Variant combinations

In the forums for the Bughouse and Crazyhouse groups, we've discussed the possibility of combining some of the variants, notable Bug/Crazyhouse with Fischer Random.

However, there's no real functional reason why you couldn't combine any two or more variants (with the exception of bughouse/crazyhouse, although even that is not impossible as I'll describe below).  Some of the combinations might be rather insane and might be played infrequently, but if the implementation is sufficiently generic, there should be no special coding needed for most combinations.

I would assume that any time you combine two or more variants, all games are played unrated.  Otherwise, all the different combinations would cause the number of ratings to increase exponentially every time a new variant is added.

Note: to combine bughouse and crazyhouse, when you capture a piece, it goes into a pool from which both you and your partner can draw for drops.  Whoever grabs it first gets it.  Imagine the fun when you need a knight to checkmate your opponent and your partner isn't watching and grabs it first.

The ultimate stupid combination, then, would be bughouse + crazyhouse + KotH + 3-check + Fischer Random.  Would probably finish in no more than 20-25 moves most of the time (the combination of drops + 3-check would be a bit crazy).  Best played at 1|0.  But I can see some clubs running live tournaments of that combination of variants (perhaps without bughouse).

Much more common, though, would probably be combining Fischer Random with one of the other variants.

I would like to add my voice and second most of what @MGleason has mentioned. (Worth the thought where the effort would be minimal, especially if we hold variant + variant unrated.)

My Line-Item Responses...
(Some Agree, some Disagree) ...

Bug and/or Crazyhouse w/ Fischer Random (960) sounds like a ton of fun. I can (easily) see it extending to the likes of both KoH and 3-Check.

For all of theses, I see an active and practical (abstract) training use. (Not without merit for implemented purpose or intent, especially as it pertains to the likes of "problem solving," etc.)

I actually do see a specific type or thought or vein of training-related function, though abstract. I see the combination of any two variants focusing on specific types of tactical thinking as having a reasonable degree of functional application or approach.

On the other hand, the combinatronics (absolute permutations thereof, per human capacity for combined reasoning) quickly diminish if you start to integrate or even attempt to integrate more than one variant's sub-set of (augmented) rules.

This is to say -- it could become confusing, and lose a specific training-related purpose.

Insane, yes.
Infrequently played, yes.

Sufficiently generic depends on the specific rule-set merge or integration.

And, yes, for each of these, no matter how much has been "abstracted" away, some minor degree of specialized coding for specific implementation becomes necessary.

(Period. I know this, factually, as a programmer, especially when it comes to deployment and someone says: "But, hey, what about this-or-that-circumstance-or-situation." Etc... We saw it with the Bughouse and/or Crazyhouse deployment here on Chess.com, etc...)

In other words... NO, what has been done before cannot necessarily or specifically be simply "merged" to create a new abstracted or automatically / arbitrarily integrated set of rules. Especially when we play Live Chess (where we primarily play these games).

Too much time to think (daily) might make current variants pointless. Too little time to think could have the same (inverse ramification) for variant + variant.

Unless or until we play or do this... difficult to say.

BUT, irrespective, some degree of specialized code becomes necessary for integration, especially optimized integration for the likes of live-chess (responsive interface, instead of loading all of the rules and then bogging down a browser with back-and-forth between client-and-server-side rules and verification, etc...)

I would generally (very strongly) agree that "variant + variant" rule-sets should be exempt from specific rating.

Then you also have the problem of finding / seeking players willing and/or able to play under custom rule-sets. (Not commonly played OTB, otherwise, as it were.)

Yeah, but, try programming this, with latency between 4 players. Sounds fun. Consider the mayhem and confusion it would cause trying to do this OTB.

With Bughouse, OTB, you grab and drop in your team member's queue.

With Crazyhouse OTB, you have to actively manage that queue, versus the quick, automatic reactionary aspect of programming something to happen via bits & bytes over the interweebs.

Make sense?

Try programming all of the logic yourself to take into account not only the interaction, but a (popular) high-volume, distributed network-noted website.

In summary: It's not nearly as easy or as simple as it sounds when sometimes you're talking in micro-seconds. Thus, assume a "2-second" increment per move, per game, such that there's "1-second" latency to allow for all of 4 clients across the globe to update properly in "real-time."

Think about it. (For about 500-ms.)

What it boils down to is latency fairness, like some little (RPG) gaming dweebs get all uptight about when their ISP sucks, or chess games are lost due to minor degrees of latency, etc.

Which ultimately leads back to one raw, basic fact: Easy and nice to say that now that we have rules implemented for certain, specific variants, that we should quickly and easily be able to extend these rules to further abstractions ... except that there are still some considerations to be handled.

Don't get me wrong: It sounds fun, but, there are issues in play you have to think about if you were to try and do it or implement it yourself.

That said ... the time it takes ... how better could you prioritize this time? If you implement another variant, you have increased support costs, etc.

Variants should probably be limited to paying client / customers, and variants upon variants further limited to premium paying clients / customers, such that the cost of development and support makes it cost-efficient, which then limits the pool of potential players, etc...

(Too many moving parts to consider, like too many rules to try and integrate makes variant + variant cost-prohibitive, despite the fun it sounds, eh?)

See, now, that's just absurd.

Consider for a moment ...

"List of chess variants" (Wikipedia)

Link: https://en.wikipedia.org/wiki/List_of_chess_variants

Wouldn't combining all of those (teh ultimate) render this observation moot?

Or is this only considering what Chess.com puts forth?

What about the Chess.com poll for the next variant?

Link: https://www.chess.com/surveys

There was one about the next variant, but I can't find it in the list. I remember voting for "blind chess" (Make a move to guess the other player's next move and your next move, since I use pre-moves so often, and Live Chess has a new "pre-move vote" interface to challenge yourself while watching other games; I thought it made sense.)

And that makes a lot of sense where all of the variant rules apply, only you have a different board start position. (Easy to integrate these two rule-sets.)

A long-winded $0.02 paid-in-full.