Kramnik-Fritz Chatter

I am rooting for Kramnik of course, but at this point it seems too hard for the human to avoid making any errors of significance throughout the entire game. The psychological effect of playing 40 great moves and then one mediocre move -- and losing as a result (or merely drawing instead of winning) is going to be too great. I don't think that Kramnik can steer all of the positions towards positional battles; chess gives too many opportunities for either opponent to complicate. I predict Fritz 3.5 - Kramnik 2.5.

When the "Mig is a shill for Chessbase" trial begins, this will be Exhibit A in his defense. After reading his piece I think I'll skip the match and attend the party.

Fritz 4-2.

Sort of a fin-de-siecle feel about this match. You feel it'll probably be the last of its kind.

What a pathetic set of rules. MM bores me to tears anyway, but with these kinds of rules, count me out completely.

So this match is all about Kramnik finding openings with not too much tactics and with technically winnable positions. Oh wait Kramnik always does that! I guess Kramnik has a very good style to play against a computer.

I guess with white he'll play 1.Nf3. With black against 1.e4 the Berlin and against 1.d4 the Slav? What do you think?

Also it appears from the rules that if Kramnik survives until move 56 he can adjourn the game and resume play the next day. I don't see anything about Fritz getting to think all night on the position but obviously Kramnik can, with Fritz to check against, plus whatever humans are willing to help him. Seems like a huge Kramnik advantage there. It's not clear to me from the rules whether he has to adjourn exactly at move 56 or not at all, or if he can pick any spot after move 56. And note that Fritz only gets tablebases up to 5 pieces.

How about "if the human wins one game" or "human wins count double"?
It is ironic isn't it?

A human-machine match can have only one reason:
To check if best machines are stronger than best men or vice-versa.

And of course both need conditions to express their best chess.
(Really not intresting to know if the 100m olympic gold can beat a 50m handicapped car)

It's wrong to decrease in any way the power of a machine.
It need hashtables and tablebases, let hashtables and tablebases.
It need openings, SECRET openings , let the openings.

And if it win the result must be accepted.

But this is very strange: finding conditions to decrease machine power but nobody who cares to find THE RIGHT CONDITIONS FOR A MAN TO EXPRESS HIS BEST CHESS LEVEL.

Surely those are not the classical 2hx40 that is for a man vs man match.
Both in equal condition they express their best level for that condition.
And not their ABSOLUTE BEST LEVEL.

Computer needs informations to give his best.
It never blunders, never needs a rest, never needs to eat.

A man need TIME and REST to avoid BLUNDERS and to give his best.

So a real equal condition is give computer what it needs (openingbook, tablebases...).
But give to the man too WHAT REALLY HE NEEDS.

An hypothesis of an equal match.
1.5h +3 minutes increment for both players. It is 3.5h x 40 moves.
There is something more important: TIME STOPPED when human has the move but is not at the table.
If he is walking, eating, sleeping, chatting,...even something sexy is allowed :-)

To make the game playable in a reasonable time human has to respect a rule: game starts at 13: 40 moves (7h) to be done before 22 (9h elapsed).

Human has then 3.5h for his moves and 5.5 (2 + 3.5 of computer'stime) for his needs.

After move 40 game is adjourned (old times !). It continues the day after again with 1h30'+3'.

I think mr. Kramnik could have really good chanches in a match like this.
And, more important, a really equal match.
Topalov (even without manager) has good chanches too.
Anand no, Anand strenght decreases with more time.

Kudos to Kramnik for getting these concessions. In negotiations, everyone has the right to ask for as much as possible and the opponent has the right to say no. Obviously Fritz/Chessbase feel they need the match more as they were willing to compromise.

The playing field was loaded in favour of the machines, especially the Kasparov-Deep Blue matches. They had 1000s of Kasparov games to work with and look for weaknesses and he had a handful of Deep Blue ones. This is getting something back for the human.

The problem for Kramnik is that with all of these concessions, then he won't have a leg to stand on if he loses.

Marcolantini- I think Anand would have a good chance vs the computer- from his comments on the Tal memorial he was really admiring the solid players and obviously that's become his style as he has matured. Also his preparation is a strength.

Topalov would be a bad choice to play for humankind as, although he has found novelties, they are not necessarily good ones. His style of play is more aiming not to play the objectively best moves, but relying on psychological pressure, unbalancing the position to constantly ask questions of his opponent, in the expectation that his superior tactical skill and fitness will pay off. This would be meat and drink to an emotionless computer which has set the benchmark in these strengths.

Since I find Mig and Polgar speaking in one voice against Kramnik, I have to disagree :) AT the least, he hasn't accused the computers of cheating :)

Except for the part about Kramnik viewing Fritz' analysis, I agree with the rules and they shd be the standard for all man-machine events. It is obvious to constrain the machine to plya on its own merit not on the collective/accumulated wisdom of humans wrt opeings and endgames. The test of the machine is of its intelligence/heuristics, calculating and pruning ability and not much data and moves it can store in the memory and hard disk. And also not take advantage of human fatigue.

I am with Kramnik and predict a 3-3 draw. That will be in everybody's interest as the series can continue.

Kapalik

This seems like a Man Machine match with an asterisk in the records book. Getting to see the computer evaluations? Options to adjourn? I think I'll spend the time doing something else.

These rules are crazy. Whose crazy idea was it to have adjournments? Can't Kramnik use the engine to play the rest of the game before it resumes? For that matter, what's to stop Kramnik simply pre-arranging the games from move 1? I'd like to see what Fischer would say about this!

Why should anyone care about this match?

I predict 3-3.

Likely Chessbase and Kramnik have negociated the final score when they signed the contract simply
because any other result would do damage to the "competitors".
Fritz-10 can't be weaker than its great predecessors and a defeat would make Kramnik look bad as a player and WCC.

3-3 is "win-win" , each side gets advertised by the other in the case of such result, a resul useful for future money making and matches.

Such is life.

I am looking forward to the games I find the whining about the rules unedifying and beside the point. Kramniik has done his best to get rules that give him the best chance of scoring a million. Its going to be a tough and interesting fight. By the way I think the current top programs would destroy any GM at Fischer Random - I wonder over 6 games if they could even get a sngle draw!

>For that matter, what's to stop Kramnik simply pre-arranging the games from move 1? >

there is a fine "random factor" inserted by chessbase programmer in Fritz' choices, "he" is not simply playing out the best evaluated move each time , Kramnik can try this but it is unlikely that he will get right much far with it

however

this idea with "open" opening books reaches the absurd, in effect Kramnk is allowed to choose out the possible for the machine middle games those queenless ones which suit him most

> I'd like to see what Fischer would say about this!>

sadly he will come with something as "They are all fu.... Jews"

Of course it would be nice, somehow, to see Kramnik win. But, I don't think he'll win even a single game. 0-2-4

Anand, Karpov, Leko and other GMs express their opinions about the Kramnik-Fritz match: http://www.chesspro.ru/index_en.html.

The correct URL is without a dot: http://www.chesspro.ru/index_en.html

If Kramnik loses, humans will never see the "World Title" again!

Why? Machines don't need to use the bathroom. Furthermore, Fritz does not accept draws...

"For that matter, what's to stop Kramnik simply pre-arranging the games from move 1?"

Are you serious?

I love watching top level games with chess.fm
I get a glass of wine, tune into chess.fm, load a chess program and enjoy the game. Is there a better way than that to spend a Saturday morning?

Mig, I've seen you and GM Benjamin in the Game Over documentary. It should be fun to have you two do those shows, and I think you'll do a great job. In fact, I cannot think of a beter team to do a broadcast for a man versus machine match.

If they do Sofia rules then it's 5.5-.5 in favor of Fritz. Objectively speaking, machines are already stronger than Kramnik, the question is whether humanity can pull one final swindle.

I'm with the pessimists; the machine will win big. Kramnick should be happy to win one game of the six. Not as bad as Hydra vs. Adams, but in general I think it is over for humans. The next World Chess Champion will be silicon.

I believe in the near future, humans will win selected brilliant games, and will learn how to dupe the computer in certain rare situations, but in a match over many games, the machines are just too too steady and mistake-free.

Apparently a whole series of fairly strong moves, with no tactical oversights, results in better chess than a mix of very strong moves and weaker moves. Humans play more like the latter, with more move-quality variance, and the occasional weaker move is their undoing against the computer.

tjallen

If Kramnik wins, not only does he get $1 million now, but he probably guarantees himself another million from an eventual rematch.

However, I rate Kramnik's chancing of actually winning around 10%, drawing around 25%, and losing around...
Well let's just say this, if he defeats the machine right after beating Topalov... wow.

On an unrelated topic, I just bought a cd of Debussy etudes for piano.

It's awful.

One thing I noticed - none of the younger players, who grew up with computers, seems to want to take them on in a match. It is just an older generation's thing to "battle with the computer for humanity's sake."

Why shouldn't a brash youngster like say Nakamura, play a match with a computer? They don't seem to want to. A well-ingrained respect/fear of the computer? The younger generation seems to see the computer more as a tool than as an opponent.

tjallen

tjallen Nakamura's strength and chess knowledge compared with Kramniks (especially-against a computer) is so small that it is just worthless to arange such a match.

Btw, I also think Chessbase pre-arranges the draws. After the Smirin-Computers match, it seemed that with a proper preparation strong humans can crush computers. I would put the differences between Smirin-Computers and Bahrain matches, to make clear what I mean.
1. The hardware in Bahrain was 8xPIII800, against Smirin it was dual 1Gz or single 2.2Ghz, depending whether engine was "deep" or ordinal one. I cannot say for sure, but I guess 8x0.8Ghz isn't much faster than a single 2.2Ghz.
2. The engines against Smirin were not-worse than Fritz: Junior, Shredder, Hiarcs, Tiger.
3. Kramnik had plenty of time to prepare for the match, he had the exact program version, many seconds etc. He was playing against a SINGLE program, while Smirin had to play against 4. Smirin's match was scheduled 1 month after the same programs beat Boris Gulko. Smirin had 1 month.
4. Smirin won the match with score 5:3, without losing a single game. The score could be more favourable for Smirin: he let some very real opportunities go.
5. Smirin won a fine tactical game against Tiger. I think that is the best game played against a computer by a human.
Let's hope Chessbase understands that another draw will make all this stuff suspicious.

I have to say the last time Kramnik leading 3-1 and then making this unbelievable piece sacrifice against a computer made me a little bit suspicious too about what was the real motivation there.

Anyway, based on their last encounter I'd say Kramnik has good possibility to win. But I would say he is in a deep trouble, because he might not have got enough data of his opponent's style to pick correct openings.

raindeer, didn't he receive the Fritz 10 some time ago? Don't you think that is enought to prepare the openings?

As people has mentioned, the rules are a bit too much on the Kramnik side.

Kramnik has had the program since May, and the final version since October.

Since Kramnik will be able to look at the Fritz book, Kramnik can play many Rybka-Fritz games and see how Rybka spanks Fritz 10 with a 80% score (see computer rating lists).

I believe with these types of rules, several other GMs will be able to draw the match as well.

Also remember that the programmer of Shredder is working for Kramnik for this match.

Suppose Fritz's team paid 25 grandmasters to play short 4-game matches against the machine, producing 50 Fritz vs GM games and 50 GM vs Fritz games, and Kramnick was allowed to study that and only that. No option to take the machine home, examine it, (and make sure Fritz 10 does not have radio communication to Topolov - joke).

Seriously, suppose as part of the commercial venture of a man vs machine match, the machine team was obligated to produce 100 GM level games, to the human opponent, for study. Would that even things up?

tjallen

To judge from the Chessbase link to Der Spiegel, Kramnik doesn't get to see the openings book until the day.

And I thought the machine was programmed in the opening to play moves with particular frequencies; so it will answer, say 1 d4 with 1....d5 half the time and 1...Nf6 half the time? If so then there's little chance to pre-arrange the games, or at least not for Kramnik to do so unilaterally: of course the conspiracy theorists will say what they like.

Is there any thought of doing a podcast of this at all?? Would be quite interesting :)

Great site Mig - thanks for all the hard work.

Titu said:

"Kramnik has had the program since May, and the final version since October."

That also happened the last time Kramnik played Fritz, and yet he could only get a draw... Even knowing that this time, Kramnik can see the opening book during the opening phase of the match, I doubt Kramnik can win more than one game knowing that Fritz' positional analysis is much better...

I was looking forward to this match until I found out, from reading this thread, that Kramnik gets to follow Fritz's analysis in the opening. Now if Kramink wins it won't really count because he is going to get help from his opponent. That absolutely ruins it for me.

> Now if Kramink wins it won't really count because he is going to get help from his opponent.>

In a way it is similar to Kramnik-Topalov after the forfeit. Only Kramnik could have won the match since Topa winning would have been tainted by the forfeit.

Here Kramnik gets to see Fritz' opening book. In effect Kramnik chooses out of the possible openings/middle-games the best ( queenless, closed, and/or simple) for him.

That's huge advantage, it disables a lot of Fritz' potential. How would have been to force Tal to play exchange-Slav or exchange-French ? What would been left for him to play with, aim to ?

Nevertheless Fritz will still have some (3-5, perhaps) moments/moves to assert himself.

That's it always with young players. They have to fight a lot and in rigged situations until they assert themselves and dislodge the older guys who cling to their place under the sun and try every dirty trick to stay there.

Go Fritz !
Such is life but you will prevail in the end.

Ovidiu - I've never played with Fritz, but my impression was that the opening book display doesn't show what Fritz *will* play, only the probabilities of each move. Am I wrong? Or in other words, is it that easy for Kramnik to choose the opening in each match?

rsfb- I may be wrong here.

I gathered from my, granted superficial, readings on this match that the "random factor" has to do with Fritz' evaluation function but it may be that it referred to the opening-book moves.

If so then Kramnik's advantage is that he can prepare in the same sense that you prepare against
another player using his former games as given by databases "most likely he will play a Nf6-gxf6 kind of Caro-Kann, albeit now and then but not
often he plays Boleslavsky-Najdorf with an early b5"..and so on.

It becomes similar (identical) to preparing for usual match were you can gear yourself to meet a given style, a typical but not 100% sure type of opening choices.

"most likely he will play a Nf6-gxf6 kind of Caro-Kann"

That's not easy to play with black if White plays accurate.

freitag that was just an example.

Ovidiu I see what you mean. Regarding the random factor, the only one which I saw mentioned in the match rules is the hash table size (I presume this refers to the transposition table commonly used in chess programming, or something similar). Other than that, I was assuming that during the opening, it was implicit that Fritz will have some random behavior too, but I don't know for sure!

Ovidiu, you're being ridiculous. Chessbase have hired a top GM (their words not mine) to help construct its opening book. I think we can be pretty sure it's not going to have the Exchange Slav in there.

I think Kramnik will probably lose. The last match was drawn - against a weaker version of Fritz on lesser hardware. And there's no reason to think that since that match, Kramnik has learnt to calculate an extra 6 billion positions per second.

But there are several factors in his favour - he has a version of the programme, and the fact some of Fritz's thinking will be 'visible'. To an extent this is an unfair advantage, on the other hand, having an opening book in the first place is an unfair advantage computers have over humans, as is tablebases.

I think the makers of Fritz must be pretty confident though, given the help they've provided to Kramnik . . .

one should read before write, it works better.

I have just read the chessbase.com article which has that :

> The exact speed of the computer and the modification to the openings book are the two unknown factors for Kramnik in this match.>

I guess that this means that he can NOT actually prepare against it. He can not study it to look for style and "flaws" ( for instance search where he could deviate and afterwards the only sensible response, for anyone not only Fritz, would lead to simplifications ).

Seeing the book during the game only avoids situations where Kramnik would think for 1/2 hour since he does not know the theory anymore while Fritz is still in the books.

However if the book is carefully prepared Kramnik may find that from all the theoretical moves given by Fritz he wouldn't choose any because they are all sharp!

There is no obligation from those who made Fritz
opening book to list there ALL good variations,
i.e. including those leading to dull game.

It is getting interesting.

Since I do love MvM matches I am very excited about this match.

Ok still a couple of questions:

1)Kramnik will have the weights given to each variation but not before the game starts. Only during the game correct? So he won't know before hand what road the computer will go down in any given opening he will just know during the game. So although this is clearly an advantage never seen before it may not be enough.

2)Will he see computer evaluations or just the openign book moves and weights? This is an important difference.

3) Will any humans be able to help fritz during the adjournemnts? I don't know if this is possible but if they could take the computer down certain lines where it would save the information in cache it would certainly be helpful. It seems everyone assumes Kramnik can use computers during the adjournament (and that may be correct I don't know) but can fritz be helped by humans?

I find it interesting that everyone's so worked up about the perceived inequalities between human and computer in the match: what's "fair" and "unfair". Isn't the entire interest in these matches precisely that the humans and computers are *not* equal? Hobbling one side or the other in the interets of a "level playing field" seems to miss much of what makes these games fascinating in the first place.

This so-called match is of course just a bad joke: Kramnik can even rest a game after 56 moves or 6 hours, go home and analyse the rest of the game all night with, yes: Deep Fritz 10!

C'mon: Even Topolav could win this match against Deep Fritz. ;-)

I only wonder why so many serious journalists, just like you Mig, cover this event and give it more credibility than it deserves.

This is going to be a great event. I think it is fair to give Kramnik access to opening database and endgame tablebases - well, because computer has access to those. It's not like Fritz worked hard on its opening repertoire or its endgames- it just uses human knowledge. And if it does that, I don't see why it isn't fair to give Kramnik the same things.

As for Kramnik having the option to adjourn the game - well, it is only fair that a human can adjourn the game if he feels tired. He did have access to Fritz 10, but that is not exactly the computer that will be playing against him since the hardware is likely to be different (and superior) in the actual games.

If anything, this match is more fair than previous Man versus MAchine matches.

> Kramnik can even rest a game after 56 moves or 6 hours, go home and analyse the rest of the game all night with, yes: Deep Fritz 10!

And not have any sleep? And remember, if that happens, the resting day between the matches will be lost, since the games are 2 days interval even in the case of adjournment.

The only advantage I see is that Kramnik could work with the computer before. The other things are not advantages, they just equalize advantages the computer has - like books, tablebases and energy.

>The only advantage I see is that Kramnik could work with the computer before. The other things are not advantages, they just equalize advantages the computer has - like books, tablebases and energy.
Posted by: freitag at November 24, 2006 14:1>

Yes, things appear to be well set for a true match. The opening books issue and even the "energy" issue as you point out.
When paying against such a merciless beast you tend to check everything twice, even what seems routine, and that is exhausting.

well, see you on palychess.com to for a "fin-de-siecle" show

There is a live betting line on this match from Pinny:

Deep Fritz -0.5 games -240
Vladimir Kramnik +0.5 games +220

For those of you who think this is a prearranged draw designed for publicity, you have a chance to put your money where your mouth is. Betting Kramnik getting half a game (draw odds essentially) at 2.2-1 odds should be a no brainer for you all. You have a chance at serious money here. A draw does seem like the PR dream for both sides, allowing the interest to continue for a Fritz 12-14 vs Carlsen match in the future.

Just to let you know where the betting public lies, the line for Fritz started about 3 weeks ago at -220 and has been bet all the way down to -240, so it seems all the money is going on Fritz, the heavy favorite (according ot the bookie and the public) at this point.

It seems to me that the whole superiority in chess question of Man vs Machine has already been answered by the conditions of this match. If the machine were given "Deep Blue" rules (going all out for the win, nothing given to the opponent about how it thinks), especially with the largest tablebases available AND the ability to refuse all draw offers, it would crush any human easily. Positions that GMs can't wait to shake each others' hands as a draw should of course be played on by the computer, down to bare kings every game. This tactic alone should produce a good number of wins against faulty humans with a high propensity for slightly inaccurate moves. It's not even close anymore.

>If the machine were given "Deep Blue" rules ... it would crush any human easily....It's not even close anymore.>

You are too discouraged. Lack of faith in one's abilities is the single most important cause of defeat in life. I won't advise you bet on any of them in such a mental condition. Wait until tomorrow to think from a fresh viewpoint.

"Deep Blue" was very strong, maybe stronger (16 millions/sec ?) than "Deep Fritz" now, and yet it did not "crush" Kaspy.
Yes it won, but "crushingly" ? No.

So much of chess has been played out( after 100 years of being played at serious level) and Kramnik is such a bore of a player that even Fritz will often run out of moves to make progress.

I am completely unsure what is going to happen. My first tendency is to think Kramnik will be crushed. But if his preparation is excellent, and his stamina, too, who knows? The man can be quite amazing, I have found over the years.

Still, if I had to make a prediction, it would be 4:2 for Fritz (Kramnik's points: 3 draws with White, 1 with Black).

> "Deep Blue" was very strong, maybe stronger (16 millions/sec ?) than "Deep Fritz" now, and yet it did not "crush" Kaspy.
Yes it won, but "crushingly" ? No.

Strength is not just speed (million positions per/sec). The heuristics, search method and board evaluation function are much more important, and that's what makes Fritz play better than Deep Blue.

The opening book rule seems pretty sound to me - when Kramnik is playing another human, the human is not allowed to look at a book while he plays. Why should the computer, and not Kramnik? Should Fritz have the advantage of scanning 100 years of opening theory, while Kramnik is on his own, having to dig it up from his head? No way. If I'm not mistaken, the creators of Hydra turned the opening book off after move 11 so that the beast would have to play it's own moves from an early stage.

Ovidiu: "You are too discouraged. Lack of faith in one's abilities is the single most important cause of defeat in life. I won't advise you bet on any of them in such a mental condition. Wait until tomorrow to think from a fresh viewpoint."

This kind of sugar-coated tommy-isms is why a fat shyster like Dr. Phil is successful. I bet you wear a lot of tie-die shirts too. Listening to you give me gambling advice is what I imagine a ELO 2400+ chess player goes through when an ELO 1900 player tries to tell him how to play chess. Eye rolling incredulousness is justified, especially when the your ability with words are as broken as your logic.

The betting play on this match (which I actually have in action) was to bet Fritz very early, when the first line came up, knowing the public was going to bet on the computer heavily. Now wait until a day before game 1 (today) and bet Kramnik on an alternative site that puts up even worse lines on heavy favorites since squares always bet the favorites late. You can take advantage of this late money by going the other way. Through arbitrage in the line fluctuations, I stand to make $220 if Kramnik ties/wins OR profit $180 if Fritz wins, risk free. I could have made more in this easy arb opportunity but my alternative betting site places caps on these sort of exotic bets. But its not bad for 10 minutes of "work".

Ovidiu: "Deep Blue" was very strong, maybe stronger (16 millions/sec ?) than "Deep Fritz" now, and yet it did not "crush" Kaspy.
Yes it won, but "crushingly" ? No."

1) Though a six game sample is very small, if Deep Blue were truly a 3.5 - 2.5 favorite over Kasparov, well, a 58.3% winner on an even money bet is "crushing" in my world. And that was way back then. I'm not talking about how Deep Blue would fare today, but how an unrestricted Deep Frtiz 10 would using "Deep Blue" RULES (no holds barred). In the present...

2) Kasparov himself said that Deep Junior which he played a few years later was much stronger than Deep Blue. We are a generation or two later from that point in time too. Note that tablebases, hardware, and most importantly chess software have advanced greatly in time since Deep Blue. I don't think there are many (any?) chess programming experts OR players who think Deep Blue circa 1996 (a friggin DECADE ago, ages in computer technology) is stronger than Deep Fritz 2006.

3) The most important part of my point you seemed to have ignored: "especially with the largest tablebases available AND the ability to refuse all draw offers". Just looking at using the best available tablebases (is it 7 pieces? 8? 9? I don't remember), if the game ever reached this stage, Fritz is * freerolling for a win * if the tablebases says it is at least a draw with perfect play from Kramnik. If somehow Fritz got into a losing position in the tablebase ending, he is a much stronger than average position to recover a draw or even a win through what is essentailly perfect play. And of course it gets all the full points without error if he was a won position in the tablebase.

4) The ability to refuse all draws even in drawish middlegame positions is practically another freeroll situation. Who do you think will make a blunder more often in reduced material positions, Fritz or Kramnik? We just saw Carlsen lose a game that was a theoretical draw, as well as many others. Fritz will never lose on time either. Even if Kramnik is given rest opportunities to minimize pure fatigue as the source of losses, if Fritz did not have to be a "gentleman" about drawish positions and played to bare kings, it will pick up many extra points. It can play on practically risk free.

The whole point is that this is not a true all-out Man vs Machine match, but a handicapped match biased towards the human by a lot. So if you are wondering "which is superior", the handicaps in this match are telling.

11 moves and it coudln't use a book? That seems much easier than Kramniks rules. I wasn't aware of that rule during the hydra match.

There is no easy way to smooth over the asymmetry of man versus machine. A computer denuded of an opening book completely would lose very quickly; a computer consisting almost only of an opening book with evals might play 35 moves in under 1 second perfectly - and then win on time, whilst transposing to a tablebase win. Idiot and parrot are certainly two extremes.

But - a *human* allowed to take every piece of opening and endgame knowledge into a match with Kramnik would undoubtedly have a huge advantage. How weak would a human player with that knowledge have to be, to beat Kramnik, say, or Fischer or Kasparov?

2200?

Probably less. . . if a 2050 rated player beat Kramnik with Fritz's data available . . . what would it prove?

Those who wish to 'blind' Kramnik in this way only wish to prove a club standard player could beat him - if that player had all knowledge of chess, condensed on and available via a computer.

Big deal.

But this - it might not be perfect - but it ought be a good match.

"By the way I think the current top programs would destroy any GM at Fischer Random - I wonder over 6 games if they could even get a single draw!"

I'd like to see it. Put Hydra back up against Michael Adams, who is (or was, anyway) strong at Fischerandom, and bill it as a rematch. I'd pay to see that.

>I'd like to see it. Put Hydra back up against Michael Adams, who is (or was, anyway) strong at Fischerandom, and bill it as a rematch. I'd pay to see that.>

I would too but you won't find Adams willing to play Hydra anytime soon.

Adams would do it if the money was there.

But who would sponsor it?

& don't forget - Hydra has a negative score against human correspondence chess. We do still rule - but probably just for now.

> Eye rolling incredulousness is justified, especially when the your ability with words are as broken as your logic.>

Stern, "the your ability with words are" not very impressive either.

However after reading your new email I have became certain that you are more successful at betting than at chess (some 500 ELO pts. difference) so what can I say instead of a "have courage" or "be optimist" as Dr.Phil..."good luck" perahps ?
I have no special expertise in fortunetelling.

>Adams would do it if the money was there.
>But who would sponsor it?

who did it when it happened ?

Stern wrote: "...the handicaps in this match are telling." Right you are. And they should have removed the handicaps to give Kramnik a chance to score a genuine victory. The way the match is set up, if Kramnik wins, who cares?

Presumably Pal, Ovidiu.

Why?

Schedule is confirmed. Games DO begin at 1500 local, 1400 GMT, 9am EST, 6am PST. GM Joel Benjamin and I will be doing live audio commentary during tomorrow's game one on chess.fm:

http://www.chessclub.com/chessfm/

Free 7-day trial membership:

http://www.chessclub.com/tryicc/register/

"The only advantage I see is that Kramnik could work with the computer before. The other things are not advantages, they just equalize advantages the computer has - like books, tablebases and energy."
Posted by: freitag at November 24, 2006 14:1

"The opening book rule seems pretty sound to me - when Kramnik is playing another human, the human is not allowed to look at a book while he plays. Why should the computer, and not Kramnik? Should Fritz have the advantage of scanning 100 years of opening theory, while Kramnik is on his own, having to dig it up from his head? No way. If I'm not mistaken, the creators of Hydra turned the opening book off after move 11 so that the beast would have to play it's own moves from an early stage."
Posted by: anonymous_coward at November 24, 2006 18:13

It seems such a pointless (and baseless) exercise to argue over what's fair and what's not in a MvM match. You can argue that opening books, tablebases, and the fatigue factor are all unfair advantages for the computer. Or, you could just as easily argue that a perfect memory and stamina are inherent strengths of a computer, not unfair at all. After all, Kramnik has access to all the same opening and endgame theory, it's just logistically impossible for him to study it all and physically impossible to remember all he studies. Hell, if Fritz wins here, next match they'll be limiting the processor speed or number of plies ahead the computer is allowed to calculate, and everyone will be calling it "fair" because, after all, humans can't calculate as quickly.

In my opinion, such conditions designed to "level the playing field" render the match meaningless in terms of answering whether man or machine is superior at chess, and all because it would be too hard for some people to stomach a negative answer. Why is this? Did they stop running the hundred metre dash when the first car able to drive 100m in 9.8 seconds was built? I believe humans can and will enjoy chess for many years to come, just not against computers.

Cheer up, remember that computers, the game of chess, and chess software are all human inventions!

Good Luck Mig and have fun! We except some entertaining 'color commentary' from Joel and all the best chess analysis coming from you! ;-)

All the Best, - Mal (Berkeley, CA)

Franlky if Kramnik plays for a draw on each game (which is probably his strategy) I just can't see how he can lose this match. Kramnik should have access to opening and endgame databases, the computer does so I dont see anything wrong with this.

I predict Deep Fritz 10 will win by a score of +6 -0 =0, yes I am predicting a total shutout. I'm totally convinced that even though Deep Fritz 10 is likely weaker than Fritz 10, it is about 50 elo stronger than Hydra and that this should be enough for a 6-0 score against the human champion.

> Franlky if Kramnik plays for a draw on each game (which is probably his strategy) I just can't see how he can lose this match.

He can lose this match by blundering.

He can lose this match by not foreseeing a tactical attack which begin 10 moves after.

He can lose this match by tiny disadvantages which accumulate during the games.

He can lose this match because he'll be playing against an opponent who supposedly has a higher rating than him.

I predict Deep Fritz 10 will win by a score of +6 -0 =0, yes I am predicting a total shutout. I'm totally convinced that even though Deep Fritz 10 is likely weaker than Rybka, it is about 50 elo stronger than Hydra and that this should be enough for a 6-0 score against the human champion.

A couple of you said that Deep Blue is stronger that Deep Fritz 10. That is completely false, Deep Blue played at around 2700 elo, you can confirm this by checking Deep Blue's moves using today's superior engines. So why did Deep Blue, a 2700 strength machine, beat Kasparov in their second match? Psychology.

For a tour! Ivanchuk is going away with the Capablanca memorial at +3 in four rounds after playing a fine game versus Bareev on the black side of a Nimzo-Indian!

> "Deep Blue" was very strong, maybe stronger (16 millions/sec ?) than "Deep Fritz" now, and yet it did not "crush" Kaspy.
Yes it won, but "crushingly" ? No.

Strength is not just speed (million positions per/sec). The heuristics, search method and board evaluation function are much more important, and that's what makes Fritz play better than Deep Blue.

rsfb- I have heard this lie so many times that by now I should believe it.

But I don't and here why : Fruit's code is known, it is open source, and it sacrifices "heuristics, search method, etc." i.e. sacrifices complexity of the "evaluation function" for the sake of doing deeper analysis.

The result ? 3rd postion on SSDF.

It is true that Fritz and Hiracs do the other way around but they dont get that much better than Fruit.

Hence, I suspect that Deep-Blue with its z-millions more than Fritz was in fact very strong, and it was likely stronger than Deep-Fritz...but Kaspy was strong too during those times.

September 10th, 2006
THE SSDF RATING LIST 107700 games played by 281 computers

Program Rating + - Games Won Oppo
Name ------ --- --- ----- --- ----
1 Rybka 1.2 256MB Athlon 1200 MHz 2924 32 -30 626 77% 2711
2 Hiarcs 10 HypMod 256MB Athlon 1200 MHz 2853 23 -22 1116 75% 2665
3 Fruit 2.2.1 256MB Athlon 1200 MHz 2847 23 -23 997 68% 2712
4 Shredder 10 UCI 256MB Athlon 1200 MHz 2837 28 -27 682 68% 2703
5 Shredder 9.0 UCI 256MB Athlon 1200 MHz 2817 22 -21 1157 67% 2695
6 Fritz 9.0 256MB Athlon 1200 MHz 2811 24

No offense Ovidiu but SSDF is outdated and too slow. CEGT and CCRL are now considered the best rating lists for chess engines. http://www.husvankempen.de/nunn/
http://www.computerchess.org.uk/ccrl/4040/

As I said earlier, just go over Deep Blue's games with today's programs and you'll see that they either find Deep Blue's moves much quicker or they find stronger moves.

Does anyone have the Live game working throuth the oficial site? CAuse I am not :(

I got it to work, just had to try different servers.

Okay – and please stop shouting – she is Marie, a French journalist who writes front-page stories for a major Paris newspaper. And yes, it is serious, quite obviously so. Wedding bells? We'll keep you briefed on the subject. Promise.

http://www.chessbase.com/newsdetail.asp?newsid=3506

Si todos estabamos muy intrigados!!!!!
Buen gusto Vlad!!!

Agreed, she's pretty.

Yeah, she's pretty, but
I presume that's not the
central reason why he's
with her. I mean, as rich
as Vlad is, he could do
even better, looks-wise.

Kramnik is making a mistake.
She will draw him in few years, 50% of his money will go to her after divorce.
Men see looks and women see where the money is.

rsfb,

you are right however Kramnik will try his best to avoid any sharp positions and will try as much as he can to steer each game into quiet, slow positions. You must admit Kramnik is a true master in such positions. See his game today. I do believe top GMs have no chances to win against comps. Paradoxically I also believe the reverse is also true: comps cannot win against top GMs if (and only) top GMs play each game with a draw in mind.

Ovidiu -- so you think she's
like another Heather Mills eh?

Would Kramnik have played 34. g4 against a human? 34. Nc7 practically draws but he wants to make something out of this dry position. Walking his King to gthe Q-side - Fritz may not see it all the way through.

Still, looks like a draw to me ...

easy draw. i wonder what fritz will play with white. e4? if so i bet kram will play petrof.

>Ovidiu -- so you think she's
like another Heather Mills eh?>

yes, from the pictures and body language, from the fact that she insisted to be present here and be advertised, "officialized", and she got her way... typical stuff

anyway, what else are money good for if not for buying pretty stuff ? but she will outcompute Kramnik in the end

Ovidiu -- LOL

but given all of Kramnik's $$$
wouldn't you buy even a prettier
model?

Ugly as hell, come on! He can find tens of prettier on any Moscow street.

I have no skin in the game either way, don't much care who wins this match or if it ends in another draw. But it's worth pointing out what seems like a gross overestimation of the importance of opening books and endgame tablebases, that permeated an earlier segment of the discussion. I'm referring to several comments that focused on:

a) whether giving Kramnik access (to Fritz's opening books and tablebases) was unfair to the machine;

b) whether such having such information would make Kramnik literally "invincible" (asserted by "ed"); and

c) the hypothetical case of a weaker human given access to such information being able to prevail against a stronger human (as posited by "Tom of Streatham Chess Club").

The final point, (c), is what might be called a reductio ad absurdum. "Tom" asserted that a human 2200 player, even a 2050 player, armed with complete opening books and endgame tablebases, would be favored to beat "Kramnik say, or Fischer or Kasparov."

Once you finish laughing, the realization sets in that it is indeed interesting to consider, what is the precise point-value of having opening books and endgame tablebases? Of course the value probably is neither constant nor linear: such knowledge might well be worth substantially more (or substantially less?) to, say, a 2400 player than to an 1800 player.

I hesitate to guess at its value to players so far above my own level, but fwiw, here's a stab. At the very highest level of play, over the course of an extended match, I'll say the data confers an advantage somewhere between 75 and 150 Elo points.

At amateur levels, I'd guess the value is somewhat similar -- maybe a little more, maybe a little less.

At very low levels (say, under 1400), it might be more, because opening mistakes commonly made at that level could be so egregious that their guaranteed punishment would often lead to checkmate or overwhelming material advantages (that even another under-1400 player couldn't fail to exploit), right out of the opening book.

Note, therefore, that I am not denying that these databases have value. I don't think I am even "belittling" their value, at least not relative to some standard that is grounded in reality. All I mean to do is correct the wildly inflated perspective that some other commenters seem to have, regarding the value of the data in head-to-head combat.

Two caveats in particular seem worth emphasizing:

1) Some commenters describe opening books in terms almost the same as endgame tablebases, i.e., "perfect", "all human knowledge," etc.

Of course, this is very far from being true. The knowledge codified in opening databases comes almost entirely from human (mainly GM) practice. As a result, all such books, even those that have been computer-checked, are full of holes -- improvements just waiting to be discovered.

Just as with a match against another human, a GM preparing for a match against a computer will delve deeply into a group of opening lines, doing his best to ferret out some of those as-yet-undiscovered improvements. It is only a small exaggeration to say that people like Kasparov, Fischer and now Topalov re-wrote established opening theory nearly every time they sat down at the board.

2) Endgame tablebases, on the other hand, do by definition contain "perfect knowledge." I believe such knowledge is of significantly greater value to a super-GM than to an amateur like me, for the simple reason that a GM is far more likely to survive into an endgame. The vast majority of my games, even against other amateurs, are decided in the middlegame (some, indeed, in the opening!). I doubt that the ability to play endgames perfectly would enhance my Elo strength by more than 10 or 20 points if I played a match against Kramnik; in other words, the tablebases would be wholly immaterial.

Even for a super-GM, however, making it into any kind of endgame, let alone a tablebase endgame, cannot be taken for granted when the opponent is a tactical monster. One of chess's best-known historical aphorisms is, "Between the opening and the endgame, the gods have placed the middlegame."

That is no less true today than when uttered a century or so ago (by Tarrasch? Tartakower?), and is no less true for super-GMs than for club players. Adams, for instance, was crushed by Hydra in middlegames time after time; the tablebases hardly mattered at all.

True, Adams didn't use "anti-computer" strategy as Kramnik did in Bahrain. But still, I would remind readers that getting the Queens off doesn't in itself bring about an "endgame"; queenless middlegames can remain quite complex and tactical. (Just ask Ilya Smirin -- if you believe as I do that Varshavsky had help from Shredder in their celebrated World Open game.)

>Ugly as hell, come on! He can find tens of prettier on any Moscow street.>

man, you go overboard for few reasons

1st. she is prettier than Danailov to whom Topalov is married

2nd. it is human nature, read your sexology and anthropology books and you will find that this pattern "looks-body against money-status" is the rule in the human ape in all cultures and times

3rd. marrying a russian girl would have been wiser
but too easy. Marrying a french girl when you are russian means that you have high status, you can make, you can afford, unusal-expensive choices.
It makes you stand apart, above from fellow russian
The same with buying a BMW insead of a Toyota to get back and forth from your 20 km away job-place daily ( you could have taken the bus as well, but then it would meant that you were "low" on hierarchy)

>I have no skin in the game either way, don't much care who wins this match or if it ends in another draw.>

The bottom line is that it is boring as usual when Kramnik plays. Just as his twin bro Leko he has spcialized in killing off any fight, any game.

Because of mere statistics and of white-black asymmetry we may see 1 or, at most, 2 worth seeing games.

Ovidiu took an extened vacation to the planet Neptune one month ago, for the duration of the Kramnik-Topalov match. That explains his last comment, in case anyone was wondering.

He's back from Neptune??

I just checked and his IP definitely traces to Neptune.

I think Kramnik has every reason to be satisfied with Game 1. He was never in danger, had everything under control, and even had some chances in the endgame.
Monday will be different, of course, when he has Black. But today's game wasn't exhausting, I think, so we can look forward to an interesting battle.

Jon,

I think that you are upset for people here not taking your long analysis of the match as serious
as you do.

Yes Adams didn't use "anti-computer" strategy against Hydra and neither does Kramnik now or did against Topalov in the Elista match.

Both play(ed) just as they usally do, and they can't otherwise. Look at Leko trying everything and keep falling back to his style.
The style of playing is brain hard-wired by the age of 20.

In the case of Kramnik this means trade off everything against Kaspy, Topa or Fritz and make next to impossible to lose. Actually you may even win as your opponent loses concentration, becomes defocused for the lack of "play", of aims to attack, to focus upon.
It worked against Topa but it won't against Fritz so the result will be dr...yawn...

Nice job with the commentary Mig. You and Joel made a great team. You seem to have kept more abreast of computers than he has. Altough he has some knowledge.

Many times the people they have on chess fm seem to have no understanding about how computers work and evaluate. One guy thought tablebases would kick in with like 12 pieces on the board!

When did they get out of book. Fritz thought for a long time after Qd3 but then seemed to get back into book. I suppose thats possible that it went out and back into book by transposition. Would Kramnik then be advised of this and have the computer screen back on?

Better Neptune than Uranus!

"Better Neptune than Uranus!"

Nice to see people here agreeing. The joys of democracy, the majority which is always right.

After the first game, does anyone else suspect Kramnik will have spent most of his preparation primarily searching for weaknesses in Fritz's endgame play - and then secondarily, searching for openings that lead to such positions that might exploit them?

Entirely possible, Tom, and not a bad idea.

>After the first game, does anyone else suspect Kramnik will have spent most of his preparation primarily searching for weaknesses in Fritz's endgame play>

Difficult to say how a specific "weakness in endgame play" would look like. We all know that one of Carlsen but I won't pin my hopes in exploiting it if I were to play him.

Kramnik likely spent most of the preparation searching for new way to take the Qs off the board. This has been his lifetime speciality, he is going to exhaust this topic in chess by the time he retires.

First, I agree that she's not up to my taste. In my opinion, she's ugly. But my opinion does not matters, and I wish them happyness.

Second, this match is almost as interesting as a WC match between... let's say Kramnik and any 2650 GM. Why? The difference between Fritz 10 and Rybka is of almost 150 points. Same difference between Topalov and, for instance, Oleg Korneev... At least 10 computer programs are playing better chess today than Fritz 10, and Rybka is so superior that it's clear that this match against Fritz is a total nonsense. Why not against Psion on Atari STF 68000 8 Mhz?

R+P vs. R, Ovidiu? If you mean that, I do agree that the possibility of reaching that ending against Magnus is remote.

"New way to take the Qs off the board." Levitation?
I assume the next World Champion will specialise in subtly arranging for Bishops of opposite colours by means yet unheard of.

"Difficult to say how a specific "weakness in endgame play" would look like."

Not really - like 20. ... f5 and 23. ... e5, for instance, or underestimating how much the knight would dominate the bishops after the rooks came off.

Well, now I can see why Madame Nao ended her support to chess !!

Of course, if Kramnik went from Mme Nao to this girl, this is certainly ... a kind of improvement.
But considering what Kramnik has to offer (he has a decent look, is tall, is clever, has some fortune and is know worldwide) I am still convinced that he's misplaying his cards.
Anyway ... on the whole, chess players are monomaniac people, not very talented with women. Just look at Kosteniuk, Skripchenko or at any woman presented as a sex symbol within the chess community !!

Ovidiu, I really don't get the point. You are saying that a man like Kramnik cannot get some woman interested in him? So basically if someone is intelligent and VERY attractive, but, unfortunately, earns good, he is doomed?

"doomed" is too strong of a word, but you are not alone as there is another guy ("stern") posting here who is also into "crushing", "no chance", etc.

The idea is that when you get to 30 you have, usually, secured some position in life and some money. This changes the "game" from what you played at 20 as a student who was living on pizza and paying his rent from the scholarship.
You have to take care and ask yourself on ulterior motives.

Kramnik is no Alehin on board and carries now this pattern off board too.

The reason why strong chess players are so unsuccessfull with women probably comes from the fact that you need to be nuts to play chess at this level and to spend 15 hours per day either playing, blitzing or analyzing.

Even at regular level, one would probably agree upon the fact that you can find a lot of strange or weird people in most chess clubs.

To summarize the whole thing, any woman I've met during my life has told me that chess clubs were the most boring and unsexy places on earth.

And when I became adult, I realized that trying to play chess seriously wasn't socially harmless. That's why I don't go to chess clubs any more, that's why I don't play more than a few games per year, and that's why I'm a basic 2200 patzer, spending a few minutes from time to time on Playchess, Chessbase.com or on a few chess sites.

On the other hand, I am married, my wife is beautiful (way above the new Mrs Kramnik or Mrs Topalov - I mean Danailov - for instance), I own my business, have 20 employees and at the age of 35, my income is located somewhere between top 10 GM and world champion... although I'm certainly far from their intellectual excellence.

If his chess style is any indication, Vlad is probably one of those "deep" persons who think you need more than a gal's internet photo to judge her desirability as a companion.

> To summarize the whole thing, any woman I've met during my life has told me that chess clubs were the most boring and unsexy places on earth.>

yes that is universal, look at the "bored to death" face of the woman caught in the pictures of the recent Topa's blindfold simultan (ar chessbase).

>And when I became adult, I realized that trying to play chess seriously wasn't socially harmless. That's why I don't go to chess clubs any more, that's why I don't play more than a few games per year>

This is a good, maybe great, point but also it is a bit of overkill. I am of your age and own like you my own business (with only 5 employees and nowhere that much money coming but with more time for other pursuits) and I still play 3 (at least, 6-7 with rapids) tournaments per year. Of course I do not advertise myself amongst business partners as a being the kind of person that could take a chess game seriously and I have stopped studying anything since 22.
But still I like to keep in touch with my former colleagues since "a promising junior" years, few of them "profd" GMs and some other good amateurs , so when I go to tournaments we meet, play, talk about old days, and I can enjoy again as when I was junior. Albeit I find myself nowdays less and less motivated (or capable ?) to fight a game and more interested in "atmosphere" and gossip.

I think chess is great in between 12-20 yr old.
Younger and you can not understand, handle, the concepts involved. Older and you can't play it anymore because "after all it is only a game and thus it can't be taken seriously" while chess has to be taken damn seriously if you are to play it good.

Ruslan, before bragging about your "beautiful wife" and "successful business", if they really exist, can you first grow up? Not even an 8-year old kid is such a cheap show-off.

Her name is Marie-Laure Germon, btw, journalist on Le Figaro. I think her name is misspelled at http://www.chessbase.com/newsdetail.asp?newsid=3128

Ruslan, maybe you should look up "full of it" in a dictionary.

Anyone who has to claim their wife is way above someone else's wife is very insecure.

Guys, don't reply to Ruslan, he's clearly a troll who wants to be fed with replies.

Back on chess, it was said after the last Kramnik-Fritz match that Kramnik had discovered exactly such an endgame weakness in Fritz, being over-estimation for White of the Ruy Exchange structure. Didn't Kramnik win a game for Black with some endgame line of the Scotch with ....dxc6?

From looking at the Berlin with Fritz a bit, I'd say this was still true, so it will be interesting to see what happens today if Fritz goes e4. I'd say the Petroff was a very poor anti-computer choice; exactly the sort of position they play superbly. Bit like the Lasker last time.

Guys does Kramnik play lines other than Berlin in Spanish? I tried to recall Kramnik's black repertoire against 1. e4. Is it only Petroff and Berlin with some rare Sicilians that he used with a little success? Does he play Caro-Cann?

he played once the open-ruy lopez against Kasparov but not likely that we will this again..
not likely the Fritz will play 1.e4 either

He also plays the Sveshnikov, of course, if you call that a rare Sicilian. He's not played the Caro a lot, if at all.

The Svesh would actually be a pretty interesting anti-comp choice. They're not usually all that good at oposite-coloured bishops positions like the Svesh so often produces. But it might be a bit too open and unbalanced.

i bet that we will see a Slav with Kramnik trying to remove every possible play and to get an endgame asap, just as in the first game, kind of Topa-Kramnik replayed..

So.

Giannis, since I finished my post with "and I'm far from their intellectual excellence", my point was not to show my great accomplishments, but to say that focusing your weapons on real life was by far more efficient than dedicating all your energy to chess.

As Ovidiu pointed out (and that was the only adult answer to my post), chess is a game. A game is here to help us grow, to give us weapons to become successfull adults. That's why he says that chess is usefull only between 12 and 20.

I've done a lot of different jobs and things in life. That's - at least partly - why now I'm able to cope with all the different problems involved with ruling my business.

What I've done has allowed me to try radically new paths already a few times, and to believe that in most circumstances I would be able to adapt myself to new situations, new people, new languages...

And I don't believe that a kid who has spent all his life in chess clubs and tournaments would have such qualities. Even if Carlsen or Karjakin are gods on the board, off the board they are nothing. Would they have to live in a world without chess tournaments they would instantly become zeros.

Like it or not, mastering the rook endgames gives few employment chances (although from Aronian-Carlsen at Tal memorial it seems that you can perfectly be rated almost 2700 and not even master basic rook endgames).

How could such a world arise? Many possibilities, just look at what did happen to checkers when computers became way stronger than humans. Checkers competitions have almost disappeared... or just try to imagine what extra weapon would give you a fine understanding of the Sveshnikov in war or crisis conditions !!!

A long time ago, I was trying to become a pro violin player (at 16, I started at 5). In France we study philosophy in A degree. Then I had this question to discuss : is art absolutely necessary in our lives? I proudly answered yes, doing my best to defend my ideas. Then a few weeks later I asked the question to my violin teacher, who asked me what I answered, laughed at me a little, and said me : "well my boy, you live in a world with so few problems that you forget what's life. What are human basical needs? Breath, drink, eat and reproduction if the 3 first points are fullfilled. Everything else is to be considered as non essential. If you have any doubt, just look at what people focus on under crisis or war conditions. When you're hungry, do you think that music is so essential?"

Same goes with chess, any art, litterature...

Last question to gg, rsfb, Giannis, who are all able to respond to a real post with real ideas with real insults : take the top 100. How many of those people have children? You personnaly know GMs? Do they have children? How many pro chess players have children? I know your answer. You probably know some GMs with children.

Then my real question : do you think that the percentage of GMs who have children is either normal, or even similar to the percentage of people who have children amongst humanity?

The genes of those who have no or rarely children have way lesser chances to be the genes of the future of humanity. In my opinion it's enough to say that those people have problems, and that the path they've taken is a false one.

For the record. Checkers is a SOLVED game, thanks to computers. You cannot win if you move second. Chess is far from being solved. And, rest assured, it will not be solved in our lifetime. Besides continuing to play chess at the old age (of 35 or more...) will help to slow down effects of dementia.

It depends on what type of "checkers" (not the proper name) you're referring to. There are so many different forms played on different sized boards with different rules. The 10x10 international version is certainly not solved, but there are a lot of draws. There are all types of man-machine draughts matches and the human players do quite well. The beauty of combinations in this board sport is absolutely stunning!

What a mistake in second game.

I forgot write, I am from Poland in Europe!