Gothic Chess

@HGMuller, I don't mean to argue, but there's no way a B and R are that close on a 10x8 board. And the Janus needing R + many pawns for equality just doesn't seem right either, based on my experience. By the time your opponent is tangled in so many knots that he can't stop A-somewhere+, Axp+, AxR, A-escapes back rank, Axanother pawn, opponent x Arch, you are well on the way to victory.

@Martin0 there is a YouTube video somewhere where the speaker annotating a 10x8 Gothic game calls attention to the stronger bishops being supported by knights in this variant. He had arrows and circles pointing out stuff. It was done by the YouTube account AncientChess but I just can't find the link to it. If you find it you will really like it I think.

I think the main issue is that on a wider board the Bishop more often hits the enemy camp in 2 places, while on a narrow board one of the forward moves tends to hit the edge. For a Rook the extra sideway moves are not very useful; you have always one forward attack on the enemy, wherever it is on any size board. A Queen has forward diagonal moves too, so I am not sure why this wouldn't benefit similarly as the Bishop. Perhaps the Queen is too valuable to centralize it in the middle game, and must stay in the back so far that the two diagonal moves hit the opponent less often.

The bonus for having a Bishop pair is about the same on 10x8 and 8x8. It is not very large (half a Pawn), which makes it difficult to measure it very accurately anyway. You already need a couple of hundred games to get an error bar of 0.25 Pawn.

This is the video. Great insights.

https://youtu.be/4dpnwrvc-KU

Well, it is not what I see when I let computer programs play out the game. When I start games from the (Gothic or Capablanca) start position with a Rook deleted for one player, and a Bishop plus a Pawn deleted for the other player, they score roughly 50% each.

Note that Larry Kaufman's statistical analysis of human GM games has shown that in orthodox Chess  a B-pair + P already is very close to R + (lone) B:

"When the side down the Exchange has the bishop pair, my data shows he needs only 1.15 pawns to make things even; perhaps this case is what Petrosian and Bronstein had in mind."

As for Janus vs Rook + 2 Pawns: if there is sufficient additional material on the board, computers would score about 80% in favor of the Janus.

We need to arrange matches between your programs and the best humans.

That would indeed be interesting. Note that my own programs (Fairy-Max, Joker80 and Spartacus) are not the strongest available; in the last "Battle of the Goths" tournament they were clobbered by Bihasa. And now there is Fairy-Stockfish, which in general is at least 500 Elo stronger than any competition in most variants it plays.

It doesn't seem to matter much what engine one uses (or what thinking time) for determining material balances, as long as the engine plays against itself. As it of course should be; a Rook is stronger than a Bishop, both between weak and strong players. This suggest that you don't really have to be a strong player to outperform a Bishop with a Rook.

One of the strong players is George Tsavdiris. He actually does quite well against computers, and can beat Joker80. He has developed a special method for doing that, which plays into a blind spot of most engines: he blocks the center with Pawns, and then lures away the opponent's super-pieces to the other side of the board than their King, offering them small material such as Pawns or Knights. And he places his own super-pieces such that they don't attack the opponent's King fortress yet, but can be quickly moved to do so. He then quickly moves all his pieces to the attack, and by the time the computer gets the danger within the horizon, it is too late to do anything about it, as his own super-pieces are cut off from the action. I don't think this works against Bihasa or Fairy-Stockfish, though.

What is the main reason for Bishop + Knight synergy bonus being that immense compared to Rook + Bishop. 

I already proposed an explanation for that above. Orthogonally adjacent move targets are worth more than others. B+N gives 16 such new contacts, R+B or R+N only 8. This also contributes to a Rook being worth more than a Bishop.

Thanks once again. 

What is the biggest synergy bonus you can think of for two pieces merged. 

I imagine that a color-bound Universal Leaper (teleports to every square of the same shade) and a color-alternating Universal Leaper would gain quite a lot when you combine them to a Universal Leaper.

If you once played with Chancellor and Archbishop, you miss them when you play classical chess. Very cool pieces. Would like to play Gothic Chess live again.

It was live on Pychess dot org then the site owner hid it for some reason. You have to go "under" the Capablanca Chess menu and do some sort of FEN string change. And all the Gothic ratings were hidden too. The inventor was rated #1 with 5 or 6 wins over the Stockfish engine at the highest setting. Yet I beat him in Janus

One man who was very enthusiastic with Archbishop was Gabriel Vicente Maura, from Puerto Rico. I read he passed away several years ago. He invented "Modern Chess" over a 9x9 board with a "Minister" moving as BN. He founded an international federation then, wrote books. I found his book, in Spanish, when I was very young, about 1980. I still have it. He developed a (very naive) mathematical theory of chess, in order to explain why his Modern Chess was more logical and natural than orthochess. He also developed the theory that xiangqi was older and superior ... just because it gave a point to a chess with 9 columns instead of 8. Despite these naive flaws, his game was quite interesting and all what he endeavoured in support to his invention deserves some admiration.

You can check here https://www.chessvariants.com/large.dir/modern.html With a lot of pain I realise that the photograph was taken by my best friend Jean-Luc, who died of cancer this year in April. RIP.

Maybe 9x9 board suits better for more major pieces. It's easier to accommodate new heavy pieces as Korchnoi said that two queens can be redundant on a 8x8 chess board in presence of weaker pieces. https://boardgamegeek.com/boardgame/27643/ministers-chess

Isn't controlling the central square in 9x9 huge advantage compared to 8x8 ?

If the bishop is of wrong color then KBN v K is declared a draw ain't it in 9x9. 

wow really I didn't know there were that many 

On a 9x9 board, Bishops never fight. The two on white's side are the opposite of black's side. So there's no point.

Really? No point. No, bad point for you, it's just the opposite. On a 9x9 board, the 4 Bishops do fight. 
I don't defend the 9x9 or anything kind of board.

It's crazy. Most of chessvariant enthusiasts are complaining that many chess lovers are narrow-mind when they want to ban chessvariants. But among chessvariants fans, some want to ban large boards, or some type of variants. This is not a different attitude. In fact some people are living thinking they have invented the new game that will surpass chess and they keep fighting against other inventors that they see as competitors. This is quite naive. I would say sad. Others, and I belong to that category, like chess variants for the fun they found in them, for the variety of situations they offer. I like big boards, 10x10, 12x12, even more. I like small ones too. I like new pieces, I appreciate the analysis some like HG can do from that. I have pleasure with that. What is the problem? I don't force anyone to like what I like.

The 8x8 board is too small for dropping new pieces. The very same person you have openly criticized has published play showing the errors in your thinking. Take Sierawan Chess for example. You can drop an Archbishop and Chancellor onto the 8x8 board. The crazyhouse extension to this variant has a forced trick mate starting in the Dunst Opening which the Gothic inventor showed to all of us in the Discord channel. After 1. Nc3 almost everyone reacts with d5 to chase this knight. But that leads to this incredible loss:

When you have a dedicated analyst that searches for defects and flaws in games, you'd think you'd show some appreciation for their work. Instead, trying to detract from their accomplishments makes you look really really small.