Author: Thorsten Czub
Date: 04:43:45 04/02/02
the version 276h was really strong.
here the original text stuff with games of from the past:
The story so far, in a few words:
Mr. Moritz Berger, a friendly force of the ChessBase-Beta-Tester-team, came to
me with a bet.
He wanted to offer me a six-pack of beer when Chess System Tal would win in a
tournament of 10 games (40/120) on his machines against his Fritz5.01 engine.
Because ChessSystemTal ranked in Paris (for some unknown reasons) a little
higher than Fritz5, we have come into a discussion before this bet, how strong
each program is.
This tournament should, at leas, show some more evidence.
Mr.Moritz Berger gave CSTal his p5/166 Mhz machine and used his p5/133 machine
for Fritz5.01. He used 128 MB RAM to get at least 58 MB Hash-Table size under
Fritz5.
The first games were 2 draws, nicely played by CSTal (I refer to the second
game). This was the beginning of our bed. From the very first beginning I
claimed that I see a bait in this bet.
Shall I take the lure and eat it ?
I said ok. So Mr.Moritz Berger continued his official match with the before
downloaded "Paris"-version 284a (from the Oxford server) and after 10 rounds the
result was a clearly 9 from 10 for Fritz.
You may understand that I cannot let this result stay in a context with my
beloved program without any comments or explanations to what has happened from
my point of view, about the weather in general, and deseases - especially
deseases of human behaviour emulating artificial intelligence software before an
important tournament in specific.
Please take the text that follows with a SMILEY. It is meant with many :-)
arround !
I followed the data that was produced by my friend ( I have to make a little
pause behind this word unless I have seen his data... ) with great care.
I replayed the game one by one and was shocked.
This was NOT the engine I have played in Paris.
It looked like a very weak program.
In Paris some programs had the "influenca espagnol", maybe a similar kind of
influenca has attacked Chess System Tal in Moritz'es wholy testing-lab ?!
After a few days CSTal lost almost any game, game after game, without defense.
Is the program wrongly styled. Is the guy a fool ?
Are his machines from east-germany ?
Something has to be done.
This cannot continue anymore.
So I began my own answer to this shameful tournament (in the end Moritz even
offered me: we don't have to publish it, Thorsten. Nobody is helped by this
disaster. maybe we should put Dieter Steinwender out of the mailing list, I
mean... --- if this helps you...) meanwhile Moritz gave his best to hide his
malicious glee behind some very social and friendly words. Pah !
Papperlapapp.
I used the same version I played in Paris and - because I don't have this
"secret" engine Fritz5.01 I used my commercial Fritz5.
I have to explain this here in a more precise tone to be as scientific as I can
be (after a currywurst and some Schnitzel this is always very difficult,
especially when the Schnitzel was a Zwiebelschnitzel):
All games were played exactly from the same opening-positions, Mr. Moritz Berger
claims to have played with CSTal (version 284a) against (his special fritz
engine) Fritz5.01 his "official match" where Fritz5.01 should have won by 9 from
10 (2 draws in the first 2 games).
I doubt that Mr.Moritz Berger's results reflect in any way the strength of CSTal
related to Fritz5.
From my point of view Mr.Moritz Berger has used a broken version (by will or
without any willingly target) to show that a broken version plays broken chess
against an average opponent program.
This reminds me on a similar situation I have read once in the past (1990?):
I remember the austrian computerchess-magazin MODUL once tested a dedicated
machine: SUPER-FORTE C and came to the result:
1.the Super-Forte C ("we bought in germany to a very cheap but nice-price") has
strange behaviour in playing chess.
2.the power-supply is not original, it is from a Go-Mo-Ku computer.
3.there is no handbook.
4.there is no piece-set.
Later in this test-review Mr.Mader claims he has the suspicion that the machine
had also a hardware-problem and fails technically.
In other words:
The machine was broken ! Were were unable to test it. The service was mean/bad !
Not a good light onto this machine and this company !
How can this happen ? Now - if you study the article less shallow , you see a
statment in the beginning that the Super-Forte C was so expensive in austria and
not reachable that the team had to watch out for a different source. But there
was a very cheap distributor in germany (bavaria). It was said that the NOVAG
"evaluation-machine" was ordered by MEPHISTO related distributor OSSI WEINER.
From my point of view, a Novag machine that leaves the house of Ossi Weiner,
cannot behave different than fail in an important NOVAG test-review. Ossi Weiner
could not have better organised it.
So : No wonder it was broken. No wonder there were no pieces and no handbook. No
wonder the adaptor was from a Go-Mo-Ku system. It is a wonder the machine did
not explode at all.
Here the situation is nearly similar.
Mr.Moritz Berger wanted to do a bet with me, and would pay me a six-pack of beer
or 3 currywurst if CSTal would only get 50 %, he said.
So the case is clear: from his point of view, CSTal had to lose, otherwise he
would not only lose his bet, but also his reputation known as Mr.Moritz
Massive-Chess-Hash-Base-Table-Berger.
Mr.Moritz Berger has the version 276h and knows about how to set the style (deep
search and capture-search to manic) and could reproduce the games IF he would
have these fast machines. Also any other guy in the world would be able to
reproduce these games with having a commercial Fritz5 and the version 276h I
used in Paris.
Due to a bug in the PGN-handling of ChessBase-product Fritz5 (:-) it is for
unknown reasons NOT saving the time, eval and search depth of subvariations !
All games were played 40/120 (later games 60 in 180 to give Fritz better
time-algorithm as Mr.Moritz Berger explained to me) on K6/200 Mhz.
[Event "Level=120'/40+120'/40+120'. "]
[Site "Luenen"]
[Date "1997.11.16"]
[Round "1"]
[White "CST 276h"]
[Black "Fritz 5.00"]
[Result "1/2-1/2"]
[ECO "B06"]
[PlyCount "70"]
[EventDate "1997.11.16"]
{8192kB, forced opening-line} 1. e4 {0} 1... g6 {0} 2. d4 {3} 2... c6
{0} 3. Nc3 {3}
3... d6 {0} 4. Be3 {3} 4... Bg7 {0} 5. Qd2 {2} 5... Qa5 {0} 6. Bd3 {2}
6... Nf6
{0} 7. f3 {177} 7... Nbd7 {0} 8. Nge2 {344} 8... e5 {0} 9. O-O {215}
9... O-O {
0} 10. Nd5 Qxd2 11. Ne7+ Kh8 12. Bxd2 exd4 13. g4 Nc5 14. Bf4 d5
15. exd5 Nxd3 16. cxd3 Nxd5 17. Nxd5 cxd5 18. Bd6 Re8 19. Nf4 Bd7 20.
Nxd5 Rad8
21. Bf4 Re2 22. Rf2 Rde8 23. Kf1 Rxf2+ 24. Kxf2 Bc6 25. Nb4 Bb5 26. a4
Bd7 27.
a5 Bb5 28. Re1 Rxe1 29. Kxe1 h5 30. gxh5 gxh5 31. Bb8 a6 32. Ke2 Ba4 33.
Nd5 h4
34. Bc7 Kh7 35. Ne7 Bd7 {adjud.by czub} 1/2-1/2
[Event "Level=120'/40+120'/40+120'. "]
[Site "Luenen"]
[Date "1997.11.17"]
[Round "2"]
[White "Fritz 5.00"]
[Black "CST 276h"]
[Result "1/2-1/2"]
[ECO "C05"]
[PlyCount "81"]
[EventDate "1997.11.17"]
{8192kB, forced opening-line} 1. e4 {0} 1... e6 {3} 2. d4 {0} 2... d5
{3} 3. Nd2 {0}
3... Nf6 {3} 4. e5 {0} 4... Nfd7 {3} 5. f4 {0} 5... c5 {3} 6. c3 {0}
6... Nc6 {
3} 7. Ndf3 {0} 7... Qb6 {3} 8. g3 {0} 8... cxd4 {3} 9. cxd4 {0} 9...
Bb4+ {3}
10. Kf2 {0} 10... g5 {3} 11. h3 {0} 11... gxf4 {3} 12. gxf4 {0} 12...
Rg8 {3}
13. Be3 {0} 13... Be7 {3} 14. Qc2 {0} 14... Nf8 {3} 15. a3 {0} 15... Ng6
{3}
16. h4 {0} 16... Nxf4 {3} 17. Rd1 {0} 17... Ng6 {3} 18. h5 {0} 18... Nf8
{3}
19. Bd3 {0} 19... Bd7 {3} 20. b4 {0} 20... f5 {3} 21. exf6 {0} 21...
Bxf6 {3}
22. Bxh7 {-0.16/11 340} 22... Nxh7 {207} 23. Qxh7 {0.38/10 49} 23... Ne7
{383}
24. Ne5 {0.41/12 0} 24... Rg7 {239} 25. Qh6 {0.19/11 185} 25... Bxe5
{295} 26.
dxe5 {0.00/11 0} 26... Rf7+ {241} 27. Nf3 {-0.28/12 231} 27... Qa6 {169}
28.
Bc5 {-0.47/11 321} 28... Qxa3 {196} 29. Rh3 {-0.63/11 202} 29... Rc8
{161} 30.
Rg3 {-0.59/10 371} 30... Rxc5 {165} 31. bxc5 {-0.22/11 115} 31... Qxc5+
{383}
32. Kg2 {-0.19/12 92} 32... Qc2+ 33. Qd2 Qf5 34. Qb4 b6 35.
Qd6 Qc2+ 36. Rd2 Qe4 37. Ra2 Rf5 38. Rxa7 Qe2+ 39. Kg1 Qe3+ 40. Kg2 Qe2+
41.
Kg1 1/2-1/2
[Event "40 / 120"]
[Site "40 / 120"]
[Date "1997.11.17"]
[Round "3"]
[White "CST 276h"]
[Black "Fritz5"]
[Result "1/2-1/2"]
[ECO "B07"]
[PlyCount "104"]
1. e4 d6 2. d4 g6 3. Nc3 c6 4. Be3 Bg7 5. Qd2 Nf6 6. f3 O-O 7. O-O-O b5 8. Bh6
Be6 9. Bxg7 Kxg7 10. d5 cxd5 11. exd5 b4 12. dxe6 bxc3 13. Qxc3 fxe6 14.
Nh3 Qc8 15. Qb3 Nc6 16. Ng5 d5 17. Qe3 e5 18. Bb5 Nb4 19. Ba4 h6 20. Qxe5 Nxa2+
21. Kb1 Nc3+ 22. bxc3 Qb8+ 23. Qxb8 Raxb8+ 24. Kc1 hxg5 25. Rhe1 Kf7 26. Bc6
Rfd8 27. Re5 e6 28. c4 dxc4 29. Rxd8 Rxd8 30. Ra5 a6 31. Rxa6 c3 32. Ra7+ Kf8
33. Ra5 Rc8 34. Bb5 Rc5 35. Ra8+ Ke7 36. Ra7+ Kd6 37. Ra6+ Ke7 38. Bd3 Re5 39.
Kd1 Nd5 40. Bxg6 Nf4 41. Be4 Rb5 42. Ke1 Rb1+ 43. Kf2 Rd1 44. g3 Rd2+ 45. Kg1
Nh3+ 46. Kf1 Rf2+ 47. Ke1 Rxh2 48. Rc6 g4 49. Rxc3 Ng5 50. Bb7 gxf3 51. Bxf3
Nxf3+ 52. Rxf3 Rxc2 1/2-1/2
[Event "Level=180'/60+180'/60+180'. "]
[Site "Luenen"]
[Date "1997.11.18"]
[Round "4"]
[White "Fritz 5.00"]
[Black "CST 276h"]
[Result "0-1"]
[ECO "A22"]
[PlyCount "107"]
[EventDate "1997.11.18"]
{8192KB, forced opening-line} 1. c4 {0} 1... Nf6 {3} 2. Nc3 {0} 2... e5 {3} 3.
g3 {0}
3... Bb4 {3} 4. Bg2 {0} 4... O-O {3} 5. d3 {0} 5... Re8 {3} 6. Bd2 {0} 6... Nc6
{3} 7. Nf3 {0} 7... Bxc3 {3} 8. Bxc3 {0} 8... d5 {3} 9. cxd5 {0} 9... Nxd5 {3}
10. O-O {0} 10... b6 {3} 11. Qa4 {-0.06/10 124} 11... Bd7 12.
Qc4 Nxc3 13. Qxc3 Rc8 14. e3 Qf6 15. Nd2 a5 16. f4 Qh6 17. Nc4 exf4 18. Rxf4 b5
19. Bd5 bxc4 20. Bxf7+ Kh8 21. Bxe8 Rxe8 22. Raf1 Qd6 23. dxc4 Bh3 24. c5 Qd8
25. Re1 Ne5 26. Rf2 h6 27. e4 Qg5 28. Rd1 Qg4 29. Re1 Qh5 30. Ra1 Rd8 31. Qc2
Rd3 32. Rf8+ Kh7 33. Rf2 Re3 34. Qc1 Qg5 35. Qc2 Nd3 36. Rd1 Qe5 37. Qxd3 Rxd3
38. Rxd3 Qxe4 39. Rd1 Qc4 40. Rfd2 Bg4 41. Ra1 Qxc5+ 42. Rf2 Bf3 43. Re1 Bd5
44. a3 a4 45. h3 Kg6 46. Ref1 Bb7 47. Re1 Qb6 48. Ref1 0-1
It does not look like Fritz5 is the superior program.
Also my results, although I used the same opening-positions, do in no way fit to
Mr.Bergers opposite results.
I can only believe in a minor advertising clue.
Of course I will further investigate negotiations with the programmer Chris
Whittington to rebuild or upload version 276h onto the Oxford-Server so that all
of you who are legal users could come into the pleasure to download it for free.
I want to subscribe to the point Mr.Matthias Wuellenweber mentioned in one of
his posts in the mad-cow-desease-infected computer-tschuess'n-newsgroup, he
wrote as an asnwer the following:
Mclane:
>> I guess I know what has happened: Moritz tested a version chris released
AS THE PARIS version although it was NOT the paris version. But the
versions >276h are broken. Therefore Moritz tested a broken version.<<
Mr.Matthias Wuellenweber quoted therefore:
"I have never won a chess game against a healthy opponent."
Alexander Alekhine.
"I have to admit, we are equals."
Friedrich Koehnlein after losing 0:6 in a private match against Adolf
Anderssen.
Thorsten, my goal, if any, in this match is that you (in private) say
something like:
"There is an unexpected and strange possibility that Fritz5 by whatever
crooked proletarian materialistic programming tricks could achieve
practical results not much worse than those of CSTal, the saviour of
idealistic mankind and bearer of the ethernal cosmic light".
Why would I want you to say that? Because I enjoy that there are different
approaches in chess programming. It makes work more interesting. Imagine
all you could do would be autoplaying CSTal 295s vs. 295t. Why do you and
Chris need the ideological prayer-wheel of the "right" and the "wrong"
approach to chess programming? As said before, the Romans understood that
it makes an interesting match to put up a swordsman in heavy armour against
a light opponent with net and trident. Well, admittedly an evaluation
function of 1100 processor cycles is a bit far on the light side.
Matthias
To these words of Mr.Wuellenweber I can only comment that I would very much like
to DO autoplaying with Fritz5, but due to a forgotten implementation (GRRRRRR!)
the ChessBase team has stopped any wild speculations about the real strength of
Fritz5 in the swedish-rating list by NOT implementing the
WINDOWS-autoplayer-device for Fritz5. Only in some internal versions of Fritz5
the autoplayer works (I guess in preparing for the Paris championship the
ChessBase team tried to use this autoplayer succesfully).
So it is up to you if the public is allowed to measure your product. We would
have nothing against this implementation! (:-)
I answered to him:
Wait until you see the results of MY match. Maybe you won't laugh or be
happy afterwards.
I don't see a reason is Moritz, testing a broken version, however this
makes him very confident that Fritz5 is stronger than CStal.
My games 276h against Fritz5 don't look like Fritz5 could win the match.
This fits into my pattern from Paris.
So wait and see.
He replied:
I am not laughing at you. Lets accept that Fritz5 is weaker than CST. I am
nevertheless interested in the question why you are so unhappy about the
fact that two programming philosophies as radically different as CST and
Fritz can both lead to the capability of playing reasonable chess with
sometimes even comparable style. Why don't you share my view that the
higher the variety of paradigms the more fascinating computer chess
becomes?
Matthias
Matthias is mainly right:
It is not important how the result of a match is. It is an enrichement to the
whole computerchess market when two programs of a different paradigm fight
against each other, or have almost the same playing strength.
I would have nothing against this statement. It was always my point of view that
we need MANY different programs from different programmers and different
companies. It is new to me that CHESS BASE has come to this opinion too. But
nevertheless, when a 180.000 NPS program like Fritz5 plays a 4000 NPS program
like CSTal, two different paradigms fight against each other. I would say it is
to early to subscribe Mr.Bergers point of view, that the fast-search paradigm
has won over the slow-artificial-intelligence approach (how else could one name
all the bugs the programmer Mr.Chris Whittington names FEATURES or INVENTIONS in
his program-versions).
This discussion has not come to an end by now.
In the moment, it looks (and Paris - seen, completely ignoring Cstals
participation, but the rest of the games played there - has given me right) like
the knowledge approach leads.
Shredder won Jakarta championship. Junior Paris. Virtual-chess2, an intelligent
program plays knowledged too, follows second. Shredder2 in a top 3rd position in
paris. Mchess7, as fourth also very near. Rebel and Hiarcs both in the top of
the ssdf-list. This all indicates that the fast-searchers have NO CHANCE at all
in the moment to influence the top-rankings anyway. The massive hardware
increase (up to DEC Alpha 767 Mhz !) that was forced by some fast-search-
companies/programs in Paris has NOT lead to the wished TOP-ranking.
The whole Paris championship was (from the very first beginning to the end)
dominated by knowledged programs. And if Rebel or Hiarcs would have
participated, it would not have developed any different. Maybe even worse for
the fast-searchers.
I will continue to play more games. And as long as Mr.Berger cannot provide any
data that my data is wrong, we have data against data.
Please understand: I don't doubt Mr.Bergers data. I only doubt that it
represents the real playing strength of both programs. There must be a
system-immanent bug in his test-lab. Wether this is in his hardware or came at
least with installing the Whittington software downloaded from Oxford. He will
find out, I am sure.
Results so far:
CStal 2.5/4
1 wins
3 draws
0 losses
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.