| |
Main
Date: 15 Jan 2009 05:36:48
From:
Subject: Traxler Gambit
|
albert alberts demonstrated the machines ineptness to evaluate it
correctly in his first MAMS book - I'll give the moves, and then it
would be interesting to review 2 things
what 'fritz' [any computer engine, even Sanny's] thinks of it
&
from a database, what are actual results or white/black scores
in the dreadfully complicated
TRAXLER GAMBIT
1 e4 e5
2 Nf3 Nc6
3 Bc4 Nf6
4 Ng5 Bc5
5 Nf7 Bf2
[where alberts says (+-3.44)
as we know Tal played it in correspondence against 10,000 Pravda
readers 68/69 with commentary by M de Zeeuw in Yearbook 67 - but I
exclude that from actual encounters.
but what happens next is albert's attempt to unravel this opening - he
notes that 'right off the bat' Fritz refuses the bishop...
anyway, what does your computer think. I'll illustrate more moves if
the conversation proceeds, and also transpose this note to its on
thread 'traxler'
Phil Innes
|
|
| |
Date: 17 Jan 2009 08:05:43
From: help bot
Subject: Re: Traxler Gambit
|
On Jan 17, 9:46=A0am, [email protected] wrote:
> But actually, the interaction between people and computers points out
> their mutual flaws.
Ah, but as for the flaws in humans, computers
ca in no way be held accountable; but when we
find weaknesses in the computers, it is always
due to them having been poorly programmed...
by humans.
> The very thing to note is that Alberts has found out the weakness of
> computer evaluations in highly unbalanced positions - and suggests
> playing those positions against them. He has scored hundreds of wins
> this way, against first tier chess engines.
One notes the absence of this "Alberts" name
from the list of players brave enough to face
Rybka, who takes on all comers, offering odds.
> In this sense, the 'exceptions' disprove the rule, and we might as
> well admit that 'Fritz' type analysis is a poor guide to such
> unbalanced positions early in the game.
A chap named "Heisman" has crafted a CD
on the Traxler, using computers to refute and
correct countless misevaluations by humans
in this complex line. Why then do some
folks still remain in de Nile?
> > =A0 What I think would have made a good
> > introduction to such a thread as this, is an
> > introduction to some of the terms tossed
> > out, seemingly at random. =A0 =A0 Terms like
> > "MAMS", "Kingstonizing" and "de Zeeuw"
> > require clarification, and one should clearly
> > define what is meant by the term "Fritz";
> For the innocent reader:
>
> MAMS =3D Man Assisted Machine Shach [Chess] which was once called
> 'Kasparov Chess'
>
> Kingstonizing =3D What Kryptonite was to Superman and Sensibility
>
> de Zeeuw =3D is a bloke's name
>
> Fritz =3D a chess engine [inc Rybka etc]
> > are you talking about three minutes of
> > cranking by Fritz 5.32, for instance? =A0Or
> > virtually unlimited time with the latest
> > available version, perhaps?
> Depends what is being tested - you can't really take a finite game and
> change the rules to infinite crank. If we are evaluating Elo, eg, then
> the current GM rate for titles is about 2.5 minutes.
All my recent rated games were restricted to
a paltry G/61, with five seconds time delay and
openings book off (which is to say, I have
forgotten everything).
> That is the basis
> current players have for their rating. Therefore, and since time is a
> rule of chess, then that might be the rate - more so, if GMs play 1t
> 40 moves in 2 hours, so should 'Fritz'.
You are discussing OTB competition; the
issue was chess analysis-- a different animal.
> > =A0 The truth is, I am surprised -- no, shocked
> > -- that it has yet to be mentioned that this
> > very famous line is more often refered to by
> > another name entirely, here in the USA.
> O yes, the MacDonalds opening. 'You want to get friend with that?'
> It hasn't really caught on in the rest of the world.
Just give us time; soon, we will have subdued
them all. In China, they now eat at our Pizza
Hut restaurants (but they like eel, not pepperoni).
> In chess we say, Bxf7+
>
> Fritz doesn't like that, especially since once the black K shifts,
> there is an obligation to move the B immediately, otherwise the
> supportive knight is undermined and a piece lost
Fritz sees all that? ; >D
> - while black's d5
> gives him a huge center for white's loss of two tempii - in other
> words, white is 1/3 of a tempi ahead.
Um, Dr. IMnes, White *voluntarily* moved
the same piece, N-f3 then N-g5, so all this
lecturing me about a loss of time is pointless.
In fact, I never do that because I saw what
happened to all those patzers who did it
against Paul Morphy (not pretty). What was
left of the victims was disposed of by packing
it in plastic bags, then feeding it to goldfish in
a small pond.
> These are what players think, and since the current writer already
> admits that computers are over materialistic, then caution is advised.
> In fact, try the line yourself against Fritz and tell us how you get
> on.
My problem with "Fritz" is that I had to clean
up a big mess on my computer, and did so by
reinstalling Windows with a disk wipe. Hence,
Fritz is dead. Kaput.
BTW, Rybka is not overly materialistic... .
> I think I mentioned he writes at Chessville, and has his own research
> site. I managed to introduce him to the gent who developed Rybka, and
> I think they are getting on just fine. Usenet relies on writers
> putting in more than summary attention to a topic otherwise you wind
> up like you know who =A0;)
Hey, I typed something into a google search
box and came up with an HTML article which
nitpicked Mr. Heisman's Traxler CD, showing
how he messed up more than a couple of
things. Even so, it was called the best
there is on the Traxler-- and note that he
relied upon computer analysis so heavily he
missed some *obvious* transpositions.
> That is a tad abstract. The point of these MAMS moves is that 'Fritz'
> often scores them radically different afterwards, actually validating
> the [M] move better than anything it can think of - since on its own
> it rejects the [M] move.
Then just set "Fritz" to look at everything.
Surely, this is adjustable?
> But the subject is interesting - I think most GMs would not play this
> opening well - and have not!
"Most GMs" wallow in deep theory, such
as that of the Marshall Gambit or the Ruy
Lopez. They are likely to avoid lines in
which Black can easily equalize early on,
as after moving the same pieces over and
over like the Duke and the Count.
> And play neither side well. So there is
> something to be learned by strong players - and also for the engine to
> 'learn', since otherwise it's evaluation is false, and demonstrably an
> unreliable guide or evaluator.
As we saw with the Pal Benko chess problem
that BF could not solve, grandmasters do not
easily solve all types of problems. In a game
against a GM many years ago, I steered for a
stupid position, thinking he might not play it
any better than me... and it worked!
This Traxler stuff is wild and wooly, and it is
precisely the sort of thing best handled by
todays computers-- but you have to give them
lots of time to calculate. Remember: BF had
a half hour to find a mate-in-three, and failed.
-- help bot
|
| |
Date: 17 Jan 2009 06:46:52
From:
Subject: Re: Traxler Gambit
|
On Jan 17, 6:32=A0am, help bot <[email protected] > wrote:
> On Jan 16, 8:56=A0am, [email protected] wrote:
>
> > Alberts goes on to say that this highly compact miniature of only 12
> > moves, with 2 human intercession [M] moves as 'essential
> > interventions' - which the computer can't see.
>
> > Now then: his Question is: is 6. Kf1 the definitive proof that the
> > Traxler Gambit is overbold?
>
> =A0 The trouble with this kind of thinking is that
> it struggles to insist that humans are better
> chess analysts than computers in wild and
> crazy positions.
I can only make a brief reply at the moment.
But actually, the interaction between people and computers points out
their mutual flaws. Just a little later in the book we see an
appreciation of the solution to a so-called Kling&Horwitz fortress,
unsolved since 1851, and always adjudicated a draw. [with an
appreciative note to John Roycroft, 1972.]
A computer actually demonstrated the win in 1992 from a game Timman -
Speelman
> =A0 In reality, the reverse is most often the case,
> with a few exceptions here and there, mainly
> in positions which highlight some nasty
> human-programmed flaw in one of the lesser
> engines.
The very thing to note is that Alberts has found out the weakness of
computer evaluations in highly unbalanced positions - and suggests
playing those positions against them. He has scored hundreds of wins
this way, against first tier chess engines.
In this sense, the 'exceptions' disprove the rule, and we might as
well admit that 'Fritz' type analysis is a poor guide to such
unbalanced positions early in the game.
> =A0 A better try would be to suggest that at
> tournament time controls, say two or three
> minutes per move, computers often fail due
> to an over-weighting toward materialism (not
> unlike our beloved USA).
Quite so.
> =A0 What I think would have made a good
> introduction to such a thread as this, is an
> introduction to some of the terms tossed
> out, seemingly at random. =A0 =A0 Terms like
> "MAMS", "Kingstonizing" and "de Zeeuw"
> require clarification, and one should clearly
> define what is meant by the term "Fritz";
For the innocent reader:
MAMS =3D Man Assisted Machine Shach [Chess] which was once called
'Kasparov Chess'
Kingstonizing =3D What Kryptonite was to Superman and Sensibility
de Zeeuw =3D is a bloke's name
Fritz =3D a chess engine [inc Rybka etc]
> are you talking about three minutes of
> cranking by Fritz 5.32, for instance? =A0Or
> virtually unlimited time with the latest
> available version, perhaps?
Depends what is being tested - you can't really take a finite game and
change the rules to infinite crank. If we are evaluating Elo, eg, then
the current GM rate for titles is about 2.5 minutes. That is the basis
current players have for their rating. Therefore, and since time is a
rule of chess, then that might be the rate - more so, if GMs play 1t
40 moves in 2 hours, so should 'Fritz'.
> =A0 The truth is, I am surprised -- no, shocked
> -- that it has yet to be mentioned that this
> very famous line is more often refered to by
> another name entirely, here in the USA.
O yes, the MacDonalds opening. 'You want to get friend with that?'
It hasn't really caught on in the rest of the world.
> =A0 As for "proof" that this line may be over-
> bold on Black's part,
The question, not the proof, is Dr. Alberts'
> has it occured to Dr.
> IMnes that theory gives Bxf7+ as the real
> test? =A0 =A0Instead of seeing whether or not
> White can survive the attack on his own
> King, there is a clear alternative in smack-
> ing around the Black King a bit with a pre-
> emptive strike by White's F-7 fighter-
> bombers.
In chess we say, Bxf7+
Fritz doesn't like that, especially since once the black K shifts,
there is an obligation to move the B immediately, otherwise the
supportive knight is undermined and a piece lost - while black's d5
gives him a huge center for white's loss of two tempii - in other
words, white is 1/3 of a tempi ahead.
These are what players think, and since the current writer already
admits that computers are over materialistic, then caution is advised.
In fact, try the line yourself against Fritz and tell us how you get
on.
> =A0 One more thing: as long as I've been
> reading rgc, I've not once read any clear
> introduction to exactly who this Alberto
> Albert Alberts is. =A0 Freaky words are just
> tossed about, as in a word-salad: MAM,
> MAMS, Seniors... Monseuirs ...and
> Mad'ams are mixed in together with a
> bunch of snippets of chess analysis,
> with no apparent organization and the
> obvious design of rubbishing computers.
I think I mentioned he writes at Chessville, and has his own research
site. I managed to introduce him to the gent who developed Rybka, and
I think they are getting on just fine. Usenet relies on writers
putting in more than summary attention to a topic otherwise you wind
up like you know who ;)
> =A0 I call it denial-- no longer just a river in
> Africa. =A0 The idea that relatively-weak
> humans must "intervene" to "help" such
> greats as Rybka is misguided; the truth
> is, what is needed is more patience in
> complex positions, along with better
> organization on the part of humans. =A0No
> more "here, best is...", only to find that
> the exact same position is treated quite
> differently when it occurs via a different
> move-order.
That is a tad abstract. The point of these MAMS moves is that 'Fritz'
often scores them radically different afterwards, actually validating
the [M] move better than anything it can think of - since on its own
it rejects the [M] move.
> =A0 By and large, the worst aspect of this
> gambit is that a player with Black can
> in no way compell his opponents to
> play the silly N-g5, nor decide which
> way things may go from there, Bxf7+
> or otherwise. =A0 White could even steer
> toward an exchange Ruy... .
Of course anything can happen, but if this happens, and it does - then
this is the result. One cannot explore specific failiings of people or
chess engines via abstractions and escape into other variations.
But the subject is interesting - I think most GMs would not play this
opening well - and have not! And play neither side well. So there is
something to be learned by strong players - and also for the engine to
'learn', since otherwise it's evaluation is false, and demonstrably an
unreliable guide or evaluator.
But good post anyway - at least you raise real objections, if not
exactly prospective ones, since they escape to other lines not
currently under evaluation.
Phil Innes
> =A0 -- help bot
|
| |
Date: 17 Jan 2009 03:32:57
From: help bot
Subject: Re: Traxler Gambit
|
On Jan 16, 8:56=A0am, [email protected] wrote:
> Alberts goes on to say that this highly compact miniature of only 12
> moves, with 2 human intercession [M] moves as 'essential
> interventions' - which the computer can't see.
>
> Now then: his Question is: is 6. Kf1 the definitive proof that the
> Traxler Gambit is overbold?
The trouble with this kind of thinking is that
it struggles to insist that humans are better
chess analysts than computers in wild and
crazy positions.
In reality, the reverse is most often the case,
with a few exceptions here and there, mainly
in positions which highlight some nasty
human-programmed flaw in one of the lesser
engines.
A better try would be to suggest that at
tournament time controls, say two or three
minutes per move, computers often fail due
to an over-weighting toward materialism (not
unlike our beloved USA).
What I think would have made a good
introduction to such a thread as this, is an
introduction to some of the terms tossed
out, seemingly at random. Terms like
"MAMS", "Kingstonizing" and "de Zeeuw"
require clarification, and one should clearly
define what is meant by the term "Fritz";
are you talking about three minutes of
cranking by Fritz 5.32, for instance? Or
virtually unlimited time with the latest
available version, perhaps?
The truth is, I am surprised -- no, shocked
-- that it has yet to be mentioned that this
very famous line is more often refered to by
another name entirely, here in the USA.
As for "proof" that this line may be over-
bold on Black's part, has it occured to Dr.
IMnes that theory gives Bxf7+ as the real
test? Instead of seeing whether or not
White can survive the attack on his own
King, there is a clear alternative in smack-
ing around the Black King a bit with a pre-
emptive strike by White's F-7 fighter-
bombers.
One more thing: as long as I've been
reading rgc, I've not once read any clear
introduction to exactly who this Alberto
Albert Alberts is. Freaky words are just
tossed about, as in a word-salad: MAM,
MAMS, Seniors... Monseuirs ...and
Mad'ams are mixed in together with a
bunch of snippets of chess analysis,
with no apparent organization and the
obvious design of rubbishing computers.
I call it denial-- no longer just a river in
Africa. The idea that relatively-weak
humans must "intervene" to "help" such
greats as Rybka is misguided; the truth
is, what is needed is more patience in
complex positions, along with better
organization on the part of humans. No
more "here, best is...", only to find that
the exact same position is treated quite
differently when it occurs via a different
move-order.
By and large, the worst aspect of this
gambit is that a player with Black can
in no way compell his opponents to
play the silly N-g5, nor decide which
way things may go from there, Bxf7+
or otherwise. White could even steer
toward an exchange Ruy... .
-- help bot
|
| |
Date: 16 Jan 2009 05:56:15
From:
Subject: Re: Traxler Gambit
|
The F FFP schrieb:-
> =A0 Were Mr. Murray not such an incorrigible patzer,
> he would easily figure out that the reference to
> ...Qh4 obviously apllies to the following variation
> which I shall lay out for him and all the other mere
> patzers here:
>
> =A0 1. e4 =A0e5
>
> =A0 2. Nf3 =A0Nc6
>
> =A0 3. Bc4 =A0Nf6
>
> =A0 4. Ng5 =A0Bc5
>
> =A0 5. Nxf7 =A0Bxf2+
>
> =A0 6. Kxf2 =A0Nxe4+
>
> =A0 7. Kg1 =A0(drumroll)
>
> =A0 =A0...Qh4
Now to prefer what the real-Fritz [RF] likes, compared to the F FFP:
8. g3 Ng3
9. Nh8 Nd4 [M]
Now, the [M] is a MAMS move; the one suggested by de Zeeuw
But I'm interested if your own Fritz goes for 'the sideline' 9. ...
Nh1. If so, on 10 Qe2 will it play d5?! another [M]
analysis of this line has been thought to be equal by move 16 [de
Zeeuw] but Fritz does not agree. So to get the 9. ... Nh1 sideline out
of the way
10 Qe2 d5 [Fritz wants d6]
11 Bb5 e4
12 Qf1 Bh3
13 Qf7 Kd8
14 Bf1 Q1
15 Nc3 Qf1
16 Qf1 and which gives black -1.1 instead as Alberts says 'equality'.
Now - the very strange thing about the computer's analysis is that the
9. ... Nd4 is a MIRACLE move 'given the observation that the risk
comes down from +4.3 [the ultime limit] to a draw after 10. Bf7. So
Fritz wants 10. Nc3 but again white will go under. The only move is',
says Alberts
10. hg3 Qg3 (=3D0.00)
11 Kf1 Qg4
12 Qg5
And the draw is a fact.
Alberts goes on to say that this highly compact miniature of only 12
moves, with 2 human intercession [M] moves as 'essential
interventions' - which the computer can't see.
Now then: his Question is: is 6. Kf1 the definitive proof that the
Traxler Gambit is overbold?
A summary of 9. ... Nh1 is that left alone Fritz does not 'see'
10. ...d5, and will continue to play until at move 12 it likes Kh1,
whereby 12. ... Bg4 wins for black.
In the next segment we will see another [M] move, but at 11 for black,
which gives Black a runaway win.
Phil Innes
|
| |
Date: 16 Jan 2009 05:03:14
From:
Subject: Re: Traxler Gambit
|
> =A0 I here introduced the concept of Traxler in
> order to discuss chess-- not engage in
> petty pedantry of what-piece-can-move-
> where. =A0 If you can not understand that, Mr.
> Murray, then Pfft to you! =A0 Get your own
> shrubbery.
>
> =A0 R
>
> =A0 =A0O
>
> =A0 =A0 =A0F
>
> =A0 =A0 =A0L
>
> =A0 =A0 =A0!
>
> =A0 =A0--PI
Let me set aside the opportunity to redress the False Frequently-False-
Phil [F-FFP], who is obviously just some jealous monk 'coming out' of
the shrubbery, albeit sideways with a joke, not unlike Taylor Kingston
would do, though he dressed like the Pope
otherwise the chess side of this will get all messed up. Sorry Mike,
we literally lost the thread, so couldn't follow further than Mike's
Fritz refusing Kxf2, preferring Kf1
If the casual reader didn't know this, then at least they won't give
up the ghost at move 6 - and to be fair to the casual reader, very
strong players have argued about the Traxler for decades.
Thank you for supporting Usenet, and the rights and wrongs of crabby
monks!
Now... back to the Games...
The real... Frequently-False-Phil etc
|
| |
Date: 16 Jan 2009 01:39:12
From: help bot
Subject: Re: Traxler Gambit
|
On Jan 16, 1:48=A0am, Mike Murray <[email protected] > wrote:
> =A0Phil bot wrote something and
> Our deepest fears realized. =A0It's contagious.
Pfft! How can one discuss chess with someone
whose sole purpose here is to emit fumes-- to
rubbish great players such as myself? Impossible.
Were Mr. Murray not such an incorrigible patzer,
he would easily figure out that the reference to
...Qh4 obviously apllies to the following variation
which I shall lay out for him and all the other mere
patzers here:
1. e4 e5
2. Nf3 Nc6
3. Bc4 Nf6
4. Ng5 Bc5
5. Nxf7 Bxf2+
6. Kxf2 Nxe4+
7. Kg1 (drumroll)
...Qh4
Now then, what's so very difficult about
this? It should be obvious... to any half-
decent chessplayer. Oh, I forgot-- I am
dealing here with rank patzers, to whom
I could likely give Rook odds. Pfft!
You are lucky I had time to stop in the
middle of writing my new book on the
origin of Andean Spanish, and reply--
though why I bother when all you lowly
patzers cannot understand real chess...
cannot comprehend the subtleties that
are only intelligible to the few, like me.
You see, it all goes back to game 3 of
Taimanov-Fischer... to the move Qh3!!
Pfft. Even Kasparov himself could not
solve the key variation, nor Fritz on high
boil setting... Pfft!
I am a Celt, Mr. Murray-- not a Murican.
You want to see if you can out-drink me,
eh? Irish whiskey, laddie-- not that corn-
pone rubbish... er, Jack Daniel-Boone or
whathaveyou. Pfft! Pfft! (hic)
Do you see that this is the main line?
That White dare not snatch the proffered
Bishop, but does well to side-step by
K-f1? Oh, why do I waste my time with
these patzers, who know not the
difference between a Traxler and two
short planks? Whose intellect is that
of two short planks... Pfft!
--PI
|
| |
Date: 15 Jan 2009 21:51:16
From: help bot
Subject: Re: Traxler Gambit
|
On Jan 15, 7:04=A0pm, Mike Murray <[email protected] > wrote:
> On Thu, 15 Jan 2009 11:38:21 -0800 (PST), [email protected] wrote:
> >Ah - 6. Kf1 Qh4 is the main alternative - in case anyone else comes up
> >with that I have some analysis - but to follow Fritz-Mike
>
> (I thought I'd replied
Murray thought? Can he think-- does he
even comprehend the meaning of the word?
Pfft!
> , but my post seems to not appear.)
To not appear, or to disappear? What Mr.
Vagueness believes may not be what is so,
but merely a manifestation of his own
vagueries and compulsion to emit fumes
from his two cranial orifices. ROFL!
> Doesn't your Knight Block the Queen from h4 ?
What is a Knight? Or a Queen, for that
matter. Can you define it? Measure it?
Or is it a mere concept, shrouded in vaguery,
and devoid of any inherent meaning per se.
Mr. Murray thinks the Queen cannot reach
square h4 in the C21st, but is this correct?
Can not my hand lift the chessman, the
Queen, and set her down wherever I wish?
What vagueness! For a shrubbery I would
refute the notion, but not to appease those
whose motives are clearly to rubbish myself
and others-- real chessplayers, not mere
commentators.
I here introduced the concept of Traxler in
order to discuss chess-- not engage in
petty pedantry of what-piece-can-move-
where. If you can not understand that, Mr.
Murray, then Pfft to you! Get your own
shrubbery.
R
O
F
L
!
--PI
|
| | |
Date: 15 Jan 2009 22:48:30
From: Mike Murray
Subject: Re: Traxler Gambit
|
On Thu, 15 Jan 2009 21:51:16 -0800 (PST), <[email protected] >
wrote:
Phil bot wrote something and
Our deepest fears realized. It's contagious.
|
| |
Date: 15 Jan 2009 12:16:40
From:
Subject: Re: Traxler Gambit
|
On Jan 15, 10:54=A0am, [email protected] wrote:
> On Jan 15, 2:36=A0pm, [email protected] wrote:
>
>
>
> > =A0albert alberts demonstrated the machines ineptness to evaluate it
> > correctly in his first MAMS book - I'll give the moves, and then it
> > would be interesting to review 2 things
>
> > what 'fritz' [any computer engine, even Sanny's] thinks of it
> > &
> > from a database, what are actual results or white/black scores
>
> > in the dreadfully complicated
>
> > TRAXLER GAMBIT
>
> > 1 e4 e5
> > 2 Nf3 Nc6
> > 3 Bc4 Nf6
> > 4 Ng5 Bc5
> > 5 Nf7 Bf2
>
> > [where alberts says (+-3.44)
>
> > as we know Tal played it in correspondence against 10,000 Pravda
> > readers 68/69 with commentary by M de Zeeuw in Yearbook 67 - but I
> > exclude that from actual encounters.
>
> > but what happens next is albert's attempt to unravel this opening - he
> > notes that 'right off the bat' Fritz refuses the bishop...
>
> > anyway, what does your computer think. I'll illustrate more moves if
> > the conversation proceeds, and also transpose this note to its on
> > thread 'traxler'
>
> > Phil Innes
>
> This does not define the Traxler itself but a variation of the
> Traxler. =A0[Not saying that you said otherwise.]
> The moves above lead to a Traxler after Black's 4th move.
> I followed the theory in the late eighties and there was a consensus
> that 5.Bxf7+ is White's strongest move.
Paul - good post, and save those thoughts - for the moment looks like
Mike and I headed down another tree - and yes book=3Doff
> I'm not surprised Fritz refuses the bishop -- it's probably just
> following its opening book. =A0Declining the bishop sacrifice is
> considered best.
sure - looks like black can force a quickish draw by repetition after
the capture
> Does Fritz refuse the bishop if you make it play without its opening
> book?
ay
but where we are headed is a refutation of Traxler.
I ask you to keep the thought, since some other and current lines go
deep, and you can encounter for example Syslov's opinion of a move 29
- and Tal twiced missed a mate,, once at 26, and another opportunity
at 27 - eventually Smyslov pointed out the solution.
Your line deserves its own thread entirely - though I think Frtiz now
prefers this one.
Cordially, Phil Innes
> Paul Epstein
|
| |
Date: 15 Jan 2009 11:38:21
From:
Subject: Re: Traxler Gambit
|
On Jan 15, 11:25=A0am, Mike Murray <[email protected] > wrote:
> On Thu, 15 Jan 2009 05:36:48 -0800 (PST), [email protected] wrote:
> >in the dreadfully complicated
>
> >TRAXLER GAMBIT
>
> >1 e4 e5
> >2 Nf3 Nc6
> >3 Bc4 Nf6
> >4 Ng5 Bc5
> >5 Nf7 Bf2
> >[where alberts says (+-3.44)
> >but what happens next is albert's attempt to unravel this opening - he
> >notes that 'right off the bat' Fritz refuses the bishop...
> >anyway, what does your computer think.
>
> After 15 minutes on my computer, Rybka discarded the Bishop capture,
> and had White up by 1.13 on 6 Kf1 with 6 Ke2 about even. =A0
>
> I played the line forward to 6 Kf1 Qe7,
Ah - 6. Kf1 Qh4 is the main alternative - in case anyone else comes up
with that I have some analysis - but to follow Fritz-Mike
> 7 Nxh8 and gave it another
> fifteen minutes. =A0At this point, Rybka considered 7 ... Bc5 (White
> plus 1) and 7 ... Bb6 (White plus 1.07) roughly equal.
OK - at longer examination Bb6 seems to be the predominant choice. But
Fritz can't see the way forward - therefore a whole succession of
moves need a [M] Mams intercession.
I will make another pause here, since there may be other efforts at
refuting the Traxler, and to give people a chance to catch up with the
main chances.
Phil Innes
> The "book" line, 7 ... d5, =A0didn't make the top two favored
> candidates. =A0I decided to play it anyway, and gave Rybka another
> fifteen to think about it.
>
> At that point, it was strongly preferring 8 exd5, having White up by
> 1.86.
>
> Probably, Rybka 3 on a faster processor would revise some of these
> numbers.
|
| | |
Date: 15 Jan 2009 12:32:30
From: Mike Murray
Subject: Re: Traxler Gambit
|
On Thu, 15 Jan 2009 11:38:21 -0800 (PST), [email protected] wrote:
>On Jan 15, 11:25�am, Mike Murray <[email protected]> wrote:
>> On Thu, 15 Jan 2009 05:36:48 -0800 (PST), [email protected] wrote:
>> >in the dreadfully complicated
>>
>> >TRAXLER GAMBIT
>>
>> >1 e4 e5
>> >2 Nf3 Nc6
>> >3 Bc4 Nf6
>> >4 Ng5 Bc5
>> >5 Nf7 Bf2
>> >[where alberts says (+-3.44)
>> >but what happens next is albert's attempt to unravel this opening - he
>> >notes that 'right off the bat' Fritz refuses the bishop...
>> >anyway, what does your computer think.
>>
>> After 15 minutes on my computer, Rybka discarded the Bishop capture,
>> and had White up by 1.13 on 6 Kf1 with 6 Ke2 about even. �
>>
>> I played the line forward to 6 Kf1 Qe7,
>
>Ah - 6. Kf1 Qh4 is the main alternative - in case anyone else comes up
>with that I have some analysis - but to follow Fritz-Mike
Doesn't the Knight block the Queen from h4 ?
|
| | |
Date: 15 Jan 2009 16:04:38
From: Mike Murray
Subject: Re: Traxler Gambit
|
On Thu, 15 Jan 2009 11:38:21 -0800 (PST), [email protected] wrote:
>Ah - 6. Kf1 Qh4 is the main alternative - in case anyone else comes up
>with that I have some analysis - but to follow Fritz-Mike
(I thought I'd replied, but my post seems to not appear.)
Doesn't your Knight Block the Queen from h4 ?
|
| |
Date: 15 Jan 2009 11:33:18
From:
Subject: Re: Traxler Gambit
|
On Jan 15, 11:25=A0am, Mike Murray <[email protected] > wrote:
> On Thu, 15 Jan 2009 05:36:48 -0800 (PST), [email protected] wrote:
> >in the dreadfully complicated
>
> >TRAXLER GAMBIT
>
> >1 e4 e5
> >2 Nf3 Nc6
> >3 Bc4 Nf6
> >4 Ng5 Bc5
> >5 Nf7 Bf2
> >[where alberts says (+-3.44)
> >but what happens next is albert's attempt to unravel this opening - he
> >notes that 'right off the bat' Fritz refuses the bishop...
> >anyway, what does your computer think.
>
> After 15 minutes on my computer, Rybka discarded the Bishop capture,
> and had White up by 1.13 on 6 Kf1 with 6 Ke2 about even. =A0
ah ok - so the discarded line by Rybka is
6. Kf2 Ne4 [when then -3.44 is reduced to -2.34]
7. Kg1 [Fritz doesn't like 7 Ke3 Qh4, 8. g3 Ng3 9. hg3 Qd4 10. Kf3 d5
11. Rh4 e4 12. Kg2 0-0 which is a line by Christophe and Moll - Fritz
can't deal with 13 Nc3 dc4! [a MAMS move, that is, a man-inserted one
- hertofore notated by [M] ] 14. Qh5 Ne7! [spotted at depth 12, says
Alberts] then 15 Ne4 Bf5 16. Nfg5 h6 17. Nc3 Bg4 18. Nh3 Qf6 19. Rg4
Qf1 20. Kh2 Rf2 21 Nf2 Qf2 draw by perpetual.
But that is to discard a 'mainline' following the bishop capture.
Instead, and to return to this post later - we can look at 6. Kf1 as
Mike Murray observes below.
Phil Innes
> I played the line forward to 6 Kf1 Qe7, 7 Nxh8 and gave it another
> fifteen minutes. =A0At this point, Rybka considered 7 ... Bc5 (White
> plus 1) and 7 ... Bb6 (White plus 1.07) roughly equal.
>
> The "book" line, 7 ... d5, =A0didn't make the top two favored
> candidates. =A0I decided to play it anyway, and gave Rybka another
> fifteen to think about it.
>
> At that point, it was strongly preferring 8 exd5, having White up by
> 1.86.
>
> Probably, Rybka 3 on a faster processor would revise some of these
> numbers.
|
| |
Date: 15 Jan 2009 08:25:08
From: Mike Murray
Subject: Re: Traxler Gambit
|
On Thu, 15 Jan 2009 05:36:48 -0800 (PST), [email protected] wrote:
>in the dreadfully complicated
>
>TRAXLER GAMBIT
>
>1 e4 e5
>2 Nf3 Nc6
>3 Bc4 Nf6
>4 Ng5 Bc5
>5 Nf7 Bf2
>[where alberts says (+-3.44)
>but what happens next is albert's attempt to unravel this opening - he
>notes that 'right off the bat' Fritz refuses the bishop...
>anyway, what does your computer think.
After 15 minutes on my computer, Rybka discarded the Bishop capture,
and had White up by 1.13 on 6 Kf1 with 6 Ke2 about even.
I played the line forward to 6 Kf1 Qe7, 7 Nxh8 and gave it another
fifteen minutes. At this point, Rybka considered 7 ... Bc5 (White
plus 1) and 7 ... Bb6 (White plus 1.07) roughly equal.
The "book" line, 7 ... d5, didn't make the top two favored
candidates. I decided to play it anyway, and gave Rybka another
fifteen to think about it.
At that point, it was strongly preferring 8 exd5, having White up by
1.86.
Probably, Rybka 3 on a faster processor would revise some of these
numbers.
|
| |
Date: 15 Jan 2009 07:54:35
From:
Subject: Re: Traxler Gambit
|
On Jan 15, 2:36=A0pm, [email protected] wrote:
> =A0albert alberts demonstrated the machines ineptness to evaluate it
> correctly in his first MAMS book - I'll give the moves, and then it
> would be interesting to review 2 things
>
> what 'fritz' [any computer engine, even Sanny's] thinks of it
> &
> from a database, what are actual results or white/black scores
>
> in the dreadfully complicated
>
> TRAXLER GAMBIT
>
> 1 e4 e5
> 2 Nf3 Nc6
> 3 Bc4 Nf6
> 4 Ng5 Bc5
> 5 Nf7 Bf2
>
> [where alberts says (+-3.44)
>
> as we know Tal played it in correspondence against 10,000 Pravda
> readers 68/69 with commentary by M de Zeeuw in Yearbook 67 - but I
> exclude that from actual encounters.
>
> but what happens next is albert's attempt to unravel this opening - he
> notes that 'right off the bat' Fritz refuses the bishop...
>
> anyway, what does your computer think. I'll illustrate more moves if
> the conversation proceeds, and also transpose this note to its on
> thread 'traxler'
>
> Phil Innes
This does not define the Traxler itself but a variation of the
Traxler. [Not saying that you said otherwise.]
The moves above lead to a Traxler after Black's 4th move.
I followed the theory in the late eighties and there was a consensus
that 5.Bxf7+ is White's strongest move.
I'm not surprised Fritz refuses the bishop -- it's probably just
following its opening book. Declining the bishop sacrifice is
considered best.
Does Fritz refuse the bishop if you make it play without its opening
book?
Paul Epstein
|
|
