It's significantly harder to track because not only are you playing 4 more things (double regular TR and 33% more than KC), you also have to remember whether each action card was played once or twice or not at all. That's going to get super messy especially as you nest. The tree expands much more quickly than with even KC with more to remember at each branch.
I still fail to see how this is more messy than keeping track of all the extra actions and buys you get from throned cards (you often need a countring help). You can just use a token to signify whether an Action card has already been played which is significantly easier than counting actions/buys.
Furthermore I think that Double Throne Room is boring as it is just two Throne Rooms in one card and too weak for 6. The mixed order makes Imperial Household a decent 6$ card and above all it makes it interesting to play with.
I don't know how much more clearly I can explain it. This has
all the tracking you need to do with TR/KC and
more, since you have more card plays. TR plays 2. KC plays 3. This plays 4. The higher branching factor means it already has more to track. Note that the fixed order variant has this problem too, but it avoids the next problem (and also doesn't require you to set aside a second card that may not actually be played until dozens of choices and card plays later).
The second problem: each single play of this card requires you to remember data for TWO revealed cards. With TR (KC), there are 3 (4) possibilities when you continue resolution:
- no card has been chosen and played yet
- a chosen card has been played once
- the chosen card has been played twice (done for TR)
- the chosen card has been played thrice (not for TR, done for KC)
But the OP's Imperial Household, you have
- no cards have been revealed yet
- card A played once
- card B played once
- card A played twice
- card B played twice
- card A and B played (AB or BA, though the order doesn't really matter)
- card A played twice, B once (AAB, ABA or BAA, but again, order doesn't matter)
- card B played twice, A once (ABB, BAB or BBA...)
- card A and B played twice each (6 orders that don't really matter, done)
That's 8 scenarios. There is
significantly more to keep track. A single play of TR/KC/IH is not too hard to track, and the increase from one to each isn't much. But the branching factor means that KC chains are a lot tougher than TR chains, and it also means that IH chains would be a lot tougher than KC.
These are just plain numbers now. 4 is bigger than 3 is bigger than 2, and 8 is bigger than 4. If you still don't understand, I don't think I can help you here.
(
Edit: I thought of a point that may help understand the chains. Remember, the effect of nesting these cards has quadratic growth. TR-TR is 4, KC-KC is 9, IH-IH is 16. It just explodes as you continue.)
I think the easiest version to do, since it would not take as much tracking but still allow for the original intended effect, would be very similar to the original text posted:
Reveal two Action cards in your hand. Play them each twice, in any order.
Granted, this does allow for some unintended repercussions, such as: Remodel as "a" and smithy as "b" in order abba. Remodel would trash smithy for gold, then both smithies get you a total of 6 cards, then remodel something you just drew. Things like that would warrant a cost of at least 7 though, and you could limit the branching a bit with a clause like: "the revealed actions cannot be Imperial Household"
Just my thoughts on the matter
The only difference here is that you said "reveal", whereas Dingan's said "choose". That doesn't really help, since we're all already assuming that the chosen cards are revealed. Moreover, the revealed action cards would have to be set aside somewhere anyway, so you should not be able to Remodel the chosen Smithy. Otherwise, the tracking gets even more ridiculous. Keeping them in hand would also cause accountability problems.
The tracking difficulties don't really have anything to do with that part of the card though.
Limiting the branching by not allowing nested IHs helps a little bit, but all you need to do is throw in TR or KC or Golem and it becomes crazy again.
faust's Reserve-ish version still seems like the most elegant solution to me, though I just realized that it actually is a lot stronger than the original card. You can actually mix up the order of chosen cards across
multiple plays of it since they all dump to the Tavern, so it's still far from perfect.
Imperial Household
+4 action
reveal two action card from your hand, gain a copy for each of them, and put one of your return tokens on each piles you gained from.
Return tokens (colored for each player):
When your play a card from a pile with your return token(s), return it to the supply and remove one return token from the pile.
Interesting, but lots of unintended consequences. This is stronger in that you can gain cards and never return them, gaining them also triggers various effects, and you can use the actions to play cards that aren't your chosen cards. It's weaker in that you may not even get a chance to play the chosen cards twice because the gained cards don't go in your hand, but maybe that's just an oversight.
Regardless, if you're going to use tokens, you can just use them as memory aids without getting into gain shenanigans. Set aside your two action cards with tokens on them. When you play a card the first time, leave the token on it to remember that it still has another play left. Remove the token when you play it the second time.
You'd need a bag full of tokens to really make this work.
I just thought of another tracking problem that
all of these variants share. If you chain multiple IHs together (even if they are broken up with TR/KC/Golem), you're going to accumulate multiple set aside cards waiting to be played for the first time. Now you have to track which set aside card is connected to which IH in the chain. I guess a solution is to set them aside face down on top of the IH that is going to play it? Ehh.