Thursday, 23 October 2014

Data driven chess

One of the side effects of the recent "Millionaire Chess" even in Las Vegas, is a greater focus on how the game can be presented. Chess is still hoping to emulate the TV Poker boom of the last decade, and the use of numbers may be the way to do it. GM Maurice Ashley is working on a system called Deepview, which collects a larrge data set from the games of the worlds top players, and then uses that information to provide the viewer with a perspective on the game. Unlike the current system of just using a chess engine to state who is winning, Deepview also looks at the players relative abilities in openings and endings, as well as how they fair in quiet or tactical positions.
While this idea is not necessarily new (I even read articles about chess statistics from the 1950's), the era of 'big data' provides the technology to make it happen. Even if it doesn't succeed in capturing the broader market of sports enthusiasts such a system would still be a great addition to the current online chess coverage.

Wednesday, 22 October 2014

You either know it or you don't

R+BvR is a rare but tricky ending that most players know about, but don't know enough about it. It turns up just often enough that I suspect it should be memorised (especially the defences). An example of this occurred at the ANU Chess Club tonight, when the following position was reached. After much shuffling around (on a 10 second increment mind you), White was unable to find the winning idea and a draw was eventually reached. After the game I thought that the position shown was a winning one, but could not remember the exact technique for converting it. I had a vague recollection that the Black rook needed to be on the third rank, so the bishop could deny it access to a key defensive file.
On checking at home it turned out my memory was right, although I doubt I would have worked it out over the board. For those who are curious the bare bones line to win this position is as follows (note: Alternative variations are left to the reader for further study)
1. Ra7 Rd1 2.Rg7 Rf1 3.Bg3! (The key idea in this ending) 3. ... Rf3 4.Bd6 Re3+ 5.Be5 Rf3 (For now it looks like nothing has changed, but both the c and g files are now denied to the Black rook) 6.Re7+ Kf8 7.Rb7 Kg8 8.Rg7+ Kf8 9.Rg4! Re3 (To prevent Bd6+) 10.Rh4 and the Black rook cannot defend with Rg3.

Tuesday, 21 October 2014

Tashkent GP

The 2014-15 FIDE Grand Prix series has got off to a compressed start, with the Tashkent GP starting almost as soon as the Baku GP event had finished. While there has been a slight change to the lineup (players in the series play 3 of the 4 events) both winners from Baku, Fabiano Caruana and Boris Gelfand have backed up for this tournament.
The whole event began this evening (Canberra time). Of the 6 round 1 games, 5 are still in progress as I write this, with Giri and Gelfand drawing in 39 moves. Curiously the website has Caruana playing under the USA flag, which is either late breaking news, or just a simple mistake.
If you want to stay up late and follow the action the website, with the usual live games plus commentary is here.

Monday, 20 October 2014

Oh the horror!

I had a quick look at the chess-db website, and it seems that they have added some new features. Chess-db is basically a huge online chess database with some bells and whistles thrown in. One such whistle is an 'Explore Blunder' option where you can be shown some of the worst blunders in a players career. I clicked on the link for my own games, and while the list is restricted to 4, it is a pretty horrible 4. In fact a couple of them are so bad that I am struggling to remember where they happened, but I'm pretty sure a few cam from some sub standard Olympiad performances.
The other feature I am going to test is the 'Play Chess against X' where a computer engine is supposed to play in the style of Player X, as determined by the analysis of their games. If it truly emulates the style of my game I expect to be hacked for a bit, and if the attack fails, hang on to win after a series of mistakes.

Sunday, 19 October 2014

Chess and chess-like games

The trend for computer games is for them to be purchased online. While this is convenient to the user (except for those with limited download capacity) it is making the traditional game store obsolete.
The biggest online store is probably Steam, and purchase a lot of games from there (including a number I have never actually played). While they have lots of first person shooters etc, they don'y have a lot of chess programs. Of course dedicated chess programs are a bit of a niche product, but the number one chess program on there is Fritz 14. This is followed by some familiar titles like Battlechess, while the chess variant Chess 2, features prominently.
Below that are what I would either call chess-like or chess-inspired games. Khet 2.0 is one example, while the venerable Archon still seems to be around. After that it is a bit of a mixed bag, where the game may be inspired by chess (eg it is a turned based strategy game) but after that it is anything goes.
As a source for games Steam is usually a little cheaper than the non-downloaded versions so if you are looking to stretch your dollar, it may be a sensible place to shop.

Saturday, 18 October 2014

Is chess a puzzle?

From the 'Lawyers will argue anything if paid enough' file comes a court case from India concerning duty paid on games. In this specific case Funskool India is trying to avoid paying 17% duty on miniature chess sets by having chess classified as a puzzle, rather than a competitive game. A previous determination stated that a product is only a puzzle if the outcome is predetermined. Clearly this does not apply to chess (as yet), although they may try and get away with this argument in regards to draughts, which is at least a draw with best play.
I doubt they will succeed in their appeal, although in a lot of cases of this sort, I am never sure whether it is the facts, or the law as written which is the most important factor.

Friday, 17 October 2014

2014 ICCF Congress - Wrap

Having attended the start of the 2014 ICCF Congress, I made the trip back to Sydney to attend the closing banquet. A quick discussion with various ICCF delegates indicates that International Correspondence Chess is doing OK, and the ICCF is doing quite well as an organisation. The congress itself was both constructive and without rancor, although the fact it was a non election year may have helped.
At the closing dinner there were a few speeches, but what really pleased me were the number of delegates who congratulated the CCLA on how we organised the congress. Thanks for this must mainly go to Brian Jones, and the extended Jones family, for putting in an enormous amount of work.
Apart from the congress there were some chess activities as well. IM Gary Lane won the traditional blitz event with 8/9, while the ICCF team defeated a Western Sydney team 6.5-2.5 in an over the board match.