Excellent Canadian Open, Spoiled

[Eric Hansen]

Re: Endgame

Jonathan: I have a hypothetical question regarding the value of tablebases in draw claims.

Let's say a master(player a) had the rook and pawn and a much lower rated(player b) had the queen and hardly any time on the clock and wanted to draw. I don't know much about claiming draws in the endgame, but lets say player b claimed a draw and went to the arbiter. And he had memorized tablebase evaluations beforehand and his position is "winnable" . So can he claim a draw to the arbiter based on his argument that it is dead winning for him according to tablebases? Would the arbiter go check tablebases or how would he proceed?

[Zeljko Kitich]

Re: Re : Re: Re : Re: Excellent Canadian Open, Spoiled

Don't play idiot with me, you are too good at this game for your own good.

When you see a bully in action what you do determine the kind of person you are.
1) You can do as if you have seen nothing and keep going. The victim may have done something to deserve it. Then you are a wimp;
2) You can jump in there and try to save the victim's life. This is risky but courageous; this is heroic but somehow foolish behaviour, especially if you have family and loved ones; :)
3) You can yell at the bully then run and call for help. Then you are a responsible citizen. Pretty good and probably well above average;
4) You can join in and go along with the bully. After all the victim is most probably a despicable individual who is getting what he deserves. Then you are a wimp and a sadist.

It is disappointing to find out that so many people on Chesstalk virtually belong to the 4th category. But it is in line with the success that the nazis had leading to WWII. It turns out that lots of ordinary looking people could easily turn into sadist butchers if given the opportunity.

too good at the game for your own good? who else could that describe :):):)

the bully is the one who is attempting inappropriate financial gain, in this case Ed Zator wants more money that belongs to others, so he is attempting to bully and take money from those players by making libelous and slanderous accusations against them of game fixing, not to mention accusations against someone because of the country he comes from

the rest (who have nothing to gain monetarily) I think are following your point 3 above and letting Ed Zator know that he can't bully people to get their money by making unfounded accusations, they are yelling at the bully just as you suggest - if you get confused just follow the money - the one looking to take someone else's money is usually the bully in this case Ed Zator wants at least $100 according to his first post

[Jean Hébert]

Re : Re: Re : Re: Excellent Canadian Open, Spoiled

If you think that this is being beaten up then you should watch what happens in downtown Windsor on a Saturday night on a fairly regular basis.

Don't play idiot with me, you are too good at this game for your own good.

When you see a bully in action what you do determine the kind of person you are.
1) You can do as if you have seen nothing and keep going. The victim may have done something to deserve it. Then you are a wimp;
2) You can jump in there and try to save the victim's life. This is risky but courageous; this is heroic but somehow foolish behaviour, especially if you have family and loved ones; :)
3) You can yell at the bully then run and call for help. Then you are a responsible citizen. Pretty good and probably well above average;
4) You can join in and go along with the bully. After all the victim is most probably a despicable individual who is getting what he deserves. Then you are a wimp and a sadist.

It is disappointing to find out that so many people on Chesstalk virtually belong to the 4th category. But it is in line with the success that the nazis had leading to WWII. It turns out that lots of ordinary looking people could easily turn into sadist butchers if given the opportunity.

[Jonathan Berry]

Re: Endgame

Thnx.
I remember Karpov wrote they (with Kasparov) played blitz (3min) K+Q vs K+R with a chess program (probably including TBs). Not always they could win :D

Six-time US Champ GM Walter Browne failed to win in 50 moves with Q vs R, and then just squeaked through the second time, both at tournament time controls versus a tablebase. It's very much an accomplishment to win with only 3 minutes. Truly great players can do many varied (chess) things well, and they continue to amaze us.

It's my fault that the tablebase stuff got introduced (even though I wasn't the first to do so), but I think it is very relevant. If the position is unwinnable, the complaint immediately fails. The complaint was difficult to understand, but seemed to contain the misapprehension that the further back the pawn is, the less likely that the defending side can draw. Not true. It's very complicated. That much at least, every master would remember from his study of endgames. That's why I introduced the drawn ECE position.

When we finally had a gamescore to work with, matters became absolutely clear, and the tablebases helped. Of course, they only help if one refrains from making unjustified conclusions, and that's what we did: no unjustified conclusions.

--

While Jean and I both dislike bullying, we'll have to remain in disagreement on the redemptive value of an apology.

[Jean Hébert]

Re : Re: Re : Re: Excellent Canadian Open, Spoiled

Defenceless? The original post was offensive and it turned out to be merit-less. At the very least, Ed Zator should apologize to Keith MacKinnon. A reader would have to be brainless to miss the implication made against KM. An apology would put EZ ahead of one of those bullies.

Offensive and meritless ? I suppose this would the first such post on ChessTalk. Just as failing to apologize when appropriate would be.

With all the evidence (repeated ad nauseum spiced with insults) and what has been said in this thread, I don't think that Keith McKinnon's reputation is in jeopardy in any way. It is simply a convenient excuse to bully. His name in this will very soon be forgotten. But at half dozen other people (not you Jonathan, of course you have been a gentleman throughout) should really worry now about their reputation if they had any to defend.

[Steve Douglas]

Re: Endgame

Thnx.
I remember Karpov wrote they (with Kasparov) played blitz (3min) K+Q vs K+R with a chess program (probably including TBs). Not always they could win :D

I seem to recall reading somewhere that when the first K+Q vs. K+R tablebase was calculated it was discovered that the longest forced line for the stronger side was much longer than previously thought. Best play for the weaker side eventually had the rook being moved far away from the king.

Several years ago my son did a science fair project for school dealing with the basics of how computers play chess. One of the experiments he had done was to set up a number of different endgame positions and have different chess engines analyze them, with and without tablebases. I showed him a couple of fortress positions that I figured would screw the computer up.

At the science fair where the parents came in to look at the student projects, one of the things he would do is offer to teach neophyte parents a basic technique of how to defend K+Q vs. K and then the parent could play it against Fritz or something else on the computer (tablebases disabled).

Steve

[Gary Ruben]

Re: Excellent Canadian Open, Spoiled

It is up to the organizers if they are going to allow last rounds between different sections, to make sure that all the results are on the level, and
to investigate any suspicious ones.

They did not do that, and that spoiled it for me.

What rule did you cite as a reason to look at something suspicious in the game? What was the remedy you requested?

Often when a player is in a position where he can't find a way to make progress he offers a draw. That position in this thread would nicely fall into that category.

[Egidijus Zeromskis]

Re: Endgame

A tablebase line is pre-calculated and presumes best play by both sides. For each move the computer simply looks up the current position and chooses the move that has the shortest path to a victory (or the longest path if it is on the short end of the stick). How "easy" or "obvious" the line is depends upon the position. A lot of chess engines won't find the line for K+Q vs. K through calculation although the technique is trivial for a human (the number of moves can take things outside of the computers "depth"). K+N+B vs. K is an example of something that many players find difficult and it's easy to go wrong. A computer using a tablebase won't.

I can't say in this case whether the exact line would be "easy" or "obvious"
for an expert or master to find. At my level (C-class) I'm not qualified to judge except to say that I wouldn't have found it, and I doubt very much EZ would have either. Obviously KM didn't find it but he tried to find it for 38 moves, and obviously his opponent had figured out the proper defensive technique: keep the pawn where it is and use it to support the rook, keeping the opponent's king away.

Steve

Thnx.
I remember Karpov wrote they (with Kasparov) played blitz (3min) K+Q vs K+R with a chess program (probably including TBs). Not always they could win :D

[Steve Douglas]

Re: Endgame

How "easy" or "obvious" is a tablebase line?

A tablebase line is pre-calculated and presumes best play by both sides. For each move the computer simply looks up the current position and chooses the move that has the shortest path to a victory (or the longest path if it is on the short end of the stick). How "easy" or "obvious" the line is depends upon the position. A lot of chess engines won't find the line for K+Q vs. K through calculation although the technique is trivial for a human (the number of moves can take things outside of the computers "depth"). K+N+B vs. K is an example of something that many players find difficult and it's easy to go wrong. A computer using a tablebase won't.

I can't say in this case whether the exact line would be "easy" or "obvious"
for an expert or master to find. At my level (C-class) I'm not qualified to judge except to say that I wouldn't have found it, and I doubt very much EZ would have either. Obviously KM didn't find it but he tried to find it for 38 moves, and obviously his opponent had figured out the proper defensive technique: keep the pawn where it is and use it to support the rook, keeping the opponent's king away.

Steve

[Kerry Liles]

Re: Excellent Canadian Open, Spoiled

While irrelevant, I'm sort of disturbed by all these references using tablebases to confirm how winnable a position is. We are humans, If someone can give the plan/idea to win such positions that would be more helpful and beneficial(and realistic). We never will have the skill of tablebases. For example, try putting a super GM game and compare it with a tablebase and you'll still see the super gm's making "mistakes" a lot.

I have the same opinion of using an engine to analyse as well. Seeing many players falling into the rybka/fritz trap and solely depending on them.

The Tablebase(s) provide an absolute answer for certain endgame positions (provided they have a small enough number of total pieces). It is somewhat satisfying to know a certain position does in fact have a win - even after (say) 59 moves. That is not necessarily something easily shown without resorting to a tablebase lookup.

I agree with computer analysis in general though - if it coughs up a winning line that is fine, but if it does NOT, it might be because it just didn't look quite far enough - the user of the computer cannot know whether there was no win or one wasn't spotted.

[Eric Hansen]

Re: Excellent Canadian Open, Spoiled

While irrelevant, I'm sort of disturbed by all these references using tablebases to confirm how winnable a position is. We are humans, If someone can give the plan/idea to win such positions that would be more helpful and beneficial(and realistic). We never will have the skill of tablebases. For example, try putting a super GM game and compare it with a tablebase and you'll still see the super gm's making "mistakes" a lot.

I have the same opinion of using an engine to analyse as well. Seeing many players falling into the rybka/fritz trap and solely depending on them.

[Jonathan Berry]

Re: Excellent Canadian Open, Spoiled

User CP
Edit Ignore List
type in the multiply-offending name
Click OK (maybe twice)

I'm on a bit of an Ignore List LOLL. Added one yesterday and one this morning for a grand total of four. So easy, yet so satisfying.

[Tyler Longo]

Re: Excellent Canadian Open, Spoiled

I would've turned my back on you too if I was Hal.

[Kevin Me]

Re: Excellent Canadian Open, Spoiled

Actually, I was not making any such inference or accusation.

Now that you've brought it up though, I point out in the wake of Anand's becoming World Champion, and huge earnings from sponsors, chess bribes in India have simply gone through the roof, way out of proportion to actual prize funds.

That is relevant here, since Bharat is visiting from India.

lol. Ed just doesn't give up.

[Egidijus Zeromskis]

Endgame

Actually, that position is lost. After the best move 1.Kb4-c5, Black mates in 45. Tablebase.

But that isn't the point. Black tried to win but didn't find it. Perfectly reasonable.

As you have already analyzed it: was there a shorter win early?

How "easy" or "obvious" is a tablebase line?