"But in chess-960, there will be practically no accumulation of experience: there are too many opening positions, and too many differences between them. And thus, the concept of the opening phase will find itself frozen, for a long time, at a childhood level." )

In practice Dvoretsky's criticism of FRC-chess960 is right. Sadly, there is "no accumulation", though accumulation is necessary to achieve the most interesting play. (General intuitive accumulation might be achievable if most players focused most of the playing on chess960, but that will not happen.)

*** IS IT TIME ? ***

-- for Hans-Walter Schmitt and the ChessTigers.de in Mainz to PRE-ANNOUNCE one of the 960 setups eleven months in advance of its annual August chess960 tournament?

ALL chess960 games that pre-announced year would use just that one non-traditional setup.

One such pre-announced year would teach us more about abstract chess than will all future years combined of yet more constantly random setup selection.

In other words, chess960 offers two GOALS....

[1] To free the players from having to study specific opening variations; but that has now been proven to lead to a low uninteresting quality of opening phase play. This was the only FRC goal that Bobby Fischer ever spoke of.

I love it. A great gift from the most important chess player in modern history.

there is a new clock out that has the 960

plus iphone apps

(In the above, entry number #145615)

I did not know the following until today:

"The Chess960 rules have officially been implemented in the FIDE rules this year, and the acceptance of the Chess960 tournaments in Mainz has been a decisive reason for FIDE to accept Chess960 in their handbook."

Thanks <Gene M> for your Dvoretsky post.

That would be a great idea. Hopefully FIDE is paying attention to it!

I'm happy to see this website fostering this amazing chess variant by creating team chess games for it.

<To conclude, I tend to agree with <YouRang>’s view that “<the castling rule> in chess960 is arbitrary and unnatural” and that MK2S rule would at least keep it fair and simple.>

Although I still think the chess960 castling rule is somewhat arbitrary, now that I've participated in a game (Chessgames Challenge: Team White vs Team Black, 2009), I can at least see some justification for it, if I look at it from this perspective:

Here is the position QRBKNRNB and I play this in a tournament on Schemingmind with QRKBBRNN (also on chess.960.cz is #873 so)

Fischer's rule, which merely retains the traditional destination squares for the castling king and rook, is better at promoting asymmetrical positions between White and Black, in the form of opposite wing castling.

In FRC when the kings start in column-c, few players would castle their king to column-e. Instead we would see relatively dull sameness of both White and Black castling to column-a, highly symmetrical and drawish.

If FIDE or ChessTigers.de announced that RNBBKNQR was the only chess960 start setup to use for at least the next decade, we could watch the growth from infancy of a whole new set of openings. (RNBBKNQR is legally reachable after 1. e3 e6)

Opening "theory" took hundreds of years to accrue for the traditional setup. But with today's legions of sophisticated grandmasters, all equipped with chess engines that crush the World Chess Champion, we could observe the growth of virgin opening theory in relative real time. It would be facinating.

Until we have substantial opening theory accrued for a second start setup, we cannot reliably judge our assumptions that are currently based on only the one traditional body of opening theory. For instance, Reuben Fine said it is unwise to move develop your queen too early. But I am confident that such advice is bad in some chess960 start setups. Fine's principles remain untested until a second chess960 setup is study for years.

Even grandmasters cannot play strong opening moves when presented with a random setup they never studied. Truly random FRC is still fun to play, but so could be a second stable setup before deep move trees are available to memorize.

For spectators, a stable second setup replayed and honed by grandmasters over the coming years would be far more interesting than continued random setups.

What we have not yet learned about chess is: What are the factors that make White's advantage relatively larger or smaller. And what does the ultimate answer teach about chess? Likely something we do not yet understand as well as we could.