Yeah I looked at the math again and we are indeed fine with just a few copies of BBM, and 3 is definitely enough. (I think 2 works too but we have 3 anyway)
Hmm, I think our Stern proctor/Soul separator line is a bit trickier to work due to March of the machines:
March of the machines must be in play when we BB combo, so it is still in play at the end of the BB combo.
Thus we do not have any artifacts in play at the end of the BB combo. So the real ones must be in our hand.
Soul Separator needs to tap for its activation and does not have haste. so we need to bounce March with a proctor trigger before making any new ones.
I'm having trouble making sustainable stack that works for remaking it between each BB computation? (need to at least make a fresh Boom keg)
Might be missing something simple but the timings seem pretty awkward
So we need to have a bunch of Panharmonicons on the battlefield. We tap a Soul Separator to exile Stern Proctor and create a Stern Proctor token, which triggers a bunch of bounce effects. We bounce our nontoken artifacts to our hand, play March of the Machines, play the artifacts to get copies, then bounce the March of the Machines and the artifacts back to our hand, with still some Stern Proctor ETB triggers left. Then, we tap a Soul Separator token to create a Tidal Barracuda token, tap a Soul Separator token to create a Desolation Giant token, play March of the Machines, and resolve the destruction effect. Afterwards, we will resolve more bounces and be able to bounce March of the Machines back to our hand.
Yeah the issue I'm seeing with that line is that we get all the panharmonicon/doubling seasonstern proctor targets at once, and can't use one to bounce the same thing twice.
Oh right, we have targeting problems. I think the issue of bouncing things multiple times can be handled by resolving multiple copies of Pull from Eternity. So we resolve a Pull from Eternity, use a Soul Separator to create a Stern Proctor token, and bounce things to hand once, and replay those that we want to. Then we can resolve another Pull from Eternity to get another Stern Proctor, and bounce things again.
I think so, we don't actually need many artifacts/enchantments each time through, I'm just not convinced we can make a lot of progress (Dual Nature's destruction trigger is annoying), but fortunately we don't really need to.
We really only need to make a boom keg and set up the triggers to do it again, which seems doable.
The question is what do we need before we can start profiting panharmonicons?
Hmmm - while a flicker applied to Panharmonicon can profit by exiling and then creating new tokens, the timing won't work out with bouncing and replaying. So, we want to bounce March of the Machines, bounce Panharmonicon, play March of the Machines, and then play Panharmonicon. Two bounces allow us to take X Panharmonicon tokens to 2X+2.
I don't think we need *too* many Panharmonicons, since if we have X copies of Pull from Eternity (that we can use on Stern Proctor, that is), and Y copies of Panharmonicon, we get X(Y+1) bounces in total.
Okay, you guys may hate this idea, but I gotta ask. What if we eliminate the rule where the opponent will cooperate with us to go infinite? So the opponent will always try to minimize the amount if damage they receive, and if that stops us from being able to reach arbitrarily large amounts of damage, the deck will be legal. So for example, giving the opponent the opportunity to stop a computation will not result in an infinite, since they will choose to stop it at a point that minimizes our potential, i.e. when we have the fewest number of creatures. So there's no "waiting for an arbitrarily long time" with the opponent.
The reason I like this is not just that it allows for potentially greater numbers, but also that it is a more elegant ruleset. Under the current rules, we basically use one number to check for legality, the minimum negative life where the opponent cooperates with us to achieve the highest score. But, if that number is finite, the actual score is achieved by calculating the minimum negative life where the opponent fights with us to reduce the score. Under the proposed ruleset, it would just be the final score that is checked for legality.
I don't like requiring cooperation from the opponent. Our opponent should have opposing goals from us, and since our goals are Maximum And Finite, the opposite goals are: Minimum Or Infinite.
Its also odd where to draw the line,
"Hey dont block this attack as you'll take more damage by letting us start a computation while still having flash"
I guess it enables us to browbeat them by threatening to go infinite, but IDK if that's actually an interesting strategy
Well, what I'm proposing is that the opponent *never* cooperates with us. The idea that this is somehow actually cooperating with us is a metagame concept, that by not cooperating they're allowing a deck to become legal, so therefore they should scheme to make the deck illegal in this challenge and therefore they win. It's *that* notion that I find odd - the players should be oblivious to the metagame aspect of which decks we will declare legal or illegal, and should just try to do the best they can. That's how our player acts anyway - they will try there best to do the most damage, which means invalidating our deck as often as possible. So the opponent should just act inversely, rather than being wise to the challenge while our player is not.
We can cast Goblin Game and Wheel of Misfortune, but the opponent will simply kill themselves without giving us a high score. That seems normal to me, more so than saying the opponent will deal arbitrarily high amounts of damage to themselves to help us out.
There won't be a line to draw, we are trying to maximize the negative of the lowest life total we can always bring the opponent to and win, no matter what the opponent does. So the rule is firm, I think.
As far as not being an interesting strategy, I don't see our strategies changing really. It would just mean no more scenarios where the opponent cooperates with us to go infinite. The actual strategy won't be related to that infinite, based on what we have seen in the past.
The current legality and scoring can be described by finding an upper bound X and a lower bound Y, and proving "It is not possible to deal more than X damage (regardless of what the opponent does), and it *is* possible to deal at least Y damage (regardless of what the opponent does)." Provided these numbers exist, then Y is taken as the score.
Is the proposed change expressible in a similarly nice way?
I'm not super comfortable with saying that a deck that casts Wheel of Misfortune "has an upper bound" on the amount of damage it can deal.
Yes, X is the largest number for which it is possible to deal at least X damage (regardless of what the opponent does). Under the proposed rules, if this number is not infinity, then X is taken as the score.
This seems more natural to me than having two numbers, one for legality and one for the score.
Wheel of Misfortune may not have an upper bound on the damage that it can deal, but I don't think it is correct, or at least not standard, to say that by putting Wheel of Misfortune in your deck, you are now capable of dealing any amount of damage to your opponent. This requires opponent cooperation, which I would like to get rid of. Certainly our score X doesn't go infinite when we put Wheel of Misfortune in our deck, and I would like to merge our score with deck legality.
I agree with FortyTwo, that the players should be opposed - and when players are truly opposed, they never cooperate. They cooperate under the current ruleset, because they share the mutual goal of wanting to go infinite - and that is because our player is trying to do their best in that particular game, and the opponent is trying to do their best in the metagame. It's that asymmetry that I dislike.
But our opponent would be doing what we as deck designers want them to do, which seems fairly cooperative to me. Even a random opponent would help us less.
Well, that's just the flip side of our player doing exactly what we don't want them to do, which seems odd as well. The basic objective is to maximize "the largest score we can always obtain, no matter the behavior of the opponent" (where score is "the negative of the opponent's life when we win the game, 0 if we don't"), so it's not really about the opponent acting a specific way, we are minimizing over all possible opponent behaviors.
The "no infinite" restriction has weird consequences, so we get things like ourselves bemoaning an infinite combo that will allow us to deal an enormous amount of damage to the opponent, while the opponent celebrates that outcome. I just think that such concerns should be "above" the players, they are just trying to play the game with the decks that have been handed to them.
jfb, what do you think?
Edit: So, the way I am looking at the opponent, is as a "game" opponent, for one particular game using one particular deck. They are not an overarching "metagame" opponent, who is trying to sabotage us as deck designers for the challenge as a whole. This makes sense with the fact that our player is definitely not a "metagame" player for our side, they're messing us up whenever they possibly can.
So, the opponent can do things that will help make the deck legal, and our player can do things that will help make the deck illegal - that's because the "no infinite" rule is just weird, but one that we need.
But the point more is that it should be bounded for any players. Which includes any of our agents that use cards in unintended ways to go infinite, and opponent agents that enable us to go infinite. Why relax that restriction on only our opponent? We can't control what our opponent choses.
The goal is that in the end the optimal play for both sides is Max vs Min, and any deck that works for the current rules will also satisfy this.
For me, it's not that we're controlling what are opponent chooses - we're calculating the highest score that we can always achieve no matter what the opponent does (which is what we are calculating currently), and if this is finite, there's no need to disqualify the deck.
Max vs Min is exactly what I am going for - our player will always try to max, the opponent will always try to min, and they will never switch to max for metagame reasons. They are just playing a single game with the decks they have, they aren't trying to make a deck legal or illegal.
But I get that you feel differently, and I respect that. We'll see how jfb feels.
I'm in favour of keeping the rules as they currently are, for a few reasons:
- I see the "maximum damage" and "no infinites" as separate goals, ("Of the decks that cannot deal infinite damage, which one can deal the most?") and in evaluating each one, we cannot make any assumptions whatsoever about the opponent's actions.
- It feels a bit "cheap" to change the rules of an ongoing challenge in a way that makes it easier for ourselves.
- In the presence of lines where the opponent could allow us to go infinite, then determining whether the opponent has any available plays that DON'T allow us to go infinite is in general a non-trivial question that may be required in determining deck legality; especially when Turing-completeness is considered. As an extreme and not very realistic example, say you can set up a Turing machine that searches for counterexamples of the Riemann hypothesis, and also give the opponent a way to halt it early. Also imagine that if the machine halts either naturally or due to the opponent's intervention, it allows us to go infinite. Then, the question of whether this situation is legal depends on whether the opponent is able to allow the machine to run forever and make the game a draw; i.e. whether the machine halts; i.e. whether the Riemann hypothesis is true.
Edit: Hmm, for that third reason - it seems like that example basically exists in the current rules? We can have a deck that runs the TM that tests for counterexamples to the Riemann Hypothesis, and have either us or the opponent have the ability to stop the computation. Then the deck will be illegal precisely when the Riemann Hypothesis is true. Of course, this isn't really a problem, since our actual decks have the ability to construct many TM's, including both obviously halting and nonhalting ones. But for the same reason, your example isn't really a problem either.
Also, speaking of sphinxes, not that it particularly matters here, but it looks like the rules might be updated so that Sphinx of the second Sun does draw on the first turn. There'll be a confirmation one way or the other with the Kaldheim rules update (~2 months)
Hmm, another issue I just noticed, without Opalescence, We need to hardcast Omniscience each worldfire?
I think a Leonin Relic-Warder effect would solve it? But that's a white creature, so oblivion ring? but how to recur the oblivion ring then? We can't just have a waiting trigger as it breaks on later iterations of Worldfire...
Hmm, is getting copies of Show and Tell a problem? We could conceivable resolve copies of Worldfire, followed by copies of Show and Tell, putting Dual Nature and Sanctum Plowbeast on the battlefield among other things, followed by copies of Wrong Turn, giving our Sanctum Plowbeasts to the opponent, followed by Starlight, gaining us life.
Private Mod Note
():
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
Hmm, I think our Stern proctor/Soul separator line is a bit trickier to work due to March of the machines:
March of the machines must be in play when we BB combo, so it is still in play at the end of the BB combo.
Thus we do not have any artifacts in play at the end of the BB combo. So the real ones must be in our hand.
Soul Separator needs to tap for its activation and does not have haste. so we need to bounce March with a proctor trigger before making any new ones.
I'm having trouble making sustainable stack that works for remaking it between each BB computation? (need to at least make a fresh Boom keg)
Might be missing something simple but the timings seem pretty awkward
Does this make sense?
I think Glint hawk + Drake familiar works, but is more cards...
Edit: no it doesn't.
We really only need to make a boom keg and set up the triggers to do it again, which seems doable.
The question is what do we need before we can start profiting panharmonicons?
I don't think we need *too* many Panharmonicons, since if we have X copies of Pull from Eternity (that we can use on Stern Proctor, that is), and Y copies of Panharmonicon, we get X(Y+1) bounces in total.
Yeah I'm happy that this section of the combos works now.
The reason I like this is not just that it allows for potentially greater numbers, but also that it is a more elegant ruleset. Under the current rules, we basically use one number to check for legality, the minimum negative life where the opponent cooperates with us to achieve the highest score. But, if that number is finite, the actual score is achieved by calculating the minimum negative life where the opponent fights with us to reduce the score. Under the proposed ruleset, it would just be the final score that is checked for legality.
What do you guys think?
Its a bit odd to enable casting Goblin Game or Wheel of Misfortune.
Its also odd where to draw the line,
"Hey dont block this attack as you'll take more damage by letting us start a computation while still having flash"
I guess it enables us to browbeat them by threatening to go infinite, but IDK if that's actually an interesting strategy
We can cast Goblin Game and Wheel of Misfortune, but the opponent will simply kill themselves without giving us a high score. That seems normal to me, more so than saying the opponent will deal arbitrarily high amounts of damage to themselves to help us out.
There won't be a line to draw, we are trying to maximize the negative of the lowest life total we can always bring the opponent to and win, no matter what the opponent does. So the rule is firm, I think.
As far as not being an interesting strategy, I don't see our strategies changing really. It would just mean no more scenarios where the opponent cooperates with us to go infinite. The actual strategy won't be related to that infinite, based on what we have seen in the past.
Is the proposed change expressible in a similarly nice way?
I'm not super comfortable with saying that a deck that casts Wheel of Misfortune "has an upper bound" on the amount of damage it can deal.
This seems more natural to me than having two numbers, one for legality and one for the score.
Wheel of Misfortune may not have an upper bound on the damage that it can deal, but I don't think it is correct, or at least not standard, to say that by putting Wheel of Misfortune in your deck, you are now capable of dealing any amount of damage to your opponent. This requires opponent cooperation, which I would like to get rid of. Certainly our score X doesn't go infinite when we put Wheel of Misfortune in our deck, and I would like to merge our score with deck legality.
I agree with FortyTwo, that the players should be opposed - and when players are truly opposed, they never cooperate. They cooperate under the current ruleset, because they share the mutual goal of wanting to go infinite - and that is because our player is trying to do their best in that particular game, and the opponent is trying to do their best in the metagame. It's that asymmetry that I dislike.
The "no infinite" restriction has weird consequences, so we get things like ourselves bemoaning an infinite combo that will allow us to deal an enormous amount of damage to the opponent, while the opponent celebrates that outcome. I just think that such concerns should be "above" the players, they are just trying to play the game with the decks that have been handed to them.
jfb, what do you think?
Edit: So, the way I am looking at the opponent, is as a "game" opponent, for one particular game using one particular deck. They are not an overarching "metagame" opponent, who is trying to sabotage us as deck designers for the challenge as a whole. This makes sense with the fact that our player is definitely not a "metagame" player for our side, they're messing us up whenever they possibly can.
So, the opponent can do things that will help make the deck legal, and our player can do things that will help make the deck illegal - that's because the "no infinite" rule is just weird, but one that we need.
The goal is that in the end the optimal play for both sides is Max vs Min, and any deck that works for the current rules will also satisfy this.
Max vs Min is exactly what I am going for - our player will always try to max, the opponent will always try to min, and they will never switch to max for metagame reasons. They are just playing a single game with the decks they have, they aren't trying to make a deck legal or illegal.
But I get that you feel differently, and I respect that. We'll see how jfb feels.
- I see the "maximum damage" and "no infinites" as separate goals, ("Of the decks that cannot deal infinite damage, which one can deal the most?") and in evaluating each one, we cannot make any assumptions whatsoever about the opponent's actions.
- It feels a bit "cheap" to change the rules of an ongoing challenge in a way that makes it easier for ourselves.
- In the presence of lines where the opponent could allow us to go infinite, then determining whether the opponent has any available plays that DON'T allow us to go infinite is in general a non-trivial question that may be required in determining deck legality; especially when Turing-completeness is considered. As an extreme and not very realistic example, say you can set up a Turing machine that searches for counterexamples of the Riemann hypothesis, and also give the opponent a way to halt it early. Also imagine that if the machine halts either naturally or due to the opponent's intervention, it allows us to go infinite. Then, the question of whether this situation is legal depends on whether the opponent is able to allow the machine to run forever and make the game a draw; i.e. whether the machine halts; i.e. whether the Riemann hypothesis is true.
Here is the current deck:
2 Xathrid Necromancer
3 Arcbond
4 Artificial Evolution
5 Wrong Turn
6 Dralnu's Crusade
7 Dralnu's Crusade
8 Dralnu's Crusade
9 Dralnu's Crusade
10 Mirrormade
11 Mirrormade
12 Desolation Giant
13 Goblin Boom Keg
14 March of the Machines
15 Tidal Barracuda
16 Spellweaver Volute
17 Pull from Eternity
18 Bloodbond March
19 Cephalid Shrine
20 Thousand-Year Storm
21 Invulnerability
22 Mana Echoes
24 Stern Proctor
25 Soul Separator
26 Sanctum Plowbeast
27 Faith of the Devoted
28 Army of the Damned
29 K'rrik, Son of Yawgmoth
30 Rings of Brighthearth
31 Dual Nature
32 Repeated Reverberation
33 Spellweaver Helix
34 Worldfire
35 Starlight
36 Restore
37 Recoup
38 Recoup
39 Panharmonicon
40 Molten Slagheap
41 Wormfang Behemoth
43 Aegis Automaton
44 Guardian of the Guildpact
45 Black Lotus
46 Show and Tell
47 Omniscience
48 Chameleon Blur
49 Leyline of Punishment
50 Words of Wisdom
51 Lion's Eye Diamond
52 Blue Sun's Zenith
53 Consecrated Sphinx
54 Widespread Panic
55 Jester's Mask
56 Sphinx of the Second Sun
57 World at War
58 Mortuary
Edit: Hmm, for that third reason - it seems like that example basically exists in the current rules? We can have a deck that runs the TM that tests for counterexamples to the Riemann Hypothesis, and have either us or the opponent have the ability to stop the computation. Then the deck will be illegal precisely when the Riemann Hypothesis is true. Of course, this isn't really a problem, since our actual decks have the ability to construct many TM's, including both obviously halting and nonhalting ones. But for the same reason, your example isn't really a problem either.
We wouldn't really want to allow our opponent to interfere with the machine anyway, as both versions of the opponent are happy with a draw.
The first two points I agree with though.
As for this deck, I think we are missing Chrome mox for the white?
And isn't Sphinx of Enlightenment strictly better than Words of Wisdom? (I swear I don't just like sphinx tribal)
Also, speaking of sphinxes, not that it particularly matters here, but it looks like the rules might be updated so that Sphinx of the second Sun does draw on the first turn. There'll be a confirmation one way or the other with the Kaldheim rules update (~2 months)
Yup, Chrome Mox somehow got off the deck! Son of a gun... so it's 59 cards.
Yeah, we can do Sphinx of Enlightenment, and switch from Tidal Barracuda to Vedalken Orrery.
So Dual Nature, coupled with either Opalescence or Panharmonicon, followed by Consecrated Sphinx and Sphinx of Enlightenment: we draw 45 cards! Not too shabby. We might want to go with less, so that we can save more opponent draws for when we can get way more Consecrated Sphinxes.
Hmm, another issue I just noticed, without Opalescence, We need to hardcast Omniscience each worldfire?
I think a Leonin Relic-Warder effect would solve it? But that's a white creature, so oblivion ring? but how to recur the oblivion ring then? We can't just have a waiting trigger as it breaks on later iterations of Worldfire...
Worldgorger dragon?
Edit: oh we just add Show and Tell on after Worldfire and cheat it in that way again. no changes needed