Diep homepage
Vincent Diepeveen
Hello,
Checkout my brandnew homepage:
http://www.students.cs.ruu.nl/~vdiepeve/
Vincent Diepeveen
vdie...@cs.ruu.nl
--
+--------------------------------------+
|| email : vdie...@cs.ruu.nl ||
|| Vincent Diepeveen ||
+======================================+
Simon Read
vdie...@cs.ruu.nl (Vincent Diepeveen) wrote:
>Checkout my brandnew homepage:
It seems to me that you're using your university account at Utrecht
as a commercial enterprise, to sell your program. You are lucky; in
England a University wouldn't like anyone using its facilities to
make money.
Simon
Vincent Diepeveen
A lot of commercial programmers use this area to promote their program.
Is that also allowed in England, or can't you read a single msg?
BTW i get a lot of 'making money emails', is that allowed; just mailing
everyone from which you see the email address in ANY newsgroup?
Why do you blame me for some small homepage,
if other people may criticise killerbooks of concurrents without getting
a single comment that their own book is the greatest killerbook
in history (Rebel 8 book, how can rebel get such a high rating if
the engine is NOT playing better chess, it has less knowledge, is tactical
little stronger and plays more passive), and you attack a small guy, who
only sells his program not because i want to make such a big money out
of it, but because there is already NOW a lot of interest in a program
that is by far the strongest correspondence/analysis program.
Diep definitely is not
strong in tournament play. Look at the Swedish rating list, my program
sometimes plays strange moves, but it also finds in most positions the
best move possible, where the strategy of most commercial programs is to
prevent their program from doing bad moves (so they don't try to find the
very best move).
>Simon
Robert Hyatt
Vincent Diepeveen (vdie...@cs.ruu.nl) wrote:
: In <32986...@news.cranfield.ac.uk> Simon Read <strea...@mail.returnmail.com> writes:
:
: >vdie...@cs.ruu.nl (Vincent Diepeveen) wrote:
: >>Checkout my brandnew homepage:
: >
: >It seems to me that you're using your university account at Utrecht
: >as a commercial enterprise, to sell your program. You are lucky; in
: >England a University wouldn't like anyone using its facilities to
: >make money.
:
: A lot of commercial programmers use this area to promote their program.
: Is that also allowed in England, or can't you read a single msg?
:
: BTW i get a lot of 'making money emails', is that allowed; just mailing
: everyone from which you see the email address in ANY newsgroup?
:
: Why do you blame me for some small homepage,
: if other people may criticise killerbooks of concurrents without getting
: a single comment that their own book is the greatest killerbook
: in history (Rebel 8 book, how can rebel get such a high rating if
: the engine is NOT playing better chess, it has less knowledge, is tactical
: little stronger and plays more passive), and you attack a small guy, who
: only sells his program not because i want to make such a big money out
: of it, but because there is already NOW a lot of interest in a program
: that is by far the strongest correspondence/analysis program.
Hmmm.. sounds like we now need a computer correspondence chess tournament,
otherwise the claim you make above makes little sense since it is founded
in speculation rather than "deed".
:
: Diep definitely is not
brucemo
Vincent Diepeveen wrote:
> Why do you blame me for some small homepage,
> if other people may criticise killerbooks of concurrents without getting
> a single comment that their own book is the greatest killerbook
> in history (Rebel 8 book, how can rebel get such a high rating if
> the engine is NOT playing better chess, it has less knowledge, is tactical
> little stronger and plays more passive), and you attack a small guy, who
> only sells his program not because i want to make such a big money out
> of it, but because there is already NOW a lot of interest in a program
> that is by far the strongest correspondence/analysis program.
I would be delighted if you would post evidence regarding how you arrive at
this conclusion about Diep's strength at long time controls.
Claiming to be best in the world "by far" is pretty brash, and should be
substantiated. Is the basis for this claim that you believe that Diep can
find as many tactics as other programs in a few hours (but can't in a few
minutes), and gains more positional strength from each additional ply than
other programs do? How can you be sure this is so? Have you played a long
match with another program?
bruce
Simon Read
vdie...@cs.ruu.nl (Vincent Diepeveen) wrote:
>A lot of commercial programmers use this area to promote their program.
>Is that also allowed in England, or can't you read a single msg?
Yes, I can read messages, thank you very much.
>Why do you blame me for some small homepage,
BLAME? Who said anything about blame? I said you were LUCKY
to have a university policy which is so helpful to you.
>and you attack a small guy,
ATTACK? Who said anything about attack? I am interested that your
university's policy seems to be different to mine.
Simon
\/\/ Don't use the email address in my header. It's robot poison bait. \/\/
Chris Whittington
vdie...@cs.ruu.nl (Vincent Diepeveen) wrote:
>
> In <32986...@news.cranfield.ac.uk> Simon Read <strea...@mail.returnmail.com> writes:
>
> >vdie...@cs.ruu.nl (Vincent Diepeveen) wrote:
> >>Checkout my brandnew homepage:
> >
> >It seems to me that you're using your university account at Utrecht
> >as a commercial enterprise, to sell your program. You are lucky; in
> >England a University wouldn't like anyone using its facilities to
> >make money.
>
> Is that also allowed in England, or can't you read a single msg?
I think the point he is making is that you are using
the *university* computer for your own commercial reasons
and that this would not be allowed in England.
You seem to be responding as if the point he made was that you
were using r.g.c.c.
Get the difference ?
>
> BTW i get a lot of 'making money emails', is that allowed; just mailing
> everyone from which you see the email address in ANY newsgroup?
>
> Why do you blame me for some small homepage,
> if other people may criticise killerbooks of concurrents without getting
> a single comment that their own book is the greatest killerbook
> in history (Rebel 8 book, how can rebel get such a high rating if
> the engine is NOT playing better chess, it has less knowledge, is tactical
> little stronger and plays more passive), and you attack a small guy,
Ah ! So its ok for you to attack anyong you like, but
counterattacks against you (which Simon Read was not doing IMO), are
verboten because 'you are a small guy'.
Cunning, eh ?
> who
> only sells his program not because i want to make such a big money out
> of it, but because there is already NOW a lot of interest in a program
> that is by far the strongest correspondence/analysis program.
Unsound logic. If you want to supply this alleged interest, then
you could give it away (like the other university person, Bob Hyatt,
who, incidentally, uses his *own* computer for his hobby / non-
comnmercial purposes).
Don't believe you, Vincent.
You want your little mittens over all that loverly lolly, don't you ?
Strongest at corespondence ?
Evidence ?
Facts ?
Games ?
Vincent, the day you stop trying to make Diep the best program
by attacking other programs and programmers will be the day you
get taken seriously.
If its good, it'll get results at AEGON and in SSDF and so on.
Attacking me or Ed or whoever doesn't make Diep any stronger.
Chris Whittington
Ed Schröder
From: vdie...@cs.ruu.nl (Vincent Diepeveen)
Subject: Re: Diep homepage
> It seems to me that you're using your university account at Utrecht
> as a commercial enterprise, to sell your program. You are lucky; in
> England a University wouldn't like anyone using its facilities to
> make money.
: A lot of commercial programmers use this area to promote their program.
: Is that also allowed in England, or can't you read a single msg?
Vincent, you missed the point.
Read the above again.
The subject was University versus money.
: Why do you blame me for some small homepage,
: if other people may criticise killerbooks of concurrents without getting
: a single comment that their own book is the greatest killerbook
: in history (Rebel 8 book, how can rebel get such a high rating if
: the engine is NOT playing better chess, it has less knowledge, is
tactical
: little stronger and plays more passive), and you attack a small guy, who
: only sells his program not because i want to make such a big money out
: of it, but because there is already NOW a lot of interest in a program
: that is by far the strongest correspondence/analysis program.
Vincent, if you don't like Rebel8 why don't you return the illegal copy
of Rebel8 you have to the "ilegal" dealer you got Rebel8 from without
paying for it. Please remove Rebel8 from your hard disk.
Misuse of the unlimited installation Rebel8 offers people is forbidden.
- Ed Schroder -
: Vincent Diepeveen
: vdie...@cs.ruu.nl
Vincent Diepeveen
In <57cbl8$v...@juniper.cis.uab.edu> hy...@crafty.cis.uab.edu (Robert Hyatt) writes:
>: Diep is by far the strongest correspondence/analysis program.
>
>Hmmm.. sounds like we now need a computer correspondence chess tournament,
>otherwise the claim you make above makes little sense since it is founded
>in speculation rather than "deed".
I don't claim that it wins games. I claim that if you select a lot of
difficult correspondence positions where there is an excellent move,
either tactical or positional or strategical, taken from both opening,
a lot taken from middlegame and endgame (1:2:1), then allow all program
to search every position many hours. Now write on a paper how much positions
every program solved (take care that the positions are so difficult that
very probably no of the programs will come close to a perfect score,
because scoring 99% or 99.1% out of a test is hard to compare).
Diep will solve most.
I use myself positions taken from Informator, and a lot of correspondence
positions.
DON'T use a standard test. These Positions are hacked into the programs,
or parameters have been set to that they solve them.
Don't use only strategical choices (BK test),
or exchange choices(certain endgame tests), or only
piece winning combinations(win at chess, although at long thinking times
diep solves them all, like all other programs in theorem will do i guess).
Use them all.
Why not allowing an objective correspondence player (with a high elo
preferably, so that he doesn't select crap moves) to select a lot of difficult
correspondence game, where a difficult
BTW how many programs solve next position: Kasparov-Deep Blue game 6,
winning combination: Bxh7. After little more than an hour of
thinking at a PP200-32, Diep (30 mb hashtables) sees this move to
be the winning move.
this game can be found:
http://www.chess.ibm.park.org/deep/blue/games/game6/s8log.B019.html
position after 19...,Nge7?
white: Kg1,Qd1,Rc1,Re3,Bb1,Bb2,Nf3,a2,b3,c4,d4,f2,g3,h4
black: Kg8,Qd7,Rc8,Rd8,Bb4,Nc6,Ne7,a7,b7,d5,e6,f7,g7,h7
White to move wins extremely tactical with Bxh7.
Simple woodcounting for programs.
I don't see why other programs (the programs i have)
don't find something tactical like this.
Just capture, recapture, few moves
lot of checks and captures, and 2 pawns up for white.
Diep needs something like an hour to find this.
So diep needs about 3600 x 15000 (nodes per second at a PP200) =
1800 x 30000 = 54 000 000 nodes.
Now look to the time other programs need to find it, and if you want to
compare supercomputers with Diep, look to the number of nodes (so DoMoves)
your program needs to find it.
Every DoMove in Diep i call a node. Number of Evaluations is only a small
part of this.
Vincent Diepeveen
vdie...@cs.ruu.nl
http://www.students.cs.ruu.nl/~vdiepeve/
Vincent Diepeveen
In <3299f...@news.cranfield.ac.uk> Simon Read <bbr...@ftc.gov> writes:
Is Simon Read another word for Chris Whittington?
why is your email not something like
si...@ftc.gov or re...@ftc.gov or sr...@ftc.gov?
What is your real name, Big Brother ==> bbroder (Broder = German for Brother?)?
Vincent Diepeveen
In <57dbam$d...@news.xs4all.nl> "Ed Schrder" <rebc...@xs4all.nl> writes:
(Most msg's of Ed Schroeder i cannot read, as the o with umlaut is not
always accepted by my mailreader, but this one i could answer with a lot
of mailreader trouble)
>: little stronger and plays more passive), and you attack a small guy, who
>: only sells his program not because i want to make such a big money out
>: of it, but because there is already NOW a lot of interest in a program
>: that is by far the strongest correspondence/analysis program.
>
>Vincent, if you don't like Rebel8 why don't you return the illegal copy
>of Rebel8 you have to the "ilegal" dealer you got Rebel8 from without
Without that dealer you would not even have written a chessprogram.
Is it true that you don't want to sell your program anymore, and just
spread 'decade' versions and 'light' versions in order to decrease
income of other chessprogrammers, hoping that when you die no one can
make money out of chessprograms, as Rebel 6 Decade, Rebel 7 Silver (although
i still don't see any difference between the engines?),
Rebel 8 light, and perhaps Rebel 9 d
>paying for it. Please remove Rebel8 from your hard disk.
I have a new computer (PP200-32), so how can i install Rebel on that,
other than
the Decade and light and future versions you plan to hand over?
When i'm at Jan Louwmans home, i'm testing your program, remember, at 10
computers at the same time, and at the 11th computer i play simultaneous blitz
against your program (which by the way is a nice feature).
I don't dislike your program,
i'm just like other testers little dissappointed that besides all the new
interface features rebel8 seems to have and the better un-human anticomputer
book, that you haven't done a thing on making a program that
plays better chess.
This criticism is also true for: Mchess, Genius, Kallisto, Fritz.
In the case of rebel8; it plays more passive, tactical little stronger.
It is interesting to see that this already gives you +60 elopoints.
>- Ed Schroder -
>: Vincent Diepeveen
>: vdie...@cs.ruu.nl
--
Ed Schröder
From: vdie...@cs.ruu.nl (Vincent Diepeveen)
: Vincent, if you don't like Rebel8 why don't you return the illegal copy
: of Rebel8 you have to the "ilegal" dealer you got Rebel8 from without
: paying for it. Please remove Rebel8 from your hard disk.
> Without that dealer you would not even have written a chessprogram.
So now a lot of insiders know the corrupted dealer in question...
Not so smart Vincent...
> Is it true that you don't want to sell your program anymore, and just
> spread 'decade' versions and 'light' versions in order to decrease
> income of other chessprogrammers, hoping that when you die no one can
> make money out of chessprograms, as Rebel 6 Decade, Rebel 7 Silver
> (although i still don't see any difference between the engines?),
> Rebel 8 light, and perhaps Rebel 9 d
What was it Chris said about you?
I fortunately forgot...
: I have a new computer (PP200-32), so how can i install Rebel on that,
: other than the Decade and light and future versions you plan to
: hand over?
Perhaps you can ask the dealer in question again?
It is ILLEGAL what you are doing.
It is ILLEGAL what this dealer is doing.
You both steal!
It is NOT allowed by dealers to install copyrighted chess programs on
other people's PC and not paying for it. Therefore I have removed
this dealer.
It's by you and corrupted dealers that copy protection remains.
And now the "unlimited installations" feature various chess companies
offer is misused *even* by dealers giving away free installs of
copyrighted chess programs.
I am speaking here in general, it also happens with other chess
programs other than Rebel.
> When i'm at Jan Louwmans home, i'm testing your program, remember, at
> 10 computers at the same time, and at the 11th computer i play
> simultaneous blitz against your program (which by the way is a nice
> feature).
Nothing wrong with that.
If someone has 100 PC's he may install Rebel8 on his 100 PC's.
However it is limited to personal use.
Don't you have no respect for copyrights?
> I don't dislike your program, i'm just like other testers little
> dissappointed that besides all the new interface features rebel8 seems
> to have and the better un-human anticomputer book, that you haven't
> done a thing on making a program that plays better chess.
This is not true. Jeroen Noomen my book editor only adds theory from
new released (on paper) opening books. All Rebel opening books have been
human alike since more than 10 years.
> This criticism is also true for: Mchess, Genius, Kallisto, Fritz.
: In the case of rebel8; it plays more passive, tactical little stronger.
: It is interesting to see that this already gives you +60 elopoints.
Thanks for the free ad.
Robert Hyatt
Vincent Diepeveen (vdie...@cs.ruu.nl) wrote:
: In <57cbl8$v...@juniper.cis.uab.edu> hy...@crafty.cis.uab.edu (Robert Hyatt) writes:
:
:
: >: Diep is by far the strongest correspondence/analysis program.
: >
: >Hmmm.. sounds like we now need a computer correspondence chess tournament,
: >otherwise the claim you make above makes little sense since it is founded
: >in speculation rather than "deed".
:
: I don't claim that it wins games. I claim that if you select a lot of
: difficult correspondence positions where there is an excellent move,
: either tactical or positional or strategical, taken from both opening,
: a lot taken from middlegame and endgame (1:2:1), then allow all program
: to search every position many hours. Now write on a paper how much positions
: every program solved (take care that the positions are so difficult that
: very probably no of the programs will come close to a perfect score,
: because scoring 99% or 99.1% out of a test is hard to compare).
:
: Diep will solve most.
While anyone can test chess engines any way they want, this particular
test is simply wrong if you use the results to draw any conclusions. I'd
be willing to bet that you can take Fischer's 100 greatest games, (not from
the book) where he sacrificed something for an attack, and give those
positions to crafty *after* the sacrifice, that Crafty could prosecute the
attack correctly and win the games Fischer won. Problem is, it likely (a)
would not make the sac and (b) would be unhappy with the position after the
sac.
It's *easy* to win a won position, it's not *easy* to first reach that
won position. That's why I suggested a correspondence tournament since
you believe Diep is "by far the strongest correspondence/analysis" program.
For that to be true, it would have to beat every other program at
Correspondence chess, otherwise it would be giving "bad advice" to anyone
that believed what it suggested.
:
: I use myself positions taken from Informator, and a lot of correspondence
: positions.
:
: DON'T use a standard test. These Positions are hacked into the programs,
: or parameters have been set to that they solve them.
Don't use *any* test set. Play real games. On ICC. If you can maintain a
2800+ rating there, you are strong... very strong. If you can't maintain
a 2300+ rating there, then you are weak (from a top-level program perspective.)
Problem suites prove little. Crafty generally doesn't do very well with them,
yet (at least for the games I get from ICC and chess.net) I haven't found a
commercial program that is significantly better than it is at playing chess.
Again, Lonnie can offer more info here, because I only have total games won
and lost by "handle" but Lonnie changes chess programs like the weather changes
in the South. :) Therefore I can't give you accurate Crafty vs Rebel8 or
Crafty vs CM5000 results, but can only say that for all the games Lonnie has
played over the past few months, Crafty wins more than it loses. I'd much
rather have that result than a great performance on BT2630 or whatever. I'm
not into solving puzzles or problems, I'm into winning chess games and finding
solutions to the steady diet of problems that humans are giving me.
:
: Don't use only strategical choices (BK test),
: or exchange choices(certain endgame tests), or only
: piece winning combinations(win at chess, although at long thinking times
: diep solves them all, like all other programs in theorem will do i guess).
: Use them all.
:
: Why not allowing an objective correspondence player (with a high elo
: preferably, so that he doesn't select crap moves) to select a lot of difficult
: correspondence game, where a difficult
:
: BTW how many programs solve next position: Kasparov-Deep Blue game 6,
: winning combination: Bxh7. After little more than an hour of
: thinking at a PP200-32, Diep (30 mb hashtables) sees this move to
: be the winning move.
My first thought is "so what?" In a game, you wouldn't "solve" this at
all. If Kasparov had an hour to think on that one position, and "if"
Bxh7 wins (Deep Blue claimed Bxh7 was a draw OTB when the game was going
on, the GM's there couldn't find a win either) I'd bet Kasparov could
/would find it with that much time. But in the game, he elected to not
try it, and I doubt *any* program would play it, including Diep because
you wouldn't have enough time... Again, what's the point of this? You
are picking on the *easy* part of chess, running down checks and the
like. I'd bet WchessX would play this move in a heartbeat knowing how
it evaluates king safety. Maybe CSTal as well although I don't have much
first-hand info about it. But unless you do this positionally, you haven't
made any great progress, because all programs are doing tactics well. It's
the non-tactical positions they are getting zapped in...
:
: this game can be found:
: http://www.chess.ibm.park.org/deep/blue/games/game6/s8log.B019.html
:
: position after 19...,Nge7?
:
: white: Kg1,Qd1,Rc1,Re3,Bb1,Bb2,Nf3,a2,b3,c4,d4,f2,g3,h4
: black: Kg8,Qd7,Rc8,Rd8,Bb4,Nc6,Ne7,a7,b7,d5,e6,f7,g7,h7
:
: White to move wins extremely tactical with Bxh7.
: Simple woodcounting for programs.
:
: I don't see why other programs (the programs i have)
: don't find something tactical like this.
:
: Just capture, recapture, few moves
: lot of checks and captures, and 2 pawns up for white.
:
: Diep needs something like an hour to find this.
: So diep needs about 3600 x 15000 (nodes per second at a PP200) =
: 1800 x 30000 = 54 000 000 nodes.
:
: Now look to the time other programs need to find it, and if you want to
: compare supercomputers with Diep, look to the number of nodes (so DoMoves)
: your program needs to find it.
:
: Every DoMove in Diep i call a node. Number of Evaluations is only a small
: part of this.
:
: Vincent Diepeveen
: vdie...@cs.ruu.nl
: http://www.students.cs.ruu.nl/~vdiepeve/
The big question is, does Bxh7 win? To this point, I've not seen anything
that says after Bxh7 black resigns. If you are selective in your forward
pruning, I'm even less convinced of your "find." If you are full width all
the way, it'd be more convincing. As I write this, I've had crafty searching
as I wrote this, searching only Bxh7, after depth=13, 12 minutes, it has a
score of -1.752 for white. It would probably search to depth 12 in this
position in a real game, and would not play this move, period. Until I see
something that convinces me it is the right move to play, I'm not going to
bother analyzing what's going on here. However, this one move has nothing
to do with beating humans of course, because the human refused to play it
as well. It's not so much about tactics now as it is about understanding.
Before you can claim you are better than everyone else, you have to some
how prove *something* by playing games...
Robert Hyatt
Ed Schröder (rebc...@xs4all.nl) wrote:
: From: vdie...@cs.ruu.nl (Vincent Diepeveen)
:
: : In the case of rebel8; it plays more passive, tactical little stronger.
: : It is interesting to see that this already gives you +60 elopoints.
:
: Thanks for the free ad.
:
: - Ed Schroder -
:
Passive is *not* a synonym for "bad". Can you spell Karpov?
Bert Gower and I noted over the years that as Cray Blitz became
stronger, it played more passive. We ran some opening book tests
on it years ago to see which "player" it would play like. After
lots of games, the amazing answer was Karpov. It always liked
the moves Karpov played, or at least 99% of the time it did. It
did not always like moves played by Kasparov, or Korchnoi or <you
fill in a GM name here>.
We didn't feel like that was bad. It *always* played from a sound
position, and you had to beat it. It would never "beat itself" as
the attacking players often do. Might be "boring" but it's won many
a world championship over the years although we all would rather remember
Capa or Tal or Fischer...
Another interesting statistic, although the number of games is low at
present. If you run "annotate" on a group of games from several GM
events, Karpov consistently generates the fewest "suggested move to
avoid a tactical blunder" type of comment. Of course, we sleep through
his games and overlook this shortcoming. :) Kasparov makes numbers of
blunders. Which is amazing. yet they don't result in losses because
most of them are overlooked by his opponent... Or they aren't bad
enough.. (ie blunder by winning two pawns rather than a whole piece,
where either will probably win the game, as in one of the DB games.)
Bob
Simon Read
[Here's a special warm christmas message to all my fans in surveillance
and counterintelligence! Your scanners will really trip out over this one:]
Fort Knox NORAD coup PLO MOSSAD CIA Ghadafi F-15 F-18 SCUD president
oil Treasury hijack DES plutonium genetic jihad NSA plutonium computer
Honduras BATF Semtex Warsaw missile cyanide Honduras cocaine MI6 BATF
plot MIG-29 pipeline kill laser IRA Libya MI5 Iraq communist krugerrand
encryption Semtex Cuba research secret Vatican kidnap homosexual diamonds
chemical assassinate tornado EFA plutonium prime minister laboratory
satellite lieutenant KGB calibre propaganda rabies heroin Kim Il Sung
Greenpeace plutonium supergun PanAm TWA800 PGP
Since you ask, Vincent, my real name is Simon Read. No, I'm not
Chris Whittington. I couldn't possibly be, since I'm going to
write a program which will beat Chess System Tal.
As for that bbroder ftc gov address, it's _poison_ for the robots
and it has nothing to do with me. Email me and I'll tell you about it.
My real email address is in other postings I have made.
I'm still curious about your university policy vis-a-vis assisting
you with non-university projects. In the current climate in
Britain, my university would try to charge me.
Simon
Chris Whittington
vdie...@cs.ruu.nl (Vincent Diepeveen) wrote:
>
> In <3299f...@news.cranfield.ac.uk> Simon Read <bbr...@ftc.gov> writes:
>
> Is Simon Read another word for Chris Whittington?
I think he's going to regret that one.
Go for him, Simon :)
Chris Whittington (runs upstairs to second PC and different
Internet connection) .......
> why is your email not something like
>
> si...@ftc.gov or re...@ftc.gov or sr...@ftc.gov?
>
> What is your real name, Big Brother ==> bbroder (Broder = German for Brother?)?
>
> Vincent Diepeveen
> vdie...@cs.ruu.nl
> http://www.students.cs.ruu.nl/~vdiepeve/
brucemo
Robert Hyatt wrote:
> Another interesting statistic, although the number of games is low at
> present. If you run "annotate" on a group of games from several GM
> events, Karpov consistently generates the fewest "suggested move to
> avoid a tactical blunder" type of comment. Of course, we sleep through
> his games and overlook this shortcoming. :) Kasparov makes numbers of
> blunders. Which is amazing. yet they don't result in losses because
> most of them are overlooked by his opponent... Or they aren't bad
> enough.. (ie blunder by winning two pawns rather than a whole piece,
> where either will probably win the game, as in one of the DB games.)
In some cases you see tactical play that isn't understandable by a computer. Play over
some of Alekhine's games, sheesh.
bruce
Rolf W. Tueschen
vdie...@cs.ruu.nl (Vincent Diepeveen) wrote:
>In <57dbam$d...@news.xs4all.nl> "Ed Schrder" <rebc...@xs4all.nl> writes:
>>
>>Vincent, if you don't like Rebel8 why don't you return the illegal copy
>>of Rebel8 you have to the "ilegal" dealer you got Rebel8 from without
-----------------------------------------------------------------------------------------------------------------
Dear Vincent,
I came across this thread by chance. As you know I'm *only* a new noname
here around.
1. But I do want to state that your answer on my endgame-position was very
helpful for me.
2. You never entered my quarrels with Schroder BV. And for that I even
thank you more.
3. Therefore I want to tell you that I'm there if Schroder BV now try to
catch you the way they did *try* me.
Some hints. I'm almost sure that Schroder BV would write some heavy emails
to you. They probably will threaten you. I would recommand in that case
that you go the same way I did. The way to publish the main points and the
fact as such here in the group.
If you now start to shiver, and perhaps give in too quick, Schroder BV
would never stop to command you around like a little doggy dog.
I quoted just the above mentioned phrase which is so typically for the
whole *thinking* of Schroder BV.
It's so mediocre and hypocritical that it makes me real sad. This
completely absurd logic they used in my case too.
There they said, well you even don't have those prgs --- AND you will
critizise me and my arguments ...
Vincent, if you try to communicate with Schroder BV via email you would be
dissapointed because whatever you write Schroder BV will firmly stay in his
crooked thinking. Myself e.g. I confirmed four or five times, Ed I'm
completely new here ... But they answered well I read this, but ... so and
so years old, and I can smell it ...
Schroder BV even proposed medical diagnosis for this behaviour after 15
years in the chess business. So I'm almost sure that Schroder is no bad guy
at all. But in reality you have to handle the situation now.
Finally I want to express my hope that longtime collegues could intervene
on the side of Schroder BV and not against Vincent. Whatever this young
expert might have said in the past --- I don't know it in detail --- he
shouldn't be under multiple bombardement now. But Schroder BV apparently
seems to have sort of *free aggression ticket* against each and everyone
who starts to critizise his own behaviour. Do these collegues read
carefully this logic seen above? It's simply not fair to let the one acting
this way but asking strangely the other for his server and so on.
And at last, because I'm really upset now. Dear Chris, don't praise my
writings if you expect on the other side that I might stay mute in front of
this behaviour of Ed. If you are not prepared to tell your friend some
needed truths, other people must do it. I read nowhere a takeback of
Schroder BV, so don't go into s.th. that you cannot judge correctly. You
read my aggressive posts but the bullshit Ed wrote behind the curtain must
have been kept from your attention selfunderstood. I thought you smart
enough to be able to differentiate these facts. Big Ed can do what he likes
but certainly younger Vincent must be shot. I can't follow this logic!
As long as you participate in this keep away from me with any commentary on
my postings, Sir Chris.
Rolf Tueschen -- with best wishes for Vincent
pit...@aol.com
Im Artikel <57evu6$7...@news.xs4all.nl>, "Ed Schröder" <rebc...@xs4all.nl>
schreibt:
>: Vincent, if you don't like Rebel8 why don't you return the illegal copy
>: of Rebel8 you have to the "ilegal" dealer you got Rebel8 from without
>: paying for it. Please remove Rebel8 from your hard disk.
>
>> Without that dealer you would not even have written a chessprogram.
>
>So now a lot of insiders know the corrupted dealer in question...
>Not so smart Vincent...
Corrupted Dealer ? What `s his name ?
>
>> Is it true that you don't want to sell your program anymore, and just
>> spread 'decade' versions and 'light' versions in order to decrease
>> income of other chessprogrammers, hoping that when you die no one can
>> make money out of chessprograms, as Rebel 6 Decade, Rebel 7 Silver
>> (although i still don't see any difference between the engines?),
>> Rebel 8 light, and perhaps Rebel 9 d
>
>What was it Chris said about you?
>I fortunately forgot...
>
>: I have a new computer (PP200-32), so how can i install Rebel on that,
>: other than the Decade and light and future versions you plan to
>: hand over?
>
>Perhaps you can ask the dealer in question again?
>It is ILLEGAL what you are doing.
>It is ILLEGAL what this dealer is doing.
>You both steal!
Dear Ed, what`s the name of this dealer ?
>
>It is NOT allowed by dealers to install copyrighted chess programs on
>other people's PC and not paying for it. Therefore I have removed
>this dealer.
>
>It's by you and corrupted dealers that copy protection remains.
>And now the "unlimited installations" feature various chess companies
>offer is misused *even* by dealers giving away free installs of
>copyrighted chess programs.
>
>I am speaking here in general, it also happens with other chess
>programs other than Rebel.
>
>> When i'm at Jan Louwmans home, i'm testing your program, remember, at
>> 10 computers at the same time, and at the 11th computer i play
>> simultaneous blitz against your program (which by the way is a nice
>> feature).
>
>Nothing wrong with that.
>If someone has 100 PC's he may install Rebel8 on his 100 PC's.
>However it is limited to personal use.
>
>Don't you have no respect for copyrights?
>
At Aegon some people making illegal copies from the tournament machines
!!! So I always remove MChess from the computer...........
>> I don't dislike your program, i'm just like other testers little
>> dissappointed that besides all the new interface features rebel8 seems
>> to have and the better un-human anticomputer book, that you haven't
>> done a thing on making a program that plays better chess.
>
>This is not true. Jeroen Noomen my book editor only adds theory from
>new released (on paper) opening books. All Rebel opening books have been
>human alike since more than 10 years.
>
>> This criticism is also true for: Mchess, Genius, Kallisto, Fritz.
>
>: In the case of rebel8; it plays more passive, tactical little stronger.
>: It is interesting to see that this already gives you +60 elopoints.
>
>Thanks for the free ad.
>
>- Ed Schroder -
Dear Ed, perhaps you should consult your lawyer .................
Best regards
-Peter
Chris Whittington
> : of Rebel8 you have to the "ilegal" dealer you got Rebel8 from without
> : paying for it. Please remove Rebel8 from your hard disk.
>
> > Without that dealer you would not even have written a chessprogram.
>
> So now a lot of insiders know the corrupted dealer in question...
> Not so smart Vincent...
>
> > spread 'decade' versions and 'light' versions in order to decrease
> > income of other chessprogrammers, hoping that when you die no one can
> > make money out of chessprograms, as Rebel 6 Decade, Rebel 7 Silver
> > (although i still don't see any difference between the engines?),
> > Rebel 8 light, and perhaps Rebel 9 d
>
> What was it Chris said about you?
> I fortunately forgot...
I didn't :)
I believe I said he was a dickhead.
But I retract that now.
Louwmann's poodle would be more apposite :)
Chris Whittington
>
> : I have a new computer (PP200-32), so how can i install Rebel on that,
> : other than the Decade and light and future versions you plan to
> : hand over?
>
> Perhaps you can ask the dealer in question again?
> It is ILLEGAL what you are doing.
> It is ILLEGAL what this dealer is doing.
> You both steal!
>
> It is NOT allowed by dealers to install copyrighted chess programs on
> other people's PC and not paying for it. Therefore I have removed
> this dealer.
>
> It's by you and corrupted dealers that copy protection remains.
> And now the "unlimited installations" feature various chess companies
> offer is misused *even* by dealers giving away free installs of
> copyrighted chess programs.
>
> I am speaking here in general, it also happens with other chess
> programs other than Rebel.
>
> > When i'm at Jan Louwmans home, i'm testing your program, remember, at
> > 10 computers at the same time, and at the 11th computer i play
> > simultaneous blitz against your program (which by the way is a nice
> > feature).
>
> Nothing wrong with that.
> If someone has 100 PC's he may install Rebel8 on his 100 PC's.
> However it is limited to personal use.
>
> Don't you have no respect for copyrights?
>
> > I don't dislike your program, i'm just like other testers little
> > dissappointed that besides all the new interface features rebel8 seems
> > to have and the better un-human anticomputer book, that you haven't
> > done a thing on making a program that plays better chess.
>
> This is not true. Jeroen Noomen my book editor only adds theory from
> new released (on paper) opening books. All Rebel opening books have been
> human alike since more than 10 years.
>
> > This criticism is also true for: Mchess, Genius, Kallisto, Fritz.
>
> : In the case of rebel8; it plays more passive, tactical little stronger.
> : It is interesting to see that this already gives you +60 elopoints.
>
> Thanks for the free ad.
>
> - Ed Schroder -
>
>
> >: Vincent Diepeveen
> >: vdie...@cs.ruu.nl
>
>
Robert Hyatt
brucemo (bru...@nwlink.com) wrote:
I agree. In fact, some "blunders" might not be blunders at all and take a
lot of work to verify as such. One that comes to mind was the game where
Kasparov scored his first win after offering DB a draw which they refused.
He played one move, and detailed analysis convincingly proved that there
was a much better move that most of the computers kibitzing saw. However,
my statement still stands... the GM's are far from perfect it seems...
Bob
Chris Whittington
pit...@aol.com wrote:
>
> Im Artikel <57evu6$7...@news.xs4all.nl>, "Ed Schröder" <rebc...@xs4all.nl>
> schreibt:
>
> >: of Rebel8 you have to the "ilegal" dealer you got Rebel8 from without
> >: paying for it. Please remove Rebel8 from your hard disk.
> >
> >> Without that dealer you would not even have written a chessprogram.
> >
> >So now a lot of insiders know the corrupted dealer in question...
> >Not so smart Vincent...
>
>
Ed posted:
>: Vincent, if you don't like Rebel8 why don't you return the illegal copy
>: of Rebel8 you have to the "ilegal" dealer you got Rebel8 from without
>: paying for it. Please remove Rebel8 from your hard disk.
>
>> Without that dealer you would not even have written a chessprogram.
>
>So now a lot of insiders know the corrupted dealer in question...
>Not so smart Vincent...
One would assume from what Ed posted above, that it is someone, in
Holland, with whom Ed used to work closely with some years ago.
Presumably he doesn't work with him anymore.
Now who could that be ?
Another clue from Ed: we can discern that the allegedly corrupted
dealer is now doing some work with Vincent.
Read on, all shall be revealed :)
Chris Whittington
>
> >
> >> Is it true that you don't want to sell your program anymore, and just
> >> spread 'decade' versions and 'light' versions in order to decrease
> >> income of other chessprogrammers, hoping that when you die no one can
> >> make money out of chessprograms, as Rebel 6 Decade, Rebel 7 Silver
> >> (although i still don't see any difference between the engines?),
> >> Rebel 8 light, and perhaps Rebel 9 d
> >
> >What was it Chris said about you?
> >I fortunately forgot...
> >
> >: I have a new computer (PP200-32), so how can i install Rebel on that,
> >: other than the Decade and light and future versions you plan to
> >: hand over?
> >
> >Perhaps you can ask the dealer in question again?
> >It is ILLEGAL what you are doing.
> >It is ILLEGAL what this dealer is doing.
> >You both steal!
>
> Dear Ed, what`s the name of this dealer ?
>
> >
> >It is NOT allowed by dealers to install copyrighted chess programs on
> >other people's PC and not paying for it. Therefore I have removed
> >this dealer.
> >
> >It's by you and corrupted dealers that copy protection remains.
> >And now the "unlimited installations" feature various chess companies
> >offer is misused *even* by dealers giving away free installs of
> >copyrighted chess programs.
> >
> >I am speaking here in general, it also happens with other chess
> >programs other than Rebel.
> >
> >> When i'm at Jan Louwmans home, i'm testing your program, remember, at
> >> 10 computers at the same time, and at the 11th computer i play
> >> simultaneous blitz against your program (which by the way is a nice
> >> feature).
> >
> >Nothing wrong with that.
> >If someone has 100 PC's he may install Rebel8 on his 100 PC's.
> >However it is limited to personal use.
> >
> >Don't you have no respect for copyrights?
> >
>
> At Aegon some people making illegal copies from the tournament machines
> !!! So I always remove MChess from the computer...........
>
> >> I don't dislike your program, i'm just like other testers little
> >> dissappointed that besides all the new interface features rebel8 seems
> >> to have and the better un-human anticomputer book, that you haven't
> >> done a thing on making a program that plays better chess.
> >
> >This is not true. Jeroen Noomen my book editor only adds theory from
> >new released (on paper) opening books. All Rebel opening books have been
> >human alike since more than 10 years.
> >
> >> This criticism is also true for: Mchess, Genius, Kallisto, Fritz.
> >
> >: In the case of rebel8; it plays more passive, tactical little stronger.
> >: It is interesting to see that this already gives you +60 elopoints.
> >
> >Thanks for the free ad.
> >
> >- Ed Schroder -
>
brucemo
Simon Read wrote:
> I'm still curious about your university policy vis-a-vis assisting
> you with non-university projects. In the current climate in
> Britain, my university would try to charge me.
>
> Simon
Personally, I wouldn't concern myself with it, as I believe that it
is none of my concern.
If you really want to rat on the guy, simply send email to his
university, rather than trying to embarass him here.
I find it odd that this kind of post is originating with someone who
is obviously peeved at other people (government agencies, bulk e-mail
advertisers) messing in his business.
bruce
Ed Schröder
From: vdie...@cs.ruu.nl (Vincent Diepeveen)
: Diep is by far the strongest correspondence/analysis program.
This is a very huge claim!
Besides Kasparov-Deep Blue game 6, 20. Bxh7?! do you have more proof?
How much scores Diep at the LCT-II test?
How much scores Diep at the NOLUT test?
Ok, I don't like these test sets either as being the absolute truth on
the playing strength of a chess program but I still believe these tests
do give an *INDICATION*
If program X scores 2200 on the LCT-II test, program X does (probably)
not belong to the top. If program X scores 2500 it *probably* belongs
to the top.
Looking forward for your results.
- Ed Schroder -
Here are the 11 NOLUT positions...
r3qb1k/1b4p1/p2pr2p/3n4/Pnp1N1N1/6RP/1B3PP1/1B1QR1K1 w - - bm Nxh6;
r4rk1/pp1n1p1p/1nqP2p1/2b1P1B1/4NQ2/1B3P2/PP2K2P/2R5 w - - bm Rxc5;
r2qk2r/ppp1b1pp/2n1p3/3pP1n1/3P2b1/2PB1NN1/PP4PP/R1BQK2R w - - bm Nxg5;
r1b1kb1r/1p1n1ppp/p2ppn2/6BB/2qNP3/2N5/PPP2PPP/R2Q1RK1 w - - bm Nxe6;
r2qrb1k/1p1b2p1/p2ppn1p/8/3NP3/1BN5/PPP3QP/1K3RR1 w - - bm e5;
rnbqk2r/1p3ppp/p7/1NpPp3/QPP1P1n1/P4N2/4KbPP/R1B2B1R b - - bm axb5;
1r1bk2r/2R2ppp/p3p3/1b2P2q/4QP2/4N3/1B4PP/3R2K1 w - - bm Rxd8+;
r3rbk1/ppq2ppp/2b1pB2/8/6Q1/1P1B3P/P1P2PP1/R2R2K1 w - - bm Bxh7+;
r4r1k/4bppb/2n1p2p/p1n1P3/1p1p1BNP/3P1NP1/qP2QPB1/2RR2K1 w - - bm Ng5;
r1b2rk1/1p1nbppp/pq1p4/3B4/P2NP3/2N1p3/1PP3PP/R2Q1R1K w - - bm Rxf7;
r1b3k1/p2p1nP1/2pqr1Rp/1p2p2P/2B1PnQ1/1P6/P1PP4/1K4R1 w - - bm Rxh6;
BTW, I tested these 11 NOLUT positions once overnight with Rebel8.
Level 30:00 fixed time.
Machine PP-200
Result: 0 (this reads as NUL) positions found!!
I am curious about the results of Diep.
Chris Whittington
TUESCHEN.MEDIZ...@t-online.de (Rolf W. Tueschen) wrote:
>
> vdie...@cs.ruu.nl (Vincent Diepeveen) wrote:
>
> >In <57dbam$d...@news.xs4all.nl> "Ed Schrder" <rebc...@xs4all.nl> writes:
>
> >>
> >>of Rebel8 you have to the "ilegal" dealer you got Rebel8 from without
>
Unbelievable !
Ed knows very well that I'll say exactly what I think without
any fear or favour. Don't imagine for one moment that Ed gets
spared.
> I read nowhere a takeback of
> Schroder BV, so don't go into s.th. that you cannot judge correctly. You
> read my aggressive posts but the bullshit Ed wrote behind the curtain must
> have been kept from your attention selfunderstood. I thought you smart
> enough to be able to differentiate these facts. Big Ed can do what he likes
> but certainly younger Vincent must be shot.
No, no, not shot :)
> I can't follow this logic!
> As long as you participate in this keep away from me with any commentary on
> my postings, Sir Chris.
Bull, rag, red.
>
> Rolf Tueschen -- with best wishes for Vincent
Don't know which one of you two allies I feel sorry for the most :)
The words kiss of death spring to mind.
Chris Whittington
>
>
>
brucemo
>
>
> Besides Kasparov-Deep Blue game 6, 20. Bxh7?! do you have more proof?
> Here are the 11 NOLUT positions...
> r3qb1k/1b4p1/p2pr2p/3n4/Pnp1N1N1/6RP/1B3PP1/1B1QR1K1 w - - bm Nxh6;
> r4rk1/pp1n1p1p/1nqP2p1/2b1P1B1/4NQ2/1B3P2/PP2K2P/2R5 w - - bm Rxc5;
> r2qk2r/ppp1b1pp/2n1p3/3pP1n1/3P2b1/2PB1NN1/PP4PP/R1BQK2R w - - bm Nxg5;
> r1b1kb1r/1p1n1ppp/p2ppn2/6BB/2qNP3/2N5/PPP2PPP/R2Q1RK1 w - - bm Nxe6;
> r2qrb1k/1p1b2p1/p2ppn1p/8/3NP3/1BN5/PPP3QP/1K3RR1 w - - bm e5;
> rnbqk2r/1p3ppp/p7/1NpPp3/QPP1P1n1/P4N2/4KbPP/R1B2B1R b - - bm axb5;
> 1r1bk2r/2R2ppp/p3p3/1b2P2q/4QP2/4N3/1B4PP/3R2K1 w - - bm Rxd8+;
> r3rbk1/ppq2ppp/2b1pB2/8/6Q1/1P1B3P/P1P2PP1/R2R2K1 w - - bm Bxh7+;
> r4r1k/4bppb/2n1p2p/p1n1P3/1p1p1BNP/3P1NP1/qP2QPB1/2RR2K1 w - - bm Ng5;
> r1b2rk1/1p1nbppp/pq1p4/3B4/P2NP3/2N1p3/1PP3PP/R2Q1R1K w - - bm Rxf7;
> r1b3k1/p2p1nP1/2pqr1Rp/1p2p2P/2B1PnQ1/1P6/P1PP4/1K4R1 w - - bm Rxh6;
>
> BTW, I tested these 11 NOLUT positions once overnight with Rebel8.
> Level 30:00 fixed time.
> Machine PP-200
>
> Result: 0 (this reads as NUL) positions found!!
>
> I am curious about the results of Diep.
This is a very cool test suite, for the reason that it is so very hard. I
believe that a few of them are solvable on micros but you may have to give
them a little more time. I'll run it through mine soon.
FYI: I ran the Kasparov position through a version of Ferret (possibly a bogus
version, possibly a great version), and found Bxh7+ with a slightly negative
evaluation after approximately 30 hours.
I figure that there must be other programs who can solve this more quickly
than Ferret, Vincent has already stated that Diep solves this fairly quickly,
has anyone else had time to try?
bruce
Ed Schröder
Dear Rolf,
When are you planning to stop your mud?
You had your show, the thread was closed.
Why do you start all over again?
I can't see the connection.
Can somebody tell me how to stop you?
Perhaps a Rolf Tueschen FAQ on my home page?
I am not in the mood for all your shit...
Please stop this...
I was not able to convince you about my good intensions.
I can live with that.
Can't you?
- Ed Schroder -
>needed truths, other people must do it. I read nowhere a takeback of
>Schroder BV, so don't go into s.th. that you cannot judge correctly. You
>read my aggressive posts but the bullshit Ed wrote behind the curtain must
>have been kept from your attention selfunderstood. I thought you smart
>enough to be able to differentiate these facts. Big Ed can do what he likes
>but certainly younger Vincent must be shot. I can't follow this logic!
>As long as you participate in this keep away from me with any commentary on
>my postings, Sir Chris.
>
Ed Schröder
From: Chris Whittington <chr...@cpsoft.demon.co.uk>
> And at last, because I'm really upset now. Dear Chris, don't praise my
> writings if you expect on the other side that I might stay mute in front of
> this behaviour of Ed. If you are not prepared to tell your friend some
> needed truths, other people must do it.
: Unbelievable !
: Ed knows very well that I'll say exactly what I think without
: any fear or favour. Don't imagine for one moment that Ed gets
: spared.
That's the right spirit Chris... :)
- Ed -
: Don't know which one of you two allies I feel sorry for the most :)
Chris Whittington
brucemo <bru...@nwlink.com> wrote:
>
> Ed Schröder wrote:
> >
> > From: vdie...@cs.ruu.nl (Vincent Diepeveen)
> >
> >
I seem to remember posting at the time of the match that CSTal
'solved' Bh7+ within a few seconds.
Note the use of the '' :)
Chris Whittington
>
> bruce
Michael F. Byrne
brucemo <bru...@nwlink.com> wrote in article <329F68...@nwlink.com>...
>[deleted[
I attended the 6th game in Philadelphia...two points ...there was wide
speculation tha Kasparov might play Bxh6 during the game...and he admitted
at the conclusion of the game that that he would have given that move a
higher priority against a *human* opponent; howver with Deep Blue tacital
abilities ..why take the chance...he had an excellent game w/o playing Bxh7
anyway....
I seemed to recall that the Deep Blue handlers stated during the course of
that game that DB actually expected Bxh7 to be played....if you folks are
reading this..can you confirm?
> FYI: I ran the Kasparov position through a version of Ferret (possibly a
bogus
> version, possibly a great version), and found Bxh7+ with a slightly
negative
> evaluation after approximately 30 hours.
>
> I figure that there must be other programs who can solve this more
quickly
> than Ferret, Vincent has already stated that Diep solves this fairly
quickly,
> has anyone else had time to try?
>
> bruce
>
Simon Read
brucemo <bru...@nwlink.com> wrote:
>If you really want to rat on the guy, simply send email to his
>university, rather than trying to embarass him here.
>
>I find it odd that this kind of post is originating with someone who
>is obviously peeved at other people (government agencies, bulk e-mail
>advertisers) messing in his business.
You're completely misunderstanding. I went out of my way to tell
Vincent I WASN'T criticising or attacking him; I was just _asking_
about university policy. Is it illegal to ask questions around here?
Yes, I pointed out the contrast between his university policy and mine,
but that was a motivation for being curious. I was not jumping to any
conclusions and assuming anything dodgy was going on. It's very dangerous
on usenet to try to read between the lines of someone's posting and
extract conclusions which aren't there. You think you could extract the
conclusion that I was criticising Vincent? Well, you could just as
easily turn the comparison the other way round and come to the
conclusion that I was criticising my own university for NOT supporting
me. Neither conclusion would be valid. Nowhere did I say anything
like "you should not be doing this."
There are in fact some government research organisations in the UK
which do allow use of their facilities for their employees to do
other commercial work, with certain provisos.
Simon
If you're a human being with a modicum of intelligence, I can be reached at
"ac cranfield read s uk" but in this order: s(dot)r_(at)c_(dot)a_(dot)u_
Robots beware! Sending advertising email to the address in the header of this
posting will get you into a _lot_ of trouble, ha ha ha!!!!
Vincent Diepeveen
In <329F68...@nwlink.com> brucemo <bru...@nwlink.com> writes:
>> From: vdie...@cs.ruu.nl (Vincent Diepeveen)
>>
>> : Diep is by far the strongest correspondence/analysis program.
>>
>> This is a very huge claim!
>>
>> Besides Kasparov-Deep Blue game 6, 20. Bxh7?! do you have more proof?
>> Here are the 11 NOLUT positions...
>> r3qb1k/1b4p1/p2pr2p/3n4/Pnp1N1N1/6RP/1B3PP1/1B1QR1K1 w - - bm Nxh6;
Ain't this position from a match Kasparov-Karpov?
These are the kind of positions i test on Diep!
Thanks for the positions!
>> r4rk1/pp1n1p1p/1nqP2p1/2b1P1B1/4NQ2/1B3P2/PP2K2P/2R5 w - - bm Rxc5;
I analyzed a little with Amindo Naarden:
{--------------
r . . . . r k .
p p . n . p . p
. n q P . . p .
. . b . P . B .
. . . . N Q . .
. B . . . P . .
P P . . K . . P
. . R . . . . .
white to play
--------------}
1. Rxc5 Nxc5 2. Nxc5 Qxc5 3. Bf6 Nd5 4. Bxd5 Qxd5 5. Qh6 Qc4+ 6. Ke1 Qb4+
7. Kf1 Qb5+ 8. Kg1 Qc5+ 9. Kh1 Rfc8 10. Qg7 mate.
So at maximum 13 ply needed to find it.
The only question is: will Diep get 13 ply, or is there a variation longer
than this?
>> r2qk2r/ppp1b1pp/2n1p3/3pP1n1/3P2b1/2PB1NN1/PP4PP/R1BQK2R w - - bm Nxg5;
>> r1b1kb1r/1p1n1ppp/p2ppn2/6BB/2qNP3/2N5/PPP2PPP/R2Q1RK1 w - - bm Nxe6;
>> r2qrb1k/1p1b2p1/p2ppn1p/8/3NP3/1BN5/PPP3QP/1K3RR1 w - - bm e5;
>> rnbqk2r/1p3ppp/p7/1NpPp3/QPP1P1n1/P4N2/4KbPP/R1B2B1R b - - bm axb5;
>> 1r1bk2r/2R2ppp/p3p3/1b2P2q/4QP2/4N3/1B4PP/3R2K1 w - - bm Rxd8+;
>> r3rbk1/ppq2ppp/2b1pB2/8/6Q1/1P1B3P/P1P2PP1/R2R2K1 w - - bm Bxh7+;
>> r4r1k/4bppb/2n1p2p/p1n1P3/1p1p1BNP/3P1NP1/qP2QPB1/2RR2K1 w - - bm Ng5;
>> r1b2rk1/1p1nbppp/pq1p4/3B4/P2NP3/2N1p3/1PP3PP/R2Q1R1K w - - bm Rxf7;
>> r1b3k1/p2p1nP1/2pqr1Rp/1p2p2P/2B1PnQ1/1P6/P1PP4/1K4R1 w - - bm Rxh6;
I'm printing the positions to the printer right now, and will try them on
Diep. Starting this night. Diep runs at home at a PP200-32.
Vincent
>> BTW, I tested these 11 NOLUT positions once overnight with Rebel8.
>> Level 30:00 fixed time.
I'll give Diep will 12:00 (12 hours) a position at maximum.
>> Machine PP-200
>>
>> Result: 0 (this reads as NUL) positions found!!
>>
>> I am curious about the results of Diep.
Me too. Lucky i don't need to solve them myself, the level in these positions
is far beyond my own capabilities! Kasparov-Karpov... ..cool.
Diep outsearching the level of its creator!
>This is a very cool test suite, for the reason that it is so very hard. I
>believe that a few of them are solvable on micros but you may have to give
>them a little more time. I'll run it through mine soon.
>
>FYI: I ran the Kasparov position through a version of Ferret (possibly a bogus
>version, possibly a great version), and found Bxh7+ with a slightly negative
>evaluation after approximately 30 hours.
>I figure that there must be other programs who can solve this more quickly
>than Ferret, Vincent has already stated that Diep solves this fairly quickly,
>has anyone else had time to try?
I heard rumors DB should also find it. Why did it play Ng6-e7 then? Bug
in the program at that time?
>bruce
Vincent
--
+----------------------------------------------------+
| Vincent Diepeveen email: vdie...@cs.ruu.nl |
| http://www.students.cs.ruu.nl/~vdiepeve/ |
+----------------------------------------------------+
Vincent Diepeveen
>Dear Rolf,
>When are you planning to stop your mud?
>You had your show, the thread was closed.
Not yet.
>Why do you start all over again?
>
>I can't see the connection.
>Can somebody tell me how to stop you?
>>Rolf Tueschen -- with best wishes for Vincent
Ed,
The non-currupt dealer, you are blaming to be corrupt,
asked me to tell you this in the newsgroup:
"I have sent Vincent an official diskversion of Rebel8, paying the
price of the program myself.
Still I don't understand why Ed himself installs versions of, for example
MchessPro 6, at his friends harddisk.
The Non-corrupt Dealer"
Vincent Diepeveen
>From: Chris Whittington <chr...@cpsoft.demon.co.uk>
>
>> And at last, because I'm really upset now. Dear Chris, don't praise my
>> writings if you expect on the other side that I might stay mute in front of
>> this behaviour of Ed. If you are not prepared to tell your friend some
>> needed truths, other people must do it.
>
>: Unbelievable !
>
>: Ed knows very well that I'll say exactly what I think without
>: any fear or favour. Don't imagine for one moment that Ed gets
>: spared.
Well Chris, What is your opinion about the Rebel super killerbook.
I played a rapid-game Rebel-Genius3. Rebel in book: far over 20 moves.
Genius in book: around 10 moves.
After his book Genius3 played: ... Qd8-b6 followed by Qb6xb2??
Although this was a typical computerline, it was of course in the
rebel-book. After getting out of book Rebel was a piece up.
I therefore stopped the game.
Probably not a killerline in the eyes of Ed. A killerline is in the
eyes of Ed a line where after getting out of book the opponent is
mated, Ed?
Perhaps i think too simplistic, I don't see difference between a
full piece up, so a mate in about 20,
instead of an already seen mate in 7 (mcpro).
I don't blame anyone writing a killerbook, but don't blame someone
else of having a killerbook if you have yourselve the biggest one?
Vincent
Chris Whittington
vdie...@cs.ruu.nl (Vincent Diepeveen) wrote:
>
> In <57ntr9$3...@news.xs4all.nl> "Ed Schrder" <rebc...@xs4all.nl> writes:
>
> >From: Chris Whittington <chr...@cpsoft.demon.co.uk>
> >
> >> And at last, because I'm really upset now. Dear Chris, don't praise my
> >> writings if you expect on the other side that I might stay mute in front of
> >> this behaviour of Ed. If you are not prepared to tell your friend some
> >> needed truths, other people must do it.
> >
> >: Unbelievable !
> >
> >: Ed knows very well that I'll say exactly what I think without
> >: any fear or favour. Don't imagine for one moment that Ed gets
> >: spared.
>
> Well Chris, What is your opinion about the Rebel super killerbook.
When you've apologised for your earlier bad behaviour (see earlier
threads), I'll consider answering you.
Chris Whittington
Tord Kallqvist Romstad
brucemo (bru...@nwlink.com) wrote:
: FYI: I ran the Kasparov position through a version of Ferret (possibly a bogus
: evaluation after approximately 30 hours.
^^^^^^^^^^
Negative evaluation??? Does Ferret believe that black is better in this
position?
Tord
: I figure that there must be other programs who can solve this more quickly
: than Ferret, Vincent has already stated that Diep solves this fairly quickly,
: has anyone else had time to try?
: bruce
Ed Schröder
From: vdie...@cs.ruu.nl (Vincent Diepeveen)
: Ed,
: The non-currupt dealer, you are blaming to be corrupt,
: asked me to tell you this in the newsgroup:
: "I have sent Vincent an official diskversion of Rebel8, paying the
: price of the program myself.
Good...
: Still I don't understand why Ed himself installs versions of, for
: example MchessPro 6, at his friends harddisk.
Well tell Jan Louwman that he is not fully informed.
I always buy atleast 2-3 copies of chess software I use.
One copy for myself, one copy for Jeroen my cooperator.
No one else get copies...
Do not let yourself being used...
That is stupid.
- Ed Schroder -
: The Non-corrupt Dealer"
: +----------------------------------------------------+
Chris Whittington
> : I played a rapid-game Rebel-Genius3. Rebel in book
> : far over 20 moves.: Genius in book
> : around 10 moves.
>
> : After his book Genius3 played: ... Qd8-b6 followed by Qb6xb2??
> : Although this was a typical computerline, it was of course in the
> : rebel-book. After getting out of book Rebel was a piece up.
> : I therefore stopped the game.
>
> : Probably not a killerline in the eyes of Ed. A killerline is in the
> : eyes of Ed a line where after getting out of book the opponent is
> : mated, Ed?
>
> he will tell you immediately from which (on paper) opening book your
> example comes from :)
>
> : Perhaps i think too simplistic, I don't see difference between a
> : instead of an already seen mate in 7 (mcpro).
> : I don't blame anyone writing a killerbook, but don't blame someone
> : else of having a killerbook if you have yourselve the biggest one?
>
> : Vincent
>
I'm in agreement here with Ed.
but also wrong, misleaded and used...
And in agreement here also. The man hiding behind you is a well
known collector of young chess programmers.
They all leave him in the end, unhappy.
Ask Ed.
He's also an unpleasant devious character. Vincent, you used to
be a reasonable guy. I used to support you in the group. But recently
you've become snide and attacking of your colleagues. Little
digs here, accusations there.
Putting two and two together, I am rather assuming that this is
due to a certain nasty, whispering influence that you've acquired.
Since he's been operating in whispering mode now for years, he's
become rather used to the power it gives him, but only when he
doesn't have to answer for it in public.
Which is why it amuses me that he now tries to operate on r.g.c.c.
behind you, rather than opening himself up directly.
Chris Whittington
> I am more interested in your results of the NOLUT positions.
>
> - Ed -
>
>
Chris Whittington
tor...@ifi.uio.no (Tord Kallqvist Romstad) wrote:
>
> brucemo (bru...@nwlink.com) wrote:
> : FYI: I ran the Kasparov position through a version of Ferret (possibly a bogus
> : version, possibly a great version), and found Bxh7+ with a slightly negative
> ^^^^^^^^^^^^^^^^^^^^^^^^
> : evaluation after approximately 30 hours.
> ^^^^^^^^^^
>
> Negative evaluation??? Does Ferret believe that black is better in this
> position?
Don't knock it !
Many programs do :)
Chris Whittington
Robert Hyatt
Vincent Diepeveen (vdie...@cs.ruu.nl) wrote:
:
: I heard rumors DB should also find it. Why did it play Ng6-e7 then? Bug
: in the program at that time?
Again, right after the game, Hsu or Murray said that DB had seen Bxh7, searched
to a drawn position, and was assuming that Kasparov would play it and go for the
draw. Whether it is a forced win has never been demonstrated that I have seen,
although it'd be nice if someone could do so...
Chris Whittington
>
> : The non-currupt dealer, you are blaming to be corrupt,
> : asked me to tell you this in the newsgroup:
>
> : "I have sent Vincent an official diskversion of Rebel8, paying the
> : price of the program myself.
>
> Good...
>
> : Still I don't understand why Ed himself installs versions of, for
> : example MchessPro 6, at his friends harddisk.
>
> Well tell Jan Louwman that he is not fully informed.
> I always buy atleast 2-3 copies of chess software I use.
>
> One copy for myself, one copy for Jeroen my cooperator.
> No one else get copies...
>
> Do not let yourself being used...
He's young. He'll learn.
> That is stupid.
Probably many more stupidities first though.
One wonders when the devious manipulator will come forward
in person to face the world.
Chris Whittington
mclane
Chris Whittington <chr...@cpsoft.demon.co.uk> wrote:
>vdie...@cs.ruu.nl (Vincent Diepeveen) wrote:
>>
>> In <57ntr9$3...@news.xs4all.nl> "Ed Schrder" <rebc...@xs4all.nl> writes:
>>
>> >From: Chris Whittington <chr...@cpsoft.demon.co.uk>
>> >
>> >> And at last, because I'm really upset now. Dear Chris, don't praise my
>> >> writings if you expect on the other side that I might stay mute in front of
>> >> this behaviour of Ed. If you are not prepared to tell your friend some
>> >> needed truths, other people must do it.
>> >
>> >: Unbelievable !
>> >
>> >: Ed knows very well that I'll say exactly what I think without
>> >: any fear or favour. Don't imagine for one moment that Ed gets
>> >: spared.
>>
>> Well Chris, What is your opinion about the Rebel super killerbook.
Rebel has a killer book ? WHO HAS SUPPOSED THIS ???
IS the guy crazy ????
If Rebel8 plays with Rebel7 book, it still plays 80 points stronger.
So WHAT ?!
REBEL has a killer-book?!
Now I am really that far that I will buy a gun and the next guy that
says such a shame shit will be shot without any further comments....
What the hell do you think your postings do tell us ABOUT YOU ????
Is any program leading the ssdf-list just this high because it has a
KILLER-BOOK. Sorry, but you seem to be crazy!
Look in the mirror when you want to see an idiot!
And post it , but please not here.
>When you've apologised for your earlier bad behaviour (see earlier
>threads), I'll consider answering you.
>Chris Whittington
>> I played a rapid-game Rebel-Genius3. Rebel in book: far over 20 moves.
>> Genius in book: around 10 moves.
>>
>> After his book Genius3 played: ... Qd8-b6 followed by Qb6xb2??
>>
>> Although this was a typical computerline, it was of course in the
>> rebel-book. After getting out of book Rebel was a piece up.
>>
>> I therefore stopped the game.
>>
If you post the game we could see if it is theory of humans, or just
computer-garbage-lines. But you have not posted it!!!
>> Probably not a killerline in the eyes of Ed. A killerline is in the
>> eyes of Ed a line where after getting out of book the opponent is
>> mated, Ed?
>>
Come on , post the game instead blaming Ed.
>> Perhaps i think too simplistic, I don't see difference between a
>> full piece up, so a mate in about 20,
>> instead of an already seen mate in 7 (mcpro).
>>
>> I don't blame anyone writing a killerbook, but don't blame someone
>> else of having a killerbook if you have yourselve the biggest one?
>>
Rebel has the biggest killer book ????
Is this a joke or do you really mean what you write ??!!!??!
It must be a joke. You could not program a chess-program with this
ideas of paranoic scholastic word games.
>> Vincent
>> --
>> +----------------------------------------------------+
>> | Vincent Diepeveen email: vdie...@cs.ruu.nl |
>> | http://www.students.cs.ruu.nl/~vdiepeve/ |
>> +----------------------------------------------------+
I don't accept any further shit like this and will not stop to call it
SHIT until you show us your games you have played or excuse for your
words, as chris has said.
You have said some more of this bullshit before, that diep is a good
program and chess system tal is a weak program. Now you say:
Rebel has a big killer book.
Are you drunken vincent?
Do you smoke cannabis ?
Unbelievable....
Rolf W. Tueschen
>:From: Chris Whittington <chr...@cpsoft.demon.co.uk:>
>>(Rolf Tueschen wrote:)
>> And at last, because I'm really upset now. Dear Chris, don't praise my
>> writings if you expect on the other side that I might stay mute in front of
>> this behaviour of Ed. If you are not prepared to tell your friend some
>> needed truths, other people must do it.
>: Unbelievable !
>: Ed knows very well that I'll say exactly what I think without
>: any fear or favour. Don't imagine for one moment that Ed gets
>: spared.
>(Schroder BV wrote commentary:)
>That's the right spirit Chris... :)
>- Ed -
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Sherlock Holmes might solve this case like this:
First, Rolf complained about the idea to stay mute in front of Schroder BV.
Second, Chris didn't oppose.
Third, Chris wrote he always wrote that he thought without fear. And *Ed*
is knowing about. And Rolf shouldn't imagine that *Ed* gets spanked. Ehh,
spared to get spanked.
Forth, Schroder BV courageously is saying *thank you* for that noble
spirit.
Fifth, Chris is thinking about the difference *kissing death/ kissing ass*.
Holmes' commentary:
If you have a body in the cellar yourself please don't talk about in
public. Or you might also getting spanked.
(In the meantime:)
*Please, Watson, thou givenst me all of Chrissies' critics about Eddie for
the past two months*.
....
*By my needle, shit, sister morphy, you couldn't find any, Watson?*
*No, sir, but if you're pleased with some material about Vincent instead?*
*Watson, stop that nonsense.*
*Pardon me, sir.*
*Now, go and get me the Encycl. Britannica, letters Ass, F and K. And the
usenet etiquette.*
(Meanwhile on the radio they played the old folksong:
*Hit me with your rhythm stick...*)
(Curtain falls without further hesitation.)
Robert Hyatt
Tord Kallqvist Romstad (tor...@ifi.uio.no) wrote:
: brucemo (bru...@nwlink.com) wrote:
: : FYI: I ran the Kasparov position through a version of Ferret (possibly a bogus
: : version, possibly a great version), and found Bxh7+ with a slightly negative
: ^^^^^^^^^^^^^^^^^^^^^^^^
: : evaluation after approximately 30 hours.
: ^^^^^^^^^^
:
: Negative evaluation??? Does Ferret believe that black is better in this
: position?
: Tord
:
A lot of people did. :)
The GM's were split while the game was in progress, but did not find
any win for white during the game. Later, I don't know, but I've not
heard anyone offer a win for white after Bxh7...
:
: : I figure that there must be other programs who can solve this more quickly
Dave Gomboc
In article <57jept$m...@news.xs4all.nl>, Ed Schröder <rebc...@xs4all.nl> wrote:
>From: vdie...@cs.ruu.nl (Vincent Diepeveen)
>
>
>This is a very huge claim!
>
>Besides Kasparov-Deep Blue game 6, 20. Bxh7?! do you have more proof?
>
Even this "proof" has been lacking..
Where is the tree of main variations stemming from 20. Bxh7 that
demonstrates the win? If Diep is constructed for analysis of
difficult positions, I expect that there is some facility in it for
extracting the most important lines of the search (this may be done in
combination with a human, I don't require that the computer understand
exactly which lines the human wants to see, just that the human can
walk the tree and select them and write them in a news posting on the
internet).
I once asked about Bxh7.. Bob Hyatt responded that DB analyzed it to a
draw OTB. If your program analyzed it to a win, then one or the other
is in error.
Dave Gomboc
drgo...@a.stu.athabascau.ca
Stefan Meyer-Kahlen
On 2 Dec 1996 19:45:19 GMT, "Ed Schröder" <rebc...@xs4all.nl> wrote:
>From: vdie...@cs.ruu.nl (Vincent Diepeveen)
>
>: I played a rapid-game Rebel-Genius3. Rebel in book
>: far over 20 moves.: Genius in book
>: around 10 moves.
>
>: After his book Genius3 played: ... Qd8-b6 followed by Qb6xb2??
>: Although this was a typical computerline, it was of course in the
>: rebel-book. After getting out of book Rebel was a piece up.
>: I therefore stopped the game.
>
>: eyes of Ed a line where after getting out of book the opponent is
>: mated, Ed?
>
>he will tell you immediately from which (on paper) opening book your
>example comes from :)
>
>: instead of an already seen mate in 7 (mcpro).
>: I don't blame anyone writing a killerbook, but don't blame someone
>: else of having a killerbook if you have yourselve the biggest one?
In the book of my program Shredder I have many lines like the one you
mentioned above. One side (hopefully not me) is grabbing a poisend
pawn and is out of book. The other side (hopefully me) is still in
book for some moves and will leave the book in a won position. If you
call those lines killer lines, well, that's up to you, but I don't
think that these lines are killer lines in the sense discussed in this
newsgroup. I add those lines to my book because often Shredder is too
stupid to win the good position after the capture, it will be simply
down a pawn. I get those lines from books on opening theory or chess
magazines. A human chess player whould do the same, looking at the
lines given in the book and knowing that the pawn mustn't be captured.
Probably he also would keep the refutation in mind. So what's wrong
about that? Yesterday I've put a very sharp line in the Sicilian
Defense in my book. The line was played at the tournament in Tilburg
and many variations end in a mate.
When you blame somebody for having killer lines in his opening book
(no matter weather you can blame somebody for that or not) you have to
take a very close look at the given line. I definetly didn't add any
line to my book to win against a special program, I add lines to my
book to win against any program.
Stefan
Enrique Irazoqui
> Vincent Diepeveen <vdie...@cs.ruu.nl> wrote
> What is your opinion about the Rebel super killerbook.
> I played a rapid-game Rebel-Genius3. Rebel in book: far over 20 moves.
> Genius in book: around 10 moves.
> After his book Genius3 played: ... Qd8-b6 followed by Qb6xb2??
Dear Vincent,
Could you post this game? I am looking for killer lines in programs and so
far I did not find this one.
Enrique
Vincent Diepeveen
It is a clear win for white. 2 pawns up in ending.
Robert Hyatt
Stefan Meyer-Kahlen (mey...@fmi.uni-passau.de) wrote:
: On 2 Dec 1996 19:45:19 GMT, "Ed Schröder" <rebc...@xs4all.nl> wrote:
: >From: vdie...@cs.ruu.nl (Vincent Diepeveen)
: >
: >: I played a rapid-game Rebel-Genius3. Rebel in book
: >: far over 20 moves.: Genius in book
: >: around 10 moves.
: >
: >: After his book Genius3 played: ... Qd8-b6 followed by Qb6xb2??
:
:
:
:
I, too, have lots of those. On many occasions Crafty exits the book with
an eval of +2 to +3, and wins easily. On other occasions it exits with -2
and loses, simply because it is following some GM game. Hopefully the
learning function is eliminating the latter, and favoring the former. :)
Bob
Vincent Diepeveen
>Chris Whittington <chr...@cpsoft.demon.co.uk> wrote:
>>vdie...@cs.ruu.nl (Vincent Diepeveen) wrote:
>>> In <57ntr9$3...@news.xs4all.nl> "Ed Schrder" <rebc...@xs4all.nl> writes:
>>> >From: Chris Whittington <chr...@cpsoft.demon.co.uk>
>>> >
>>> >> And at last, because I'm really upset now. Dear Chris, don't praise my
>>> >> writings if you expect on the other side that I might stay mute in front of
>>> >> this behaviour of Ed. If you are not prepared to tell your friend some
>>> >> needed truths, other people must do it.
>>> >
>>> >: Unbelievable !
Just few questions:
a) what is your own rating
b) Suppose that i start with a knight up. Do you know that this is
an easy to win game?
If not, then your 'unbelievable' expression is something i understand.
If you DO know that with a knight up win is easy, you should become
more objective.
Vincent Diepeveen
vdie...@cs.ruu.nl
Vincent Diepeveen
1> r3qb1k/1b4p1/p2pr2p/3n4/Pnp1N1N1/6RP/1B3PP1/1B1QR1K1 w - -
Diep v 1.57.13 finds Nxh6 after 51 minutes and 59 seconds at 10 ply.
Branching factor really sucks in this positions. Diep almost dies because
of extensions even at a PP200-32.
2> r4rk1/pp1n1p1p/1nqP2p1/2b1P1B1/4NQ2/1B3P2/PP2K2P/2R5 w - -
It is now analyzing at this position. At the time i left it was
busy searching 12 ply. From ply 9 diep gives
Bf6 (Qb5+ Ke1 Bb4 Kf2 Nxf6 Nxf6 Kg7 Nh5+ with score 0.00)
Let's pray when i come back this evening!
3> r2qk2r/ppp1b1pp/2n1p3/3pP1n1/3P2b1/2PB1NN1/PP4PP/R1BQK2R w - -
4> r1b1kb1r/1p1n1ppp/p2ppn2/6BB/2qNP3/2N5/PPP2PPP/R2Q1RK1 w - -
5> r2qrb1k/1p1b2p1/p2ppn1p/8/3NP3/1BN5/PPP3QP/1K3RR1 w - -
6> rnbqk2r/1p3ppp/p7/1NpPp3/QPP1P1n1/P4N2/4KbPP/R1B2B1R b - -
7> 1r1bk2r/2R2ppp/p3p3/1b2P2q/4QP2/4N3/1B4PP/3R2K1 w - -
8> r3rbk1/ppq2ppp/2b1pB2/8/6Q1/1P1B3P/P1P2PP1/R2R2K1 w - -
9> r4r1k/4bppb/2n1p2p/p1n1P3/1p1p1BNP/3P1NP1/qP2QPB1/2RR2K1 w - -
10> r1b2rk1/1p1nbppp/pq1p4/3B4/P2NP3/2N1p3/1PP3PP/R2Q1R1K w - -
11> r1b3k1/p2p1nP1/2pqr1Rp/1p2p2P/2B1PnQ1/1P6/P1PP4/1K4R1 w - -
>
>ANSWERS (per bm)
>-------
> 1) Nxh6
> 2) Rxc5
> 3) Nxg5
> 4) Nxe6
> 5) e5
> 6) axb5
> 7) Rxd8+
> 8) Bxh7+
> 9) Ng5
>10) Rxf7
>11) Rxh6
>
>Regards. ~~ ~~
> ~~
> ~~ ~~
Vincent Diepeveen
Robert Hyatt
Vincent Diepeveen (vdie...@cs.ruu.nl) wrote:
: In <57v314$q...@juniper.cis.uab.edu> hy...@crafty.cis.uab.edu (Robert Hyatt) writes:
:
: >Vincent Diepeveen (vdie...@cs.ruu.nl) wrote:
: >:
: >: I heard rumors DB should also find it. Why did it play Ng6-e7 then? Bug
: >: in the program at that time?
: >:
: >
: >Again, right after the game, Hsu or Murray said that DB had seen Bxh7, searched
: >to a drawn position, and was assuming that Kasparov would play it and go for the
: >draw. Whether it is a forced win has never been demonstrated that I have seen,
: >although it'd be nice if someone could do so...
: >
: >
:
: It is a clear win for white. 2 pawns up in ending.
:
By who's analysis?
Don Fong
In article <57v314$q...@juniper.cis.uab.edu>,
Robert Hyatt <hy...@crafty.cis.uab.edu> wrote:
>Vincent Diepeveen (vdie...@cs.ruu.nl) wrote:
>:
>: I heard rumors DB should also find it. Why did it play Ng6-e7 then? Bug
>: in the program at that time?
>:
>
>Again, right after the game, Hsu or Murray said that DB had seen Bxh7, searched
>to a drawn position, and was assuming that Kasparov would play it and go for the
>draw. Whether it is a forced win has never been demonstrated that I have seen,
>although it'd be nice if someone could do so...
a few months ago Danny Mozes gave an analysis on his "chess
treasure" site (http://www.netvision.il/~ChessTreasure/)
that seemed to show Bxh7 winning. it would have been swell to see
Deep Blue get smashed that way, but as it was it was pretty awesome
to see Deep Blue get crushed the other way too.
at least the DB side knew when to resign that game.
--
don fong ``i still want the peace dividend''
http://got.net/~dfong/
brucemo
Stefan Meyer-Kahlen wrote:
> When you blame somebody for having killer lines in his opening book
> (no matter weather you can blame somebody for that or not) you have to
> take a very close look at the given line. I definetly didn't add any
> line to my book to win against a special program, I add lines to my
> book to win against any program.
>
> Stefan
Or against anyone else, I assume.
I don't think that including a known bust in your book means that you have
a killer line. There are many lines in ECO that end in +-, I would have
absolutely no qualms about putting any of them in my book.
A killer book is a book that is tailored against the middlegame play of a
specific opponent, in my opinion.
bruce
Ed Schröder
From: hy...@crafty.cis.uab.edu (Robert Hyatt)
Vincent Diepeveen (vdie...@cs.ruu.nl) wrote:
: I heard rumors DB should also find it. Why did it play Ng6-e7 then?
: Bug in the program at that time?
> Again, right after the game, Hsu or Murray said that DB had seen Bxh7,
> searched to a drawn position, and was assuming that Kasparov would play
> it and go for the draw. Whether it is a forced win has never been
> demonstrated that I have seen, although it'd be nice if someone could
> do so...
How about the 2 following variations...
20. Bxh7+ Kxh7 21. Ng5+ Kg8 22. Qh5 Nf5 23. cxd5
a) 23.. Nxe3 24.dxe6 fxe6 25.fxe3
b) 23.. Ncxd4 24.Rxc8 Rxc8 25.dxe6 fxe6 26.Qh7+ Kf8 27.Nxe6+ Nxe6 28.Qxf5+
Both variations look good for white.
Maybe Vincent can display the main variation + score?
And what about CST who finds Bxh7+ in seconds?
Chris?
- Ed -
brucemo
Vincent Diepeveen wrote:
>
> In <57ntr9$3...@news.xs4all.nl> "Ed Schrder" <rebc...@xs4all.nl> writes:
>
> >From: Chris Whittington <chr...@cpsoft.demon.co.uk>
> >
> >> And at last, because I'm really upset now. Dear Chris, don't praise my
> >> writings if you expect on the other side that I might stay mute in front of
> >> this behaviour of Ed. If you are not prepared to tell your friend some
> >> needed truths, other people must do it.
> >
> >: Unbelievable !
> >
> >: any fear or favour. Don't imagine for one moment that Ed gets
> >: spared.
>
> I played a rapid-game Rebel-Genius3. Rebel in book: far over 20 moves.
> Genius in book: around 10 moves.
>
> After his book Genius3 played: ... Qd8-b6 followed by Qb6xb2??
>
> Although this was a typical computerline, it was of course in the
> rebel-book. After getting out of book Rebel was a piece up.
>
> I therefore stopped the game.
>
> Probably not a killerline in the eyes of Ed. A killerline is in the
> eyes of Ed a line where after getting out of book the opponent is
> mated, Ed?
>
> full piece up, so a mate in about 20,
> instead of an already seen mate in 7 (mcpro).
What's the line?
bruce
Chris Whittington
vdie...@cs.ruu.nl (Vincent Diepeveen) wrote:
>
> In <E1t9K...@news.prima.ruhr.de> mcl...@prima.ruhr.de (mclane) writes:
>
> >Chris Whittington <chr...@cpsoft.demon.co.uk> wrote:
> >>> In <57ntr9$3...@news.xs4all.nl> "Ed Schrder" <rebc...@xs4all.nl> writes:
> >>> >From: Chris Whittington <chr...@cpsoft.demon.co.uk>
> >>> >
> >>> >> And at last, because I'm really upset now. Dear Chris, don't praise my
> >>> >> writings if you expect on the other side that I might stay mute in front of
> >>> >> this behaviour of Ed. If you are not prepared to tell your friend some
> >>> >> needed truths, other people must do it.
> >>> >
> >>> >: Unbelievable !
>
> a) what is your own rating
Answer (a)
I got it up to 217 BCF about 25 years ago, and then stopped.
That's around 2340 UK ELO.
Highest published was around 190 BCF or so.
> b) Suppose that i start with a knight up. Do you know that this is
> an easy to win game?
Answer (b)
Bog off matey.
Chris Whittington
> If not, then your 'unbelievable' expression is something i understand.
> If you DO know that with a knight up win is easy, you should become
> more objective.
>
> Vincent Diepeveen
> vdie...@cs.ruu.nl
>
brucemo
Robert Hyatt wrote:
>
> Vincent Diepeveen (vdie...@cs.ruu.nl) wrote:
> :
> : I heard rumors DB should also find it. Why did it play Ng6-e7 then? Bug
> : in the program at that time?
> :
>
> Again, right after the game, Hsu or Murray said that DB had seen Bxh7, searched
> to a drawn position, and was assuming that Kasparov would play it and go for the
> draw. Whether it is a forced win has never been demonstrated that I have seen,
> although it'd be nice if someone could do so...
I let mine go for 30 hours on a P6/200 and it found the move but it was still
slightly minus. Oh well.
I wouldn't be surprised if someone else found it faster, because this seems like a
position in which some obvious extensions that I don't use would help. I let
"Extreme Chess" (Fritz) have a go at it for a day or so on a P5/133, and it didn't
find it (it liked 1. a3, like everyone else, including Kasparov), and that was my
best bet, I think. Maybe some of the others can find it.
Remember that what started this thread off onto this tangent was the claim by
Diepeveen that his program was way better than everyone else at long time
controls, and he used a relatively quick solution of this position to support that
claim.
I read Diep's web page, the first line is "Diep is without a doubt the strongest
chess analysis program at infinite level (few hours a move)". I normally don't
bang on people in here about their programs, but this claim is WAY too audacious
and needs challenging.
In the web page (http://www.students.cs.ruu.nl/~vdiepeve/homepage/about.html),
Diep claims that his program is better because it has a more extensive evaluation
function than other programs -- the hypothesis advanced is that more knowledge
reduces strength drop-off at increased search depths. This is an interesting
idea.
My analysis is that Diep many claims in his web page, some minor and some
about as major as you can get, that the burden of proof is definately upon him to
substantiate these claims, that little or no evidence is advanced to substantiate
these claims, and as a consequence these claims are irresponsible and can easily
be nothing more than wishful thinking.
I would welcome any evidence to the contrary. In the mean time, caveat emptor.
bruce
brucemo
Chris Whittington wrote:
>
> tor...@ifi.uio.no (Tord Kallqvist Romstad) wrote:
> >
> > brucemo (bru...@nwlink.com) wrote:
> > : FYI: I ran the Kasparov position through a version of Ferret (possibly a bogus
> > : version, possibly a great version), and found Bxh7+ with a slightly negative
> > ^^^^^^^^^^^^^^^^^^^^^^^^
> > : evaluation after approximately 30 hours.
> > ^^^^^^^^^^
> >
> > Negative evaluation??? Does Ferret believe that black is better in this
> > position?
>
>
> Many programs do :)
Whoops, I forgot that I already posted something about the 30-hour
pseudo-solution.
Yeah, it thinks black is better, and did during the game, too. So did Fritz,
which was doing analysis on-site, and I gather that it came under some ridicule.
I will try to explain why my program would rather play black, although when you
read this please bear in mind that I haven't verified this by taking apart my
program, and I do realize that what I'm saying is stupid, but that's computer
chess for you, at least my variety. It's embarassing to do this, but maybe
someone will see my point.
In the given position, white's bishop is on b1, which is a hard square for a
program to evaluate. You want your bishops off the back row, because you don't
want to impede the rooks, and because you want more potential mobility for the
piece, but obviously in this position the truth is that b1 is a great place for
the bishop because it is out of the way, doesn't impede a rook, is going to
operate on b1..h7 regardless, and can be part of a Bb1 + Qc2 (or Qd3) battery.
My program does a bad job evaluating Bb1.
White's other bishop is on b2, which is a fine place for a bishop, but it's
biting into a ram, albeit one that's being levered. This piece is probably fine,
and my program would probably understand this.
White has moved pawns in front of his king, which is a no-no. Furthermore he's
moved the g2 pawn and there is no bishop replacing it there. Of course black
can't take advantage of this, but we're talking stupid, remember? My program
would blow it here, this may be most of the source of the problem with evaluating
this position.
I don't really understand the rest of the pawn structure, it looks like white can
make a majority q-side, but then d4 becomes static, and the b2 bishop really does
start to look bad. My program may or may not understand any of this, it probably
understands at least some of it.
Black's knight should be on f6 rather than e7, and that bishop is going to get
chased around to c7 possibly, but otherwise black is structurally fine. If you
take white's pieces off the board, and just look at black's pieces, they are all
on perfectly acceptable squares, and black has no static weaknesses whatsoever.
His bishop is even "good".
The problem with this is that it doesn't correspond well to the truth of the
position, in which black is really very well restrained, and white is ready to
attack on both wings at the same time.
My program evaluates one position at a time, very quickly. I try to put as much
knowledge into the leaves as I can, and in fact my program knows a lot of things
it didn't a few years ago, and goes several times faster now. But the fact
remains that one position evaluated is like one brain cell, it can't do a lot,
but hopefully if you collect a bunch of them together you will create something
that is capable of appearing intelligent -- you hope that the whole will be
greater than the sum of its parts, or at least will be approximately equal to the
sum of its parts. Occasionally, though, all of the brain cells are dumb in the
same way, and you end up creating something that drives way too slow in the left
lane of the freeway, and you just want to shoot it, and everyone laughs at you.
bruce
brucemo
>
> This is a very huge claim!
>
> Besides Kasparov-Deep Blue game 6, 20. Bxh7?! do you have more proof?
>
> How much scores Diep at the NOLUT test?
>
> Ok, I don't like these test sets either as being the absolute truth on
> the playing strength of a chess program but I still believe these tests
> do give an *INDICATION*
>
> If program X scores 2200 on the LCT-II test, program X does (probably)
> not belong to the top. If program X scores 2500 it *probably* belongs
> to the top.
>
> Looking forward for your results.
>
> - Ed Schroder -
>
> Here are the 11 NOLUT positions...
> r3qb1k/1b4p1/p2pr2p/3n4/Pnp1N1N1/6RP/1B3PP1/1B1QR1K1 w - - bm Nxh6;
> r4rk1/pp1n1p1p/1nqP2p1/2b1P1B1/4NQ2/1B3P2/PP2K2P/2R5 w - - bm Rxc5;
> r2qk2r/ppp1b1pp/2n1p3/3pP1n1/3P2b1/2PB1NN1/PP4PP/R1BQK2R w - - bm Nxg5;
> r1b1kb1r/1p1n1ppp/p2ppn2/6BB/2qNP3/2N5/PPP2PPP/R2Q1RK1 w - - bm Nxe6;
> r2qrb1k/1p1b2p1/p2ppn1p/8/3NP3/1BN5/PPP3QP/1K3RR1 w - - bm e5;
> rnbqk2r/1p3ppp/p7/1NpPp3/QPP1P1n1/P4N2/4KbPP/R1B2B1R b - - bm axb5;
> 1r1bk2r/2R2ppp/p3p3/1b2P2q/4QP2/4N3/1B4PP/3R2K1 w - - bm Rxd8+;
> r3rbk1/ppq2ppp/2b1pB2/8/6Q1/1P1B3P/P1P2PP1/R2R2K1 w - - bm Bxh7+;
> r4r1k/4bppb/2n1p2p/p1n1P3/1p1p1BNP/3P1NP1/qP2QPB1/2RR2K1 w - - bm Ng5;
> r1b2rk1/1p1nbppp/pq1p4/3B4/P2NP3/2N1p3/1PP3PP/R2Q1R1K w - - bm Rxf7;
> r1b3k1/p2p1nP1/2pqr1Rp/1p2p2P/2B1PnQ1/1P6/P1PP4/1K4R1 w - - bm Rxh6;
>
> BTW, I tested these 11 NOLUT positions once overnight with Rebel8.
> Level 30:00 fixed time.
> Machine PP-200
>
> Result: 0 (this reads as NUL) positions found!!
>
> I am curious about the results of Diep.
We haven't heard anything from Diep, but here's something that should show that these
positions aren't impossible. I believe I can show that #11 is solvable in reasonable
time, and I'm going to try some of the other ones soon, too (position #10 is
promising, if I remember right). I think you should have gone for one hour per
position, Ed, you might have gotten a couple of them.
Here's a snippet from the original post on these, from two and a half years ago:
> From: m...@cnam.cnam.fr (Marc-Francois Baudot)
> Subject: Pierre Nolot's solutions to the 11 positions
> Date: 29 Jul 1994 11:55:22 GMT
> Organization: Conservatoire National des Arts et Metiers, Paris, France
>
> [snip]
>
> # Position: 11
> # Move: W
>
> r.b...k. Teichmann - NN, Zurich 1920
> p..p.nP. This was a trap, as the game was spectacularly won because
> ..pqr.Rp of black's weak play. Several authors have published this game
> .p..p..P without checking anything. For instance, Neishtadt in his
> ..B.PnQ. excellent book "Lecons de tactique" (sorry, I don't know
> .P...... the title of the english version) gives :
> P.PP.... "1.Rxh6!! the g pawn will be helped by the h pawn 1...Nxh6
> .K....R. (1...Rxh6 2.Bxf7+ Rxf7 3.g8Q+) 2.Qg5 Nf7 3.Qd8+!! Nxd8
> 4.h6 and there is nothing to prevent h6-h7+ 1-0"
> Right, nothing can prevent h6-h7+, but it is not decisive!
> 4...Qd4!! 5.h7+ Kf7 6.g8Q+ Ke7 7.h8Q Kd6 8.Rg7 Qxd2!! 9.Qxd8 Rc5
> 10.Rxd7 Bxd7 11.Qxa8 Rb4 12.Qxa7 Qe1 13.Kb2 Qc3 1/2 as white cannot
> escape the black checks.
Here's my program's output (P6/200) starting at ply 12 (earlier plies snipped to save
space). Rxh6 was found in ply 13 after 14 minutes 24 seconds, resolved as -2.05 pawns
after 15:41, and ended up being down a pawn at the end. I haven't looked at the
variations produced, other that to verify that most of the important ideas present in
the snippet above are seemingly present in the later PV's below.
bruce
--------
PV 00:02:43.665 12 -382 [wrong] Bxe6 Nxe6 Rf6 Qe7 Qg6 Ba6 Rf5 b4 Rg3 Nf4 Qg4 d5
PV 00:08:16.624 13 -332 [wrong] Bxe6 Nxe6 Rf6 Qe7 Qg6 Ba6 Rf5 b4 Rg3 Nf4 Qg4 d5
PV 00:09:35.757 13 -277 [wrong] Bxe6 Nxe6 Qh4 Qc5 Rf1 Nfg5 Rxh6 d5 exd5 cxd5 Qg3
Kxg7 Rg6+ Kh7 Qxe5
PV 00:14:23.902 13 -227 [right] Rxh6
PV 00:15:41.083 13 -205 [right] Rxh6 Nxh6 Qg5 Nf7 Qd8+ Nxd8 h6 Qe7 h7+ Kxh7 g8=Q+
Kh6 Qh8+ Qh7 Rh1+ Nh5 Rxh5+ Kxh5 Qxh7+ Rh6 Be2+
Kg5 Qg7+ Rg6
PV 00:20:00.195 14 -155 [right] Rxh6 Nxh6 Qg5 Nf7 Qd8+ Nxd8 h6 Qe7 h7+ Kxh7 g8=Q+
Kh6 Qh8+ Qh7 Rh1+ Nh5 Rxh5+ Kxh5 Qxh7+ Rh6 Be2+
Kg5 Qg7+ Rg6
PV 00:22:19.225 14 -143 [right] Rxh6 Nxh6 Qg5 Nf7 Qd8+ Nxd8 h6 Qe7 h7+ Kxh7 g8=Q+
Kh6 Qh8+ Qh7 Rh1+ Nh5 Rxh5+ Kxh5 Qxh7+ Kg5 Bxe6
Nxe6 Qf5+ Kh4
PV 00:33:13.246 15 -93 [right] Rxh6 Nxh6 Qg5 Nf7 Qd8+ Nxd8 h6 Qe7 h7+ Kxh7 g8=Q+
Kh6 Qh8+ Qh7 Rh1+ Nh5 Rxh5+ Kxh5 Qxh7+ Kg5 Bxe6
Nxe6 Qf5+ Kh4
PV 00:40:05.729 15 -59 [right] Rxh6 Nxh6 Qg5 Nf7 Qd8+ Nxd8 h6 Qd4 h7+ Kf7 g8=Q+
Ke7 h8=Q Kd6 Bxe6 Ndxe6 d3 Kc7
PV 01:09:11.569 16 -26 [right] Rxh6 Nxh6 Qg5 Nf7 Qd8+ Nxd8 h6 Qe7 h7+ Kxh7 g8=Q+
Kh6 Bxe6 Ndxe6 Qh8+ Qh7 Qf6+ Qg6 Rh1+ Nh5 Qh8+
Qh7 Qf6+ Qg6
PV 03:39:05.772 17 -134 [right] Rxh6 Nxh6 Qg5 Nf7 Qd8+ Nxd8 h6 Qd4 h7+ Kf7 Bxe6+
Ndxe6 g8=Q+ Ke7 h8=Q Qxe4 d3 Qf5 Qhh7+ Qxh7 Qxh7+
Kf6 Qh8+
Jouni Uski
I think, that final conclusion from different sources
was, that Bxh7 is draw (ALSO MY OWN RESULT).
Jouni Uski
Rolf W. Tueschen
mcl...@prima.ruhr.de (mclane) wrote:
R.W. Tueschen:
>>> >> And at last, because I'm really upset now. Dear Chris, don't praise my
>>> >> writings if you expect on the other side that I might stay mute in front of
>>> >> this behaviour of Ed. If you are not prepared to tell your friend some
>>> >> needed truths, other people must do it.
Chris:
>>> >: Unbelievable !
>>> >
>>> >: Ed knows very well that I'll say exactly what I think without
>>> >: any fear or favour. Don't imagine for one moment that Ed gets
>>> >: spared.
Vincent:
>>> Well Chris, What is your opinion about the Rebel super killerbook.
Thorsten:
>Rebel has a killer book ? WHO HAS SUPPOSED THIS ???
>IS the guy crazy ????
>REBEL has a killer-book?!
>Now I am really that far that I will buy a gun and the next guy that
>says such a shame shit will be shot without any further comments....
> Sorry, but you seem to be crazy!
>Look in the mirror when you want to see an idiot!
>And post it , but please not here.
>Rebel has the biggest killer book ????
>Is this a joke or do you really mean what you write ??!!!??!
>It must be a joke. You could not program a chess-program with this
>ideas of paranoic scholastic word games.
>You have said some more of this bullshit before, that diep is a good
>program and chess system tal is a weak program. Now you say:
>Rebel has a big killer book.
>Are you drunken vincent?
>Do you smoke cannabis ?
>Unbelievable....
----------------------------------------------------------------------------------------------------------
Idiot, bull, paranoic, drunken, cannabis.
Thorsten think about your model function with all your experience since
'79.
Hope you're not engaged as *Wadenbeisser* for Eddie. You could do him no
worse favour if you're supporting his always ready goasts/nightmares. If
you know what I mean.
Maybe you are right in your judgement, but please think of the worldwide
public you publish for. And the children everywhere. What if you're not
informed about Eddie's latest performances?
I would like to read more of your easy satirical store instead. :)
Rolf Tueschen alias Thorsten II. :)
Tord Kallqvist Romstad
brucemo (bru...@nwlink.com) wrote:
I don't laugh any more. I tested my own program in this position, and it agrees
with Ferret: Black is slightly better! By the way, my program plays 20.a3.
Tord
: bruce
Chris Whittington
"Ed Schröder" <rebc...@xs4all.nl> wrote:
>
> From: hy...@crafty.cis.uab.edu (Robert Hyatt)
>
> Vincent Diepeveen (vdie...@cs.ruu.nl) wrote:
>
> : I heard rumors DB should also find it. Why did it play Ng6-e7 then?
> : Bug in the program at that time?
>
> > Again, right after the game, Hsu or Murray said that DB had seen Bxh7,
> > searched to a drawn position, and was assuming that Kasparov would play
> > it and go for the draw. Whether it is a forced win has never been
> > demonstrated that I have seen, although it'd be nice if someone could
> > do so...
>
>
> 20. Bxh7+ Kxh7 21. Ng5+ Kg8 22. Qh5 Nf5 23. cxd5
>
> a) 23.. Nxe3 24.dxe6 fxe6 25.fxe3
> b) 23.. Ncxd4 24.Rxc8 Rxc8 25.dxe6 fxe6 26.Qh7+ Kf8 27.Nxe6+ Nxe6 28.Qxf5+
>
> Both variations look good for white.
>
> Maybe Vincent can display the main variation + score?
>
> And what about CST who finds Bxh7+ in seconds?
> Chris?
I've been keeping quiet because I don't have the PGN game score :(
If anybody posts it me, I'll try the latest version of CST ...
Then we can compare with Droop :)
Chris Whittington
>
> - Ed -
>
>
Vincent Diepeveen
>Vincent Diepeveen (vdie...@cs.ruu.nl) wrote:
>: In <57v314$q...@juniper.cis.uab.edu> hy...@crafty.cis.uab.edu (Robert Hyatt) writes:
>:
>: >Vincent Diepeveen (vdie...@cs.ruu.nl) wrote:
>: >:
>: >: I heard rumors DB should also find it. Why did it play Ng6-e7 then? Bug
>: >: in the program at that time?
>: >:
>: >
>: >Again, right after the game, Hsu or Murray said that DB had seen Bxh7, searched
>: >to a drawn position, and was assuming that Kasparov would play it and go for the
>: >draw. Whether it is a forced win has never been demonstrated that I have seen,
>: >although it'd be nice if someone could do so...
>: >
>: >
>:
>: It is a clear win for white. 2 pawns up in ending.
>:
>:
>
>By who's analysis?
>
Mine combined with Diep. GrandMaster class analysis thus.
Why don't you try to defend it:
I play
1. Bh7,Kg8 2.Ng5
Now there are 2 possibilities
a) Kg6
b) Kg8
After b 2...,Kg8 i play 3.Qh5
Vincent
Vincent Diepeveen
In <01bbe111$a3c4d4e0$f3ec...@10.0.1.1> "Enrique Irazoqui" <en...@lix.intercom.es> writes:
>> Vincent Diepeveen <vdie...@cs.ruu.nl> wrote
>
>> What is your opinion about the Rebel super killerbook.
>> I played a rapid-game Rebel-Genius3. Rebel in book: far over 20 moves.
>> Genius in book: around 10 moves.
>
>> After his book Genius3 played: ... Qd8-b6 followed by Qb6xb2??
>
>Dear Vincent,
>
>Could you post this game? I am looking for killer lines in programs and so
>far I did not find this one.
I aborted the game, but i'll look up the line for you this evening.
If you have an auto232player, then you see these lines regurarly.
For example: Fritz3 plays next line sometimes when playing blitz against
Diep:
1. d4,d5 2. Bf4,Nf6 3. c3
or
1. d4,d5 2. Nf3,Nf6 3. c3
Up to version 1.57.1 Diep didn't know what to do in this position
It usually played 3...,Nc6 after which e3 followed by Bb5 gave fritz
a nice position where programs usually go wrong.
>Enrique
Vincent Diepeveen
In <32A4BA...@nwlink.com> brucemo <bru...@nwlink.com> writes:
>Robert Hyatt wrote:
>>
>> Vincent Diepeveen (vdie...@cs.ruu.nl) wrote:
>> :
>> : I heard rumors DB should also find it. Why did it play Ng6-e7 then? Bug
>> : in the program at that time?
>> :
>>
>> Again, right after the game, Hsu or Murray said that DB had seen Bxh7, searched
>> to a drawn position, and was assuming that Kasparov would play it and go for the
>> draw. Whether it is a forced win has never been demonstrated that I have seen,
>> although it'd be nice if someone could do so...
>
Well, i have a lot of years to prove this... ...just busy 3 years, and
already Diep solving 3 out of the 11 nolut positions. Not bad for a 3year old.
Vincent
Don Fong
In article <58269h$9...@news.xs4all.nl>, Ed Schröder <rebc...@xs4all.nl> wrote:
>From: hy...@crafty.cis.uab.edu (Robert Hyatt)
>
>
>: I heard rumors DB should also find it. Why did it play Ng6-e7 then?
>: Bug in the program at that time?
>
>> Again, right after the game, Hsu or Murray said that DB had seen Bxh7,
>> searched to a drawn position, and was assuming that Kasparov would play
>> it and go for the draw. Whether it is a forced win has never been
>> demonstrated that I have seen, although it'd be nice if someone could
>> do so...
>
i'm no master, but it seems to me that the consensus of the
post-game analysis is that Bxh7 wins for White. many people have
published lines purporting to show a win. AFAIK, no one except
the DB team claimed it was a draw. it would be nice if they
(DB team) would publish their drawing line, assuming they still
believe in it.
Vincent Diepeveen
>it didn't a few years ago.
Without going too much in details: how many times a second can your program
evaluate at this moment (and at what hardware)?
>remains that one position evaluated is like one brain cell, it can't do a lot,
>but hopefully if you collect a bunch of them together you will create something
>that is capable of appearing intelligent -- you hope that the whole will be
>greater than the sum of its parts, or at least will be approximately equal to the
>sum of its parts. Occasionally, though, all of the brain cells are dumb in the
>same way, and you end up creating something that drives way too slow in the left
>lane of the freeway, and you just want to shoot it, and everyone laughs at you.
>
The problem of the dumb cells is that you need complete knowledge.
The knowledge must contain all areas a little.
If you forget to evaluate open files, then poor you! It will prefer to
push a pawn instead of trying to get a rook on the only open file there is,
as it does get a lot of bonuses when a pawn is a field further, but doesn't
get a thing for that rook (except mobility bonuses).
>bruce
The advantage of many dumb cells is that you can have 2nd order dumb cells
that can create order in the dumb cells giving you more oversight.
Also the majority of dumb cells will always correct a wrong cell.
Another thing: thousands of small bonuses are worth way more than a queen.
Slowly this happens to Diep: it sees very quickly compensation, which looks
very human.
I guess this is the reason it finds for example Rxf7 at 10 ply in the
NOLUT positions. Diep does not use many extensions. Kallisto for example
is way faster and also searches brute force, and much deeper. Still it
has problems seeing these kind of combinations as only few
evaluation parameters must do the job.
Only very few extensions Diep uses, and
a mating extension like certain programs have Diep doesn't have.
Also i think kings evaluation of Diep still sucks compared to
for example bishop code.
Already now the sum of all the small bonuses and penalties give you a good
evaluation, which doesn't care that much about material, and does give
quit a good mobility, although the partial calculation of it is not that
difficult (but systemtime consuming).
Vincent Diepeveen
>> How much scores Diep at the NOLUT test?
>> Ok, I don't like these test sets either as being the absolute truth on
>> the playing strength of a chess program but I still believe these tests
>> do give an *INDICATION*
all on a PP200
>> Here are the 11 NOLUT positions...
>> r3qb1k/1b4p1/p2pr2p/3n4/Pnp1N1N1/6RP/1B3PP1/1B1QR1K1 w - - bm Nxh6;
51 minutes, 10 ply.
>> r4rk1/pp1n1p1p/1nqP2p1/2b1P1B1/4NQ2/1B3P2/PP2K2P/2R5 w - - bm Rxc5;
Diep needs 16 ply for this one, too long.
>> r2qk2r/ppp1b1pp/2n1p3/3pP1n1/3P2b1/2PB1NN1/PP4PP/R1BQK2R w - - bm Nxg5;
too difficult, although score after moving Nxg5 is only -0.3 for white,
and the main variation (Bxg5) is about 0.4
>> r1b1kb1r/1p1n1ppp/p2ppn2/6BB/2qNP3/2N5/PPP2PPP/R2Q1RK1 w - - bm Nxe6;
18 ply, or i need to adjust some parameters. Nxe6 is +0.59 but
the move Diep wants to play it optimistically evaluates +0.80
>> r2qrb1k/1p1b2p1/p2ppn1p/8/3NP3/1BN5/PPP3QP/1K3RR1 w - - bm e5;
This one is also in LCT test. 16 ply is needed to find this, Diep doesnt
get 16 ply in this position.
>> rnbqk2r/1p3ppp/p7/1NpPp3/QPP1P1n1/P4N2/4KbPP/R1B2B1R b - - bm axb5;
This is a strange position: after few seconds Diep wants to play axb5
on 6 and 7 ply. After this it gets a fail low. Score of axb5 drops to
about -2.80 and Diep switches to a different move. I analyzed this
position, and i'm not able to win this position with the black pieces.
Just like Diep gives: axb5 Qxa8 and next move white plays h3 and
captures the bishop if knight leaves. How to win this? Diep doesn't
see it, and i don't see it. Diep wins this position against me over and
over. I don't see the winning idea not to tell winning variation of
black. After axb5 black just looses a rook for some vague compensation.
>> 1r1bk2r/2R2ppp/p3p3/1b2P2q/4QP2/4N3/1B4PP/3R2K1 w - - bm Rxd8+;
This position should be no problem in future. Just give that rook at h8
a huge penalty for not participating in the game. Currently this is evaluated
'fuzzy logically' (so only a modest penalty).
After 16 seconds Diep wants to play Rxd8, as it thinks Rxd8 better than
without chances loosing the game. After few minutes at 8 ply however it
sees that Rb7 is a draw (score absolute draw: 0.00), and chooses for Rb7
>> r3rbk1/ppq2ppp/2b1pB2/8/6Q1/1P1B3P/P1P2PP1/R2R2K1 w - - bm Bxh7+;
This is tricky one! 14/15 plies for Diep i analyzed. So not within few hours.
I need to make branching factor better first in this kind of position,
or i need to add mate extension!
What programs containing mating extension do find this move?
>> r4r1k/4bppb/2n1p2p/p1n1P3/1p1p1BNP/3P1NP1/qP2QPB1/2RR2K1 w - - bm Ng5;
This one is too deep for Diep. At least 17 plies.
>> r1b2rk1/1p1nbppp/pq1p4/3B4/P2NP3/2N1p3/1PP3PP/R2Q1R1K w - - bm Rxf7;
This is an easy one for Diep. 10 ply and 10 minutes and 39 seconds.
>> r1b3k1/p2p1nP1/2pqr1Rp/1p2p2P/2B1PnQ1/1P6/P1PP4/1K4R1 w - - bm Rxh6;
Diep also finds this one at 13 ply. Diep was calculating at
11 ply after 10 minutes when i left home.
So Diep solves 3 positions. The positions solved it solved quite quickly.
More memory will certainly overcome the branching factor at the great depths.
Few positions can be solved when king safety becomes better or when
a mating extension is used.
>>
>> BTW, I tested these 11 NOLUT positions once overnight with Rebel8.
>> Level 30:00 fixed time.
>> Machine PP-200
>>
>> Result: 0 (this reads as NUL) positions found!!
>>
>> I am curious about the results of Diep.
>
>We haven't heard anything from Diep, but here's something that should show that these
>positions aren't impossible. I believe I can show that #11 is solvable in reasonable
>time,
Yes, Rxf7 is a children trick.
So is Rxh6. It is hard to see of course because of the delaying moves,
but the idea: Rxh6, Nxh6 Qg5, Nf7 Qd8, Nd8 h6 and h7 is quite easy.
Another thing that just can be brought back to simplistic tactics.
Why doesn't rebel see this? Fritz? I mean: there must be a bug in the
programs if they don't find #11. After few moves you simply put mate,
or win a bunch of material.
Has this to do with forward pruning, and/or dubious improving alfa with
nullmove or improving alfa and beta with hashtables?
Vincent.
and I'm going to try some of the other ones soon, too (position #10 is
>promising, if I remember right).
Yes, 10 minutes for Diep.
Robert Hyatt
Vincent Diepeveen (vdie...@cs.ruu.nl) wrote:: >>
: >> :
: >> : I heard rumors DB should also find it. Why did it play Ng6-e7 then? Bug
: >> : in the program at that time?
: >> :
: >>
: >> Again, right after the game, Hsu or Murray said that DB had seen Bxh7, searched
: >> to a drawn position, and was assuming that Kasparov would play it and go for the
: >> draw. Whether it is a forced win has never been demonstrated that I have seen,
: >> although it'd be nice if someone could do so...
: >I let mine go for 30 hours on a P6/200 and it found the move but it was still
They aren't "solved" at all, until you can play the move OTB. Probably several
programs can solve some of those positions at 12-24 hours per move. But that
helps not one iota in a game OTB of course, unless perhaps it is correspondence.
brucemo
Ed Schröder wrote:
>
> From: vdie...@cs.ruu.nl (Vincent Diepeveen)
>
> : Diep is by far the strongest correspondence/analysis program.
>
> This is a very huge claim!
>
> Besides Kasparov-Deep Blue game 6, 20. Bxh7?! do you have more proof?
>
> How much scores Diep at the NOLUT test?
>
> Ok, I don't like these test sets either as being the absolute truth on
> the playing strength of a chess program but I still believe these tests
> do give an *INDICATION*
>
> not belong to the top. If program X scores 2500 it *probably* belongs
> to the top.
>
> Looking forward for your results.
>
> - Ed Schroder -
>
> r3qb1k/1b4p1/p2pr2p/3n4/Pnp1N1N1/6RP/1B3PP1/1B1QR1K1 w - - bm Nxh6;
>
> Level 30:00 fixed time.
> Machine PP-200
>
> Result: 0 (this reads as NUL) positions found!!
>
> I am curious about the results of Diep.
Position 10 was harder than position 11, but is still solvable "overnight" on
a modern microcomputer. Ferret found this one in ten hours fifty-two minutes
on a P6/200, in ply 15.
Scientific honesty compels me to state that it may not have held this through
ply 16, as it was failing low when I stopped it. I didn't want to wait
another day or two for the fail-low to resolve.
I'm going to try position #5 next.
bruce
----------------
> From: m...@cnam.cnam.fr (Marc-Francois Baudot)
> Subject: Pierre Nolot's solutions to the 11 positions
> Date: 29 Jul 1994 11:55:22 GMT
> Organization: Conservatoire National des Arts et Metiers, Paris, France
>
> # Position: 10
> # Move: W
>
> r.b..rk. Van der Wiel - Ribli Amsterdam 1980.
> .p.nbppp Ribli is a solid player, but he didn't see this one
> pq.p.... coming :
> ...B.... 15.Rxf7!! Rxf7 16.Bxf7 Kxf7 17.Qh5+ Kg8 16.Qe8+ Bf8
> P..NP... 19.Nd5 Qxd4 20.Nxe7+ Kh8 21.Rf1 Qf6 22.Rxf6 and white on
> ..N.p... on move 29.
> .PP...PP This is propably the easiest one, with number 5.
> R..Q.R.K
PV 00:09:34.906 12 -36 [wrong] Nf5 Bg5 Nxd6 Ne5 Bb3 h6 Qd5 Bf6 a5 Qc7 Nxc8
Raxc8 Rad1
PV 00:29:08.544 13 -68 [wrong] Nf5 Bg5 Bb3 Nc5 Nxd6 Nxb3 cxb3 Be6 Nd5 Qc6
Nf5 Rfe8 Nfxe3 Bxe3
PV 00:39:30.989 13 -26 [wrong] Bb3 Bg5 Nd5 Qa7 Qh5 Qxd4 Rxf7 Rxf7 Ne7+ Kh8
Ng6+ Kg8 Ne7+
PV 01:28:26.119 14 -28 [wrong] Bb3 Bg5 Nd5 Qa7 Qh5 g6 Qxg5 Qxd4 Nxe3 Qxb2
Nf5 Qe5 Nh6+ Kg7 Nxf7
PV 03:43:06.769 14 -10 [wrong] Bxf7+ Rxf7 Rxf7 Bf6 Rxf6 Nxf6 Qd3 Ng4 Kg1
Qxb2 Rb1 Qa3 h3 Ne5 Qxe3
PV 04:31:45.686 15 -9 [wrong] Bxf7+ Rxf7 Rxf7 Bf6 Rxf6 Nxf6 Qd3 Ng4 Kg1
Qxb2 Rf1 g6 Rb1 Qa3 h3 Ne5
PV 10:52:06.991 15 40 [right] Rxf7
PV 11:40:57.135 15 77 [right] Rxf7 Rxf7 Bxf7+ Kxf7 Qh5+ Kg8 Nd5 Qxd4 Qe8+
Bf8 Ne7+ Kh8 Rf1 Qf6 Rxf6 gxf6 Kg1 Kg7 Nxc8
Ne5
Ed Schröder
>
>: I heard rumors DB should also find it. Why did it play Ng6-e7 then?
>: Bug in the program at that time?
>
>> Again, right after the game, Hsu or Murray said that DB had seen Bxh7,
>> searched to a drawn position, and was assuming that Kasparov would play
>> it and go for the draw. Whether it is a forced win has never been
>> demonstrated that I have seen, although it'd be nice if someone could
>> do so...
>
>
>
>a) 23.. Nxe3 24.dxe6 fxe6 25.fxe3
>b) 23.. Ncxd4 24.Rxc8 Rxc8 25.dxe6 fxe6 26.Qh7+ Kf8 27.Nxe6+ Nxe6 28.Qxf5+
>
>Both variations look good for white.
>
>Maybe Vincent can display the main variation + score?
>
>And what about CST who finds Bxh7+ in seconds?
>Chris?
>
I did an overnight analysis with Rebel8.
With "Power Analyse" I entered Bxh7+ as only move to analyse.
At ply=11 Rebel claims a pawn up for white.
But surprisingly at ply 12 and 13 the score dropped to -0.16 and -0.01
- Ed -
brucemo
Don Fong wrote:
>
> In article <58269h$9...@news.xs4all.nl>, Ed Schröder <rebc...@xs4all.nl> wrote:
> >From: hy...@crafty.cis.uab.edu (Robert Hyatt)
> >
> >Vincent Diepeveen (vdie...@cs.ruu.nl) wrote:
> >
> >: I heard rumors DB should also find it. Why did it play Ng6-e7 then?
> >: Bug in the program at that time?
> >
> >> Again, right after the game, Hsu or Murray said that DB had seen Bxh7,
> >> searched to a drawn position, and was assuming that Kasparov would play
> >> it and go for the draw. Whether it is a forced win has never been
> >> demonstrated that I have seen, although it'd be nice if someone could
> >> do so...
> >
> >How about the 2 following variations...
>
> i'm no master, but it seems to me that the consensus of the
> post-game analysis is that Bxh7 wins for White. many people have
> published lines purporting to show a win. AFAIK, no one except
> the DB team claimed it was a draw. it would be nice if they
> (DB team) would publish their drawing line, assuming they still
> believe in it.
I've read here that Deep Blue analyzed it to a draw during the game.
Figuring out that your opponent can force a draw is what you do the move
before you discover that you are mated.
So unless you are saying that the Deep Blue guys claimed it was a draw AFTER
the game, it was probably a mistake.
Remember, that if a program is searching D plies deep on a position, that for
any position N plies into its PV, the program is really searching D minus N
plies deep on THAT position, and if N is large, even if D is also large, D
minus N is dinky. So a mistake near the end of the PV is common, and in
these cases a mistake is often the difference between drawing and losing,
since the lines are so forcing that there is no way to avoid the position you
have critically mis-evaluated.
I think Bob is going down the wrong line here, this argument should not
proceed so rigidly, I would really hate for people to get the idea that, if
Bxh7+ wins for white, Diep is somehow proven to be the best long time
control program in the world.
My goal is to get Mr. Diepeveen to submit evidence to substantiate his claim,
either that or to stop making this claim. I have seen little evidence so
far, certainly not enough that we should declare him to be world champion by
consensus. We don't know that Diep has ever won a game at a long time
control.
bruce
Tim Mirabile
Chris Whittington <chr...@cpsoft.demon.co.uk> wrote:
>"Ed Schröder" <rebc...@xs4all.nl> wrote:
>> And what about CST who finds Bxh7+ in seconds?
>> Chris?
>
>I've been keeping quiet because I don't have the PGN game score :(
>
>If anybody posts it me, I'll try the latest version of CST ...
>
>Then we can compare with Droop :)
[Event "ACM m"]
[Site "Philadelphia, USA"]
[Date "1996.??.??"]
[Round "6"]
[White "Kasparov,G"]
[Black "Deep Blue"]
[Result "1-0"]
1. Nf3 d5 2. d4 c6 3. c4 e6 4. Nbd2 Nf6 5. e3 c5 6. b3 Nc6 7. Bb2 cxd4 8. exd4
Be7 9. Rc1 0-0 10. Bd3 Bd7 11. 0-0 Nh5 12. Re1 Nf4 13. Bb1 Bd6 14. g3 Ng6 15.
Ne5 Rc8 16. Nxd7 Qxd7 17. Nf3 Bb4 18. Re3 Rfd8 19. h4 Nge7 20. a3 Ba5 21. b4 Bc7
22. c5 Re8 23. Qd3 g6 24. Re2 Nf5 25. Bc3 h5 26. b5 Nce7 27. Bd2 Kg7 28. a4 Ra8
29. a5 a6 30. b6 Bb8 31. Bc2 Nc6 32. Ba4 Re7 33. Bc3 Ne5 34. dxe5 Qxa4 35. Nd4
Nxd4 36. Qxd4 Qd7 37. Bd2 Re8 38. Bg5 Rc8 39. Bf6+ Kh7 40. c6 bxc6 41. Qc5 Kh6
42. Rb2 Qb7 43. Rb4 1-0
--
Tim Mirabile <t...@mail.htp.com> - http://www.webcom.com/timm/
Visit my homepage for information on USCF & FIDE rated chess on Long Island.
TimM on the Free Internet Chess Server - telnet://fics.onenet.net:5000/
ICD/Your Move Chess & Games - http://www.icdchess.com/
The opinions of my employers are not necessarily mine, and vice versa.
Vincent Diepeveen
>Vincent Diepeveen (vdie...@cs.ruu.nl) wrote:
>: In <32A4BA...@nwlink.com> brucemo <bru...@nwlink.com> writes:
>:
>: Well, i have a lot of years to prove this... ...just busy 3 years, and
>: already Diep solving 3 out of the 11 nolut positions. Not bad for a 3year old.
>:
>
>They aren't "solved" at all, until you can play the move OTB. Probably several
>programs can solve some of those positions at 12-24 hours per move. But that
>helps not one iota in a game OTB of course, unless perhaps it is correspondence.
>
For example: position 10 Rxf7. 10 minutes at a PP200.
I'm busy implementing mating extension. As soon as this works
properly, then even at todays hardware it'll play it at tournament level.
Also every week i am busy implementing new knowledge.
Perhaps i allow to solve few problems in the Fritz/Rebel/Genius style?
Within few years, even a cheap computer will allow the great audience
to solve al of these problems.
The problem of correspondence level is the small hashtable.
It works great, but the bigger that hashtable is, the smaller the branching
factor becomes. About 1/3 of these problems Diep has at depth > 11 ply
a branching factor of 2.5-3.0, even with a hashtable which is very small
compared to the number of nodes searched (huge loading factor).
The combination nullmove/hashtable seems to work really great!
Robert Hyatt
: >
: >Vincent Diepeveen (vdie...@cs.ruu.nl) wrote:
: >
: >: I heard rumors DB should also find it. Why did it play Ng6-e7 then?
: >: Bug in the program at that time?
: >
: >> Again, right after the game, Hsu or Murray said that DB had seen Bxh7,
: >> searched to a drawn position, and was assuming that Kasparov would play
: >> it and go for the draw. Whether it is a forced win has never been
: >> demonstrated that I have seen, although it'd be nice if someone could
: >> do so...
: >
: >How about the 2 following variations...
: >20. Bxh7+ Kxh7 21. Ng5+ Kg8 22. Qh5 Nf5 23. cxd5
: >a) 23.. Nxe3 24.dxe6 fxe6 25.fxe3
: >b) 23.. Ncxd4 24.Rxc8 Rxc8 25.dxe6 fxe6 26.Qh7+ Kf8 27.Nxe6+ Nxe6 28.Qxf5+
: >
: >Both variations look good for white.
: >
: >Maybe Vincent can display the main variation + score?
: >And what about CST who finds Bxh7+ in seconds?
: >Chris?
: >
: >- Ed -
:
:
: I did an overnight analysis with Rebel8.
: With "Power Analyse" I entered Bxh7+ as only move to analyse.
:
: At ply=11 Rebel claims a pawn up for white.
: But surprisingly at ply 12 and 13 the score dropped to -0.16 and -0.01
:
: - Ed -
:
My last comment on the subject... If you sit down with a *strong* GM,
and study this *in depth*, you'll find that the immediate Bxh7+ does not
work.
Just to lend a little "mystery" here, I'd like to dispell one rumor about
Garry in this game. If anyone thinks he "wimped out" and played a safe
move (a3), again sit down with a strong GM. With a lot of study, you
*might* be surprised about a3. :) He didn't play Bxh7 because he was
afraid of the tactics, he didn't play it because it doesn't work. *yet*...
Bob
Peter W. Gillgasch
Vincent Diepeveen <vdie...@cs.ruu.nl> wrote:
> In <581e5p$q...@juniper.cis.uab.edu> hy...@crafty.cis.uab.edu (Robert
> Hyatt) writes: > >By who's analysis? >
>
> Mine combined with Diep. GrandMaster class analysis thus.
[ 2 ply "analysis" snipped. Not worth being quoted. ]
Whoah !!! What a convincing analysis !!! Thanks for sharing :)
-- Peter
Walter Ravenek
> My goal is to get Mr. Diepeveen to submit evidence to substantiate his claim,
> either that or to stop making this claim. I have seen little evidence so
> far, certainly not enough that we should declare him to be world champion by
> consensus. We don't know that Diep has ever won a game at a long time
> control.
>
> bruce
As far as I know the games of the last open dutch computer chess tournament
have not been posted in this newsgroup. Diep played scored 5/11 if I am
correct. I'll try to dig up the results and post them separately.
--
_/_/_/_/_/_/_/_/_/_/_/_/
_/ Dr. Walter Ravenek _/
_/ Dept. of Chemistry _/
_/ Vrije Universiteit _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
_/ De Boelelaan 1083 _/ phone: +31 20 444 7617 _/
_/ 1081 HV Amsterdam _/ fax : +31 20 444 7643 _/
_/ The Netherlands _/ email: rav...@chem.vu.nl _/
_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
Chris Whittington
t...@mail.htp.com (Tim Mirabile) wrote:
>
> Chris Whittington <chr...@cpsoft.demon.co.uk> wrote:
>
> >"Ed Schröder" <rebc...@xs4all.nl> wrote:
>
> >> And what about CST who finds Bxh7+ in seconds?
> >> Chris?
> >
> >
> >If anybody posts it me, I'll try the latest version of CST ...
> >
> >Then we can compare with Droop :)
>
> [Event "ACM m"]
> [Site "Philadelphia, USA"]
> [Date "1996.??.??"]
> [Round "6"]
> [White "Kasparov,G"]
> [Black "Deep Blue"]
> [Result "1-0"]
>
> 1. Nf3 d5 2. d4 c6 3. c4 e6 4. Nbd2 Nf6 5. e3 c5 6. b3 Nc6 7. Bb2 cxd4 8. exd4
> Be7 9. Rc1 0-0 10. Bd3 Bd7 11. 0-0 Nh5 12. Re1 Nf4 13. Bb1 Bd6 14. g3 Ng6 15.
> Ne5 Rc8 16. Nxd7 Qxd7 17. Nf3 Bb4 18. Re3 Rfd8 19. h4 Nge7 20. a3 Ba5 21. b4 Bc7
> 22. c5 Re8 23. Qd3 g6 24. Re2 Nf5 25. Bc3 h5 26. b5 Nce7 27. Bd2 Kg7 28. a4 Ra8
> 29. a5 a6 30. b6 Bb8 31. Bc2 Nc6 32. Ba4 Re7 33. Bc3 Ne5 34. dxe5 Qxa4 35. Nd4
> Nxd4 36. Qxd4 Qd7 37. Bd2 Re8 38. Bg5 Rc8 39. Bf6+ Kh7 40. c6 bxc6 41. Qc5 Kh6
> 42. Rb2 Qb7 43. Rb4 1-0
>
I'll leave it running overnight and see if there is any change.
Chris Whittington
brucemo
mclane wrote:
> Now I am really that far that I will buy a gun and the next guy that
> says such a shame shit will be shot without any further comments....
>
> What the hell do you think your postings do tell us ABOUT YOU ????
> Is any program leading the ssdf-list just this high because it has a
> KILLER-BOOK. Sorry, but you seem to be crazy!
> Look in the mirror when you want to see an idiot!
> And post it , but please not here.
> I don't accept any further shit like this and will not stop to call it
> SHIT until you show us your games you have played or excuse for your
> words, as chris has said.
>
> You have said some more of this bullshit before, that diep is a good
> program and chess system tal is a weak program. Now you say:
> Rebel has a big killer book.
> Are you drunken vincent?
> Do you smoke cannabis ?
Maybe you could tone down the personal language, threats of violence,
profanity, and enraged ranting, just a bit.
bruce
brucemo
Vincent Diepeveen wrote:
> Yes, Rxf7 is a children trick.
> So is Rxh6. It is hard to see of course because of the delaying moves,
> but the idea: Rxh6, Nxh6 Qg5, Nf7 Qd8, Nd8 h6 and h7 is quite easy.
>
> Another thing that just can be brought back to simplistic tactics.
> Why doesn't rebel see this? Fritz? I mean: there must be a bug in the
> programs if they don't find #11. After few moves you simply put mate,
> or win a bunch of material.
Rxh6 supposedly is not a win, it is a draw. Maybe you could provide a
variation if you think it is a win.
bruce
brucemo
> Ed Schröder wrote:
> >
> > From: vdie...@cs.ruu.nl (Vincent Diepeveen)
> >
> > : Diep is by far the strongest correspondence/analysis program.
> >
> > This is a very huge claim!
> >
> > Besides Kasparov-Deep Blue game 6, 20. Bxh7?! do you have more proof?
>
> > r3qb1k/1b4p1/p2pr2p/3n4/Pnp1N1N1/6RP/1B3PP1/1B1QR1K1 w - - bm Nxh6;
> > r4rk1/pp1n1p1p/1nqP2p1/2b1P1B1/4NQ2/1B3P2/PP2K2P/2R5 w - - bm Rxc5;
> > r2qk2r/ppp1b1pp/2n1p3/3pP1n1/3P2b1/2PB1NN1/PP4PP/R1BQK2R w - - bm Nxg5;
> > r1b1kb1r/1p1n1ppp/p2ppn2/6BB/2qNP3/2N5/PPP2PPP/R2Q1RK1 w - - bm Nxe6;
> > r2qrb1k/1p1b2p1/p2ppn1p/8/3NP3/1BN5/PPP3QP/1K3RR1 w - - bm e5;
> > rnbqk2r/1p3ppp/p7/1NpPp3/QPP1P1n1/P4N2/4KbPP/R1B2B1R b - - bm axb5;
> > 1r1bk2r/2R2ppp/p3p3/1b2P2q/4QP2/4N3/1B4PP/3R2K1 w - - bm Rxd8+;
> > r3rbk1/ppq2ppp/2b1pB2/8/6Q1/1P1B3P/P1P2PP1/R2R2K1 w - - bm Bxh7+;
> > r4r1k/4bppb/2n1p2p/p1n1P3/1p1p1BNP/3P1NP1/qP2QPB1/2RR2K1 w - - bm Ng5;
> > r1b2rk1/1p1nbppp/pq1p4/3B4/P2NP3/2N1p3/1PP3PP/R2Q1R1K w - - bm Rxf7;
> > r1b3k1/p2p1nP1/2pqr1Rp/1p2p2P/2B1PnQ1/1P6/P1PP4/1K4R1 w - - bm Rxh6;
> >
> > BTW, I tested these 11 NOLUT positions once overnight with Rebel8.
> > Level 30:00 fixed time.
> > Machine PP-200
> >
> > Result: 0 (this reads as NUL) positions found!!
> >
> > I am curious about the results of Diep.
Position 1 solved in one hour fifty-six minutes on a P6/200. I tried position five for a
day or so, it didn't solve.
So Ferret has gotten 1, 10, and 11 in finite time :-) I'm trying #4 now, but I don't feel
optimistic.
bruce
-------------
PV 00:03:17.464 10 -119 [wrong] Qd4 c3 Bxc3 Nxc3 Qxb4 Nxe4 Qxb7 Rb8 Qxb8
Qxb8
PV 00:06:20.096 11 -119 [wrong] Qd4 c3 Bxc3 Nxc3 Qxb4 Nxe4 Qxb7 Rb8 Qxb8
Qxb8 Bxe4
PV 00:15:42.855 12 -143 [wrong] Qd4 c3 Bxc3 Nxc3 Qxb4 Nxe4 Qxb7 Rb8 Qxb8
Qxb8 Bxe4 d5
PV 01:56:21.248 13 -93 [right] Nxh6
PV 02:31:03.632 13 -67 [right] Nxh6 c3 Nf5 cxb2 Qg4 g6 Kh2 Qd7 Nh4 Kg8 Nxg6
Bg7 Nf6+ Nxf6 Qxe6+ Qxe6
PV 05:00:06.391 14 -44 [right] Nxh6 c3 Nf5 cxb2 Qg4 g6 Kh2 Qd7 Nh4 Rd8
Nxg6+ Rxg6 Qxg6 Qg7 Qh5+ Qh6 Qxh6+ Bxh6 Rb3
> From: m...@cnam.cnam.fr (Marc-Francois Baudot)
> Subject: Pierre Nolot's solutions to the 11 positions
> Date: 29 Jul 1994 11:55:22 GMT
> Organization: Conservatoire National des Arts et Metiers, Paris, France
>
> # Position: 1
> # Move: W
>
> r...qb.k Kasparov-Karpov, 20th game 1990
> .b....p. 26.Nxh6!! c3 (26... Rxh6!? is not sufficient:
> p..pr..p 27.Nxd6 Qh5 (best) 28.Rg5! Qxd1 29.Nf7+ Kg8 30.Nxh6+ Kh8
> ...n.... 31.Rxd1 c3 32.Nf7+ Kg8 33.Bg6! Nf4 34.Bxc3 Nxg6 35.Bxb4 Kxf7
> Pnp.N.N. 36.Rd7+ Kf6 37.Rxg6+ Kxg6 38.Rxb7 +-) 27.Nf5! cxb2
> ......RP 28.Qg4 Bc8 (if 28... g6!? 29.Kh2! wins : 29...Qd7 30.Nh4 Bc6
> .B...PP. 31.Nc5! dxc 32.Rxe6 Nf6 33.Nxg6+ Kg7 34.Qg5 Nbd5 35.Ne5 Kh8
> .B.QR.K. 36.Nxd7 +-) 29.Qh4+ Rh6 30.Nxh6 gxh6 31.Kh2! Qe5 32.Ng5 Qf6
> 33.Re8 Bf5 34.Qxh6 (pretty, but there was a mate in 6 :
> 34.Nf7+ Qxf7 35.Qxh6+ Bh7 36.Rxa8 Nf6 37.Rxf8 Qxf8 38.Qxf8+ Ng8
> 39.Qg7#) 34...Qxh6 35.Nf7+ Kh7 36.Bxf5+ Qg6 37.Bxg6+ Kg7 38.Rxa8 Be7
> 39.Rb8 a5 40.Be4+ Kxf7 41.Bxd5+ 1-0
> A very deep combimation, that Tasc R30 or Genius 2 running on a pentium would
> take between a few months and a few years to find.
> The best Novag computer, the Diablo 68000, finds 26.Nxh6 after 7 months
> and a half (Pierre Nolot has let it run on the position for 14 months and one
> day, until a power failure stopped an analysis of over 80 000 000 000 nodes!)
> but for wrong reasons : it evaluates white's position as inferior and
> thinks this move would enable it to draw.
Robert Hyatt
brucemo (bru...@nwlink.com) wrote:
: mclane wrote:
:
<repeated references to fecal matter snipped> :)
:
: Maybe you could tone down the personal language, threats of violence,
: profanity, and enraged ranting, just a bit.
:
: bruce
If he did, it would look just like what I quoted above... that is, it
would be empty... :)
Ed Schröder
From: hy...@crafty.cis.uab.edu (Robert Hyatt)
: I did an overnight analysis with Rebel8.
: With "Power Analyse" I entered Bxh7+ as only move to analyse.
:
: At ply=11 Rebel claims a pawn up for white.
: But surprisingly at ply 12 and 13 the score dropped to -0.16 and -0.01
:
: - Ed -
> My last comment on the subject... If you sit down with a *strong* GM,
> and study this *in depth*, you'll find that the immediate Bxh7+ does
> not work.
I disagree, it has been proven so many times that computers do sometimes
better than GM's. Kasparov overlooking a mate in 6 combination or so
while computers see this in a few seconds. Lots of examples.
> Just to lend a little "mystery" here, I'd like to dispell one rumor
> about Garry in this game. If anyone thinks he "wimped out" and played
> a safe move (a3), again sit down with a strong GM. With a lot of
> study, you *might* be surprised about a3. :) He didn't play Bxh7
> because he was afraid of the tactics, he didn't play it because it
> doesn't work. *yet*...
For the moment I also believe that a3! was a better move... :)
- Ed -
>Bob
Chris Whittington
PP200 Chess System Tal
Bxh7 +1.24 1 sec, soon score fell to +0.29
Ng5 +0.66 66 secs
Bxh7 +1.08 1060 secs
Bxh7 +0.82 4678 secs
changed its mind to something else (unknown) in middle of night
changed back again to Bxh7 by the morning:
Bxh7 +1.71 21014 secs
CST doesn't solve or prove anything, it prunes too much and is too
speculative for that. All you can say is that it has 'feelings' about
positions.
From my experience with it, a piece sac which registers a score of +1.5 or
less is quite likely not going to work.
A piece sac which registers a score of +4.0 or +5.0 is likely to be going
to work.
A rising score, iteration by iteration is a good sign.
Given the rising score towards the final iteration, and the relatively low
sacrifice score (+1.71), my Chess System Tal intuition tells me that this is
probably going to be a draw from Bxh7+
I quite like CST results for this position. Its kind of human, and parallels
the human reaction to the position.
Yeah, Bxh7+ !!
hmm, maybe not so sure ..... ??
Yeah, well go for it, there looks like there's something there, and
we like having fun ....
Chris Whittington
Robert Hyatt
Chris Whittington (chr...@cpsoft.demon.co.uk) wrote:
: Chris Whittington <chr...@cpsoft.demon.co.uk> wrote:
I agree 100%. WchessX is quite good at this sort of thing, and I see it
against Crafty a lot. It often fails miserably of course, because Crafty
is a computer and not a human, but I have no doubt that against humans,
WchessX would be really deadly doing what it does. Your king safety/etc
code is doing well to like Bxh7, because the king has to take a hike out
into the crowded streets. Which can't be bad, except in a few positions
like this one.
Bxh7 is a mistake, but not one I'd discourage Crafty from making if it
were capable of choosing this based on king safety and piece/attack
coordination...
Enrique Irazoqui
> Vincent Diepeveen <vdie...@cs.ruu.nl> wrote
> What is your opinion about the Rebel super killerbook.
> I played a rapid-game Rebel-Genius3. Rebel in book: far over 20 moves.
> Genius in book: around 10 moves.
> After his book Genius3 played: ... Qd8-b6 followed by Qb6xb2??
You didn't post this game yet. In the meantime I played 71 games R8-G3 and
nothing like this line is being played. The mystery case of the cook that
never was?
Enrique
brucemo
Chris Whittington wrote:
> I quite like CST results for this position. Its kind of human, and parallels
> the human reaction to the position.
> Yeah, Bxh7+ !!
> hmm, maybe not so sure ..... ??
> Yeah, well go for it, there looks like there's something there, and
> we like having fun ....
I think this is a perfectly reasonable attitude. It's not mine, but that's not a
problem. I'm glad someone is trying seriously to teach a program to speculate.
bruce
brucemo
Vincent Diepeveen wrote:
>
> In <57v314$q...@juniper.cis.uab.edu> hy...@crafty.cis.uab.edu (Robert Hyatt) writes:
>
> >:
> >: I heard rumors DB should also find it. Why did it play Ng6-e7 then? Bug
> >: in the program at that time?
> >:
> >
> >Again, right after the game, Hsu or Murray said that DB had seen Bxh7, searched
> >to a drawn position, and was assuming that Kasparov would play it and go for the
> >draw. Whether it is a forced win has never been demonstrated that I have seen,
> >although it'd be nice if someone could do so...
> >
> >
>
Please post a line.
bruce
Tim Mirabile
"Ed Schröder" <rebc...@xs4all.nl> wrote:
>From: hy...@crafty.cis.uab.edu (Robert Hyatt)
>
>Vincent Diepeveen (vdie...@cs.ruu.nl) wrote:
>
>: I heard rumors DB should also find it. Why did it play Ng6-e7 then?
>: Bug in the program at that time?
>
>> Again, right after the game, Hsu or Murray said that DB had seen Bxh7,
>> searched to a drawn position, and was assuming that Kasparov would play
>> it and go for the draw. Whether it is a forced win has never been
>> demonstrated that I have seen, although it'd be nice if someone could
>> do so...
>
>How about the 2 following variations...
>
>20. Bxh7+ Kxh7 21. Ng5+ Kg8 22. Qh5 Nf5 23. cxd5
>
>a) 23.. Nxe3 24.dxe6 fxe6 25.fxe3
>b) 23.. Ncxd4 24.Rxc8 Rxc8 25.dxe6 fxe6 26.Qh7+ Kf8 27.Nxe6+ Nxe6 28.Qxf5+
>
>Both variations look good for white.
Where is the win after 22...Qe8?
mclane
"Enrique Irazoqui" <en...@lix.intercom.es> wrote:
>> Vincent Diepeveen <vdie...@cs.ruu.nl> wrote
>Enrique
Rebel8 is not that strong because it has a killer-book.
Everybody who claims this is a liar!
Rebel8 is that strong, because ed has programmed the engine that
strong. This is proofable by anybody who has SOME chess-knowledge in
his mind.
The less knowledge you have in mind, the later you will find out.
Some austrian statisticians will never find out (they really think you
have to play 5000 games to say something about chess programs!!!
If this would be true, we wouldn't know anything about the paradise
and apples and girls ...).
Some people with really less knowledge think Rebel8 has a BIG
KILLERBOOK. So the whole discussion can be used as a degree-measure
for intelligence.
mclane
Chris Whittington <chr...@cpsoft.demon.co.uk> wrote:
>PP200 Chess System Tal
For many chess-players that had to play against CSTal it was a shock
that CST is unable to PROVE if something is so or so.
I told the humans at the last AEGON-tournamentz about, when they
wanted to analyze the game afterwards with CSTal again, that we
CANNOT PROVE.
They all said: But it is a computer !!!
I answered: Yes, it is a computer-program, but Chris Whittington has
programmed it, so the program can only feel chess, and not compute
chess.
BTW:
The male players had the same problems to understand this, like the
female players had.
>From my experience with it, a piece sac which registers a score of +1.5 or
>less is quite likely not going to work.
>A piece sac which registers a score of +4.0 or +5.0 is likely to be going
>to work.
>A rising score, iteration by iteration is a good sign.
>Given the rising score towards the final iteration, and the relatively low
>sacrifice score (+1.71), my Chess System Tal intuition tells me that this is
>probably going to be a draw from Bxh7+
"my Chess System Tal intuition tells me" ---- very nice expression,
Chris!!
>I quite like CST results for this position. Its kind of human, and parallels
>the human reaction to the position.
>Yeah, Bxh7+ !!
>hmm, maybe not so sure ..... ??
>Yeah, well go for it, there looks like there's something there, and
>we like having fun ....
Thats typical for CSTal , indeed.
If we cannot say exactly what a position is, good for white, draw or
won by this move, we should not say: Oh - it is won by this move.
If we don't know it accurate, we should not suggest or tell anybody: I
know it exactly.
That's what CSTAL does: it feels chess. With all advantages and
disadvantage.
If it is enough to kill Genius on a championship with a quality-sac,
then it's really enough!
>Chris Whittington
mclane
hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
>Stefan Meyer-Kahlen (mey...@fmi.uni-passau.de) wrote:
>: On 2 Dec 1996 19:45:19 GMT, "Ed Schröder" <rebc...@xs4all.nl> wrote:
>: >From: vdie...@cs.ruu.nl (Vincent Diepeveen)
>: >
>: >: Well Chris, What is your opinion about the Rebel super killerbook.
>: >: I played a rapid-game Rebel-Genius3. Rebel in book
>: >: far over 20 moves.: Genius in book
>: >: around 10 moves.
>: >
>: >: After his book Genius3 played: ... Qd8-b6 followed by Qb6xb2??
>: >: Although this was a typical computerline, it was of course in the
>: >: rebel-book. After getting out of book Rebel was a piece up.
>: >: I therefore stopped the game.
>: >
>: >: Probably not a killerline in the eyes of Ed. A killerline is in the
>: >: eyes of Ed a line where after getting out of book the opponent is
>: >: mated, Ed?
>: >
>: >Just give Jeroen Noomen a call, he is a walking opening encyclopedia,
>: >he will tell you immediately from which (on paper) opening book your
>: >example comes from :)
>: >
>: >: Perhaps i think too simplistic, I don't see difference between a
>: >: full piece up, so a mate in about 20,
>: >: instead of an already seen mate in 7 (mcpro).
>: >: I don't blame anyone writing a killerbook, but don't blame someone
>: >: else of having a killerbook if you have yourselve the biggest one?
>:
>: In the book of my program Shredder I have many lines like the one you
>: mentioned above. One side (hopefully not me) is grabbing a poisend
>: pawn and is out of book. The other side (hopefully me) is still in
>: book for some moves and will leave the book in a won position. If you
>: call those lines killer lines, well, that's up to you, but I don't
>: think that these lines are killer lines in the sense discussed in this
>: newsgroup. I add those lines to my book because often Shredder is too
>: stupid to win the good position after the capture, it will be simply
>: down a pawn. I get those lines from books on opening theory or chess
>: magazines. A human chess player whould do the same, looking at the
>: lines given in the book and knowing that the pawn mustn't be captured.
>: Probably he also would keep the refutation in mind. So what's wrong
>: about that? Yesterday I've put a very sharp line in the Sicilian
>: Defense in my book. The line was played at the tournament in Tilburg
>: and many variations end in a mate.
>: When you blame somebody for having killer lines in his opening book
>: (no matter weather you can blame somebody for that or not) you have to
>: take a very close look at the given line. I definetly didn't add any
>: line to my book to win against a special program, I add lines to my
>: book to win against any program.
>:
>: Stefan
>:
>:
>:
>:
>:
>I, too, have lots of those. On many occasions Crafty exits the book with
>an eval of +2 to +3, and wins easily. On other occasions it exits with -2
>and loses, simply because it is following some GM game. Hopefully the
>learning function is eliminating the latter, and favoring the former. :)
>Bob
Some peolle cannot accept that the whole killer-book-discussions have
only one destination:
to damage the fame of some programmers.
When Mchess5 was leader of the ssdf-list: Mchess 5 had a BIG KILLER
BOOK.
Now Rebel8 is the leader of the ssdf-list and some guys with
sparrow-brain say: Rebel8 has a MUCH BIGGER KILLER-BOOK.
Stefan is right. Bob is right. We all know these lines.
The only reason why certain people will not accept this is:
they have INTERESTS to tell these lies.
They don't want to understand because they have their interests
in damaging someones name and product.
If - in a few months, another program is leading the ssdf-list, some
ill guys will come out of the mud and will say: Program XYZ has the
biggest killer book ever seen.
I bet 50 DM that they will come out if another program leads the list.
mclane
brucemo <bru...@nwlink.com> wrote:
>Vincent Diepeveen wrote:
>bruce
Yes Vincent, give us a line!!
It is a children-trick. So - you should be able to give us the line...
Or shall I ask the young brother of my neighbour, it is a child?!
mclane
TUESCHEN.MEDIZ...@t-online.de (Rolf W. Tueschen) wrote:
>mcl...@prima.ruhr.de (mclane) wrote:
>R.W. Tueschen:
>>>> >> And at last, because I'm really upset now. Dear Chris, don't praise my
>>>> >> writings if you expect on the other side that I might stay mute in front of
>>>> >> this behaviour of Ed. If you are not prepared to tell your friend some
>>>> >> needed truths, other people must do it.
>Chris:
>>>> >: Unbelievable !
>>>> >
>>>> >: Ed knows very well that I'll say exactly what I think without
>>>> >: any fear or favour. Don't imagine for one moment that Ed gets
>>>> >: spared.
>Vincent:
>>>> Well Chris, What is your opinion about the Rebel super killerbook.
>Thorsten:
>>Rebel has a killer book ? WHO HAS SUPPOSED THIS ???
>>IS the guy crazy ????
>>REBEL has a killer-book?!
>>Now I am really that far that I will buy a gun and the next guy that
>>says such a shame shit will be shot without any further comments....
>> Sorry, but you seem to be crazy!
>>Look in the mirror when you want to see an idiot!
>>And post it , but please not here.
>>Rebel has the biggest killer book ????
>>Is this a joke or do you really mean what you write ??!!!??!
>>It must be a joke. You could not program a chess-program with this
>>ideas of paranoic scholastic word games.
>>You have said some more of this bullshit before, that diep is a good
>>program and chess system tal is a weak program. Now you say:
>>Rebel has a big killer book.
>>Are you drunken vincent?
>>Do you smoke cannabis ?
>>Unbelievable....
>----------------------------------------------------------------------------------------------------------
>Idiot, bull, paranoic, drunken, cannabis.
>Thorsten think about your model function with all your experience since
>'79.
It depends on the text I am referring. I never tried to give a model
function, one reason that gives evidence for that is, that I AM MCLANE
and NOT WHAT YOU CALL ME. MCLANE is nobody. So he has no
model-function.
But these words Vincent is trying to tell us, are enough. And all
experience, from 79 or earlier, cannot help or hide this bullshit.
>Hope you're not engaged as *Wadenbeisser* for Eddie. You could do him no
>worse favour if you're supporting his always ready goasts/nightmares. If
>you know what I mean.
I don't know "Eddie". Do you speak of ED ?
Ed does not need me. And I am not speaking for ED. But I don't like
people attacking programmers work of many months just because they are
jealous or unconfident that their programs are not bought much.
In my opinion many many people in this newsgroup are very idle and are
only interested in their affairs, their programs....
>Maybe you are right in your judgement, but please think of the worldwide
>public you publish for. And the children everywhere. What if you're not
>informed about Eddie's latest performances?
Häh ? Who is Edidie ?? And which performance do you speak about?!?
Which children ??????? Please tell me what you speak about.
Do you think Vincent is a child ? I have seen him, he is older than a
child. But maybe you are right....
>I would like to read more of your easy satirical store instead. :)
>Rolf Tueschen alias Thorsten II. :)
HELP - Heavens - Uff.
Disclaimer:
Mclane is NOT RT !!!!!
Please Rolf: let these jokes!!
mclane
brucemo <bru...@nwlink.com> wrote:
>mclane wrote:
>> Now I am really that far that I will buy a gun and the next guy that
>> says such a shame shit will be shot without any further comments....
>>
>> What the hell do you think your postings do tell us ABOUT YOU ????
>> Is any program leading the ssdf-list just this high because it has a
>> KILLER-BOOK. Sorry, but you seem to be crazy!
>> Look in the mirror when you want to see an idiot!
>> And post it , but please not here.
>> I don't accept any further shit like this and will not stop to call it
>> SHIT until you show us your games you have played or excuse for your
>> words, as chris has said.
>>
>> You have said some more of this bullshit before, that diep is a good
>> program and chess system tal is a weak program. Now you say:
>> Rebel has a big killer book.
>> Are you drunken vincent?
>> Do you smoke cannabis ?
>Maybe you could tone down the personal language, threats of violence,
>profanity, and enraged ranting, just a bit.
>bruce
I will try, but do you understand why I am that upset. Can you feel
with me, have you read what he says, do you have tried out Rebel8 and
do you have seen the book in action ?!
Maybe than you could better understand my feeling. Sorry.
mclane
hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
>brucemo (bru...@nwlink.com) wrote:
>: mclane wrote:
>:
><repeated references to fecal matter snipped> :)
>:
>: Maybe you could tone down the personal language, threats of violence,
>: profanity, and enraged ranting, just a bit.
>:
>: bruce
>If he did, it would look just like what I quoted above... that is, it
>would be empty... :)
Again, I don't understand why you always and alyways let liars repeat
their lies, and critizise people that are upset about those lies.
You are americans, aren't you ?!
In US it is not important if somebody is telling the truth or is
saying lies.
E.g. in the O.J. Simpson case, the lawyers could do whatever they
wanted, they did a SHOW and do not try to find out the truth.
Thats the difference between US and Germany. Here we do not lie that
much, and we do not like liars that much.
In US it is always a show. And the show must go on.
chessman
On Fri, 6 Dec 1996 23:10:15 GMT, mcl...@prima.ruhr.de (mclane) wrote:
snip
>You are americans, aren't you ?!
oh oh..here we go again.. another slam on Americans....
>In US it is not important if somebody is telling the truth or is
>saying lies.
>E.g. in the O.J. Simpson case, the lawyers could do whatever they
>wanted, they did a SHOW and do not try to find out the truth.
that's right ..the OJ case symbolizes everything about what is right
wrong with Amercians...NOT...I don't think so....
>
>Thats the difference between US and Germany. Here we do not lie that
>much, and we do not like liars that much.
I always suspicious about someone who says "Here we do not lie.."
right off the bat ..that's a lie in itself....wait...there's a
qualifier..."....much"....
so if you "...don't lie that much" in Germany ...people will like
you...
is that the message you're trying to get out....
it looks like the show in Germany is *long running*...:)
anyway..you can take any country in the world and find good people and
bad people...people who tell the truth and people who lie....I don't
believe *telling the truth* is a national treasure of any one
country...just as surely I don't believe Americans have the monopoly
on lying...we both can look at our political leaders of years past and
find instances of lying..... so when you decide...welcome back to
reality .
Robert Hyatt
brucemo (bru...@nwlink.com) wrote:
: Chris Whittington wrote:
:
: > I quite like CST results for this position. Its kind of human, and parallels
: > the human reaction to the position.
: > Yeah, Bxh7+ !!
: > hmm, maybe not so sure ..... ??
: > Yeah, well go for it, there looks like there's something there, and
: > we like having fun ....
: I think this is a perfectly reasonable attitude. It's not mine, but that's not a
:
: bruce
Where in the hell have you been when WchessX plays? :)
talk about speculation... :)
*dangerous* speculation too... that works a lot of the time...