Charge of the Light Brigade 2: Electric Boogaloo

Spammer alert!

I have deleted the comment, which tried to link to a casino site.

Additional (fourth) fraction:

DARKS

• Scout: Fairy (move) + Guard (capture) = fmFfscW
• Assassin: Bishop (move) + King (capture) = mBcK
• Manticore: Knight (move) + King (capture) = mNcK
• Minotaur: Rook (move) + King (capture) = mRcK
• Hydra = Centaur: Knight + King = NK

The main theme of this faction is close combat (royal attack). I was inspired by the fourth classic hybrid - the crowned knight, without any ranged attacks.

This faction develops the idea embedded in the pawns - movement other than attack. Since 80% of the moves in chess are movement without attack, then for the most part, bishops, knights and rooks are similar to the classic ones. But at the same time, they all specialize in close combat.

The balance of this faction is still being tested, I hope everything will work out well.

An interesting idea. It probably needs further testing, indeed. But besides that the pawns are very strong. I'm not sure if the rest of the army balances it! With spartan chess I wanted to say that the army in question is already balanced against humans/Persians. Trouble with that comes from the fact the some pieces repeat, affecting the personality of the armies!

Scouts are the strongest of the pawns. But these pawns, like the new Guards, do not have a double initial move.

On the contrary, I have concerns that the army is not strong enough, since it has no long-range pieces, and no ability to create linear pressure (on the other hand, all their pieces are major, in terms of checkmating). If this turns out to be true, it will be possible to add an initial double move to the scouts.

Reminds me of Prince Igor’s Order. It seems the inventor himself called it the most boring battalion.

Apparently yes, the game is boring only in close combat. In addition, the army turned out to be quite weak - with a lag of about 4 pawns.

It's a pity that I have not yet managed to select a fourth army for the Centaur, since the classical mechanics (pure diagonal, orthogonal, hippogonal pieces) seem to be exhausted.

DARK ELVES, version 2:

• Scout = Berolina Pawn: diagonal move, vertical capture = fmFfcWifnmF
• Minotaur: Bishop (move) + Rook (capture) = mBcR
• Manticore: Knight (move) + King (capture) = mNcK
• Medusa: Rook (move) + Bishop (capture) = mRcB
• Hydra = Centaur: Knight + King = NK

I think about updating Heroes also:

• Paladin (Human Hero): King (no changes)
• Druid (Elvish Hero): King (move) + Bishop (capture)
• Shaman (Orcish Hero): King (move) + Rook (capture)
• Warlock (Dark Elvish Hero): King (move) + Knight (capture)

All this needs to be tested. Is it possible to add special rules for the royal peace in Fairy Max?

fmFfcWifnmF? You meant fmFfcWifnwA?

Sounds weak, reminds me of Carbuncles.

Wouldn't Druids and Warlocks make the game more drawish?

Those different royal pieces will probably make the regular King more OP in the endgame.

What special rules did you have in mind?

1. Different royal pieces for both armies
2. Standard Kings move, but capture like bishop, rook or knight for elvish, orcish or dark elvish heroes (kings).

Sounds weak, reminds me of Carbuncles.

There is an option to strengthen the pawns - add a horizontal capture, we'll look at the tests.

Wouldn't Druids and Warlocks make the game more drawish?

We'll see in the tests.

Those different royal pieces will probably make the regular King more OP in the endgame.

Yes, it looks like the Paladin (standard king) can checkmate them single-handedly. On the other hand, you still have to live to checkmate; it’s very interesting how the long-range attacks of the druid and shaman will look in the endgame (the shaman will generally be able to cut off other heroes horizontally or vertically). 

That should be possible, and in fact not harder than any different-army setup. If the value of a piece is defined as -1, the piece will be considered royal. It can have any move that other pieces can have.

I tested the dark elf army in Fairy Max a little.

With Berolin pawns, this army is weak, about 20% win rate (at least 2 pawns behind). With heavy Berolin pawns (additional horizontal capture, but no initial double move), the win rate is already about 55%, that is, close to equal.

After that, as an experiment, I added a change to the royal piece - a knight capture instead of a king one. And the win rate dropped to 24% (minus 2 pawns). This is interesting.

Perhaps the same will happen with the elven and orcish armies, I need to check. Is the king's capture (not the move, but the capture) really that much superior to the capture of a knight, bishop or rook?

I am increasingly inclined to think that when analyzing the value of each piece, it is necessary to evaluate the moves of movement separately from the moves of capture. In the case of the king, his royal moves of movement are weak, because they are slow, but the royal attack is apparently quite strong. In the case of the bishop, on the contrary, the moves of movement are very strong (two frontal diagonals), but the attack is weak. The queen solves these problems, having both strong moves of movement of the bishop and a strong attack of the rook. Therefore, the queen is stronger than the sum of the bishop + rook.

Although on the other hand, the royal capture did not look so strong in the first version of the dark elf army, when it was massively in all figures. So it's all quite strange.

Yes.

This is a reply to https://www.chessvariants.com/index/listcomments.php?id=62446.

Thank you! I deleted my question before your answer because I managed to figure it out myself. "royal=K" is key option.

I found out that:

Paladin (classic king) can checkmate alone Druid (bishop capture) and Warlock (knight capture), but cannot checkmate Shaman (rook capture) unless the Shaman is already cornered (as in the case of checkmate with two knights).

Only Paladin can single-handedly checkmate other heroes.

Thus, an orthogonal attack is stronger than a diagonal one, even at a distance of 1 cell, even in the center of the board. But a diagonal move is, on the contrary, stronger than an orthogonal one (for this reason, Ferz is stronger than Wazir). With diagonal moves, the Shaman can easily escape from the Paladin, who is forced to attack only diagonally. But the Druid cannot escape from the Paladin with any moves, who can only attack him orthogonally.

In general, the Paladin (King) can checkmate the (Ferz-move + Ferz-capture), but not the (Ferz-move + Wazir-capture). He can also checkmate both the (Wazir-move + Ferz-capture) and the (Wazir-move + Wazir-capture).

Ferz-move >> Wazir-move (much stronger, apparently 1.5 times - the diagonal step is longer than the orthogonal step)

Wazir-capture > Ferz-capture

I think you cannot draw the conclusion that orthogonal capture is better just from the fact that mKcB can be checkmated by K, and mKcR not. This is for a large part coincidence. If the basic move of the royal had been that of a Kirin (FD) instead of K, replacing its captures by B or R would have had the opposit effect.

An example of a game where Wazir's orthogonal capture is completely useless is checkers :)

Therefore, the context of how the pieces move is important - and within the framework of classical chess (and similar variants), the Wazir attacks stronger than the Ferz. And it's not just that the Wazir has more attacks on the edge of the board, but also the overall higher attack density (i.e., a shorter attack range) looks like an advantage.
The attack range of the Ferz is 1.5 orthocells, while the attack range of the Wazir is 1 orthocell.

Tests of the King's pieces showed that the Knight's capture is much weaker than the King's - for the same reason, the attack range of the Knight is 2.5 orthocells, while the King's is 1-1.5 orthocells (1.25).
However, the Bishop is not much inferior to the King in terms of capture, and the rook may even be slightly superior. That is, the Druid and Shaman look fine compared to the Paladin.

As for the Warlock, there is a noticeable drop - but it is unclear whether this is a feature of the Fairy Max engine, or whether the Knight's capture is objectively so weak.

Paladin:
k:-1 1,34 -1,34 1,7 16,7 15,7 17,7 -1,7 -16,7 -15,7 -17,7

Shaman:
k:-1 1,34 -1,34 1,6 16,6 15,6 17,6 -1,6 -16,6 -15,6 -17,6 16,1 -16,1 -1,1 1,1

Druid:
k:-1 1,34 -1,34 1,6 16,6 15,6 17,6 -1,6 -16,6 -15,6 -17,6 15,1 17,1 -15,1 -17,1

Warlock:
k:-1 1,34 -1,34 1,6 16,6 15,6 17,6 -1,6 -16,6 -15,6 -17,6 14,5 31,5 33,5 18,5 -14,5 -31,5 -33,5 -18,5

Maybe the Warlock should be given an extended attack - AND instead of N (all cells in a radius of exactly 2), and then it will look equivalent to the royal one (K or FW).

I am not convinced. 'Concentration' of attacks is certainly good, but I always thought this would be defined in terms of orthogonally adjacent attacked squares. The attacks of the Wazir are diagonally adjacent, and there is no special reason to think that is useful. With orthogonally adjacent attacks you can attack a Pawn and the square it could move to, so it cannot escape.

I see one advantage of a piece with W capture, though: if it is in time to stop a passer from promoting, it can annihilate the passer by attacking it from the front, thus blocking the square it could escape to rather than attacking that. In test on large boards (like 12x12) I discovered that it was very hard to compensate the R-B imbalance with Pawns. No matter how many extra Pawns the Bishop side got, he kept losing by about the same amount. I ascribed that to the ability of the Rook to annihilate isolated passers, while the Bishop can only prevent their advance (and then remains bound to stopping that passer forever). With so many Pawn on the large board most end-games started with scattered isolated Pawns, and the Rook would always win those end-games against Bishop or Knight. The extra Pawns were just cannon fodder.

As for mKcB and mKcR versus mKcN: the latter could be much weaker because it is slow, and cannot catch up with passers. But for mKcR it is enough that it can get behind the passer on the same file to stop it from promoting. And mKcB can advance its attack on the Pawn file by two squares per move by moving diagonally.

I once tested mKcN and mNcK. The former was about the same value as N and K. But the latter was about half a Pawn stronger! This surprised me, but then I realized it had the best of both worlds: it has the larger speed of the Knight, and the concentrated attacking power of the King, and thus the ability to annihilate Pawns once it catches up with those.

I tested the Warlock mKcAND in Fairy Max and got about the same odds as the standard King (mKcK).

AND has up to 16 cells but 2-3 (average 2.5) orthocells distance. K has up to 8 cells but 1-1.5 (average 1.25) orthocells distance. At least for the royal pieces, attack range is of great importance.

I tested different types of kings for different armies, and saw the following:
- replacing the capture of the king with the capture of the rook gives +1 pawn
- replacing the capture of the king with the capture of the bishop loses -1 pawn
- replacing the capture of the king with the capture of the knight loses -2 pawns

This roughly corresponds to the evaluation of the pieces in the endgame:
- rook 5 pawns
- king 4 pawns
- bishop 3 pawns
- knight 2 pawns

In the endgame, as I understand it, the power of capturing a piece comes first, and not the power of movement, for this reason the Wazir becomes stronger than the Ferz.

I guess it looks something like this:

Rook = 5 pawns (move) / 5 pawns (capture)
Bishop = 4 pawns (move) / 3 pawns (capture)
Knight = 4 pawns (move) / 2 pawns (capture)
King = 2 pawns (move) / 4 pawns (capture)

Movement is key in the opening, and capturing is key in the endgame. In the middlegame, it's about even.

For non-royal pieces, movement plays a big role, as they need to develop and become active.

Royal pieces hide from attacks for most of the game and start to play actively in the endgame, based on this, the capture plays a big role.

Wazir as a standalone piece is weak because it has slow movement. But Wazir as an add-on piece is much more valuable because the slow movement add-on doesn't matter much for pieces with good speed, while the orthogonal close capture fields add-on (which is the closest range, which is good) is very important.

Perhaps the difference between movement and capture as separate components of pieces will help us better understand why the queen is worth more than the sum of its parts, which are usually looked at by the number of squares they can capture, ignoring their range. Perhaps the queen would be close to the sum of the rook and bishop if it only had the rook's movement. And it is the bishop's extra moves that give it superiority over their simple sum.

I will be researching these points to prove them with test results.