Comments by SamTrenholme

How about this: The rules are identical to FIDE chess, but every piece can only move a maximum of two squares. Pawns can not move two squares; only one. Every piece can leap two squares over pawns and other pieces. Castling is normal--this is the only time the rook can move three squares.

The next question is this: What castling rules should we use?

• Eurasian Chess (preset)
• Great Shatranj [D version] (preset)
• W-Flyer chess (Use this preset)
• Modern chess (preset) (it’s the first one he mentioned)
• 007 Detente Chess - Balanced (Use this preset)
• Knightmate (use this preset) (I’m not sure Muller will play, though, and if he doesn’t, the next suggestion in the queue is Ajax random Chess preset)

Here is the preset, which took me all of five minutes to make. Since Fergus said “program”, I think that means “the preset needs to enforce the rules”:

/play/pbm/play.php?game%3DBalanced+Capablanca+Chess%26settings%3Ddefault

I’m calling this setup “Balanced Capablanca Chess” because computer analysis of some Capablanca positions shows this particular starting array of pieces to be quite balanced; White only has a 3% edge and draws are at 12%. As a point of comparison, in FIDE Chess, white has a 10% edge and draws are at 30%.

I’m also doing something I’ve never done before: I’m signing up for the tournament. It’ll be good to play something besides my own Schoolbook here on the server!

Fergus is intending to make sure those games have rule enforcement.

Yes, and I think it is fair for Fergus to require this. Right now, I’m seeing a lot of discussion but little consensus. Here is one list I made of candidates and another list Fergus made. We already have listed six variants.

If we instead interpret “First-come first-serve” as the first six proposed variants where the rules are enforced, so far the following presets enforce rules:

• Eurasian Chess
• Modern Chess
• Balanced Capablanca Chess (a.k.a “Capablanca Chess, Aberg setup” or “Duniho’s Capablanca Variant”)

In the interests of moving forward, I think, since Fergus is organizing this tournament, he should make the judgment call on what six games the tournament has. I like the enforcing rules requirements; if you want your variant in this tournament, you need to figure out how to program game courier to play the variant.

In terms of making a real Game Coueier preset, Fergus has some excellent documentation:

/play/pbm/devguide.html

I think it’s perfectly fair to demand anyone who wants their variant played here to either make a full rules-enforcing game courier preset or to get someone else to make said preset.

I think it’s very important all submissions use a Game Courier preset that enforces the rules. This raises the bar and makes it so people who really want to see their variant in this tournament need prove it by working at getting it here.

Someone wrote:

George is very right about the minor fudging of pieces and/or places in the initial setup not being actual variants.

That’s a subjective, not an objective judgment. Indeed, the Wikipedia article on Displacement chess calls rearrangements of the pieces in the opening a variant or variation. We have pages for a number of different opening setups using the Capablanca pieces, for example, as well at least one Grand Chess alternate setup. Each different setup has a different opening book and different themes and motifs.

That said, I agree it isn’t very original to simply rearrange the pieces in the opening. That said, the Capablanca setup I’m proposing is one which I decided to use after having my computer analyze various Capa opening setups for about a week.

In terms of what’s a variant, I could just as easily say that “all games that Fairy-Max can play are pretty much the same” or even “all games that Zillions can play are pretty much the same”. As I’ve pointed out before, there really not that many different types of Chess pieces out there (or, at least, Chess pieces with a simple move) and Betza covered pretty much all of the possible pieces in his 1990s research.

Indeed, I see Chess, Capablanca Chess (Janus Chess, Carrera Chess, Schoolbook Chess, etc.), W-rider chess and what not as different versions of the same game.

• Each participant is given a block of six games to play. The time control for the games is the same: 30 days to finish all of your moves, no grace time, no extra time. Adequate warning is given for when the games start so players don’t lose time making their first move.
• In two months, all of the games will be finished. We will then have two more rounds that will be the same.
• Should this site ever be down, we will make sure players are given appropriate extra time on their clocks. Fergus: It would be a very good idea to implement a universal way to have it so, should the site go down, you can report this in Game Courier (just let Game Courier know when the site went up), and all players who had the move when the site went down will be automatically given their time back, as well as having all games lost on time correctly reinstated as incomplete games. I had an unpleasant experience with Game Courier a couple of years ago because my opponent lost on time because the site went down; we were unable to correctly reinstate the game.

One thing we can do is have it so, if one side feels they have a significantly stronger position and the other side is just dragging the game out, make it feasible to adjudicate games so they don’t last for months.

My idea is 18 hours/two weeks for most games and 12 hours/two weeks for HyperModern Shatranj (since this game otherwise would have a somewhat slower pace than the other five proposed games); after two months, either side can request an adjudication.

These shorter time controls will guarantee that games will be decided within two months; allowing adjudication will guarantee we don’t have a game last six months because someone with, say, a bishop and a pawn refuses to resign against his opponent with a rook and two pawns.

I really don’t want to see this tournament last over six months.

Any other suggestions?

Nick: What do you consider a “blitz” time control?

Time control and number of games: There has been a lot of discussion about the the number of games, such as this posting, this posting, this posting, this posting, this posting, this posting, this posting, and this posting.

I want to see this tournament done in six months. My idea: 12 games, 24 hours per move with two weeks spare time for vacations and what not. Adjudication after two months. What do other people think? How should we guarantee that the tournament is finished up in 6 months?

Pawn: pawns move 1 square orthogonally forward, toward the opponent's back rank. They capture, by replacement, 1 square diagonally forward.

Pawn: pawns move 1 square orthogonally forward, toward the opponent's back rank. They capture, by replacement, 1 square diagonally forward. Also can move 2 squares orthogonally forward in its first move if there is no piece directly in front of the pawn