Check out Grant Acedrex, our featured variant for April, 2024.


[ Help | Earliest Comments | Latest Comments ]
[ List All Subjects of Discussion | Create New Subject of Discussion ]
[ List Latest Comments Only For Pages | Games | Rated Pages | Rated Games | Subjects of Discussion ]

Rated Comments for a Single Item

Later Reverse Order Earlier
Switch-Side Chain-Chess. Optionally swap sides with your opponent upon completing a "chain". (8x8, Cells: 64) [All Comments] [Add Comment or Rating]
Emphyrio wrote on Sat, Aug 19, 2017 07:24 PM UTC:Average ★★★

Hi.

I always enjoy new chess variants (especially as, as a composer, I've been occasionally making compositions with some feeric conditions, mostly for SPGs), so as soon as I read something about switch side chain chess on chessbase site, I DLed the android app to test it :). Here are my thoughts so far :

It's definitely original and one can clearly see the interest for AI given, even though chain detection is very easy to code, evaluating the positions is probably much more difficult to code with that possibility to switch sides (although it's obvious that when one side has the possibility to create a chain, it's a tremendous advantage - and white definitely seems to have a big edge with the current rules).

About the feeling and enjoyment of playing this :it reminds a lot of loser's chess (giveaway chess). Despite the fact that loser's chess is at the contrary veryyy AI friendly (don't know if some coders bothered solving it, but it's probably possible). You can lose extremely fast in the opening, but it is somewhat interesting to see how to play around these fast losses.

About the usefulness for chess composition : can be interesting for SPGs (shortest proof games) and retro analysis for sure. Not so much for mates/helped mates/inversed mates etc since chains require the presence of too much material.

About the software : wondered who you asked to code this.. It's passably aesthetic and the UI is ok, but the AI is inexistent.. I ve never seen it ever do a switch, and obviously not work around them either. And even when playing normal chess he is atrocious. That's the main reason for my average evaluation.

About possible changes to prevent very quick games, and give it a feel closer to chess rather than loser's chess : limiting the number of successive switches depending on the move you're at seems unnecessarily complicated. You could for instance consider a chain that involves a pawn still on its starting square as non valid. That would be a big change for sure, but keep it simple and interesting.

 

To finish, my fastest game so far against that weak program, as black :

1.e3 b6

2.Qf3 Bb7

3.Qxb7?! c6

4.Qxa8?? Qc8 (SB)

5.Qxb8 Kd8 (SW) (SB)

6.Ba6 d6 (SW) (SB)

7.Qxc8#

 


George Duke wrote on Fri, Apr 17, 2015 03:37 PM UTC:Excellent ★★★★★
When I judged Excellent one year ago, there were unmentioned reasons why.  One, it's a good original Mutator.  So far I have not found something like new checkmate condition in other CV of three-in-a-line, or other geometric pattern, which would be reminiscent of a Chain, some other designer did.  Anyway Chain is not a win condition as in those few CVs to achieve a line-up of pieces for checkmate. So Chaining is quite original.

Two, Chains themselves are flexible in that they can be legally added to up to a point, and do have clarity.  The combination of Moves to get ongoing Chains lacks the clarity. Here also lack of clarity, stated with irony in the article:

 "If the ticks are even on each side, the players have their starting colors. Otherwise they control the opponent's army at the time."

Having to think whose turn it is that way makes it a fantasy CV, not serious AI experiment or replacement candidate.  Three, the write-up is excellent in detail -- heh like one by Knappen or by Paulowich -- having real move examples, partial scores, constructed positions, to back points up.  

The majority of CVs have somewhat too strong pieces over-all, and Chain as option would put to good use the piece-types in the shadows in subvariants of CVs with Rose, Carreran Centaur (bn) and Champion (rn), or Unicorn (that is hundreds of existing CVs).  http://www.chessvariants.org/large.dir/unicorn.html -- there Switch Side Chain Unicorn Chess would be challenging CV, damping the long-range pieces, because low value pieces including Pawns are equally part of Chain(s). But circular Rose itself should be the favorite for Switch Side Chain because blockable, not like unaesthetic all-spots leaper, it can close and form a chain from all the way around to the backside.

💡📝Azlan Iqbal wrote on Tue, Apr 14, 2015 06:57 AM UTC:Excellent ★★★★★
Unfortunately, you have a "poor" understanding of SSCC. Perhaps you would like to illustrate how White can win against Black from the starting position as you imply. The point about SSCC puzzles is valid (and has been suggested to me before but not in exactly the same way).

Jörg Knappen wrote on Tue, Apr 14, 2015 06:54 AM UTC:Poor ★
This poor goes to the "game" described as a two person game. It isn't. The player who starts with white has the full control of the game and the player who starts with black is a poor bystander bound to be declared the loser by his opponent.

Why? White is in control of creating the first chain. He can deliberately wait until black is also ready to create a chain. Now the following goes on: White creates a chain, switches, black creates a chain, switches, white creates or modifies a chain, switches, and so on, until a checkmate is reached.

You can save the good ideas in this game by reformulating it as a puzzle or solitaire game (The solitaire player solves the puzzle, when he can reach checkmate with an unbroken chain of chains; otherwise he fails). To make the puzzle more interesting; vary the initial position (Fischer Random, Random pawn, both).

Reaching checkmate by a chain of chains may also be a nice fairy chess problem condition.

💡📝Azlan Iqbal wrote on Tue, Apr 14, 2015 06:51 AM UTC:Excellent ★★★★★
Perhaps it should go to the system developer that allows people to rate their own comments. To what end, I'm still not sure.

Jörg Knappen wrote on Tue, Apr 14, 2015 06:47 AM UTC:Poor ★
This poor goes to the author who talks a lot about ethics, but always rates his own creation "excellent".

💡📝Azlan Iqbal wrote on Tue, Apr 14, 2015 02:22 AM UTC:Excellent ★★★★★
Well, that's certainly not the kind of chess I like to play; but to each his own.

>I still don't see why you should be able to switch twice in a row for this. You just wait until your turn comes up. You could then already do this on a single switch. If you are the first to be able to switch, in such a way that he cannot immediately switch back, you just deplete his (future) time on 'thinking' about the move that will cause the switch. Basically the first opportunity to switch will be a win. I don't think you could do much to prevent white making a chain first: 1. d3 d5? {wait 1:59:58} 2. e3, game over...

Assuming the players agree to time switching rules (which is not a requirement of SSCC), after 1. ... d5, you're saying White waits until he has two seconds left and then switches with 2. e3? Well then he has the black pieces (with plenty of time) and needs to make a move. It's still his turn. Let's say he makes a casual move with the Black pieces that does not switch, how can he be sure White won't switch back quickly with 3. f3! and return the favor?

Regardless, certain lines of play can be avoided if your opponent has the habit of trying to win on time alone and has no real interest in the actual game. Or different time controls, such as the "unfriendly" fixed time per move or, as you suggested, non-switching time controls can be implemented. Again, even with chess, time control rules are highly variable. Depending on the situation and players, some time controls are better than others.

💡📝Azlan Iqbal wrote on Sun, Apr 12, 2015 10:36 PM UTC:Excellent ★★★★★
First of all, let me just say that in light of your most recent post, I find it ironic that in your first post you should consider a fixed-time-per-move time control "unfriendly".

When you say you and your friends played "to the limit of the rules", did this include kicking the table slightly with your foot to break your opponent's concentration or something to that effect? Because I do remember playing against such people and it told me they had no real skill at the game and had to rely on other methods to give themselves an advantage. Perhaps you and your friends also all agreed to draw in advance of tournaments or final rounds. I have also encountered things like this in my school days.

Once again, time controls are an optional part of SSCC and there can be many different implementations of it. It cannot be compared to switching one move before checkmate (which is legal under any circumstances). Even under time switching rules, the players can play to avoid situations where two consecutive switches are possible which might allow one player to switch and then wait nearly two hours before switching at the last second so his opponent loses on time. Or they could risk it in the hope the opponent might not see those chains (quite possible as well) to gain some other kind of advantage at the regular chess component of the game. I would say this can indeed make the game more interesting... even against someone like you.

💡📝Azlan Iqbal wrote on Fri, Apr 10, 2015 11:08 PM UTC:Excellent ★★★★★
I disagree. Not everything in a game can be regulated and ethics is important in every sport. Besides, the vast majority of SSCC games (and regular chess) are not played for money or world championship titles. They are played for fun between two people. Time controls are not even a requirement. That does not mean you should wait until your opponent becomes too tired to continue playing and resigns just to go to sleep; therefore you win. You are expected to make a move within a reasonable amount of time. 

Even in a tournament setting, it is unethical to agree beforehand to a draw and then just play out a few moves before shaking hands and claiming the game is drawn. Many players can do this repeatedly to share prize money but they do not. If your philosophy is to "make use of all conceivable methods allowed by the rules to win", I suspect you will not have that many people willing to play with you. Regardless, and once again, in SSCC time controls are optional and even there the appropriate settings can be agreed upon before the game.

George Duke wrote on Tue, Feb 4, 2014 05:44 PM UTC:Excellent ★★★★★
(II) So this is basically subvariant of old Rotation Chess, http://www.chessvariants.org/mvopponent.dir/rotation.html and http://www.pathguy.com/chess/Rotation.htm -- that pre-dates Neto's 14-year-old Mutators. There are thousands other ways "to elevate the complexity without affecting the theoretical size of the game tree," http://en.chessbase.com/post/a-new-challenging-chess-variant, and all deserve batch or individual consideration.

In SSCC should serial Switching be allowed? The example shows spectacularly Player A moving 14 times in a row to ultimate checkmate.  The '5...d4!'
really highlights onset of one of many possible quasi-Fool's Mates inherent in present SSCC, not a great move by and of itself.  In other words, wresting control for 14 plays in a row so early is either a blunder by Player B or cry for designer to rein things in.  How about enhancement in further subvariant: Switching up to twice or thrice and four times dis-allowed?

Appealing about the Rotation CVs is the ability to catch up. By comparison, other CVs that have that quality somehow are Rococo and Chess With Promoters and many 3d CVs.  SSCC certainly dispenses with opening theory by its extreme single Rules addition.  Also to credit of Rotational Chesses is tilt to strategic planning from tactical planning, favouring biologic intelligence over computer, as M.A.M. Iqbal describes. Since two or more empty spaces are desirable, keeping the same familiar small board 8x8 may be main cause of the runaway takeovers by one player or the other for 10 or 12 moves to Checkmate.  Improvement to SSCC might come by enlarging board to GM Reshevsky-endorsed Zonal form, http://www.chessvariants.org/shape.dir/zonal/zonal.html, or to Morley corridor form, http://www.chessvariants.org/shape.dir/morley.html.

George Duke wrote on Mon, Feb 3, 2014 10:12 PM UTC:Excellent ★★★★★
(I)  Switching sides is not entirely unused.  For example, fourteen years ago  to the day mathematician Neto has two back-to-back suggestions under More Mutators section: 

•Optional-Side(N) : After the Nth turn, the 2nd player can choose to switch sides rather than respond.

•Switch(N) : At each set of N turns, players switch sides.

The year 2000 article 'Mutators' was groundbreaking as much for that term itself: http://www.chessvariants.org/newideas.dir/mutators.html.  Other cases of switching sides in fully-defined CVs are less extreme, more often many pieces changing sides at once, not the whole team.  It is convenient  Neto puts the two above up as alternative Mutators widely applicable. That is because many (hundreds) other candidate conditions are surely possible to trigger the right to switch sides than this SSCC chooses to invoke, namely defined Chain around two or more empty squares.  The surrounded two+ adjacent-spaces criterion is not necessarily more compelling than some others. Each different signal to allow optional Switch, with or without Chain, would constitute separate CV.  Which among several alternatives -- to be listed in follow-up comment (one example: three same-side pieces in a line) --
would be more aesthetic or present better heuristic?

Now the chain of minimum 6 of SSCC may seem rather cumbersome, yet may be acceptable since the Switch should not be too easy to achieve.

11 comments displayed

Later Reverse Order Earlier

Permalink to the exact comments currently displayed.