Comments by ChessShogi
Hectochess now works, albeit with a few sprites missing. I just need to upload the sprites and diffusemaps for the Chanpion and Wizard.
Edit: Also, Hectochess doesn't have the 3d Queen, Marshall, and Cardinal for some reason...
I saw you added the 3d Kanjis back to Chu Seireigi. I went through and tested everything, and here are the errors I found
Incorrect 3d Kanji Readings
- Unpromoted Great Leopard (should say 大豹)
- Promoted Knight (should say 天馬)
- Promoted Silver (should say 走狼)
Other
- Promoted Lance, Gold seem to crash/softlock the game
- Promoted Copper General (大豹), Promoted Running Wolf (奔猪) assets doesn't load
- Promoted Kirin (角行), Promoted Phoenix (飛車) indistinguishable from unpromoted counterparts (kanji should be colored red)
P.S.
I was surprised that they didn't have the same movements. I wondered whether a different adjective might be better to distinguish them: e.g. cunning fox and vigilant rabbit, Same for the whale with chu shogi.
That was because I found the Wa Running Rabbit to be too similar to the Lance, and knew that it would add too much defense after testing with the Old Kite. So I gave the Running Rabbit the ranging moves from its Taikyoku counterpart while omitting the stepping moves. The reason it has the same name as the Wa Running Rabbit is because it is one of only two Taikyoku pieces to be a rabbit, and I thought the running kanji fit better. As for the whale, its backward bias was a problem, so I replaced that move with the current one.
You also have to add a line to /play/jocly/dist/browser/jocly-allgames.js to make Jocly notice the files that you placed in /play/jocly/dist/browser/games/chessbase/.
I've done that. It shows up on the Other Jocly games panel, but when I try to load it, it doesn't work. I get the following error:
hectochess-model.js:257 Uncaught (in promise) TypeError: JocGame.LetsTwist is not a function
at ZobristInit (hectochess-model.js:257:20)
at Model.Game.InitGame (hectochess-model.js:354:3)
at JocGame.GameInitGame (jocly.game.js:3:7385)
at JocGame.Init (jocly.game.js:3:1572)
at e (jocly.core.js:1:1233)
at eval (jocly.core.js:1:2511)
I have uploaded François Houdebert's files for the Hectochess implementation to the site. I guess now it's just a matter of waiting for the server cache to clear, unless I need to do something else when uploadnig a new Jocly game to the site.
Much better.
The only improvements I can think of are reintroducing the 3d pieces, naming the game Chu Seireigi instead of Chu Seireigi Shogi (for the same reasons as with normal Seireigi), and maybe having the hand spaces all colored solid brown. But for the latter the wood is colored similarly enough that it doesn't really matter whether you do it or not.
Also, could you update that SVG for the laser cutter? I might want to make a plywood set in the future. But for now, the cutouts I made will do.
The forced promotion issue is fixed for White's forward-only pieces (Pawn, Lance, Ram's-Head Soldier, Running Rabbit, Knight, Flying Swallow) but not for Black's.
I actually found a better replacement for the Old Kite with the Running Rabbit (fBfR).
Thankfully, this is the last time I will need to make any changes.
Here are all the errors I found in your draft:
Piece Movement/Promotion/Drop Errors
Old Kite should be replaced with Running Rabbit, which slides in any forward direction (fBfR)
For Pawn, Lance, Ram's-Head Soldier, Running Rabbit (currently Old Kite), Knight, Flying Swallow, White's promotion is always forced, while Black's promotion is never forced
When captured, White Kirin is not being placed in Black's hand
Board Setup Errors
White King and White Great Elephant are swapped from where they should be in the initial setup.
Other
The grid seems a little off, but you probably already knew that.
Here's a link to the jocly seireigi and hectochess implementations. It would require to code evaluation function to improve the level of the engine which still needs to be done, and will probably require some expertise. Personally, I don't mind the computer being beatable.
The link is broken, it seems... Actually, do include the Chu Seireigi implementation. Testing with it will be valuable.
In general, I try to associate a visual with a movement whatever the game in jocly. I don't know if it's the same for everyone, but I have to adapt a lot when the movements of a known piece change. That's why I used the trident that had been envisaged for the whale in chu shogi. I realized that the movement differed from that of the whale in chu shogi when I was making the documentation, and it seemed preferable to me to avoid confusion.
Okay, makes sense.
It would be interesting to add mnemomic sprites as an alternative. I don't know if you'd be interested in making such a file.
Here's an SVG seireigi set if you want to test the creation of a physical game. Example in plywood (compact / cheap and quick to make but rather simple), the color code for laser cutter is red/cutting, blue engraving recto, yellow engraving verso.
Maybe, once I finalize everything.
I am currently testing the KNAD move for the Lion. Unfortunately, I am no carpenter, but your set looks good (though you did forget the Coppers in the plywood example.)
Since this comment is for a page that has not been published yet, you must be signed in to read it.
Looks like you fixed everything, save for the hand spaces of course, but I won't bother you on that too much until you are ready to test the solution.
I would at least like the ones for Hectochess and Seireigi. Ideally, the files should be placed in such a way so that I know which files go where.
You may include Chu Seireigi if you wish, but I am doing one last update to make it easier to defend against a Lion:
- Renaming Prancing Stag to Running Rabbit (fBfR), which promotes to Prancing Stag (fBsRvW)
- Strong Bear moving the same as Chu Shogi Drunk Elephant
I had your Jocly implementation play out several games, and even though it played rather poorly (for obvious reasons), it was clear that there needed to be more defense.
P. S. I'm surprised you changed the Whale's symbol to a trident. I guess it makes sense if you take into account Greek mythology. I think the whale tail symbol you used before would look more accurate, but it really comes down to personal preference here.
Thanks. Hopefully we can also upload your implementations of Hectochess and Seireigi to the site soon.
I have tested the new moves and will update you on bugs to be fixed. Here are the errors I found as of the writing of this comment.
-------------------------------
Piece Movement/Promotion Errors
-------------------------------
Whale is missing its backward slide
White Golden Bird is missing its sideways orthogonal leaps
For some reason Flying Swallows that are dropped from the hand (not the ones present initially) have their move reversed
-------------------------------
As for the trouble with the hand spaces, maybe ask H. G. Muller about that? He is the one who made the original model if I remember correctly. It shouldn't be that hard for him to figure something out.
As far as I can tell, you should only need a second column of hand spaces on each side, but of course, it's probably not as simple as I am making it out to be.
I saw your Jocly implementation for Chu Seireigi. It works much better than it used to.
I went through and tested everything, and here is a comprehensive list of all known errors that I found:
-------------------------------
Piece Movement/Promotion Errors
-------------------------------
Ram's-Head Soldier and Prancing Stag are not forced to promote when reaching the last rank (these pieces must promote on last rank)
Strong Bear (Starting and Promoted) still has its old move (should move one square diagonally or sideways)
Whale is missing its forward orthogonal step
Flying Swallow is not forced to promote when reaching the last two ranks (this piece must promote on last two ranks)
White Golden Bird is missing its sideways orthogonal leaps, Black Golden Bird is missing its (1,-2) leap from White's perspective
Kirin promotes to Lion (should promote to Bishop) and White Kirin moves has extra (2,2) leap in all directions (should move as in Chu Shogi)
Phoenix promotes to Queen (should promote to Rook)
-------------------------------
Board Setup Errors
-------------------------------
There currently aren't enough spaces in the hand to accommodate all the droppable piece types (there are 19 in total). You will need a second column of hand spaces on each side to account for this.
Black Running Leopard and Black Running Wolf are swapped from where they should be in the initial setup.
As far as I can tell, the only typo I made was not changing Great Stag to Treacherous Fox in the move description.
When I first saw this game, I didn't think much of it, since I was more focused on larger Shogi variants, especially Tenjiku Shogi. However, now I have a few games against Jocly under my belt, and wow, this game completely blew me away. It is an absolute joy to play, despite its size and complexity (which melts away after a couple games). However, it is not flawless.
The Lion-trading rules are a bit complex, and making the Lion contagious (Like Maka Dai Dai Shogi's Deva, Dark Spirit, and their promoted forms) would make the rule much simpler while also achieving the same effect. However, this isn't really much of a problem, and may in fact be the better choice.
The real problem that I have with this game is that modern "innovations" have made the game more complicated than it needs to be. The repetition rules are quite complex, so much so that most computer programs for Chu Shogi that I know don't implement them, which is a trait borrowed from Xiangqi. The King Baring rule is completely unnecessary, as it does not add anything to the game that the combined effects of the other rules do not achieve. There is no evidence that it existed in the Edo period, so I'm not sure why someone thought it would be a good idea to mention this.
However, despite these problems, Chu Shogi is still easily among the best games of its kind. If you like Chess variants, you should give it a try.
Chu Seireigi is an attempt to combine elements of Chu Shogi with the ruleset of modern Shogi. It also has the benefit of not needing any special rules to preserve its quality, fixing all the problems with the modern "innovations" for Chu Shogi that I mentioned above. Players are disincentivized from trading off the Lions in many cases because they would just go into the player's hands, making them even more dangerous. The repetition rule is simply that of Shogi (draw, except perpetual check loses), and the drops make King-baring extremely rare. However, this comes with the unfortunate downside of having to remove the multi-move and orthogonal step from the Lion's move, as otherwise, it would be too strong, even if only the multi-move was removed. To compensate for this, the Lion also moves as a Bishop (in effect making it a Bishop+Squirrel compound).
They are stored in
/home/chessvariants/public_html/play/pbm/sets/
Thanks.
Do you know how to use SCP or SFTP?
Do you still have the passwords I sent you by email?
I have not used either SCP or SFTP before, but I should be able to figure out how to use WinSCP.
I still have the passwords you sent me. I figured out pretty quickly that it was best to save the email.
Edit: I have updated the set files, and now know the basics for uploading via WinSCP.
Since I am an editor now, I should be able to update my set files for the Seireigi games myself (I am in the process of updating the larger games for the final time).
However, I am not sure where these set files are stored, nor am I sure how to upload the new versions to the site. Any help for future uploads would be greatly appreciated.
For now, here is the .zip file with the up-to-date files:
/membergraphics/MSchuseireigi/seireigi-graphics-set-files-gc-final.zip
As you've seen I'm not a very good player.
Well, at least you put up a decent fight.
Generally, the best practice for assigning names to pieces is to make them easy to pronounce (or fun to say, as in bifferbubberz).
So Sloppy Knight would work better.
Resegnation is a fact of life that cannot be prevented by any rules. The best rules can do is put a sanction on doing it, and the worst imaginable sanction is an individual game is that you would lose. Which is exactly the idea, and thus has zero value as a deterrent. You will have to find a better solution for when one of the palyers forfeits (e.g. loses on time because he simply walked away).
I have edited the resignation rule to remove the constraint on resignation.
Resegnation is a fact of life that cannot be prevented by any rules. The best rules can do is put a sanction on doing it, and the worst imaginable sanction is an individual game is that you would lose. Which is exactly the idea, and thus has zero value as a deterrent. You will have to find a better solution for when one of the palyers forfeits (e.g. loses on time because he simply walked away).
I have edited the resignation rule to remove the constraint on resignation.
This sounds like a neat idea, but I struggle to make sense of the board directions even with the diagram showing where the orthogonals and diagonals are.
Resignation is not allowed if one of the opponents (or both of them) doesn’t accept it.
This seems rather arbitrary.
Otherwise, the page is good enough to be approved.
I guess the ASCII diagrams aren't terrible. The interactive diagram helps a lot.
I am mostly referring to the physical game.
My reasoning is that if the two sides are indistinguishable, or there is no other way to tell which checkers have made a complete tour, then a player could easily mistake a checker which made a complete tour for one that had not and bring said checker back into the game.
In that case, the two sided checkers would still be helpful, as they tell you which checkers are out of the game, by having the checkers that have completed the tour have one side up, and all other checkers having the other side up.
25 comments displayed
Permalink to the exact comments currently displayed.
Here are the errors I didn't find before or are new:
Incorrect 3d Kanji Readings (all should be colored red)