50 moves rule

Endgame analysis using tablebases, EGTB generation, exchange, sharing, discussions, etc..

50 moves rule

Postby Elmika » Wed Oct 15, 2008 3:19 pm

Why is this forum getting more and more contaminated by this stupid 50-moves-rule??

This rule has absolutely nothing to do with chess. It´s only there to keep human over-the-board-players alive in some cases. It´s just an aid like the chess clock. Without the clock a single game could take a whole week. And the 50-moves-rule has the same use as the clock, but it´s a worse invention. Why 50 moves?? Or why n moves? This makes no sense. Another rule is needed for practiacal play. But at least in theoretical play (for example endgame books) or tablebases research it makes ABSOLUTELY no sense to use a 50-move-rule or a n-move-rule. This destroys the game of chess!
For human over-the-board-play the 50-moves-tablebases doesn´t have any use. It´s not allowed to use them and in preparation or training endgames it´s absurde and useless to try to understand when a special endgame is won in 49 or 51 moves. Remind that a move can be stronger than another even if it takes a few more moves to the win, but leaves the opponent without any chance! For correspondence chess the rule is useless also.
For training, analysis and theoretical research (these 3 are the main fields for the tabelbases I think !!!) to use the rule makes ABSOLUTELY no sense. When I am training the endgame KRP-KBP with blocked rook pawns or KNN-KP, I like using tablebases. And I try to understand the endgame. A mistake of the defender could loose, but I´m not interested how many moves it would take to mate. I want to understand WHY it looses. This is impossible with the 50-rule. For research I like useing the tablebases. I want understand the winning method in cases like KBwBb-KN and I´m not interested whaeater it would be possible in 48 or 57 moves.
Remains only computer/ engine chess. Why should one use a n-move rule and not the real tablebases ?? I really don´t understand this. By the way it destroys the real truth and beauty in chess!
I would prefer to have one single interesting 7-man DTM or DTC than the whole set 7-man-50, even if the filesize would be 1000 times the one of the 50-case !!!!
I´m really disappointed about this deployment. I thik, it´s definitely a step in the wrong direction and wastage of resources!
If you disagree further: Think about the endgame KBN-K with a 14-move-rule. Do you really think this is interesting and of any use ??
Posts: 9
Joined: Sun Jul 13, 2008 10:26 pm

Re: 50 moves rule

Postby syzygy » Wed Oct 15, 2008 4:41 pm

Elmika wrote:Why is this forum getting more and more contaminated by this stupid 50-moves-rule??

Some reasons:
- some people are absolutely convinced that tablebases should take into account the 50-move rule, because in our world the game of chess includes the 50-move rule.
- some people find it interesting to discuss the theoretical side of generating tables that take into account the 50-move rule.

Personally I find your arguments very strong, and some of them I have used myself in other threads. However, in a way I find the extra complications that the 50-move rule brings with it interesting too.

Therefore I would personally prefer to have "DTZ50+"-tablebases that store dtz50 for positions won/lost under the 50-move rule and (not fully accurate, but to keep it simple:) 50+dtz for positions that are won/lost, but drawn under the 50-move rule. Such a table gives correct win/loss information for all positions, both for chess with and for chess without the 50-move rule, it gives optimal play for chess with 50-move rule, and it gives optimal play for chess without 50-move rule.

Remains only computer/ engine chess. Why should one use a n-move rule and not the real tablebases ??

I've had my engine play on a chess server and I've seen it walk into the 50-move rule with DTZ-tables (in suicide chess, btw). Not a big deal at all (in fact it was very funny), but hey why not solve it when I see a nice way to do it.

I´m really disappointed about this deployment. I thik, it´s definitely a step in the wrong direction and wastage of resources!

I don't think you should feel disappointment in people discussing something that you happen not to be interested in. Discussions on this forum about the 50-move rule will not stop other people from writing a 7-men DTZ-generator. And even if it does, this is about people spending some of their spare time on things they find interesting. So I think you should reconsider using words like "contaminated".

Another thing is that full scale 7-men generation (in my view) is still many many many years off. The infrastructure for distributing those isn't there and probably won't arrive before the next internet bubble.
Posts: 162
Joined: Sun Aug 03, 2008 4:02 pm

Re: 50 moves rule

Postby kronsteen » Wed Oct 15, 2008 7:59 pm

50 move rule is also currently ignored by composers, which makes full utility for them to use TBs ignoring 50-move rule, which are DTM, DTC, DTZ (and WDL).

I find also that 50-move rule (if played with) raises very interesting theoretical aspects of endings such as kqpkq, knnkp, etc... in fact, the rule can change completely the attacker's behavior in knnkp, which can sometimes have to stalemate opponent's king in order to force a pawn push and reset 50-move rule clock, instead of constantly blocking the pawn.

As the rule is a practical move and can be discarded by theorists if they wish, it makes no harm to chess theory. TBs without 50-move rule are developed for 6-men, and easiest ones (WDL and DTZ) are a project for 7-men, so every possible effort on TBs without 50-move rule is made, which should make happy those wanting to ignore it.
Posts: 88
Joined: Fri Aug 01, 2008 11:20 am
Location: France

In praise of the 50-move rule ...

Postby guyhaw » Thu Oct 16, 2008 3:34 pm

... which is officially ignored (see the PCCC Codex) in Composition except for Retrograde problems.

The 50-move rule has been an interesting challenge - leading to the creation of the DTR/DTZR metric some years ago, some interesting dialogue here, and some corrections to past thinking.

If it leads to some wonder-coder producing something better than the Yakov/Bourzutschky approach to EGT generation, that will be excellent - but I think it's very much odds against.

My own 'wish' is to see better EGTs created about 4- to 6-man Chess, and to see higher-grade knowledge gleaned from them. Dealing with a notional k-move rule is part of that.

Posts: 489
Joined: Sat Jan 21, 2006 10:43 am
Location: Reading, UK

Re: 50 moves rule

Postby Kirill Kryukov » Thu Oct 16, 2008 4:28 pm

Some of the reasons I am interested in tablebases that consider 50-move rule:

1. Suppose I analyze GM's games with an engine. I want to see the best moves, blunders, and whether the endgame is won or lost in terms of practical play. Suppose Nalimov's tablebase tells me "this is win in 230 moves". In fact there may be no such win available in practical play, because of the 50-move rule. So why do I need to see that "win in 230" if it is not related to any possible outcome of the game in any way? What I want to know is whether the position could be won over the board or not, even if it required God's play. Tablebases are God's play anyway, so this part is OK.

2. 50-move rule is absolutely necessary for running engine-engine matches. Without such rule some engines will continue to play almost forever. (or until all possible positions will be repeated 2 times, and a draw by repetition will be claimed, on move 1500000 :-)). All computer chess tournaments use 50-move rule. Whether 7-men tablebases give any advantage in such events remains to be seen, also whether the tablebases that don't consider 50-move rule are worse or better in tournament play than those that consider the 50-move rule. But at least it does make sense to try to build such tablebases and to try to use them.

3. Chess as we know it today includes the 50-move rule. It is used OTB, in computer chess events, on internet servers. Chess without this rule may be interesting, but we don't play it, our engines don't play it, there are no games played without this rule today. To think about what could happen if there was no 50-move rule is similar like to think what may happens if pawn could move backwards for example. It may be curious, well, there are hundreds of chess variants for those looking for something more than chess. I know that composers prefer this chess variant to actual game of chess. There is also a big population of those who prefer checkers, go or shogi.

By the way, I recognise the importance of the "composers camp" among the potential users of the tablebases. This is why my view is that the next generation tablebase generator should support both types of metrics. Whether both metrics are better stored in same file or different files is not clear at the moment. At least I agree that both kinds of metrics should be ultimately available for community, even if it takes many years to generate the more interesting tablebases.

But I think it is a very narrow-minded position to call the 50-move rule stupid or contaminating. Its existance is a practical necessity in human chess, and even more so in computer chess.
User avatar
Kirill Kryukov
Site Admin
Posts: 7380
Joined: Sun Dec 18, 2005 9:58 am
Location: Mishima, Japan

Return to Endgame Tablebases

Who is online

Users browsing this forum: No registered users and 2 guests