Question about PME position

I was looking back to Month 1 on PME. They then give an example from Kramnic vs Topalov to explain the thinking and state-

"It is a' PEM' where black has 2 Knights an a rook against 2 Rooks and a Pawn, It is an endgame position, so the material of the 2 players is equivalent"

This looks like an omission. The table supplied does not show this as a PME in fact it does not refer to 2 Knights at all. There would seem to be an omission in the table. Whilst intuitively I may have come to the same conclusion, I would not have been confident with my own assumption.

hoopy wrote:Think I have another example.
I was looking back to Month 1 on PME. They then give an example from Kramnic vs Topalov to explain the thinking and state-
...

The text critisized move 29) Rc1 and suggested 29) b4. Fritz select Rc1 as his #2 move but b4 to his #18. Rybka and Cyclone think almost exact the same. Well, Kramnic seems to be a little better then our teachers ;-). I will never understand why these IMs and GMs dont check their opinions by computer. Maybe thats the reason why they dont get better themself ?

Actually they don't claim b4 to be an objectively better move at all, rather they argue that it is perhaps the best 'practical' chance as demonstrated through their own games in that position.

I'm not qualified to give an opinion on whether they are right or wrong in this specific case as the chess being played is well over my level but I certainly can agree that many times in real chess, especially when behind, a player's best choice is to complicate matters and play a move which although objectively weaker when analysed with a computer can give a human opponent far more trouble over the board. A great many of Tal's sacrifices for example have now been shown to have been unsound, completely irelevent of course, as he wasn't playing Deep Blue he was playing fallible humans and his attacking combinations created problems that were way over the heads of his contempories to solve. Was he wrong? I don't think so.

I think they make their point pretty clearly, after 29.Rc1 it becomes more difficult to advance the white queenside pawns thus reducing white's counterplay. Objectively under the light of computer analysis it may be better than 29.b4 but if it reduces whites counterplay in real human terms it makes black's task easier.

So I don't think the ICS team are wrong at all in what they say.

As for your statement about maybe they should look at computers more I would argue the complete opposite. You should look at the Rybka analysis and pawn score far less. That won't help you much in a real game, because you won't be playing a computer you will be playing a real person and a computers moves are made purely as a result of calculation to a certain depth with little or no positional understanding at all. We simply cannot play like computers because we are not calculating machines and therefore rely on strategies and positional knowledge about which the computer knows nothing.

If we go back to the game in question the ICS team are providing the student with a method whereby they can evaluate positions of material equivilence, i.e sides have different material but of the same approximate value.

They give four criteria for evaluating such positions, one of which is the existence of mobile pawn majorities and the ability to create a passed pawn.

Regarding 29. b4 that move is entirely consistent with their evaluation criteria and white in having a mobile pawn majority on the queenside should play to promote that pawn.

That's why 29.b4. Whether a computer evaluates Rc1 as being 29/100ths of a pawn better is neither here or there. The ICS guys with their criteria of evaluation have given us amatuer players a clear guide and plan of play through the evalution criteria which will give you practical chances in a real game.

Just having Rybka state that 29.Rc1 is better by 29/100ths of a pawn gives me, an amatuer player no help or insight whatsoever. How is that going to give me a plan of action that I can use in a future game in a similar situation?

Remember humans navigate the terrain of the 64 squares by way of a map and compass, a computer just calculates things out to brute force.

If you try and think like a computer or make computer moves without their perfect tactical calculation skills you will fail, whereas if you use a map, signposts etc which is what ICS are trying to teach you may make it to your destination.

Kramnik ( Elo 2759) did chose Rc1 like these 3000+ Elo chessengines. And Kramik is not behind at Move 29, the position is equal ( says Rybka & Co ). He would have been a little behind after b4. Kramnik did know what a Majority attack is and how it works. I think he did see the weakness of b4. I am shure he did not play this move just to show us what a bad move is ;-). My experience is: as better the Player as more often he uses the best ( or #2) computermove.
Rybka is programmed by an IM and using positional algorithms like all good chessprograms. Brute force programs are not good, the computers are to slow for that. 0.29 is not much for me, but for players like Kramink it is.

With the words of our Teacher:
"So, is it a waste of time to look for the best move in every position? My answer is firmly NO."
...
"The more frequently you will find and use the best moves, the higher your chess level will be."

Some points, firstly Kramnik lost

Secondly, the ICS course isn't aimed at Vladimir Kramnik it is geared towards amatuer players looking for a positional strategy that they can adopt in their games, i.e a plan of action based on positional factors in a game of material equivilence.

So the ICS team highlight that 29.b4! is a better 'practical' chance in accordance with their criterias of evaluation.

I'm presuming you are rated at less than 2759 so how does Rybka's 29.Rc1 tell you specifically how to play the rest of that game? Imagine if you were at the board and were allowed advice from either Rybka or ICS on this next move what would be more helpful in practical real terms?

Knowing that the best move according to Rybka was 29.Rc1 because its 29/100ths of a pawn better according to its analytical computatations or would you rather be given the four criterias of evaluating a position of material equivilance according to the ICS team?

I would choose the ICS tips EVERY time. They give you a plan. The fourth criteria is to make use of any mobile pawn majority to create a passed pawn. That right there is something I can work towards, I have a goal.

29.Rc1 may be 29/100ths of a pawn better according to Rybka analysis but it doesn't give me any information or insight as to what I should do next.

That is why the ICS team say 29.b4! is a better 'practical' choice, meaning that if YOU play 29.b4 in that situation with a plan of trying to advance your queenside pawns you will have a better chance of saving or maybe even winning the game because you are following a plan.

How is that hard to understand?

To use an analogy its like travelling in a foreign country in which you don't know the language or the area.

Suppose you take a hire car and are given clear directions to get to a town 50 miles away that you want to go to.

Now the guide tells you the most sensible route, the one you are least likely to get lost on but not necessarily the quickest. The locals instead take a short cut route through some unmarked back lanes with no signs and through some treacherous terrain.

Now imagine you know that the shortcut begins by turning right at the next junction and then following the road until you come to a dirt path on the left. You know that if you were a local and knew those roads it would be quicker to go down that dirt road and then through the back lanes but in 'practical' terms would that be sensible seeing as once you are down that road you have no idea where next to go?

Would you want to just take a chance and drive down these country backroads miles from any town or village or would you just stick to the highway because you know you will get to your destination safely without risking getting completely lost?

That is the difference between following general chess principles, the positional strategies and ideas laid formulated by the greatest chess minds over the past 150 years or by choosing a move based purely on computer analysis without you understanding exactly what to do next.

If you insist on blindly following computer recommendations you will never seriously improve at chess because computers don't play like humans.

Despite losing 29.Rc1 might have been the best move for Kramnik to play, but he's one of the greatest players of all time and has been studying the game for probably almost 30 years.

The ICS course isn't aimed at him its aimed at us, hence 29.b4 can be a better and more practical move for us.

I like and enjoy these lessons, but b4 is not the best move. Our teachers are not the best players of the world and they make errors. so what.
I am very suprised about the fears against computeranalyses. As far as i see russian chessschools suggest playing against computers to become better than 2200. ( see: Modern methods for training a chess player at http://www.convekta.com/softscho/l1/lesson_3.html ). This might be the reason why the Russian-SuperGM's play so often like Fritz & Co.

Begin of joke ->

Bilbo wrote:Some points, firstly Kramnik lost

Some points, firstly our Teacher where never close to Worldchampionchip
Second Kramnik lost later (after he made some errors).
<- End of joke

Speculation based on computeranalysis: The goal of Rc1 is to attack the knight on b6 with Rc6 after a7-a5 and/or (later) double the rocks on the 7 th rank and blocking the black rock from access to open files, what deminish the role of the coordinating piece.

Open files and ranks for the rocks, espiacily rank #7 are not mentioned in this text. ( will come later, i am shure. No one can say, open files and ranks for rocks are of no importance in PME's. The faktors named her are just the first! 4, later to come: isolated pwans, minority attack (just the opposit! of whats mentioned here with queensidemajoritys!) and so on)
Just think what would have been the logical move, if the writer did not write about queensidemajoritys but the importance of open files ( compare to half open ) and conquering the 7 rank with rocks and blocking the activety of the coordinating piece.

Thats one problem of positional understanding. The set of positional aspects/factors is not fixed on the few from Steinitz or the 4 mentioned here, there are now, as far as i see more than 100.

Its looks logical, after highlighting the importance of queensidemajoritys and majorityattacks to play b4

and

Its looks logical, after highlighting the importance of xy and yz to play W.

If you want to make a special move look logical, you simply mention the importance of strategie #75 in the beginning of you text and then ...
To find postmortem arguments for some different moves is not that difficult. But to play as good as Kramnik does.

Just because 3 extrem good programs support the decission of Kramnik, dont make this move necessary a bad move. You dont know what to play after Rc1? but do you know what to play after b4?
Begin joke ->
That is the difference between following general chess principles, the ( duzends ) positional strategies and ideas laid formulated by the greatest chess minds over the past 150 years ( like Kramnik ) or by choosing a move based purely on 4 principles.
If you insist on blindly deny computer recommendations you will never seriously improve at chess because the best players play ( more often ) like computers.
The ICS course isn't aimed at bad moves, its aimed to bring us to 2300 Elo
<- End joke

Again the ICS team did NOT say that 29.b4 was objectively the best move according to computer analysis they said that it was the 'practical' choice making use of white's advantage in the position (potential to create a pased pawn) and giving him some counterplay.

A computer's analysis is not the de facto standard for weighing up a position. Mihal Marin one of the world's greatest chess writers analyses this game and concludes that after 29. Rc1 a5 30 Rc6 Nd5 that Black IS better yet Rybka gives this position as 0.09 in favour of White.

Who is right? Marin or Rybka?

Why not try and win for black after 29.b4 using your beloved Rybka? I think you will find that the combined pressure from both the b7 and d1 rooks on the pinned d7 knight along with a passed pawn on b4 (b4 a5 a3 axb4 axb4) means that black cannot make any progress.

There are many occasions in chess where objectively the best move in a purely analytical sense is not the best practical move to play.

At least 29.b4 complicates the game, gives white some counterplay and thus interferes with black's plans. He has to keep an eye out for that pawn.

But please, as you are so sure Rybka is right in declaring 29.b4 to be a bad move then provide the analysis for your forced win for black.

I think the example ICS chose is poor because it's way too complicated! They could of chose a simpler example that would be a bit more clear.

If memory serves, the example is mainly meant to show the idea of the coordinating piece, and it seems to be good for that. I wouldn't read too much into it. The point is, if you have two rooks and pieces versus one rook and pieces, try to trade one of your rooks (and if you are the one with a single rook, don't trade it). However, fanat may be right that they could have picked a simpler example.

A discussion of computer versus human evaluation at my blog:
http://chessconfessions.blogspot.com/2009/06/computers-in-chess.html

(The post just says 'The computer is both useless and indispensable for postmortems' but the fun happens in the comments).

And my favorite quote, from a chess publisher:
"Chess Stars books are not a digest of old books dressed with some computer analysis here and there. They attempt to mark new ways and steer the fashion into other directions. For that they count on human evaluations. Computers must associate every position with a number. On the contrary, a strong GM can sense a bad coordination and rule out lines that the engine claims to be good for him. Thus he can save weeks or months of futile analyses. So everyone can probe endless lines on his personal PC, but eventually one or two games at top level will tell the verdict about the principle variation. Or maybe they already have?!"

Wow you guys have been hammering this one. Very good stuff. For my part I don't really think it matter whether b4 or Rc1 is better or worse. In the context of the lession I don't care. What counts is
1. The concept of the coordinating piece in PME thinking.
2. The thinking process.
Perfection for humans in chess is impossible but you can go along way in perfecting the thought process that goes into move selection.

I stay away from computer analysis when looking at these positions. I prefer to do the work myself.

Chess a game between two flawed humans who can't remember their opening theory and have this irritating tune running through their head as they analyze the same line over and over again ...

In the next tournament when the clock is ticking Fritz won't be there ... some of course will have pocket fritz ...

BobbyBlunder wrote:Chess a game between two flawed humans who can't remember their opening theory and have this irritating tune running through their head as they analyze the same line over and over again ...

LOL, well put.

This is my mind during a tournament:

Bilbo wrote: Again the ICS team did NOT say that 29.b4 was objectively the best move according to computer analysis they said that it was the 'practical' choice making use of white's advantage in the position (potential to create a pased pawn) and giving him some counterplay.

A computer's analysis is not the de facto standard for weighing up a position. Mihal Marin one of the world's greatest chess writers analyses this game and concludes that after 29. Rc1 a5 30 Rc6 Nd5 that Black IS better yet Rybka gives this position as 0.09 in favour of White.

Who is right? Marin or Rybka?

Marin :As of February 2009, his FIDE rating is 2556

Rybka: 3250
Kramnik: 2759

Who is better?

Bilbo wrote: Why not try and win for black after 29.b4 using your beloved Rybka?

Ok, you play the moves of Marin and i play the moves of Rybka. But you dont need me for this game. Simply play b4 against rybka on your own computer ( if it is not tooo old ) and tell us the result. I think it would be fair if you get 10 min more then Rybka each move.

Blue Devil Knight wrote:A discussion of computer versus human evaluation at my blog:
http://chessconfessions.blogspot.com/2009/06/computers-in-chess.html

(The post just says 'The computer is both useless and indispensable for postmortems' but the fun happens in the comments).

And my favorite quote, from a chess publisher:
"Chess Stars books are not a digest of old books dressed with some computer analysis here and there. They attempt to mark new ways and steer the fashion into other directions. For that they count on human evaluations. Computers must associate every position with a number. On the contrary, a strong GM can sense a bad coordination and rule out lines that the engine claims to be good for him. Thus he can save weeks or months of futile analyses. So everyone can probe endless lines on his personal PC, but eventually one or two games at top level will tell the verdict about the principle variation. Or maybe they already have?!"

This might be of interest:
http://www.chessbase.com/news/2006/world_champions2006.pdf

Anand for example is analysing his games with a computer:
Anand said that use of computer technologies to practice and train, to develop new opening plans, analyse complex positions and solve difficult endings have completely revolutionised the way chess is played ( see: http://www.chessbase.com/newsdetail.asp?newsid=5109 )

I dont know any SGM who is not using computers for analsysing chesspositions, is ther any?

I'm not sure if you're even being serious or not but your idea that Kramnik because he has a higher rating is better able to judge a game over the board than one of the world's most renowoned theoreticians can do retrospectively and with computer analysis is ridiculous.

Marin may not be as good in over the board tournament play as Kramnik but he's certainly capable of correctly analysing the game.

Rybka itself is often wrong in analysis. Only tonight at my local club I played white against the Bogo Indian defense the following variation.

1.d4 Nf6 2.c4 e6 3.Nf3 Bb4+ 4.Nd2 b6 5.a3 Be7

Be7 is a mistake because it gives up the center after 6.e4 Bb7 Bd3 and if you look in Fritz opening database Be7 is simply never played in theory, the only and correct move is Bxd2 played 208 times compared to Be7 0 times.

Yet Rybka only gives white a 0.09 advantage after 5...Be7 whereas the theory books have this as a big advantage for white += already.

Clearly Rybka doesn't know everything.

There is nothing wrong at all with the move 29.b4. The position is still equal even according to Rybka, there is just now a -0.28 advantage to black which if you actually play out the next several moves in Rybka soon drops to about -0.13 and a drawn position.

As for the rest of your posts who has said that computers arn't a valuable tool? Everybody should make use of computers that's a complete no brainer but being a slave to every evaluation will get you nowhere in chess.

If the only correct move in a postion was the one Rybka recommends there would be less than half a dozen openings and at grandmaster level every single game would be played out the same.

So in summation you are simply dead wrong about the ICS team making a mistake with 29.b4. It's not a mistake, its an alternative move recommendation in line with their criteria of evaluation that is holding for white. Just run it through Rybka and try and force a win for black as I said before.

Finally computers and humans have totally different ways of evaluating a position. You simply cannot play like a computer because you cannot evaluate like it does by calculating millions of variations per second.

Humans have to make use of positional and strategic factors to find an appropriate plan and that is what ICS is teaching. Thus 29.b4 was their choice because it was in line with their criteria of evaluation and positional assessment.

They were working to a different plan than Rybka and Kramnik. There is nearly always room for more than one plan in chess that the whole reason why we bother to play it.

But good luck with improving your chess based on getting Rybka to tell you the best moves AFTER the game and not accepting advice from any grandmaster lower than 2753. Hope that works out for you.

Bilbo wrote:I'm not sure if you're even being serious or not but your idea that Kramnik because he has a higher rating is better able to judge a game over the board than one of the world's most renowoned theoreticians can do retrospectively and with computer analysis is ridiculous.

Marin may not be as good in over the board tournament play as Kramnik but he's certainly capable of correctly analysing the game.

Susan Polgar ( 2577 ) is certainly capable of correctly analysing the game:
http://susanpolgar.blogspot.com/2006/10/game-8-analysis-topalov-scores-1st-win.html
for Susan: 32.a4?! A positional mistake!
but 29.Tc1 is ok

No need to be angry:
"But good luck with improving your chess based on getting Marin to tell you the best moves AFTER the game and not accepting advice from any grandmaster higher than 2550. Hope that works out for you."

Just found an alternative:
http://www.chessclub.com/resources/event/elista06/round8.html
Amador Rodriguez ( <2500 ) suggest 29.Kf1 followed by 30.Ke2 one of the better moves in Rybkas candidates.

What does 32.a4? have to do with 29.b4? It's a completely different position.

They are completely different moves. It's clear that Polgar basically agrees with the ICS team, saying after 29.Rc1 that white would have been better trying to exchange rooks or pawns. Well how does he go about exchanging pawns? 29.b4 obviously seeing as black HAS to play 29...a5 next to free his rook up. After 30.a3 black has nothing better than 30...axb4 31.axb4 which is the line that ICS AND Polgar are recommending.

I think you are digging a hole here

Bilbo wrote:What does 32.a4?

Thats the error in the opinion of Susan Polger. The Questionmark shows it.

Bilbo wrote: It's clear that Polgar basically agrees with the ICS team, saying after 29.Rc1 that white would have been better trying to exchange rooks or pawns.

No she is not using past tense and its not after 29. Rc1 but after the response 29. .. a5. By the way is 29.) b4 supose to exchange the rooks? Hm, then i realy did not understand this move. That Rc1 would work in this direction i do understand ( i think / i hope )

I think you are clueless quite honestly. After 29.Rc1 a5 Polgar criticises Kramnik for not trying to exchange either a rook or pawns. Note she puts this criticism after 29.Rc1, not before or after any of his other moves. ...a5 is blacks move not whites so its clear she is making this comment regarding white's last move Rc1.

If. b4 is played prior to a5 then white can force a pawn exchange. If b4 is not played prior to a5 then white can simply no longer play b4 on the next move because of axb4.

32.a4 is a completely different move to 29.b4. If you think that moving a different pawn at a different time in the game in a different position is the same thing then you really don't know much about chess at all my friend.

Have you even once looked at the actual board during this discussion and seen with your own eyes the position after 29.b4 a5 30.a3 axb3 31.axb4???

It's clearly preferable to Kramniks actual strategy in this game, White has a passed pawn, tremendous pressure on the pinned knight and now has real counterplay.

Black is still probably slightly better but in PRACTICAL terms his task is now much harder because he has to keep an eye on whites passed b pawn.

I'm seriously thinking you might be an idiot no offense of course

Just play out the positon after 29.b4 and see what black can do. Ok so he might be 29/100ths of a pawn better according to Rybka now but it doesn't mean anything in real terms, his chances of winning the game are slim because the passed b pawn and pressure on black's pieces means he can't make any progress.

I did not read the whole bunch of text of this topic, just between the lines, and i do attest that best players are not automatically the best teachers.

It is exactly the same at school, teaching requires pedagogy and omit volontarily some too much complex things for your attendance, and focus on the basics, solid roots.

Guess it is the same for this game, Kramnik did calculated some move he thought was good for him at this time, but maybe he would choose another one tommorrow, depending on his mood, on the necessity to win a full point in the tournament or to draw, on the pressure, etc...

On the other side, some other chess players are very good at analysing games, quietly. To my opinion, the rating is not so important, if the player is not completely dumb (as we are, but we are not GMs (...yet)), and is talented in analysis, he could give some precious advices on the game played.

A proof, is that top players do have 1 or more chessmates or trainers that are underrated compared to them, nevertheless they need their opinion because playing and analysing, preparing, teaching is not the same.

Take Dan Heisman for example, this guy is 2300, far from top list, indeed his chess lessons are famous.

And for ICS team, i think the same, maybe they are not top players, but for sure they are good enough to teach us.

No need for name calling Bilbo!

This topic brings out the emotions almost as much as discussion of openings.

Back to the topic...

If you had to choose between getting postmortem help from a good human teacher or Rybka which would you do? A computer spits out variations, a human spits out those of course but more importantly ideas. Which are more likely to generalize to different positions? Which is more likely to give you advice that will be practical at your level of chess?

Of course we don't have to choose between the two. The computer is both useless and indispensable for postmortem analysis.

Computers are great in sharp positions. If they see a mate in 20 they will make that the top move, even if there is a longer win that is ridiculously easy and simple for the person winning (e.g., trade queens, simplify, etc), a win with very little risk.

That is just one example of computers stupidly (by human standards) going for top move, regardless of things like ease of practical play for humans, limiting counterplay, and flexibility. Human brains are slow, pattern-using machines that can only explore a tiny sliver of the game tree. Computer brains are fast, symbol manipulating tools that explore vast swatches of the game tree in milliseconds. These differences in computational strategies have practical consequences for how the two machines think about the game.

So, of course computers are helpful, and of course they can beat Kramnik. But I'd still rather have Kramnik as a coach if I had to choose.

[quote="AoxomoxoA"]

Bilbo wrote:I think you are clueless quite honestly. After 29.Rc1 a5 Polgar criticises Kramnik for not trying to exchange either a rook or pawns. Note she puts this criticism after 29.Rc1, not before or after any of his other moves. ...a5 is blacks move not whites so its clear she is making this comment regarding white's last move Rc1.

No whenever she is thinking a move is wrong she writes her comments directly after this move and not later after the oponets move. She suggest to lock know for an exchange of rooks after a5.

32.a4 is a completely different move to 29.b4. If you think that moving a different pawn at a different time in the game in a different position is the same thing then you really don't know much about chess at all my friend.

I dont know much about chess, thats the reason i am at this school.
I just mentioned a4? to schow that susan polger thought a4 is an error, there she is setting an ? There is no ? after Rc1

( I am glad that i am still "your friend", its only a game you know ;-)

Have you even once looked at the actual board during this discussion and seen with your own eyes the position after 29.b4 a5 30.a3 axb3 31.axb4???
..
I'm seriously thinking you might be an idiot

I did put this position in several engines, Rybka, Cyclone, Fritz and Zappa Mexico, all engines about 2900+. They judge the position -0.4 till -1.0 ( what would mean a almost clear win for black )
I guess its because the passed pawn is already blocked and an isolated pawn. As i learned at our school, knights are perfect defenders and they are very good in blooking ( good old Nimzowitsch ). It looks to me, that a majorityattack with b4 turns out to become a white - made successful minorityattacK for black. A minorityattack try to create an wheak isolatet pawn on the opponents (queen) side, block it, and then conquer it. An alternativ goal of an minorityattack can be: to get control over an open file. Both Goals are fullfilled with the movesequence you suggest. With one piece more, black have the the local superiority of the forces ( see: making decissions ) and the black King is closer. This might be the reason why 29.Kf1 and the Ke2 is a good move sequence?

And please dont get angry, i am just an Idiot, but i hardly try to get better :-)

Ok first of all, I apologise for calling you an idiot, I tend to get carried away when I'm posting sometimes. No hard feelings.

Back on topic, the idea of 29.b4 is to exchange pawns and give white a passed pawn (and hence counterplay) on the queenside. This means that for the rest of the game black cannot ignore that pawn because if it gets pushed a little more he could be in trouble. You are right that white cannot advance it unless Black played poorly but it still puts a shackle to some extent on black's kingside plans.

As for black being able to blockade and win the pawn, I don't think they can. Remember white has two rooks which are incredibly powerful combined if black wasted time in trying to win that pawn white could gain a powerful inititive with his rooks and probably invade the black postion and go on to win.

Also black's position after 29. b4. a5 30.a3 axb4 31. axb4 is very restricted. One knight is pinned to his king and attacked by both rooks, meaning in turn that black's king is tied to the defense of the knight. I think we all agree that if a rook exchange occurs that will benefit white as his rook will be stronger than the two knights so black has no real way to advance. As long as white maintains his double attack on the pinned knight with his rooks its hard for black to do anything, the position is a draw and I'd say probably not too difficult a draw to make even for a lower rated player.

29.Rc1 may well be an objectively better move, but as I see it is more risky. Firstly after 29...a5 any possibility of creating a passed pawn on the queenside has been greatly reduced for white, also the pinned knight is no longer under as much presser.

It's easy to see how these factors allowed Kramnik to go wrong. Once the knight had been unpinned and was able to jump to d5 white had no advantages in the position. There was no passed pawn for black to worry about, his knights and king were no longer pinned and tied down to the defense of each other and so he could play for the win.

Had Kramnik followed up Rc1 with the correct plan he would possibly have had a slightly better chance of winning the game than after 29.b4 with perfect play (hence the slight 29/100ths of a pawn preference by Rybka) although with correct play by black its still a draw. But Rc1 was a more risky move and led to lines that even a player of 2758 or whatever blundered in and lost the game.

That is why the ICS team said 29.b4 is a more 'practical' move.

Rc1 was not a mistake if followed up correctly and nobody is saying it is. Objectively it is probably the strongest move (I guess I'm not qualified to say). But 29.b4 is a fine move too that is probably safer for white but less ambitious, (I'm basing this on going through the future moves using Rybka, after another dozen moves or so the position just gets repeated and a draw would likely be agreed).

Its just your assumption that because Rybka scores a move more highly that every other move is inferior is just dead wrong. In chess there are rooms for many plans and styles of play. Clearly the ICS staff will have analysed this game (and all others) using computers to check their analyses and with Andrei Istratescu being rated 2630 he is certainly more than capable of instructing us.

In fact arguably the best chess writers in the world today, Mark Dvoretsky, John Watson and Jeremy Silman are not even Grand Masters.

Imagine that, the great Dvorestky trainer of countless world champions and he's just a lowly IM (not because he wasn't good enough but because he gave up his competitive career to teach).

This is the same for guys like Mihal Marin too is an excellent chess writer and analyst.

this conversation just makes me want to play some chess!! good stuff gents.

I think our teachers simply do not trust in engines opinion, here i quote a paragraph from month 7 ("e- and f- Pawn Phalanx"):

"In the game Botvinnik - Capablanca, 1938, after the first 14 moves, the position from the right was reached with Black to move.
Here, the great Capablanca made a grave strategical error. It is very interesting that the strongest chess engine gives his next move as the best:
14...c4?
The move seems very logical (as Capablanca's chess in overall):
-it blocks the white c3- and d4- pawns on the same color with the bishop;
- it sets a strong square for the knight on b3 (Black plans Na6-b8-c6-a5-b3) after which the a4-pawn is damned.
However, this move is wrong because:
- the black's knight maneuver is too long in a position which is far from being totally blocked;
- actually, the knight on b3 will be out of play and surely not on a strong square (the knight will not attack anything from there):
-Black doesn't anticipate next White's plan."