MTG Salvation

So the first part was only having Bone Harvest instead of two copies of our green creature. That way, we have to let it come in to tuck both Child of Alara and our green creature, before the Verdant Succession triggers resolve. So in our case, we'd have to leave Dark-Dwellers championed under Boggart Mob when Child goes off, so Dwellers comes back and tucks the creatures. That uses up one.

The other one was for our connecting creature. We had to crack the explosives beforehand to put Verdant Succession triggers on the stack to search it out after we play it. I'm not sure if that works with Jade Bearer, though, because we'd still get the ETB abilities from both it and a Mimic Vat token beforehand, enough to make a new Abolisher. Though I suppose we would also have to wait until after those Abolisher triggers resolve to get Jade Bearer back in the first place, so that should work!

So, here's my tentative list off your idea:



I wasn't able to find a safe 7th stage, because white and green creatures are both off-limits, as are 5-drops at the Sea Snidd level, but there's certainly room for one if we can find one. Natural Order at least looks safe in this deck, since even though we can stack the triggers however we want, there's no way to have Basilica Bell-Haunt in our graveyard after a Worldfire when the potentially unsafe triggers resolve.



Edit: J_kibbs, that's a really solid start! Something you might want to add is Rings of Brighthearth and Mach of the Machines. That way, every time you populate, you can copy the ability N times with Rings of Brighthearth, and make even more copies with the last one.

That was one of places where Verdant Succession helped. When we cast Natural Order, we'd sacrifice Nylea's Disciple, getting N Verdant Succession triggers. Each one would let us get out a Disciple while our devotion was equal to the number of Bloodbond Marches and Verdant Successions we had out. Then, after the transition, we used the Natural Order resolving to search out Disciple, gaining only 2 life.

Of course, that also required no Thousand-Year Storm, now I think on it...

Edit: Just saw the rest of this thread. Dual Nature is unsafe because we can also bounce Child of Alara in the megastage, instead of killing it. Then, we do all our necessary preprocessing and recast Child, so now we can kill it without resolving the whole transition.

Basilica Bell-Haunt looks like it could work, though! Do we have a way to account for the extra copy of Engineered Explosives and such in the megastage transition?

@J_kibbs: the problem is that extra tokens there won't matter. Even though you have these tokens, which in turn generate more mana, you aren't turning all that mana into more Doubling Seasons, at least not meaningfully. You're still making somewhere around 2^(number of Doubling Seasons) mana from all your White Sun's Zenith casts. That's not a bad thing to have, but the replicate copies you get from it are so dwarfed by stuff you're doing later on in the deck that there's no way to count it.

Let me put it this way. Right now, the first chance you get to turn all that extra mana from your Zeniths into Doubling Seasons is when you replicate Cackling Counterpart. But even if you're generating 2^N mana, where N is the number of Doubling Seasons you have out, all those replicated copies are STILL only doing as much as having 1 extra token copy of Precursor Golem out. Since you're making more Precursor Golems than there are atoms in the universe every time you cast a Cackling Counterpart, the extra copy kinda gets lost in the shuffle.


@Deedlit: I think I'm still missing something, because I don't see why we need the Dwellers to enter twice. Could you walk me through how your hyperstage transition works, exactly? I'm probably missing some detail involved.

Edit: Did not see your post. Hmm. This might mean moving over to Worldpurge. We should be able to use up the green mana by having Acorn Harvest be the flashback sorcery, no?

@J_kibbs: hmm, I see what you're saying. The mana definitely scales with Doubling Season creation. The problem is that there isn't an ideal place to put all that mana once you've got it, because the mana isn't what's limiting you.

Imagine for a moment you had just enough mana to cast/flashback Cackling Counterpart 1000 times. That's helpful for sure, but if you only have a way to get Cackling Counterpart back from exile 100 times, then you can only cast/flashback it 200 times total, no matter how much mana you generate. The rest of it just gets wasted.

Right now, your end goal is "make a gigantic amount of Doubling Seasons". The way you make lots of those (as well as lots of Mana Echoes, Thousand-Year Storms, etc.), is by casting Cackling Counterpart. And yes, every time you cast it, you can pump all your mana (proportional to the number of Doubling Seasons) into replicate. So if you have N Doubling Seasons, each replicated copy will make 2^N more Doubling Seasons. You have N replicate copies, which is enough to add a second arrow.

BUT, you also have all these Precursor Golems. Each Precursor Golem token copy you have puts a separate trigger on the stack, and each of those triggers ON ITS OWN makes N Cackling Counterpart copies targeting Doubling Season. So in other words, you can get the same amount of benefit from generating all that mana and feeding it into Djinn Illuminatus, just by creating one more copy of Precursor Golem. If you can find a way to make two copies, you'll actually be doing even better than you are now.

That's where the inefficiency is coming from in the list. It's a really common problem in these decks, and hard to spot when it happens, because it doesn't /look/ inefficient. You can still see how Djinn Illuminatus and the extra mana will make the damage total way higher at the end. But because it's only running alongside some bigger combo, it winds up disappearing into the margins when you try to calculate your end result.


If I could make a suggestion for your deck: try making a version that starts by just converting mana into enchantments. We have some... convoluted ways to do that which you don't need. A simpler way might be something like this:

Start by cutting Mana Echoes (which would go infinite this time) for Dual Nature, and add a copy of Copy Enchantment. That's a trick from our big write-up at the top of this thread. You play Copy Enchantment and don't copy anything. Then, the Dual Nature triggers combine with Doubling Season to make lots of copies, all of which can become whatever enchantment you want. You also don't have to worry about Dual Nature exiling your tokens anymore, because it's looking at a card named Copy Enchantment.

That way, every time you play a Copy Enchantment, you get a bunch of Dual Nature triggers. They all resolve one at a time, each one making 2^N copies of Doubling Season. With N Dual Natures, that's two layers already.

Then, to turn mana alone into copies of your enchantments, use Allay, Mortuary, and Rememberance. For 3 mana, you can play Allay, getting a bunch of Thousand-Year Storm triggers. With N Thousand-Year Storms, you can kill Copy Enchantment N times. Then, Mortuary puts that Copy Enchantment back into your library. Finally, Remembrance lets you search your library for Copy Enchantment and put it in your hand. You replay it, and you get more triggers, making more Doubling Seasons, Dual Natures, and Thousand-Year Storms.

So, for just 3 generic mana, you get:

N copies of Allay, one for each Thousand Year Storm (there are probably ways to get full mileage out of the copies here, I just haven't found one). Each of those become:
N triggers of Dual Nature, each of which get doubled by:
N copies of Doubling Season.

That's three arrows on its own. Then, you can find a way to turn all those Cackling Counterparts into ways to produce colorless mana (say, by making copies of Emrakul's Hatcher, where the tokens also get doubled by Doubling Season...), you can probably get close to 6 or 7 arrows at the point in your deck where you have only 4 right now.


@Deedlit: I'm not sure I follow why we need Panharmonicon back. With your version, we can't champion Vedalken Orrery anymore, so what good do multiple rebuilds get us?

Also, Wormfang Behemoth worries me because it lets us save a lot of cards though the gigastage transition, including some potentially unsafe ones like Engineered Explosives. If we can make it safe though, it might be worth seeing if we can swap the order of life/mana to save cards.

@J_Kibbs: That's a sweet deck! I'm gonna go along with what Deedlit has said here. When these decks get this complicated, it's hard to figure out what counts and what doesn't, because frankly none of us have any intuition for how these numbers work.

What we've found works best in our deck, and probably would work best with yours, is to set up this simple "chain" of resources. You can see what I mean in my old standard combo with Polyraptor, which I wrote up here.

With that deck, the objective is to make as many Polyraptors, the final source of damage, as possible. Then, I have a sequence of different combos that each turn one of some resource into an amount of some other resource equal to the number of Polyraptors I've already created. One point of toughness on Forerunner of the Empire creates N Polyraptors, where N is the number of Polyraptors I've already created. One green mana (well, more than one, but you get the idea) turns into N points of toughness on Forerunner of the Empire, each of which turn into N Polyraptors, one after the other. One energy counter turns into N green mana, each of which turns into N points of toughness, each of which turns into N Polyraptors...

You get the idea. The key here is that every new group of cards I'm adding has two things going for it: it directly creates whatever thing the last group I added was using up, and the amount it creates is directly tied to the number of Polyraptors (key being that this thing I'm tying it to is the bottom level) I've created so far. That both makes my life easier because it's easier to follow what's happening in the deck (and easier to correct for infinites: if one of these modules is broken and goes infinite, I can just slot it out of the deck and the rest of it is intact), and because I can guarantee that each new group I add adds a layer, and isn't just spinning my wheels.

Two other notes: planeswalkers usually don't work out well in these decks. Every version of our deck, all the way back to 2009, has played either Vedalken Orrery or Leyline of Anticipation because sorcery-speed is the bane of our existence. Planeswalkers are pretty much guaranteed to be sorcery speed, which makes it damn near impossible to fit in any shenanigans AFTER the stage of your deck where you use them. And you absolutely don't have to win the game that turn! Every deck I've ever seen for this challenge does, simply because the amounts of damage involved are so absurdly high, but there's nothing /requiring/ you to do so.

In fact, if you were to build a deck that required your opponent to still be alive after dealing the damage, I'd love to see it regardless of how much damage it deals.


@Deedlit: I'm still trying to fix my deck, but no real updates.

What made Anaba Ancestor work was the combination of Chalice of the Void and Myr Welder. Myr Welder meant you had to actually resolve down to a Clockwork Gnomes section to continue with the combo, while Chalice of the Void meant that doing so had a real cost associated with it: any time you made that transition, you had to at some point cast a Hurkyl's Recall to do it. Neither one was enough to make the Ancestor safe on its own.

Darn, you're right. And that knocks out one of my other ideas as well. I'm going to keep thinking on the changeling deck, but in the meantime we should probably move over to your list.

In terms of restricting zone-changing... the exile restriction isn't actually what made Muzzio safe in the Chalice deck: what made it safe was Chalice countering Hurkyl's Recall. So I don't know if that would help here.

I think there probably is a way to order triggers such that you don't need Tidal Courier, and can just use a Bearer token from Mimic Vat to bounce the Abolisher and a cast to kill it. Or vice versa. But I don't know what it is.

Even though Dark-Dwellers can get back Engineered Explosives, you need a Myr Welder to put the Explosives back, and you have no way of generating that in the middle of the megastage transition.

The plan I'm gunning for is to use Hunting Pack after all the Smites resolve, generating a huge amount beasts after all the important parts of the Megastage transition, so they stick around as we start rebuilding our artifacts from the triggers below. Then we don't bother generating all the Psychic battles immediately. We generate some constant number that we need to kick-start a hyperstage, then cast Acorn Harvest. We get several copies of Smite the Monstrous, and all of them get to generate N copies, because of Mirrorwing. THat should be enough to generate all the artifact copies we need to convert our mana. But it's done after casting the new Harvest, not before.

I don't think you have to resolve to the next one down for Mirrorwing Dragon to work. We have to add Copy Enchantment anyways, because we're trying to cast 2 instants with Volute. Once we have that ability, we can put one on Hunting Pack and two on Smite the Monstrous. The Hunting Pack one gets us N creatures covered by the Dragon, the first Smite gets us N copies of Smite the Monstrous, which can then repeatedly bounce and replay Mirrorworks (we can kill Mirrorworks, make a token with Vat, and then Muzzio the nontoken back during the transition, that's not very expensive). That gives us N copies of our other artifacts like Mimic Vat and Mirror of Fate, which is enough to turn all our mana into enchantments. Then, the second Smite works the same way as our current setup.

Hmmm, I see the problem. I had assumed we could just resolve down through extra Psychic Battle triggers to regenerate artifacts, but yeah that doesn't actually work.

Sadly, Populate won't do it: once we start running through the megastage, we can also make token copies of Goblin Dark-Dwellers, and make token copies of that instead. Keying off creature types also probably won't work, because Changeling Hero has them too. Of course, we could then go back to the Brudiclad, Telchor Engineer version, which can still probably reach 6 stages...

Thousand-Year Storm seems like the easiest way to solve this. If we can find another way to force us to use up a Muzzio in the hyperstage, it should become safe again. I'm not sure how to do that, though.

Edit: one suggestion and one rules question. If we use Mirrorwing Dragon instead of Precursor Golem, then Smite the Monstrous will get a copy for each Hunting Pack token we make.

Also, if I have a Copy Enchantment copying Bloodbond March while Verdant Succession is out, and the Copy Enchantment dies, do I search for a card named Bloodbond March or a card named Copy Enchantment?

We also need Simian Spirit Guide or some equivalent to play the Eureka at the start.

Okay, here's a weird one: does Reign of Chaos work?

We can use it to target any blue creature, but only if there's an island out. For us to have an island, we need to have already activated a Sea Snidd to turn Bayou into an island. So it goes: tap the Snidd to make an island, do a mini-gigastage to cast Reign of Chaos, get N Psychic Battles on Sea Snidd, use them to bounce/kill Old Man of the Sea and the Snidd as needed?

It seems like it should work, but I don't think we've ever tried a stage where the battle triggers can be separated from the tap ability.

As for Skull of Orm... don't we really need a way to make copies of Psychic Battle? otherwise I'd say go for it, cause we can already make as many as we want of Bloodbond March, but with no spell copiers Psychic Battle is the only thing building up our hyperstage and megastage.

In any case, that gives us:


That's gonna get us somewhere in the realm of f_{w^4+w8+2}(3), I'm guessing. Assuming we can get it off the ground. Recasting Reforge the Soul more than once will be much more difficult this time, since we need to somehow generate a hasty Millikin to get a flashback spell back.

Verdant Succession is pretty important. There's basically 0 flexibility in the Centaur Safeguard slot. Nylea's Disciple, Tireless Missionaries, they all have enters-the-battlefield abilities that go infinite with the changeling. And I haven't found another way to get back Centaur Safeguard.

A few small issues with the decklist: we both have cut Panharmonicon because it's unsafe with champion, but without it we need 3 copies of Spider Spawning, not 2. So that cuts one space out of our room. My list also needs Reforge the Soul, because without either of Consecrated Sphinx or Thousand-Year Storm, there's no way we're getting through the setup with just Words of Wisdom.

And sadly, spell copier effects are still unsafe there. The problem is that the copies won't get countered by Chalice of the Void, so they'll still resolve. And if we can resolve a resolve a Hurkyl's Recall through a Chalice on 2 during the Hyperstage transition, we can bank a Muzzio into the next level down, which lets us go infinite.

I'm not sure if it changes the calculus for the megastage transition yet, but we'll also need to put Chalice on 3 temporarily, to keep Phantatog around. If that does cause a fizzle, we can just go back to Allay. Worldfire will exile it, but I'm pretty sure we can cast it from our graveyard with Goblin Dark-Dwellers and buy it back. It's not like flashback where it gets exiled no matter what.

For the eighth stage, one card I've had in my sights for a while is Sea Snidd. It can generate any color of mana, and is immune to everything but Xathrid Gorgon. If we can find a way to turn white mana into "target blue creature", that could be number 8.

It'd be a lot easier if the gigastage keyed of white mana instead of black, and we could use "target nonblack creature" to differentiate at this level, but as far as I can tell there is not a single sorcery cmc>3 in the game that you can innately cast from the graveyard using white mana plus some amount of blue, green, and generic.

Okay, so here's the Anaba Ancestor List:




Past versions of this deck have relied on making the graveyard, exile, and library essentially interchangeable zones. That’s been useful to us in the past, but it makes Ancestor go infinite. This version prevents that by severely limiting the ways different cards can move between zones. Here are the basics:

• Any creature (or card that has become a creature) dying can easily be moved from the graveyard to exile via Mimic Vat.
• Cards in exile can always be moved into the library, by activating a Mirror of Fate token. Additionally, if they are imprinted under a Mimic Vat, they can be moved back into the graveyard with the following process: kill a green enchantment like Bloodbond March, imprinting the enchantment on the same Mimic Vat and triggering Verdant Succession. Mimic Vat dumps the old card back in the graveyard. Activate Mirror of Fate, putting the green enchantment back in the library. Use a Verdant Succession trigger to put it back on the battlefield.
• Cards in the library can be moved into the graveyard by tapping a Millikin. This requires a hasty token copy of Millikin, which constrains how often you can to this early on. Crucially, this is the only way to get back instants and sorceries post-exile: you shuffle them back in with Mirror, then mill them with Millikin.
• Finally, noncreature artifacts can be exiled from the graveyard with a hasty Myr Welder token. This is the only way to get back an Engineered Explosives after cracking it: you exile it with a hasty Myr Welder, shuffle it back in with Mirror of Fate, and pull it back out with a Muzzio, Visionary Architect activation.

The early stages of the combo are mostly the same, with the exception that we now counter spells using Chalice of the Void, not Cephalid Shrine. Most of the time, Chalice will be set to 4. This conflicts with Bloodbond March (which we can get back with Verdant Succession instead), and Vedalken Orrery (which we only have to copy at times when Chalice isn’t set to 4).

For the stage creature, we use Clockwork Gnomes. We can’t use a 2-drop or 3-drop, because if it were, we would be able to use Chalice on both it and one of Millikin and Myr Welder, and we can’t allow the deck to generate Bloodbond March triggers for either creature that cheaply.

The Hyperstage creature is a 2-drop, so for each transition, we have to safely go from a Chalice-on-4 configuration to a Chalice-on-2 configuration, and back, to rebuild our token base. That means resolving Hurkyl’s Recall twice. You also can’t resolve any of the spells in the hyperstage transition while Chalice of the Void is on 2: both Hurkyl’s Recall and Goryo’s Vengeance cost 2 mana.


First, we activate the Anaba Ancestor, targeting Changeling Hero and putting N Psychic Battle triggers on the stack. We use the Battle triggers to repeatedly bounce Changeling Hero, blinking Goblin Dark-Dwellers. Goblin Dark-Dwellers casts Goryo’s Vengeance each time, returning Muzzio, Visionary Architect (a copy with haste).

Crucially, we have 2 Muzzios. When we reanimate one nontoken, the other should already be out. That means every other Vengeance cast gets us 3 Muzzio activations: one from the new nontoken, one from the token made with Mimic Vat from the old nontoken, and one from the token we get when the new nontoken gets legend-ruled. We use two to profit one Gnomes Token, and the last one to get a Millikin token. We use the Millikin token to mill Goryo’s Vengeance, allowing us to do it again.

To get back the second Muzzio, we use an additional Battle Trigger from the Vengeance above to kill Millikin again, without bringing it back or tapping it. Then, the next Dark-Dwellers cast brings back the nontoken Muzzio, legend-ruling the token. We tap the token Millikin to mill Vengeance again, then activate the nontoken Muzzio to get back the nontoken Millikin. We are now back where we started, minus two Psychic Battle triggers on Anaba Ancestor and plus one Clockwork Gnomes token.



For the Hyperstage Transition, you need Engineered Explosives on the battlefield, at least 5 Psychic Battle triggers in the Anaba Ancestor batch, an untapped Muzzio, and a lot of floating mana.

First, we resolve a Battle trigger, blinking Dark-Dwellers and casting Hurkyl’s Recall. Before it resolves, we replay Changeling Hero, championing Vedalken Orrery instead. Hurkyl’s Recall resolves, and all our artifacts (including Chalice and Explosives) get bounced.
Next, we resolve another trigger, bouncing the Hero and getting back Vedalken Orrery. From there, we replay Mirrorworks, Mimic Vat (make a token, 2 Vats) Karn, Silver Golem, Mirror of Fate, etc. We cast Chalice for 2, then play both Millikin and Anaba Ancestor into it, getting Bloodbond March triggers. We play Explosives for 2 (we can add extra colorless to the X to prevent it getting countered) and crack it, killing both Millikin and Ancestor and putting them on both our vats. We make token copies of each, then get them back with Bloodbond March.

Note: This is also where any additional copies of Ancestor come in. If we have untapped Ancestor tokens lying around, we can activate them to create a normal stage chain before cracking Explosives. While we only have 2 Mimic Vats and 1 token Mirror of Fate, we can chain them by using spare Psychic Battle activations in our stack to blink Changeling Hero, using it to blink Mimic Vat and Mirror of Fate as needed to get more tokens.

So far so good, but now Explosives is in our graveyard. We can only get it back from the library, and we can only put it in the library from exile. To exile it, we need to activate a Myr Welder token, which we can’t possibly have at this point: Hurkyl’s Recall bounced them all, and we have no way to target Welder to get one back.

In other words, we have to resolve back down the stack to an earlier Gnomes activation to proceed!

We only have one other resource to use up, the untapped Muzzio. Its role is resetting Chalice. We have to bounce Chalice again before we reach the Psychic Battle triggers from the next Hyperstage down, otherwise we won’t be able to set it to 4. But we can only bounce it with Hurkyl’s Recall, which costs 2. Chalice is on 2 now, so Recall will get countered. We solve this by animating Chalice with Karn. Chalice becomes a 0/0, dies, and gets exiled under a Mimic Vat. We use a Mirror of Fate to put it back in the library, then activate our last Muzzio to put the Chalice back on the field, now with 0 counters.

After that, we use a battle trigger to bounce the nontoken Ancestor again, then use the last two to repeat the Hurkyl’s Recall/Vedalken Orrery process.

After that, we can use Goryo’s Vengeance to get back Muzzio again, but we still won’t have any Clockwork Gnomes. No Gnomes, no hasty Millikins to bin Vengeance again. This will actually fizzle the entire combo.


For the Megastage transition, we let Child of Alara kill Changeling Hero, getting back Orrery and letting us tuck Centaur Safeguard back in the deck to find it again. That means Hero is dead after the megastage transition. We need a Bloodbond March trigger to get it back, which means that at some point Chalice had to be set to 5. Putting Chalice on 5, then back on 4 without letting triggers from when it was on 5 resolve requires at least 2 Anaba Ancestor tokens, so that should be safe.


Whew, that's a lot of stuff. If all that works, and that's a reasonably big "if", that's 7 stages above the gigastage, 2 layers for Reality Spasm and Precursor Golem (we can't use Tribal Unity with no enchantment copiers), plus the same 2 layers from combat+draw, plus World at War. There may be a way to cram an eighth stage in there, but I haven't been able to find one.

J_kibbs: That would be infinite, yes. The formal rule we have is basically, "you give me your deck, and I pick a number. If there is a line of play in your deck that can deal more damage than the number I picked, no matter what my number was, it goes infinite".

In other words, there may be ways you can set up your opening hand that are only finite. But if there is even one way to go infinite, even if it's not the one you intended, the deck is disqualified.

As your deck becomes more complex, it will help it if you can break it down into distinct stages. That way, it's not one group of 60 crazy cards, but 10 groups of 6 crazy cards. If you look at our list over the last several pages, as the cards inside it change, the shape of the list as we post it here doesn't. You've got a group of 12 engine cards to start, then a module to make copies of enchantments, and then a series of modules that feed into each other in different ways.

Another advantage of this structure is, when you make it that way, if one of those modules has a problem and goes infinite in a way you didn't realize, you don't have to do as much thinking to get it all working again - you just cut that module and replace it with another one

One last thing: as you look at improvements, something you want to think about is whether what you're doing is meaningfully impacting the total. Let's say you added a creature like Thromok the Insatiable and a way to grant haste to your creatures. That would square the amount of damage you dealt at the end, which sounds great. But the numbers we're talking about, like 2^^^^5 and such, are so large that we actually can't describe the difference between that number and that number squared. We just discount differences that small. So even though you would have vastly improved the deck, you won't have changed the damage total as far as this challenge is concerned.

I think that's why your new deck doesn't deal as much damage in the last one. You're optimizing, and doing a great job at finding new interactions, but those new interactions aren't happening in the right order or with the right size to matter for the final total.


@Deedlit: I haven't done that thorough of a search for infinites because I'm still vetting this Anaba Ancestor deck, but that deck looks good for the most part! Thanks for the clarification on Soul Foundry.

One small issue: if we use Broken Ambitions, we can't have Sylvan Echoes/Sentry Oak as our draw level in the late game. Kithkin Mourncallers/Militia's Pride should be an easy swap though.

My Anaba Ancestor list only has one stage more than your list and the same number of layers, so it looks like we are (finally!) converging on a damage total for the gigastage deck, somewhere in the range of f_{w^4+wX+5}(n), 5<X<8. Not bad for government work


Edit: I think creatures we can kill with Explosives are still unsafe in that list. We can still use the topmost layer to make Bloodbond March triggers, then crack Explosives to make a token copy. That should only hurt the stage configuration at the top, though, not the meat of the deck.

Always great to see more minds working on this challenge!

I haven't been able to go all the way through your write-up yet, but I have a few suggestions. Feel free to ignore these if I'm actually just missing a major interaction:

1. I see a lot of effects that are doubled up in your deck. You have Angelic Chorus, but also Essence Sliver to gain ever greater amounts of life. You have Thousand-Year Storm, Djinn Illuminatus, and Riku of Two Reflections to make copies of spells. This is usually less efficient than finding a way to only use one, and then spending the remaining slots on cards to repeatedly copy it.

Take spell casting, for example. When you cast a Cackling Counterpart, you get X triggers from Riku (one for each Riku), Y triggers from Thousand-Year Storm, and Z triggers from Djinn Illuminatus. But since all those triggers go on at the same time, only one of them will really matter for your total. One of them (probably Thousand-Year Storm or the Djinn) will dominate, and the other ones won't matter. It'll be like that card slot was wasted.

If you look at the current list for our challenger (page 72, I think?), this is why it's almost all singletons. We're trying to cram as many different effects in as possible. And while there are a few that are similar (Child of Alara and Worldfire), we've found some way to exploit the small differences between them, so they do a different job.

2. While I love the Moonmist/Azor's Gateway combo, I'm not sure it's necessary here. You have Mana Echoes, and you are making a LOT of tokens. And with Mycosynth Lattice out, you don't have to worry about colors. Let me put it this way: most of the mana you're spending will either be on Mirage Mirror/Mizzium Transreliquat, or on copy making abilities. Every copy trigger you have costs 2 mana, and each creature entering the battlefield will make way, WAY more than 2 mana. So I'm not sure why that bit is important. Even later on, the mana from Mana Echoes will dwarf the amount from just tapping Azor's Gateway.

3. Speaking of Mycosynth Lattice... cards like it seem useful but can secretly be real traps in this challenge. It closes off a lot of options for adding more cards on top of your deck to add more of those juicy juicy arrows. For example, it would be really nice to have something like Aegis Automaton in your deck. You could use it to bounce your creatures back, then replay them to make lots of copies, Intruder Alarm untaps, etc. Then, you could have some way to produce a lot of white mana (Serra's Sanctum, perhaps?). Then, finally, you have some way to turn combat steps into untapping your Serra's Sanctum. But you can't, because the color of mana doesn't matter to your deck.


Every deck in this challenge that I've ever seen boils down to the same basic principle. They're all really good at taking some resource (Draw-7s in your deck), and turning that into lots of some other resource (combat steps, I think), and turning that into lots of ANOTHER resource (creature tokens?), and so on, until eventually you're turning something into damage. The more of those resources you can fit between whatever you're starting with and "damage", the more you'll deal. Mycosynth Lattice makes that harder, because it cuts off a lot of resources before you start. You can't count white mana, blue mana, red mana, etc. separately.



Switching back to our challenger, Deedlit, two things:

1. Your most recent list looks pretty neat, but I'm not sure how we reuse Perpetual Timepiece. That seems fixable, but it's gonna be tricky. I also don't see a way to actually bounce or killGodtoucher in the hyperstage. What am I missing?

2. I've been tinkering with that Anaba Ancestor/Chalice of the Void idea we had a while back, and I might have found a way to make it work. I'm still vetting it myself, because the stack has gotten /really/ complicated in the hyperstage, but if I don't find anything I'll post my new list and how it works some time next week!


I also think we should all at some point take a look at what we can do in Standard now. I haven't put much serious thought into it, but Thousand-Year Storm seems ripe for abuse. We probably won't be able to match the insanity that was last rotation's deck, but that was a fluke anyways.

So, one issue with the Ultracombo write-up is that, because the deck involved has been in flux the whole time, the startup is no longer accurate even to the most current Ultracombo list. In that case, I think that Allay thing is an actual bug in the setup, which would reduce the amount of damage it can deal.

We also have not been good about explaining in simple terms how our new discoveries work, myself included. I'm gonna try to write out most of those concepts, but I'm not going to touch on up-arrow notation or the fast-growing hierarchy, that side of the math. For that, you should check out one of the write-ups.

So, basically every deck we talk about today, including the Ultracombo and the gigastage we're working on now, start with the same basic set of engine cards, which is the first section we put in the lists. The initial setup just tries to get them out while using as few resources as possible. Going through the list:

• Omniscience lets us ignore casting costs from the hand, so we don't have to worry about getting white mana for Opalescence or black for Bloodbond March.
• Vedalken Orrery lets us play things at instant speed, which is important because we're trying to build a massive stack of triggered abilities and it's nice to play spells in response.
• Opalescence and either March of the Machines or Karn, Silver Golem make noncreature artifacts and enchantments into creatures, to make them easier to copy.
• Cowardice and Horobi, Death's Wail mean that any time we target a creature with a spell or ability, no matter what that thing would normally do, we instead either kill that creature or bounce it to our hand. If we want to bounce it, we stack the Cowardice trigger above the Horobi one. If we want to kill it, we do the opposite.
• Psychic Battle lets us get extra mileage out of each ability: because our opponent's deck is all lands and we control the order of our library, we always win the clash, so each Psychic Battle lets us reselect the target of that spell or ability, which triggers Cowardice and Horobi again. So, for instance, if we had 5 Psychic Battles, Vedalken Orrery, a Coal Stoker, and a Prodigal Pyromancer token with haste, we could tap Prodigal Pyromancer, targeting Stoker. The first activation would bounce Stoker, we replay it, and get RRR. Then the first Psychic Battle trigger resolves, we retarget the ability to the replayed Stoker, and bounce/replay it again, going up to 6 mana.
• Cephalid Shrine lets us control whether any given spell resolves: basically all our cards are singletons, and we can always refuse to pay 0 mana and get the spell countered. This is important, because sometimes we want to have access to a spell before all the triggers that happen when you cast it resolve. For instance:
• Bloodbond March, especially when we create many copies, lets us cast a creature, get several Bloodbond March triggers, counter the creature, and use the first trigger to bring it back to the battlefield. It's just like if we cast the spell, but now, if it dies, all we have to do to get it back is wait until the next one of those Bloodbond March triggers resolves, and it will come back.
• Mimic Vat is our main token-generating tool in the deck: Any time we need to make a copy of something, we can kill it, imprint it on an untapped Mimic Vat, tap the Mimic Vat to make a token, and then find some way to get the nontoken back on the battlefield. For instance, if it's a creature, we might transfer it from exile to the graveyard, then let a Bloodbond March trigger from when we cast it earlier resolve, bringing it back.
• Mirror of Fate and Perpetual Timepiece, as long as we can keep making copies, let us pretend that the graveyard, exile, and our library are completely interchangeable: for instance, to get that imprinted creature back in our graveyard without imprinting something else on a tapped Mimic Vat, we could activate a Mirror of Fate to tuck that creature into our library, then tap a Perpetual Timepiece to mill it into our graveyard. For some past decks, we've instead used Izzet Guildmage and Pull from Eternity for this step, particularly when we want to make it harder to put things specifically back in the library: We just cast Pull at the very beginning, then keep copying it with Guildmage's ability as needed. It's the same basic principle.
• Finally, we have some other mechanism for making tokens of cards we can't profitably make with Mimic Vat. In past decks, that was Dual Nature, but nowadays we use Mirrorworks to prevent infinites. That lets us make lots of copies of artifacts by bouncing and replaying them, so we never run out of new Mimic Vats, Mirror of Fates, or Perpetual Timepieces.

Right now, to copy enchantments, we do something like this: spend a green mana to flashback Ray of Revelation, which kills the enchantment. We imprint it on a new Mimic Vat, and activate it to make a token copy. Then, we sacrifice a Mirror of Fate token to shuffle both the enchantment and Ray of Revelation into our library, and tap a Perpetual Timepiece to mill them both. Then, we activate a Skull of Orm to return the enchantment from our graveyard to our hand, and replay it. Now we're right back where we started, except we have an additional copy of the enchantment, we've spent 16G, if you count paying for Mirrorworks triggers, and used up 4 artifact tokens: Mimic Vat, Mirror of Fate, Perpetual Timepiece, and Skull of Orm. We can also generate green and generic mana by bouncing artifacts, like Mana Vault and Simic Keyrune. While it's expensive to do that at first, once we've made enough copies of Psychic Battle this way, we can do this many, many times from just one ability that reads "target artifact". That also means we can't use an enchantment as a limiting factor later, because we can just use some artifacts to make more.


The other thing that threw me when I was getting started was how token stages work. So, let's say we're targeting artifacts with Rust Tick. When we activate Rust Tick's ability and get a lot of Psychic Battle triggers, we can use each trigger to do one of three things: we can bounce another artifact and replay it to make token copies with Mirrorworks, we can bounce the nontoken Rust Tick and replay it, giving us a big stack of Bloodbond March triggers, or we can kill the nontoken Rust Tick, imprint it on a Mimic Vat, and make a new hasty token copy. But if we do that, it's stuck in exile. The only we we have to get it back, at least without using something higher up, is by resolving a Bloodbond March trigger. But we can't make new Bloodbond March triggers to do it, because that requires casting it and we can't cast Rust Tick from exile. So we have to resolve down until the top of the stack is a Bloodbond March trigger from Rust Tick. That in turn requires that we cast it earlier, which means we bounced it with a Psychic Battle trigger below THAT.

So the end result is that, any creature that can tap to target itself and also something we need can make this giant stack of alternating batches of Psychic Battle and Bloodbond March triggers, where each one feeds into the other. When we run out of Psychic Battle triggers at the top of the stack, we use the last one to kill Rust Tick, make a new token copy with Mimic Vat, and bring it back with the first Bloodbond March trigger below it. When we run out of Bloodbond March triggers in that batch, we just use the top Psychic Battle trigger below it to bounce Rust Tick and replay it for more Bloodbond March triggers. Each different batch of triggers on the stack is another layer, so with N Rust Tick tokens, we get N recursive layers (well, 2N, but we round down for convenience). For instance, to match the number from the old Megacombo, we would only need to make 408 hasty Rust Tick tokens. That closes off artifacts as a resource we can use to limit ourselves later.

Two important things: the only way we have of giving anything haste is Mimic Vat, so the nontoken Rust Tick can't do anything on its own. And because we can't keep making Bloodbond March batches after killing our Tick, we can't build a loop. Put another way, to make a new Rust Tick, we need a Psychic Battle and a Bloodbond March trigger next to each other. We can only use one Psychic Battle trigger per batch, and each batch costs one Tick activation, plus the cost of a Bloodbond March trigger below it. That came from a separate Tick, but we can use each one, so think of that as costing 1/n, where n is the number of Bloodbond March triggers that activation made. So to get 1 Rust Tick back out, we have to spend 1+1/n Rust Ticks. The difference is tiny, but enough to make it so we can't loop.


Hyperstages, megastages, and gigastages all basically work the same: you spend some resource (a token, 3 life, a black mana) to put a bunch of triggers on the stack. Most of those can generate a lot of the next resource down (the hyperstage makes lots of Rust Ticks, for instance), but the last one lets you destroy all of the resource you generated to make a new, replacement one. That deletion step is important, because otherwise you can save a Rust Tick until lower down and use it to pay for the costs to keep going, and go infinite. Because that's not something you can just make with one ability, each one of these casts some subset of instants and sorceries as its trigger. Our hyperstage uses a token creature that can tap to make Goblin Dark-Dwellers enter the battlefield, which in turn lets us cast instants with converted mana cost 3 or less as a trigger. Not sorceries, I'll get to that later.

For instance, if instead of using Rust Tick at the bottom we used Metallurgeon, we could have the hyperstage start casting Battle Cry, untapping all the token copies we made for much bigger stages. Then, when we need to delete all our Metallurgeons, we would instead cast Rebuild, which would kill all our tokens, bounce Vedalken Orrery which is important so that we can't do any trick stack shenanigans in the middle, and also some artifact that will let us generate more of our resource, but can't otherwise be reset. We use 0-mana artifacts, which we can't bounce with Cowardice: Cowardice only works on creatures, and our artifact-animating spells use converted mana cost to set toughness, so a 0 mana artifact becomes a 0/0 creature and dies. In the past, the hyperstage used mana, so we bounced a mox. The new version uses a token, so we bounce Engineered Explosives, which we then crack to kill some creature that lets us regenerate our hyperstage. We've tried both having it directly kill the creature, and having it kill something else that can somehow generate a new token copy. Right now, we do the latter, but I have hopes that we can get the former working. At the moment, this step closes off white creatures (can untap with Battle Cry), goblins (we can search them out easily to replace them), humans (that's how we specifically target our hyperstage creature, Moggcatcher, creatures with flying (we target humans by giving them flying, because there isn't a convenient creature that can just target a human), and creatures with +1/+1 counters (we give humans flying by giving them a +1/+1 counter, then giving creatures with +1/+1 counters flying). Engineered explosives also closes off creatures with converted mana cost less than or equal to the number of colors of mana we can generate before reaching them in our list, because we can kill them with Explosives instead. 3-drops, however, are always safe, because even if we can kill them with explosives, doing so would also kill our Mimic Vats so we can't imprint them.


The megastage keys off of instants of any cost, not just 3 or less. It uses Spellweaver Volute to make it so, any time we cast a sorcery, we get many copies of some instant that lets us generate more copies of the token we use in the hyperstage. Generally speaking, anything that can target the token, the intermediate creature we use to make copies of it, or Goblin Dark-Dwellers itself is fair game here. That's also why we can't use sorceries in the hyperstage, or have any sorceries with cmc 3 or less in the deck: otherwise we could cast a sorcery with Dark-Dwellers, which triggers volute, which lets us replay Dark Dwellers, which lets us trigger Volute again, etc. The only sorcery we can set this off with is by flashbacking Acorn Harvest, which costs 3 life.

Then, to delete all our hyperstage creature (Moggcatcher at the moment), we use the same instant to kill Child of Alara. That destroys our Orrery, our Omniscience, all our enchantment copies (though we keep our mana), our hyperstage and stage tokens, and is also the only way so far to kill Centaur Safeguard. The death trigger gains 3 life, but we also get a trigger to search our library for a card with the same name and put it directly onto the battlefield, via Verdant Succession. We can't respond to that trigger, because we can't save Vedalken Orrery, but we run two copies, so we just alternate which one is on the battlefield and in the library/graveyard/exile. This also means that we can't have access to 3 colors of mana at this point: even if it kills Mimic Vats, we could keep playing Explosives for 3 and cycling between the Safeguards to gain life and go infinite. We have a workaround to allow 3 colors, but it's janky and hard to make the gigastage safe afterwards. Right now, we use Smite the Monstrous, so past this point any creature whose power can be boosted to 4 or greater is unsafe.


The gigastage keys off black mana, and uses sorceries as its effects. We cast the sorceries with Spellweaver Helix, which can imprint multiple cards thanks to Panharmonicon. We flashback Spider Spawning while the second copy is imprinted on a Helix, and get to cast both a spell that gains life (Archangel's Light, for instance), and Worldfire as the reset. As a general rule, the last two of the big stages (or only two) will either start with a mana-based stage and finish with a life-based stage, using Worldpurge to clear all permanents and empty the mana so you can't save it, or start with a life-based stage and finish with a mana-based stage, using Worldfire to clear all permanents and hard set your life total to 1.

Either way, the thing that sets these cards apart is they remove any permanents, not just nonlands. So we use Worldpurge to get lands that gain life (like Radiant Fountain) off the battlefield, then we replay them to gain the life that we paid into the gigastage back. With Worldfire, we just use a land that produces black mana: Bayou usually so that it can also help cast Eureka at the very beginning.

The rest of the cards in the mega- and gigastage are there to make sure we can get our other permanents back after we blow up the world. That's extra important for Worldfire because we can't save them in our hand.


After that, we spend the rest of the combo trying to fit as many stages in as possible. While there are some exceptions (if it's not itself a stage creature, just an enabler that helps connect one stage creature to another, it can have a cmc we can reach with Engineered Explosives), those stages have to be built on creatures with converted mana costs of 3, 4, or 5+, with 4 not being allowed once we allow a fourth color of mana. They can't be white (or Battle Cry untaps them) or green (Verdant Succession) makes them too easy to get back out). They must have power less than 4. They can't be humans, goblins, have flying, or use +1/+1 counters. And each stage adds a new restriction.

For instance, we start with a Zombies stage to generate black mana. Then we use a red mana stage that also targets creatures with toughness 2 or less. Then we use a merfolk stage. Then we use Old Man of the Sea to rule out creatures with power 2 or less, so all remaining stage creatures must have power exactly equal to 3. Then we wrap up with Xathrid Gorgon.


In general, we keep adding stages until we either run out of deck slots or run out of stage creatures (different decks hit different limits first). After that, we try to add layers that generate lots of that last stage creature. Past decks have used Reality Spasm, which can untap the Mimic Vat with the creature imprinted, but right now we use Tribal Unity, which pumps all our Old Man of the Sea copies until they're big enough to target Xathrid Gorgon, and is more efficient.

Depending on how many slots are still open, we either end with World at War or Words of Wisdom. If we have more space, we find some way to turn a combat step into a lot of Tribal Unity (for instance, one white mana lets us play Sphinx's Revelation for many copies with Thousand-Year Storm, each of which can draw Tribal Unity, and we get lots of white mana each attack step with many copies of Druid's Repository), then cast World at War as many times as possible to get lots of combat steps. Because World at War won't do anything after we leave our first postcombat main phase, we can't use those extra combat steps to feed more World at Wars, so that's where we run out of steam.

If we don't have the space for all that, we use Consecrated Sphinx to turn our opponent drawing one card into us drawing many cards two at a time, each of which can let us recast Tribal Unity. That's why our only card draw spells make each player draw a card: there's no way to refill our opponent's deck, so we can only do that at most 53 times. Depending on how easy it is to set everything up, we either use Reforge the Soul, which we can usually cast 4-5 times before it would kill our opponent without dealing any damage, or Words of Wisdom, which is much more efficient. I'm not sure exactly how the ultracombo's top layers worked, but I think it was roughly similar.

Deedlit, it might be a good idea to add links to this comment, as well as yours on the first stage and lijil's on the hyperstage, and any others you know about that I don't, to the OP as a starting point.




In other news, I've found at least one cut we can make to the deck, as long as we keep Academy Rector around: Skull of Orm! We can make token copies of green enchantments just fine (kill them, make token with a vat, Mirror them to the library, search them out with Succession), which is enough for the bottom stage: we can just copy Bloodbond March. Then, we can kill Academy Rector to make copies of nongreen enchantments later on.

Though actually... how are we getting Rector back now? Ugh. That setup may not work anymore.