✗ NCV ︱ Dragonanga

apparently 8th = 5th because the board is technically 14 squares

5th is the correct rank. As stated in the Notes: cross centre to promote.

Just played a game of Open Testing, the 4 player point system was active for some reason which is not supposed to be part of the game. Didn't affect the gameplay at all but it was slightly confusing.

I just finished the game. I have another one pending will see if it's still active. It looks like it is.

The game works okay anyway, but the point system should be off.

Also yeah having the Alfils look like bishops instead of elephants is actually insanely confusing... does kinda ruin gameplay a bit.

Would be nice to update the open tester to the correct ruleset for better quality tests.

I voted for the game. The vote thing is hard to find. 

I prefere the alfil 

Hello and thank you for posting a New Custom Variant! We're glad to see such a thought-out and well-tested submission, those are quite a rarity these days. Unfortunately, our extensive testing has revealed that your variant has some quite intricate yet severe issues I'd like to put greater detail upon:

Firstly, let's analyze the concrete structures: you cannot really avoid moving the e-pawn at the higher-levels, yet to avoid moving the c-pawn you can and very easily. If you have two pieces on the squares highlighted, it's pretty much guarantee no piece but king can attack you, since it's very easy to avoid a gang-up attack on one of those squares, especially if you fight for spatial dominance. Let's memorize this important point and keep going:

https://www.chess.com/variants/custom/game/30211588/0/0

https://www.chess.com/variants/custom/game/30143180/0/0

Here we see why taking the free pawns and pieces is bad: you can't just take the edge pawns without losing a dragon bishop if the other player knows these ideas. The elephants and xiangqi horses are rather hard to use for attack early on, but they do constitute a great defensive resource. Let's keep going and see one more game:

https://www.chess.com/variants/custom/game/30143451/0/0

Here we see what happens if you try to play centrally against a queenside attack: you'll ultimately get an inferior position. Elephants here constitute a too powerful attacking resource if you counter-attack the center, enough to obtain initiative and extra material.

So, what can be concluded from the above?

1. Any attack on a-b-c-d files must involve pushing c-pawn sooner or later to support your material or in the endgame, and due to the fact of how weakening it is it's not an option, so playing this strategy is not truly viable.

2. Any attack on d-e-f files must involve pushing the center, and elephants perfectly control the center and as such your attack will at most give you an equality with no initiative or you'll lose both of the above. As such, playing this strategy is also truly not viable.

3. Trying to target the opposite edges of the board with the dragon bishops loses because you'll get your dragon bishops trapped, again due to the elephants mainly. 

4. Trying to play passively or to lock out a position doesn't work either nor does playing symmetrically and trying to emphasize on the move order: spatial advantage and extra space are way too important in this variant.

So, what are we left with? Ultimately the only playable plan is to push on the e-f-g-h files, with the rest ultimately losing/almost losing if the opponent is very well-versed at tactics and playing for space. Next, understand the implication of this that results in the gameplay issues:

1. No strategy variety here means no imbalance of endgame structures, and endgames are already very restricted due to the perpetual checks.

2. No strategy variety here implies no piece imbalances, as every piece serves a specific purpose in that strategy, roughly: Elephant -> Xiangqi Horse -> King -> Dragon Bishop, and the only imbalances existing only relate to the direct winning tactical ideas.

3. No strategy variety here implies that despite the move variety is great here it's still too restricted, and also implies much worse beginner-friendliness as only one plan works after all.

In addition to all of the above, perpetual checks are not uncommon in the endgames which isn't a good thing by itself, even further limiting the strategy variety. 

So, some things like adding Sideways Pawns and a few dead pieces for more promotion options, making one horse a knight and the ideas the make all the other strategies more viable would really help here, but those changes require extensive testing. Either way, this is a great variant yet it suffers from the issues described above, and that leads to a whole complex of interrelated problems that worsen the gameplay. So, unfortunately, we're forced to:

It is to bad but I can see what they mean.

It appears that you guys played the variant quite poorly. Perhaps you need some better testers.   anyway I'll still be hosting open challenges. If any of your testers want to try to beat me they are welcome but I'm sure they will fail

1. In almost all of your games, your testers made obvious mistakes, so I'm not sure what your point about us is. We could retest with you, but please review our testing games thoroughly first.
2. Please do not quote entire posts.

No actually.

"you cannot really avoid moving the e-pawn at the higher-levels," -- simply untrue. I often never push the e pawn as it highly weakens your royal fairy. It's the d pawn you want to push, though including e push to finish the bathtub formation is also viable.

"If you have two pieces on the squares highlighted, it's pretty much guarantee no piece but king can attack you, since it's very easy to avoid a gang-up attack on one of those squares, especially if you fight for spatial dominance." -- also untrue, and this is why the king is so valuable. Like the queen in chess, it's an essential defender for your position. If the other side puts enough pressure to double attack one of your fairy's sides, the king is likely to be forced off the board for another piece, in which case the remaining king will dominate the endgame.

"Here we see why taking the free pawns and pieces is bad: you can't just take the edge pawns without losing a dragon bishop if the other player knows these ideas." -- this is intentional design. Being greedy with the dragon bishops is designed to be very tricky so that it can get trapped. One of your tester games even falls right into the elephant trap but the player with the trap available completely overlooked it. Additionally, if you'd read my strategies forum post in the Dragonanga club, you'd know that the elephants have 4 main purposes: 2 are obvious (defending c3/6 and f3/6, and attacking the c2/7 weak spot), and the other 2 are less obvious (to sacrifice for pawn advantage, to blow through the centre).

"Here we see what happens if you try to play centrally against a queenside attack: you'll ultimately get an inferior position. Elephants here constitute a too powerful attacking resource if you counter-attack the center, enough to obtain initiative and extra material." -- this game was (bluntly put) a poor Queen side expansion. The structure was sloppy and priority was not spent on space but instead on piece development (fianchettoing the bishops is useless since you are targeting the weakest piece on the board which is a trap in the first place, not to mention the elephant can make two hops to attack you).

Sure I'm salty a bit, and rightly so. You guys spent so long just to test the game poorly and not understand it at all.

not sure what you're referring to.

I think the game should be retested by both the first testers and by some people who know the game well.

Well than why wont bsrti play me???

big sad that CGA team isn't very serious 

I'm not sure what the CGA team expects from members of the community who enjoy contributing with NCVs ideas.
I mean, what is the synergy you hope to have with us? Ask yourself that after knowing that NCVs can be completely discarded both for not being creative enough and for not being "correctly exact" at all levels.

I think you are doing an incredible job and I am sure that no one understands the game like you do on an exact and theoretical level. Just as no one knows better than you the technical, financial or operational implications that our ideas may have.
But if you hope that people, with the, although very plausible, also very limited interface to make a custom, to present NCVs wich are original and fun enough, and wich are also perfectly tested as Capablanca, Ed trice or Yasser Seirawan would do, and doing it just for fun, well, I don't think that will happen very often. Perhaps you would save money and trouble if you only gave the tool to titled players.

I don't think it would be a very good idea for the CGA team to hire people to create NCVs and expect them to be creative and unique, since a person has a limited creative capacity and quite possibly, at some point, they will tend to do things very similar to each other. I think it's a great idea to leave the creativity part of the job to the community, since thousands or tens of thousands of heads are involved.
On the other hand, when it comes to finding forced lines, permutations, finding positional advantages, or anything else related to accuracy and balance of positions at high levels of skill, that's sure to be exactly what you can do better than anyone else.

If you have the chess strength enough to, as an example, find the forced line that ensures the advantage of one side, surely you could also try to find some relevant change to the position to improve it or make it more balanced, such as changing some piece for another or modify the arrangement of pieces, squares or any other element on the position, or perhaps simply make a suggestion.
But if you decline NCVs because of the person who thought of them in the first place was not able to find, calculate and solve all of those flaws, not even those that can be quite easily fixed to those that most of us would never in life be able to, then what is the point of giving this tool to any user?

Or maybe the problem is that you just need to find a polite and friendly way to say “your custom, regardless of how many flaws it has, is of no interest to our team to try to fix it”. If this is the case, I don't think there is much of a problem, it's just a community management issue. But if this is not the case, then it seems to me that the project does not have a very clear direction. If it's so easy to fixed this NCV to make a perfectly acceptable variant, then why not to do it? As I said, if you expect only having to do the testing that you're currently doing to check for general gameplay issues, and after not finding them, then start to work with the technical implications, I don't think this tool is intended for the right audience.

At the end of the day, there is always the question of who is the target audience for chess variants and 4PCV