I notice that right now when creating a custom game you can only set it up such that a player can only promote a pawn to a particular piece, or to any piece that a pawn could promote to in regular 2 player chess.
How about if when creating a custom game you could check off multiple options for what you want players to be able to promote to? So for instance you could just check off N and B, so that a player could promote to a knight or bishop, but not a queen, or rook.
>> It's worth being tested. I'm pretty sure that it's easy to find 5-10 players ready to give a try. I'm in. 100-150 games should be enough, and it's certainly enough to put 4/1 for test purpose, so that could be done within 1-2 days in fact 
I'm down for some testing if you want to help me test my idea as well 
>> How about if when creating a custom game you could check off multiple options for what you want players to be able to promote to?
+1
>> One of the things I don't like about queens in the endgame is that it can on its own completely paralyze another player if his king is weak. For that reason I think it's the opposite of what you describe, removing queen promotion leads to more active play. Having a weak king is less of a disadvantage lategame if people can't get queens that they check you with forever.
I completely agree with you, the queen in the endgame is a disaster. Such endgames very often arise and turn into a pursuit of double check and a game of nerves (until someone makes a mistake). I also do not like such situations.
But at the beginning and middle of the game, this is the main piece, thanks to which you can make a successful mate attack, but which is just as easy to lose during the attack. The queen is the main catalyst for the attack at the stages of 4 and 3 players with a large number of pieces. If there weren’t a queen at all, then the stage of 4 players would have lasted a very long time (most likely to the endgame). I do not know whether it is good or bad, but this is a completely different game. This is what kind of activity I had in mind.
>> How is this a disadvantage of having rook promotion?
Inadequate compensation after active actions and in case of loss of the queen. If there was only one queen for the whole game, which is at the beginning, then I think there would be a whole new queen hunting strategy (to destroy a unique and dangerous piece) and a passive queen game strategy (which it would be nice to leave for the endgame and keep somewhere safe near the king). That's what I mean by passivity.
>> One of the things I don't like about queens in the endgame is that it can on its own completely paralyze another player if his king is weak. For that reason I think it's the opposite of what you describe, removing queen promotion leads to more active play. Having a weak king is less of a disadvantage lategame if people can't get queens that they check you with forever.
I completely agree with you, the queen in the endgame is a disaster. Such endgames very often arise and turn into a pursuit of double check and a game of nerves (until someone makes a mistake). I also do not like such situations.
But at the beginning and middle of the game, this is the main piece, thanks to which you can make a successful mate attack, but which is just as easy to lose during the attack. The queen is the main catalyst for the attack at the stages of 4 and 3 players with a large number of pieces. If there weren’t a queen at all, then the stage of 4 players would have lasted a very long time (most likely to the endgame). I do not know whether it is good or bad, but this is a completely different game. This is what kind of activity I had in mind.
>> How is this a disadvantage of having rook promotion?
Inadequate compensation after active actions and in case of loss of the queen. If there was only one queen for the whole game, which is at the beginning, then I think there would be a whole new queen hunting strategy (to destroy a unique and dangerous piece) and a passive queen game strategy (which it would be nice to leave for the endgame and keep somewhere safe near the king). That's what I mean by passivity.
There are multiple interesting topics, so I’ll try to separate them.
Another idea is to change the way the queen works. This could be done in several ways, for example restrict its range to X squares in each direction.
Sincerely, ANY idea that would change the BASIC FUNDAMENTAL HISTORICAL way the pieces work (to limit for example queen's force, etc.) doesn't appeal to me. Maybe I'm too much conservative here (not my usual way), but to switch from 2p to 4p chess should be a natural thing. I remember that in 2018 when I joined I found that the absence of en-passant was shocking. (It was implemented later eventually). It's already tricky enough to see a new board with 64 pieces and 160 boxes (against 32 and 64) and the fact that it's not the last standing who wins, but the guy who has more points, and that to eat is not as silly as it is in 2p chess where the only thing that counts is the checkmate. If you have to add that, Eh, guys, here the queen isn't really the thingy that you thought to be comfortable with... Well, from my perspective it's not a good way to proceed.
I personally think that there are only 2 ways to address the issue correctly:
- to remove the queen promoting; OR to make it much harder (even 11th-14th line wouldn't shock me!); OR maybe limit the NUMBER of queens to 1 for any given player: if you had your queen eaten earlier, you may promote to queen, otherwise you have to choose something else, that would completely remove the issue in the Mittelspiel and to dramatically reduce it in the Endspiel;
- to oblige people "pay" for the queen they do with the corresponding number of points that is removed from them: 9 (or even more, 15?). In this case it would often become irrelevant to make a queen if your difference with the next player is narrow and if you are not 100% sure to be able to checkmate afterwards. And also it would introduce an interesting parameter: if you are sure that your queen would be eaten immediately (by a rook for instance), that would often make promoting stupid as you offer 18 points (9+9!) against 5, i.e. reduce the difference by 13 against a questionable advantage of depriving the enemy of one rook (but yes in some cases it would be worth, if for example it was his only rook!).
>> Sincerely, ANY idea that would change the BASIC FUNDAMENTAL HISTORICAL way the pieces work (to limit for example queen's force, etc.) doesn't appeal to me.
In general one can apply one of the two following mindsets: you either do everything to keep the rules similar to those of 2pc, or you try to make the rules that work the best for 4pc. The longer I play 4pc, the more I switch from the first mindset to the second. 2pc and 4pc are not the same thing at all. The main thing that seems to be hindering the development of the game, is people's wish to stick to things they are used to no matter if it makes sense or not. And I'm not saying we have to do this or that, but that it deserves to be tested.
About the case of queens specifically, one could argue that in 2pc the queen is restricted to moving up to 7 squares in each direction, whereas in 4pc it can move up to 13 squares, so it's not the same. The same counts for rooks and bishops obviously.
Well, I agree indeed (even if I'm uncomfortable). One should also have in mind that historically, even if any further change in 2p chess rules is now highly unlikely, during centuries there have been A LOT, and some huge (different pieces that appeared/disappeared, the way they move had changed, the castle was invented, en-passant, etc.).
Moreover, to limit the moves length would indeed solve many issues, but there are still 2 problems:
- exactly like for the pawn promotion, the precise length will also be questionable and challenged indeed (whereas with "the last parallel", nobody can say anything),
- during the game, especially in blitz, a lot of cases will happen, even to experienced players, where you THINK that it's Ok where in fact the length is not enough. And to lose under these circumstances will quite obviously make people very frustrated, especially to old buddies who remember about the old times 
Maybe it could be set up so that you can adjust the move length when creating a custom game. Maybe it could be set up so that the default is to have the move length of the queen is only limited by the board length, but with the ability to make it shorter.
That could be interesting though. Making a queen becomes very theoretical, but still feasible. It would be less radical than to remove promoting completely.