hit back... } Nxe4 {cstal evaluation is +1.63} 20. Qxe4 Re8 21. Qd3
Nb4 { many moves have gone since the first sac. Nothing is forced. But this is the problem: a search based program can only defend against something, that can be SEEN in the tree. Here nothing can be seen, therefore XXXX2 is not overprotecting much. + 0,76 evaluation} 22. Qd1 {evaluation +0,79 after this move. Chess System Tal feels well in this position because it knows about the fact that the king is exposed heavily and it is just a few steps to die here, other programs counting material would see it different} Qf5 23. Bf3 Nc2 {+1,12} 24. Rxe8+ Rxe8 25. Bd2 Nxa1 {CSTals eval says +1,73 } 26. Qxa1 Qg6+ {and still the problem is : the king is exposed. Moves and moves have gone, but the main problem that was build with Bxg2 is on the board. CSTal increases evaluation to +2.68 and higher because another defending piece - the queen on a1 - is far away from the main-point of the game} 27. Ng5 {horizont}
h6 {CSTal says +3,09} 28. h4 Qd3 {+4,47 !!} 29. Qc1 hxg5 {+4,42} 30. hxg5 Re5 {found late, before Rd8 was considered with 5.03, now Re5 was played with +5,80} 31. Bh5 {Martin still cannot believe, I guess.} Qe4+ {+8,51 says CSTal.}32. f3 Qh4 {+7,93} 33. f4 Re4 34.
Qh1 Qf2+ {Martin must have nerves like iron CSTal says +10,48 pawns} 35. Kh3 Rd4 36. Qa8+ {no lost game without revange-check! CSTal says +12,62 and plays...} Bf8 37. Bxf7+ Kxf7 38. Qb7+ Be7 {+13,77 pawns up and here Martin has enough... } 0-1

Wasn't this an unbelievable game ???

XXXX2 seems to be correct.
I wonder if chess system tal can beat rybka here

XXXX2 - CSystem Tal, Revange for Paderborn 95
[D]rn1r2k1/p4ppp/1p3n2/2b2q2/2P1p3/7N/PPQNBPKP/R1B1R3 w - - 0 1

Analysis by Rybka 2.3.2a 32-bit :

17.Nd2xe4 Qf5xe4+ 18.Qc2xe4 Nf6xe4 19.Be2-f3 f7-f5 20.Nh3-g5 Bc5-b4
± (1.13) Depth: 8 00:00:00
17.Nd2xe4 Qf5xe4+ 18.Qc2xe4 Nf6xe4 19.Be2-f3 f7-f5 20.Nh3-g5 Bc5-b4 21.Re1-e2
± (1.13) Depth: 9 00:00:00 3kN
17.Nd2xe4 Qf5xe4+ 18.Qc2xe4 Nf6xe4 19.Be2-f3 f7-f5 20.Nh3-g5 Nb8-c6 21.Ng5xe4 f5xe4 22.Bf3xe4 Bc5-b4
± (1.20) Depth: 10 00:00:00 14kN
17.Nd2xe4 Qf5xe4+ 18.Qc2xe4 Nf6xe4 19.Be2-f3 f7-f5 20.Nh3-g5 Nb8-c6 21.Ng5xe4 f5xe4 22.Bf3xe4 Rd8-d6 23.Bc1-f4
± (1.10) Depth: 11 00:00:00 45kN
17.Nd2xe4 Qf5xe4+ 18.Qc2xe4 Nf6xe4 19.Be2-f3 f7-f5 20.Nh3-g5 Nb8-c6 21.Ng5xe4 f5xe4 22.Bf3xe4 Rd8-d6 23.Bc1-f4 Rd6-f6
± (1.18) Depth: 12 00:00:01 92kN
17.Nd2xe4 Qf5xe4+ 18.Qc2xe4 Nf6xe4 19.Be2-f3 f7-f5 20.Nh3-g5 Nb8-c6 21.Ng5xe4 f5xe4 22.Bf3xe4 Rd8-d6 23.Bc1-f4 Rd6-f6
± (1.17) Depth: 13 00:00:03 179kN
17.Nd2xe4 Qf5xe4+ 18.Qc2xe4 Nf6xe4 19.Be2-f3 f7-f5 20.Nh3-g5 Nb8-c6 21.Ng5xe4 f5xe4 22.Bf3xe4 Rd8-d6 23.Bc1-f4 Rd6-f6
± (1.19) Depth: 14 00:00:06 343kN
17.Nd2xe4 Qf5xe4+ 18.Qc2xe4 Nf6xe4 19.Be2-f3 f7-f5 20.Nh3-g5 Nb8-c6 21.Ng5xe4 f5xe4 22.Bf3xe4 Rd8-d6 23.Bc1-f4 Rd6-f6
± (1.25) Depth: 15 00:00:14 704kN
17.Nd2xe4 Qf5xe4+ 18.Qc2xe4 Nf6xe4 19.Be2-f3 f7-f5 20.Nh3-g5 Nb8-c6 21.Ng5xe4 f5xe4 22.Bf3xe4 Rd8-d6 23.Bc1-f4 Rd6-f6
± (1.18) Depth: 16 00:00:28 1472kN
17.Nd2xe4 Qf5xe4+ 18.Qc2xe4 Nf6xe4 19.Be2-f3 f7-f5 20.Nh3-g5 Nb8-c6 21.Ng5xe4 f5xe4 22.Bf3xe4 Ra8-c8 23.Bc1-g5 Rd8-f8
± (1.12) Depth: 17 00:01:03 3677kN
17.Nd2xe4 Qf5xe4+ 18.Qc2xe4 Nf6xe4 19.Be2-f3 f7-f5 20.Nh3-g5 Nb8-c6 21.Ng5xe4 f5xe4 22.Bf3xe4 Ra8-c8 23.Bc1-g5 Rd8-f8
± (1.16) Depth: 18 00:01:52 6496kN
17.Nd2xe4 Qf5xe4+ 18.Qc2xe4 Nf6xe4 19.Be2-f3 f7-f5 20.Nh3-g5 Nb8-c6 21.Ng5xe4 f5xe4 22.Bf3xe4 Ra8-c8 23.Bc1-g5 Rd8-f8
± (1.11) Depth: 19 00:03:28 12470kN
17.Nd2xe4 Qf5xe4+ 18.Qc2xe4 Nf6xe4 19.Be2-f3 f7-f5 20.Nh3-g5 Nb8-c6 21.Ng5xe4 f5xe4 22.Bf3xe4 Ra8-c8 23.Bc1-g5 Rd8-f8
± (1.14) Depth: 20 00:06:43 23303kN

(, 11.07.2007)


XXXX2 - CSystem Tal, Revange for Paderborn 95
[D]r1q1r1k1/p4ppp/1pn5/2b5/2P1Q3/7N/PP2BPKP/R1B1R3 w - - 0 1

Analysis by Rybka 2.3.2a 32-bit :

21.Qe4-f4 Nc6-d4 22.Bc1-e3 Nd4-c2 23.Be2-f3 Nc2xa1 24.Re1xa1 Bc5xe3 25.f2xe3
+- (1.71) Depth: 7 00:00:00
21.Qe4-f4 Nc6-d4 22.Bc1-e3 Nd4-c2 23.Be2-f3 Nc2xa1 24.Re1xa1 Bc5xe3 25.f2xe3
+- (1.71) Depth: 8 00:00:00 3kN
21.Qe4-f4 Nc6-d4 22.Bc1-e3 Nd4-c2 23.Be2-f3 Nc2xa1 24.Re1xa1 Bc5xe3 25.f2xe3
+- (1.72) Depth: 9 00:00:00 12kN
21.Qe4-f4 Nc6-d4 22.Bc1-e3 Nd4-c2 23.Be2-f3 Nc2xa1 24.Re1xa1 Bc5xe3 25.f2xe3 Ra8-b8
+- (1.53) Depth: 10 00:00:00 21kN
21.Qe4-f4 Nc6-d4 22.Bc1-e3 Nd4xe2 23.Re1xe2 Bc5-d6 24.Qf4-h4 Qc8-c6+ 25.f2-f3 Ra8-d8
+- (1.61) Depth: 11 00:00:01 66kN
21.Qe4-f4 Nc6-d4 22.Bc1-e3 Nd4xe2 23.Re1xe2 Bc5-d6 24.Qf4-h4 Qc8-c6+ 25.f2-f3 h7-h6 26.Qh4-g4
+- (1.61) Depth: 12 00:00:03 162kN
21.Qe4-f4 Nc6-d4 22.Bc1-e3 Nd4xe2 23.Re1xe2 Bc5-d6 24.Qf4-h4 h7-h6 25.f2-f3 Re8-e6 26.b2-b3
+- (1.66) Depth: 13 00:00:08 400kN
21.Qe4-f4 Nc6-d4 22.Bc1-e3 Nd4xe2 23.Re1xe2 Bc5-d6 24.Qf4-f3 Qc8xc4 25.Ra1-d1 Bd6-e5 26.b2-b3 Qc4-b5 27.Be3-f4
+- (1.67) Depth: 14 00:00:17 779kN
21.Qe4-f4 Nc6-d4 22.Bc1-e3 Nd4xe2 23.Re1xe2 Bc5-d6 24.Qf4-f3 Qc8xc4 25.Ra1-d1 Ra8-d8 26.Re2-d2 Bd6-e7 27.b2-b3 Qc4-b5
+- (1.76) Depth: 15 00:00:42 1866kN
21.Qe4-f4 Nc6-d4 22.Bc1-e3 Nd4-c2 23.Be2-f3 Bc5xe3 24.f2xe3 h7-h6 25.Bf3xa8 Qc8xa8+ 26.Qf4-f3 Qa8-b8 27.b2-b3 b6-b5
+- (1.85) Depth: 16 00:02:04 6016kN

(, 11.07.2007)

[mclane]

i am not a database fan.
reason is historical.
when we started with dedicated chess computers and later chess programs,
there was no database available. when there WAS a database company available, we did not used it for other historical reason. so my games are not in a database. they are often single games, saved in programs format or PGN.

i do not work in database way, more in little tiny puzzle pieces.

i hate databases and statistics.

when i see those big lists i feel bored and chess is only 1/2 or 1/4 as interesting .

it gives the feeling that the games in the database have no value anymore. they are suddenly reduced to a number. like a telephone book with lots of names, and the people behind the named beeing reduced to their telephone number.

i hope you understand what i want to say.

chess for me is about beauty. beauty is nothing you can store in a database. its something unique !

[mclane]

CSTAL bluffed. it would have no chance against rybka ...
maybe only in specific situations of king attacks where you need to have special knowledge about materiam imbalance. IMO CSTal is better in this .

CSTAL was fishing in the mud when playing those games and saccing.
thats the difference between it and other programs. it plays for fun, not for accuracy.

therefore you are right, the games were not RIGHT. they were unsound.

but in those days it was sometimes succesful.

it would be nice to have an UCI version of the CSTAL engine.
and even better... to have the program developed.

[ArmyBridge]

I understand you Thorsten thanks anyway

[George Tsavdaris]

the patch for WIN makes the version from 2.0 to 2.03.

I also see Chess System Tal 2.50 here:
http://www.rebel.nl/ctresult.htm

It seems it is newer than 2.03(since 2.50>2.03), but i haven't seen you speak about it before. Only about 2.0 and 2.03.
So what is this 2.50?

Also do you know where i can buy CSTal II +2.03 and with book?

[fern]

T:
I asked you the DOS patch, time ago, but it seems you forgot it. Could this time oblige and indulge me?
Thanks
Fernando

[mclane]

the patch for WIN makes the version from 2.0 to 2.03.

I also see Chess System Tal 2.50 here:
http://www.rebel.nl/ctresult.htm

It seems it is newer than 2.03(since 2.50>2.03), but i haven't seen you speak about it before. Only about 2.0 and 2.03.
So what is this 2.50?

Also do you know where i can buy CSTal II +2.03 and with book?

in the work of transferring CSTAL into a new release we changed few
things. nalimov was compressed (CSTAL 2 uses uncompressed bases)
also the STYLE and DATABASE menues that were in CSTAL DOS had been transferred to windows too.
CSTAL got bitboard engines. Whittington rewrote the original engine new in bitboard technic, that made it faster.

the whole GUI was capable to operate many engines, and even had a switcher/chooser who made many brains possible.

this was all before UCI was invented and before stefan meyer kahlen did similar things in Shredder.
therefore the engine format for foreign engines in CSTAL >2
is more winboard standard than UCI.

with version 2.5 we wanted that PROGRAMMERS (not normal customers)
can download the GUI for free e.g. to make their own engines
compatible with our GUI.

this was all before ARENA or other GUIs made it possible.

so version 2.5 was only free for download if you are a programmer.

there were a few versions...
the transfer of the OLD CSTAL2.03 engine into bitboards made progress...
but then chris got the offer from purple software to sell his whole company including CSTAL source code to them.

so although CSTAL was close to the final 3rd release, it never came out.

purple went bancrott.
other people bought the source code.

there have been a few legal "clones" of CSTAL under other names.

as far as i remember there still is offer on ebay from time to time.
i have no overview WHICH version it is.

but it was never the unfinished 3rd release. it was always the CSTAL2.03
under different name or even cut (without big opening book etc.).


if somebody wants patches for CSTAL DOS or CSTAL WIN he can send me an email.
i will then send the patch via email.

the patch will only run if you have a legal version of CSTAL.

concerning the never finished bitboard CSTAL version # >2.03 or >2.5:
its a pity that this project never ended.

the search algorithms of CSTAL 2.03 are IMO today too old to be in competition with the common chess programs.
NOT the evaluation functions.

it was 1999.
today we have 2007. 8 years of programming have changed a lot concerning search algorithms.

one would need to transfer CSTALs search into something that is close to state of the art to make it fight back IMO.

i am still searching for people who would like to continue the project.
IMO it would be worth trying.