The State of TO

[Heliomance]

3.5 veteran here, now moved to 5e. I was wondering - what's the state of Theoretical Optimisation in 5e? What notable builds and exploits have been found? Highest damage possible? Breaking the action economy? Is there a new Pun-pun? I realise that 5e is by all accounts far better balanced than 3.5 ever was, but I'm sure people have managed to break it anyway.

[da newt]

As you stated, it seems 5 is much more balanced and harder to 'break.' For a good idea of what can be done, maybe the below is a good place to start for a plethora of optimized builds and such.


https://forums.giantitp.com/showthread.php?583957-An-Eclectic-Collection-of-Fun-and-Effective-Builds

[AvatarVecna]

3.5 veteran here, now moved to 5e. I was wondering - what's the state of Theoretical Optimisation in 5e? What notable builds and exploits have been found? Highest damage possible? Breaking the action economy? Is there a new Pun-pun? I realise that 5e is by all accounts far better balanced than 3.5 ever was, but I'm sure people have managed to break it anyway.

Wish-->Simulacrum shenanigans, Shapechange shenanigans, coffeelocks...I think that's all the egregious stuff?

[Heliomance]

As you stated, it seems 5 is much more balanced and harder to 'break.' For a good idea of what can be done, maybe the below is a good place to start for a plethora of optimized builds and such.


https://forums.giantitp.com/showthread.php?583957-An-Eclectic-Collection-of-Fun-and-Effective-Builds

That looks decidedly practical optimisation, rather than the sheer unholy rules abuse of proper TO :P


Wish-->Simulacrum shenanigans, Shapechange shenanigans, coffeelocks...I think that's all the egregious stuff?

...coffeelocks?

[Bobthewizard]

A few things that are broken in 5e. Google should help you find more details

Nuclear wizard
Coffeelock
Simulacrum chains
Malleable illusions with at-will silent image (oops, this should say illusory reality)
Conjure woodland beings for pixies

To a lesser degree
Shepherd druid with conjure animals
Chronurgy's arcane abeyance

[Xihirli]

The thing about all of these, though, is that they don't come online until at least level 10.

...Except for Shepherd Druid Conjure Animals.

The big low level imbalance is the Moon Druid, who can, twice per short rest, become a level 5 barbarian starting at level 2.

[Waazraath]

In line with the posters above: it's almost non-existant, the few overpowered options are already mentioned and with wish/simulacrum being the biggest offender.

[kazaryu]

That looks decidedly practical optimisation, rather than the sheer unholy rules abuse of proper TO :P



...coffeelocks?

coffee locks is a warlock/sorcerer combo. take the 'aspect of the moon' invocation so you don't need sleep. then over several short rests use sorcerers flexible casting feature convert warlock spell slots into sorcery points, and then into spell slots. over and over again. then just...don't take a long rest so they don't get clears. so like, take a 3/2 split (minimum level the combo comes online). every hour you can generate 4 sorcery points, which can be converted into 2 level 1 spell slots (you can only have 2 points at a time, so you can't use all 4 at once). meaning that over a single 24 hr resting period you can generate 48 lvl 1 spell slots if you spend the full 24 hours short resting (technically only 46, as the actual conversation takes a few seconds each time. but you get the point). as you level the potency of spell slots you can create also levels. with another sorcerer level you can now control 3 sorcery points, and therefore make 2nd level spells slots, etc,

[Chaos Jackal]

There's nothing at the level of Pun-Pun, however that doesn't mean there's nothing at all.

Simulacrum chains and coffeelocks (sorcerer/warlock multiclass with an invocation that allows them to remain sleepless abusing the warlock's short rest slot recovery by repeatedly short resting, rather than long resting, in order to keep converting their short rest slots to sorcery points) are commonly brought up. The nuclear wizard is another case, although nowhere near as egregious. Glyph of warding can allow for some ridiculous interactions that break both action economy and concentration. Some readings of shapechange might allow you to maintain buffs, though that's not very widely acceptable. Animate dead taken to the extreme allows for an absolutely huge undead army, especially when partnered by the Necromancy subclass or when a warlock manages to use it (short rest slots and animate dead aren't nice to the world). As a matter of fact, the Necromancy subclass itself can theoretically reach infinite hit points.

And there's more. 5e is by no means well-balanced or perfectly written. "Not 3.5" isn't exactly that high of a bar to overcome in terms of brokenness. Things are far tamer, yes, but there are still things that slipped through the cracks.

[Willie the Duck]

Coffeelocks definitely hold a lot of attention because of the theoretical limits of their abilities. Whether (and at which levels) they really outperform others is much more debated. At 5th level (when they come online) having neigh infinite 1st level spells is kinda 'nice, but have you seen what the 5th level normal caster can do with their 3rd level spells?' and at the highest levels people rarely burn through all their spells per day anyways. In the messy middle, though, they can be impressive (albeit potentially annoying to the rest of the group, as you are tracking time worse than an old-school gamer tracking time on their torches).

Wish-Simulacrum was probably the first exploit found. And, almost as quickly, the first exploit banned by almost every group everywhere. This one really seemed to hit the 'too obvious to be impressive' chord with a lot of people, and I haven't really seen it discussed in many actual games or even in most optimization discussions.

Nuclear wizard takes advantage of the specific wording of Magic Missile in the edition effectively stating that you roll damage once for all darts (and thus many damage-add effects apply to them all). It gets a little more play in TO discussions, but also is one of those things where plenty of gaming groups will nix the foundation upon which it is based (i.e. 'I don't care about your parsing of the language on this, no one has ever picked magic missile as a spell unless they wanted to roll a fist full of D4s, so that's how we're going to run it.').

Sorcadins (sorcerer-paladins), sorlocks, pallocks/hexadins (paladin-warlock, especially hexblade warlocks), and pretty much any portmanteau of the charisma-focused classes are well-explored territory. Coffeelock is a prime example of a sorlock. The rest are usually some form of a gish build. Most of these gishes aren't really OP in any way (although they can have some amazing nova potential, so if your DM typically runs only one combat per day...), so much as 'gish actually done well.'

Moon Druid, Conjure woodland beings for pixies, Shapechange spells, and all the summon effects are part of a broad rules convergence which basically boils down to: 'replace your character with a monster entry'-effects are very powerful*, and 'add a monster manual entry (and their additional actions per turn) to your total action economy'-effects are also very powerful**.
*especially since, if you are reduced to 0 hp in that form, you simply turn back into your normal form and take any leftover damage, meaning you have less disincentive not to turn into a glass cannon creature
**especially since other rules have been changed such that having multiple weak opponents gang up on you are more of a threat in this edition than the past couple.

Other than that, it is mostly specific subclasses being slightly over-tuned. Hexblade in Xanathar's Guide raised the bar for warlocks and a couple of different ones in Tasha's Cauldron did the same for Clerics, Sorcerers, and so on.


And there's more. 5e is by no means well-balanced or perfectly written. "Not 3.5" isn't exactly that high of a bar to overcome in terms of brokenness. Things are far tamer, yes, but there are still things that slipped through the cracks.

Y'know, I think what has really changed is the culture. Sure, optimization is still a game many play. However, I think people are just a little less impressed with it than they once were. Maybe the same thing has happened with 3e since the early 2000s.

[Rukelnikov]

That looks decidedly practical optimisation, rather than the sheer unholy rules abuse of proper TO :P

The only ones I remember that are close to proper TO are building computers (or sonars, or sometimes other things) with Magic Mouth, and Chain Simulacrum (infinite copies of a high level wizard)

[Heliomance]

A few things that are broken in 5e. Google should help you find more details

Nuclear wizard
Coffeelock
Simulacrum chains
Malleable illusions with at-will silent image
Conjure woodland beings for pixies

To a lesser degree
Shepherd druid with conjure animals
Chronurgy's arcane abeyance

...what's so good about malleable illusion with silent image? Malleable illusion looks great with Creation, or Simulacrum, but I'm not remotely seeing how a modifiable Silent Image is in the realms of TO.

[Eldariel]

Some (https://www.reddit.com/r/DnD/comments/hu2pps/i_accidentaly_created_punpun_in_5e_or_how_to/) people (https://www.reddit.com/r/DnD/comments/hu2pps/i_accidentaly_created_punpun_in_5e_or_how_to/) have (https://www.reddit.com/r/powergamermunchkin/comments/e1auza/pun_pun_5e/) tried (https://www.reddit.com/r/powergamermunchkin/comments/i2z553/how_to_become_a_god/) to do some stuff. Ultimately, Simulacrum chain lets you get an arbitrary number of level 20 PCs, which is more or less the apex of what you can do. And Shapechange and True Polymorph give you arbitrary anything. Basically, it's 9th level spells to really break stuff in this game (though Magic jar and Planar Binding do let you break stuff in a 3esque fashion earlier but 5e monsters are generally less impressive than their 3e counterparts so the results aren't as devastating), though lower level options can do a fair bit but most of it is, while absurdly powerful for normal games, still playable. This edition has fewer infinite comboes and such and even nova damage/round tends to cap in thousands.

[Unoriginal]

3.5 veteran here, now moved to 5e. I was wondering - what's the state of Theoretical Optimisation in 5e?

Well, there was a lot of white room theorycrafting for a while, and it still exists, but from what I've seen it has lost a lot of steam and it continues to lose it.



Y'know, I think what has really changed is the culture. Sure, optimization is still a game many play. However, I think people are just a little less impressed with it than they once were.

Probably because of the aforementioned white room theorycrafting. There is only so many way to say "you've cherry-picked the circumstances or the datas you used and excluded the rest to get those results, so it isn't as impressive as you're trying to convince us it is".

[Amechra]

Part of the reason that 3.x had (and probably has, honestly) a thriving TO community is that 3.x had a TON of books with a TON of player and player-adjacent options, and it didn't rely heavily on DM adjudication (because there was a poorly-thought-through rule for everything in 3.x).

5e has a miniscule fraction of those options, and relies heavily on "DM May I?" for a lot of its rules. TO stops being fun when it goes from "here's what happens if you do what the rules say without common sense getting in the way" and turns into "well, if we allow this houserule..."

[KOLE]

The worst offenders are Moon Druids at early levels, from there it mostly takes 9th level spells for real shenanigans.

Coffeelocks were the talk for a while- but it really takes some DM handwaving to be really possible. Sorcadin and Sorlock multiclasses tend to rule the meta for actual “Munchkin” Optimization.

Some subclasses stand out as strong for their level or compared to other subclasses- Swashbuckler, Arcane Trickster are hard to best for rogues, Hexblade is a tough contender for Warlocks, the new Twilight Cleric is pretty wild, and I personally feel Bladesinger is overtuned- but none of these things outright break the game, just nudge balance slightly one way or the other.

[JNAProductions]

Coffeelocks aren't an issue, in my opinion, till level 6 at the earliest. If you go Sorc 5/Warlock 1 (exact level progression to taste) and get a long break... You've got unlimited 3rd level slots when everyone else is rocking 3rds as well.

For reference, if you have two weeks of travel, which is 336 hours, and assume you can manage an average of a short rest every one and a half hours, that's 224 SP, or 44 3rd level slots.

It gets worse as you increase in level, capping out at level 10 (or, more likely, 12-if you take three levels of Warlock). At that point, even if you're a Sorc 9/Warlock 3 and therefore don't have any 6th level slots, you've got unlimited 5ths, and regen 4 SP worth of spells a rest. That same two weeks of travel now gets you 128 5th level slots.

I do agree that unlimited 1st and 2nd level slots aren't a huge deal, even at levels where they're at their most impactful. But unlimited 3rds when other people might have 4ths, and unlimited 5ths at any level... Busted.

[Gtdead]

Aside from Simulacrum chains which are legal by RAW but no one cares about them, most other things have an asterisk because the spells allow direct DM intervention that kind of dissuades people from thinking about it too much. Example:

Conjure Animals is considered one of the strongest dpr spells in the game.

If you decide to summon wolves, which is one of the more reliable ways to do damage, the answer is that the DM will chose not to give you wolves.
Then you decide to combine Giant Poisonous Snakes with some Obscurement Effect, which is even worse. Again the same answer, DM won't give you snakes.
Ok, so I will upcast Conjure Animals to level 9 and summon Owls and then have them go kamikaze by flying up high in the sky and going prone, falling down on enemies and dealing 320d6 falling damage. Guess what, now DM will ban owls too.

It's not that there aren't cool things to find. It's that it's not worth discussing them, because guess what.. Feats are optional, DM chooses what animal is available, DM rules that there is no Snow in this planet so you can't create a simulacrum etc. You could find some cool stuff in the powergamermunchkin subreddit, although it's not super active.

[Amechra]

I think the fact that pretty much everyone in this thread is talking about CO and not TO speaks volumes.

To put it in videogame terms, CO is beating Super Mario World by being super good at the game. TO is beating Super Mario World by manipulating the game state so that you warp directly to the credits (https://www.youtube.com/watch?v=RU4LfM-Y4kg).

[Unoriginal]

Part of the reason that 3.x had (and probably has, honestly) a thriving TO community is that 3.x had a TON of books with a TON of player and player-adjacent options, and it didn't rely heavily on DM adjudication (because there was a poorly-thought-through rule for everything in 3.x).

5e has a miniscule fraction of those options, and relies heavily on "DM May I?" for a lot of its rules. TO stops being fun when it goes from "here's what happens if you do what the rules say without common sense getting in the way" and turns into "well, if we allow this houserule..."


I think the fact that pretty much everyone in this thread is talking about CO and not TO speaks volumes.

To put it in videogame terms, CO is beating Super Mario World by being super good at the game. TO is beating Super Mario World by manipulating the game state so that you warp directly to the credits (https://www.youtube.com/watch?v=RU4LfM-Y4kg).

Well, unlike 3.X, 5e does not even pretend that anything can be done without DM-buy-in.

[Gtdead]

I think the fact that pretty much everyone in this thread is talking about CO and not TO speaks volumes.
.

I think this has more to do with the forum dwellers' general inclinations. I'm fairly sure that not many people here would have the patient to argue how Level 20 Arcana Cleric can use Divine Intervention to cast Shapechange inside an Antimagic Field, or Nystul Magic Aura's shenanigans, or argue if Genie Warlocks can choose a Ring of Three Wishes as their Genie Vessel.

[Amechra]

Well, unlike 3.X, 5e does not even pretend that anything can be done without DM-buy-in.

That is quite literally my point, yes.

[Chaos Jackal]

I think this has more to do with the forum dwellers' general inclinations. I'm fairly sure that not many people here would have the patient to argue how Level 20 Arcana Cleric can use Divine Intervention to cast Shapechange inside an Antimagic Field, or Nystul Magic Aura's shenanigans, or argue if Genie Warlocks can choose a Ring of Three Wishes as their Genie Vessel.

Neither Divine Intervention nor the Genie's vessel are TO though.

The former, regardless of what cleric is using it, is fully subject to the DM's whims, by very clear RAW; while everything is, obviously, subject to DM approval (hence why nobody does simulacrum chains), in this case the text itself says that the DM chooses. You can't do anything based on the theory, because the theory doesn't allow you to do anything even in a white room. You could as easily assume that your DM is gonna give you a magic item at lv1 that sets all your stats to 19, or that you can use wish to instantly kill the BBEG wherever they might be at that point. Hell, wish has more theorycrafting potential, because at least you get to choose some effect, even though the DM is practically obliged to screw you over the more you ask with said effect. Divine Intervention doesn't even do that.

You could argue maybe that Genie warlock picking a magical item as their vessel is possible, but in the same manner you could argue that picking an amulet as your spellcasting focus allows you to choose an amulet of health. I guess there is some discussion to be had (whether "ring of three wishes" is a distinct item and you can't pick it unless the text specifically mentions it, rather than just a "ring with comparment"), but ultimately I believe that there's no real basis for anything here. The theory, again, allows nothing. But unlike Divine Intervention, that's a more personal view; I might be swayed.

On the other hand, a necromancer getting an arbitrary number of hit points through aid or heroes' feast and Inured to Undeath? Yes, that's a case where the theory clearly allows something, even if the DM might not. So does using glyph of warding to cast every buff in the game on yourself with no action and no concentration; again, possible in theory, even if the DM might not allow that.

[sithlordnergal]

3.5 veteran here, now moved to 5e. I was wondering - what's the state of Theoretical Optimisation in 5e? What notable builds and exploits have been found? Highest damage possible? Breaking the action economy? Is there a new Pun-pun? I realise that 5e is by all accounts far better balanced than 3.5 ever was, but I'm sure people have managed to break it anyway.

So, I started in 3.5, and while there are a few things that break 5e, there's nothing on the level of 3.5 broken. A few fun examples:


Order of Scribes/Tempest Cleric: Order of Scribes lets you change the damage type of a spell you cast to the damage type of a different spell you know of the same level. I.E. if you know Fireball, Lightning Bolt, and Erupting Earth, you can have Fireball deal Fire, Lightning, or Bludgeoning damage. Tempest Cleric has a Channel Divinity that lets you maximize Lightning and Thunder damage once per short rest. So instead of rolling 8d6, you could have Fireball deal 48 Lightning damage instead. It also has synergy with the Crusher feat, since you have access to Bludgeoning damage at almost every spell level.


Moon Druid: Probably one of the strongest Druids in the game. Wild Shape is crazy strong at all levels, and even when it falls off its still a viable option. Druids retain their full caster abilities, and have an excellent spell list to fall back on.


Hexblade Warlocks: Best dip class in the game for any class that uses Charisma, bar none. Doesn't matter if you're a Bard, Sorcerer, or Paladin, 1 or 2 levels of Hexblade Warlock will make that class better. Heck, with just 1 level of Warlock you gain:


Medium Armor and Shield proficiency
Proficiency with all Martial Weapons
Access to the strongest cantrip in the game
The ability to use Charisma for your weapon attacks
One 1st level spell slot that recharges on a Short Rest
A Curse ability that recharges on a Short Rest that gives you bonus damage, increases crit range of all attacks to 19-20, and heals you if the cursed target dies
2 spells known, and access to 1st level Warlock spells, a lot of which are really good
Access to the Shield spell



Meanwhile a 2nd level in Warlock lets you have two 1st level spells that recharge on a Short Rest and access to 2 Warlock Invocations. These Invocations can do everything from letting you add your Charisma Modifier to your Eldritch Blast damage, effectively turning it into a Heavy Crossbow that deals Force damage, give you 120ft Darkvision that lets you see through Magical Darkness, or let you cast spells like False Life, Mage Armor, or Disguise Self at will.

Another bonus to Warlock is that your short rest spells stack with regular spell slots. So a 4th level Bard with 2 levels of Warlock will have six 1st level spell Slots that can be used to cast any spells they like.


Fairy/Rune Knight: You can go from a Small creature to Huge at level 5, you can become Gargantuan at level 17 or 18, I forget which.

[Greywander]

...what's so good about malleable illusion with silent image? Malleable illusion looks great with Creation, or Simulacrum, but I'm not remotely seeing how a modifiable Silent Image is in the realms of TO.
Since no one answered this, I will. Misty Visions, a warlock invocation (which can now be obtained via a feat, so you don't need to multiclass) allows you to cast Silent Image at will. Pretty cool on its own. But on an Illusionist wizard, it becomes fairly busted. Now, Malleable Illusions isn't the feature I would have called out, but rather Illusory Reality. You can, at will, make a 15x15x15 foot illusion, then make that illusion real, and, with Malleable Illusions on top of that you can change the real illusion once. For example, you can create an illusion of a bridge, make it real, cross it, then change the illusion to collapse the bridge behind you (you could also just drop concentration on Silent Image, but there are other shenanigans you could get up to by changing the illusion to something else). Misty Visions + Illusory Reality is what really breaks things. Though I'm sure this isn't anything impressive by 3.x standards.

[Chaos Jackal]

Since no one answered this, I will. Misty Visions, a warlock invocation (which can now be obtained via a feat, so you don't need to multiclass) allows you to cast Silent Image at will. Pretty cool on its own. But on an Illusionist wizard, it becomes fairly busted. Now, Malleable Illusions isn't the feature I would have called out, but rather Illusory Reality. You can, at will, make a 15x15x15 foot illusion, then make that illusion real, and, with Malleable Illusions on top of that you can change the real illusion once. For example, you can create an illusion of a bridge, make it real, cross it, then change the illusion to collapse the bridge behind you (you could also just drop concentration on Silent Image, but there are other shenanigans you could get up to by changing the illusion to something else). Misty Visions + Illusory Reality is what really breaks things. Though I'm sure this isn't anything impressive by 3.x standards.

What actually breaks things is combining Malleable Illusions and Illusory Reality with 6th-level major image casts. You've got an army of massive, permanent, concentration-less illusions that you can move around, transform and make real at-will. You can practically do nearly anything.

[PhoenixPhyre]

What actually breaks things is combining Malleable Illusions and Illusory Reality with 6th-level major image casts. You've got an army of massive, permanent, concentration-less illusions that you can move around, transform and make real at-will. You can practically do nearly anything.

Except Malleable Illusions requires an action. Per object. And you can only make inanimate objects part of the illusion real, and only for 1 minute, costing a bonus action each time. And causing it to move is an action per image as well. And none of this can cause damage or impose status effects.

[LudicSavant]

I think the fact that pretty much everyone in this thread is talking about CO and not TO speaks volumes.

To put it in videogame terms, CO is beating Super Mario World by being super good at the game. TO is beating Super Mario World by manipulating the game state so that you warp directly to the credits (https://www.youtube.com/watch?v=RU4LfM-Y4kg).
Yeah, they’re terms from the old 3.5e WotC CharOp boards; practical op would basically refer to stuff that was very conventionally strong, while theoretical op was stuff that was so broken you were hardly even playing D&D anymore (a la Pun Pun).

TO generally describes a level of shenanigans that extends well beyond things like sky-high damage or extremely versatile toolkits.


As you stated, it seems 5 is much more balanced and harder to 'break.' For a good idea of what can be done, maybe the below is a good place to start for a plethora of optimized builds and such.


https://forums.giantitp.com/showthread.php?583957-An-Eclectic-Collection-of-Fun-and-Effective-Builds

While there’s certainly a lot of helpful stuff there, I should note that said thread is more about so-called practical op (stuff you might use at a real table with fellow optimizers/tactical players) than so-called theoretical op (stuff like Pun Pun that even optimizers wouldn’t want to actually play). As such the thread won’t include things like Simulacrum loops, glyph demiplanes, body-swapping shenanigans, slot generator exploits, or the like.

In other words, as strong as something like a Nuclear Wizard or Jorasco Physician is, it’s still not nearly as strong as some of the things I *could* theoretically do, but wouldn’t in a real game so don’t bother talking about much. Simulacrum loops are probably the best-known example.

There may not be as many of those things as there were in 3.5, but they’re still definitely around. There’s just not really a dedicated place they’re all compiled like the “campaign smashers” list on the old 3.5e CharOp boards (that I know of).

[Greywander]

One thing I just remembered in terms of actually broken stuff is if you can push your stats to 30 (possible using the stat-boosting tomes, or using the optional rule for spending epic boons on ASIs). Between Unarmored Defense and Aura of Protection, you can double up on stat bonuses to your defenses. As long as your stats are <= 20, you tend to stay in the normally expected range for those bonuses, but when you can push your stats to 30 you explode past the normal limits. With Unarmored Defense you would have a naked AC of 30, and the barbarian can add a shield on top of that going up to 35 AC with a shield +3. With Aura of Protection and all stats at 30, you're getting +20 to all your saves, +26 for the ones you're proficient in.

These numbers might not sound like much to someone coming from 3.x, but let me put things into perspective. Let's assume we have a CR 30 monster (the highest published CR at the moment) with 30's in their stats and a proficiency bonus of +9. They get a +19 to their attack rolls, which means they only have a 25% chance to hit AC 35. Likewise, their save DC will be 27, which means we only have a 35% chance to fail a save if we're not proficient, and a 0% chance to fail a save in which we are proficient. That's right, with a +26 to saves, it is impossible to fail a DC 27 save. And that's before we stack on any other bonuses to AC or saves. Like the Staff of Power, which gives +2 to both, or the Ring of Protection, for +1 to both. There's also spells like Shield of Faith and Bless. Attack rolls still hit on a 20 (5% chance), regardless of AC, so a Cloak of Displacement/Invisibility will drop those odds to 1 in 400 (0.25% chance).

Obviously, this isn't doable until 20th level or unless you have easy access to stat-boosting tomes. But it's still kind of ridiculous how broken you can become while still being within the rules.

[Kane0]

What I don't often see mentioned are Epic Boons, which there are no real limits on. Even more so than magic items, since those tend to have attunement.

[Bobthewizard]

...what's so good about malleable illusion with silent image? Malleable illusion looks great with Creation, or Simulacrum, but I'm not remotely seeing how a modifiable Silent Image is in the realms of TO.

Sorry. I was at work and typing off the top of my head. It's illusory realty with silent image. Make a 15' cube of something real every round.

[Unoriginal]

What I don't often see mentioned are Epic Boons, which there are no real limits on. Even more so than magic items, since those tend to have attunement.

Because getting them is not a choice a player can make, unless we're talking about post-lvl-20 stuff.

[Psyren]

That looks decidedly practical optimisation, rather than the sheer unholy rules abuse of proper TO :P

By your definition then, there is extremely little of what you might term "proper TO" in 5e.

Several of the TO pillars have been heavily eroded:

Skill checks - no longer really exist, the GM decides if an ability check is even warranted rather than the system attempting an exhaustive list of use cases, AND they set the DC, and the most you can do is try to justify your skill proficiency being used there.

Buff stacking - gutted by the new Concentration mechanic, and even for those spells not subject to it, the use of specific durations rather than time/level.

Magic Items - firmly in the DM's purview, even crafting now. Stacking items is much harder due to attunement.

General Spell Power - greatly nerfed. Several pillars of TO like Planar Binding, Polymorph, or Gate have been heavily altered. Concentration applies here too, e.g. to things like summons or battlefield control effects, preventing you from blanketing an encounter space.

[LudicSavant]

Buff stacking - gutted by the new Concentration mechanic, and even for those spells not subject to it, the use of specific durations rather than time/level.

There's ways for a character to get multiple Concentration effects going at the same time (such as with glyph demiplanes or simulacrum loops). This is the kind of silly stuff that I would consider in the realms of TO, rather than things I would do in a real game.


General Spell Power - greatly nerfed. Several pillars of TO like Planar Binding, Polymorph, or Gate have been heavily altered. Concentration applies here too, e.g. to things like summons or battlefield control effects, preventing you from blanketing an encounter space.

Gate has a pretty bonkers new use to it in 5e. The reason a lot of people don't notice it is because they keep thinking of Gate as a tool for summoning minions, and thus miss the forest for the trees.

What use is that? Forget about calling allies, you set up a total deathtrap somewhere, speak the name of someone you don't like, and they get Gated right into it. No save.

[Gtdead]

Neither Divine Intervention nor the Genie's vessel are TO though.

The former, regardless of what cleric is using it, is fully subject to the DM's whims, by very clear RAW; while everything is, obviously, subject to DM approval (hence why nobody does simulacrum chains), in this case the text itself says that the DM chooses. You can't do anything based on the theory, because the theory doesn't allow you to do anything even in a white room. You could as easily assume that your DM is gonna give you a magic item at lv1 that sets all your stats to 19, or that you can use wish to instantly kill the BBEG wherever they might be at that point. Hell, wish has more theorycrafting potential, because at least you get to choose some effect, even though the DM is practically obliged to screw you over the more you ask with said effect. Divine Intervention doesn't even do that.

Divine Intervention text offers an appropriate use for the feature that the user can literally ask for. The DM should not ban it because by RAW, it's appropriate.

I don't know all the renditions of Pun-Pun, but Sharrukh is setting locked, so it's subject to DM's whim as well. I think your level of "prejudice" against DI is similar to this technicality.



You could argue maybe that Genie warlock picking a magical item as their vessel is possible, but in the same manner you could argue that picking an amulet as your spellcasting focus allows you to choose an amulet of health. I guess there is some discussion to be had (whether "ring of three wishes" is a distinct item and you can't pick it unless the text specifically mentions it, rather than just a "ring with comparment"), but ultimately I believe that there's no real basis for anything here. The theory, again, allows nothing. But unlike Divine Intervention, that's a more personal view; I might be swayed.

The rule specifically allows the player to choose a magical tiny object. The wish ring is even appropriate fluff wise, because it grants you a measure of your patron's power that is thematic. On what grounds exactly is the DM going to say no? On power level considerations, which is the reason why things get banned.

So theorize away with the knowledge that the DM is never going to let you do that.



On the other hand, a necromancer getting an arbitrary number of hit points through aid or heroes' feast and Inured to Undeath? Yes, that's a case where the theory clearly allows something, even if the DM might not. So does using glyph of warding to cast every buff in the game on yourself with no action and no concentration; again, possible in theory, even if the DM might not allow that.

Yep, infinite Create Magen too. Necromancer is fun.


Edit: Fixed a misunderstanding.

[Greywander]

On the other hand, a necromancer getting an arbitrary number of hit points through aid or heroes' feast and Inured to Undeath? Yes, that's a case where the theory clearly allows something, even if the DM might not.
I disagree. I understand that it could be interpreted that way, but I feel like that's a bad faith interpretation looking to exploit the rules, not unlike the peasant rail gun. I've seen a number of badly written rules, and I'm not afraid to call those out and implement houserules to patch them up, but I don't believe this is one of those cases. It's not just a case of the rules saying or even implying one thing and a DM simply not allowing it, I don't think the rule actually says what you're saying it does. But perhaps this all comes down to an argument of semantics. My entire reason for disagreeing with you here is this:

That a reduction in max HP is not the same as the expiration of an increase in max HP.

A "reduction" is an effect, not the lack of an effect. If you don't agree with that statement, then you're probably not going to agree with my interpretation of Inured to Undeath. Inured to Undeath prevents effects that would reduce your max HP from being applied to you, but it has no interaction with effects that increase your max HP, like Aid. When Aid expires, that's not a reduction in max HP, it's simply the loss of an effect that increases max HP.

TL;DR, infinite magen is fine, but Aid stacking for infinite HP is not.

[kazaryu]

Coffeelocks aren't an issue, in my opinion, till level 6 at the earliest. If you go Sorc 5/Warlock 1 (exact level progression to taste) and get a long break... You've got unlimited 3rd level slots when everyone else is rocking 3rds as well.

For reference, if you have two weeks of travel, which is 336 hours, and assume you can manage an average of a short rest every one and a half hours, that's 224 SP, or 44 3rd level slots.

It gets worse as you increase in level, capping out at level 10 (or, more likely, 12-if you take three levels of Warlock). At that point, even if you're a Sorc 9/Warlock 3 and therefore don't have any 6th level slots, you've got unlimited 5ths, and regen 4 SP worth of spells a rest. That same two weeks of travel now gets you 128 5th level slots.

I do agree that unlimited 1st and 2nd level slots aren't a huge deal, even at levels where they're at their most impactful. But unlimited 3rds when other people might have 4ths, and unlimited 5ths at any level... Busted.

you have to be at least lvl 3 warlock in order to fully coffeelock, since you need the aspect of the moon invocation, which requires the tome pact boon.

otherwise you *have* to long rest in order to not die. (that is, unless your DM handwaves you being able to sleep without long resting). this means that, for you, the earliest it can be a problem is lvl 8. (sorc5/warlock3).

but, that aside: infinite 1st level spells is still a problem, thats infinite healing words/cure wounds. infinite shield/absorb elements...there are probably others im not thinking of atm, but really i don't think i need them.

[LudicSavant]

*snip*My entire reason for disagreeing with you here is this:

That a reduction in max HP is not the same as the expiration of an increase in max HP.

A "reduction" is an effect, not the lack of an effect. If you don't agree with that statement, then you're probably not going to agree with my interpretation of Inured to Undeath. Inured to Undeath prevents effects that would reduce your max HP from being applied to you, but it has no interaction with effects that increase your max HP, like Aid. When Aid expires, that's not a reduction in max HP, it's simply the loss of an effect that increases max HP.

TL;DR, infinite magen is fine, but Aid stacking for infinite HP is not.

So, I think that this doesn't actually lead to infinite HP, but not due to the semantic argument you present. The rules (for the most part) are written in common English, and in common English, the word "reduction" can refer to any time that a bigger number becomes a smaller number, period.

So why do I think it still doesn't lead to infinite HP? Because the cause of those extra HP is still Aid, even if the duration ended, and effects from the same named source don't stack. Doing some quick checking around, this also seems to be the consensus on RPG Stackexchange (https://rpg.stackexchange.com/questions/113636/does-the-aid-spell-permanently-increase-hp-for-necromancy-wizards-with-the-inure).

Mind, at my own table I haven't allowed Aid to get the permanent bonus even the one time, but that has everything to do with my actions as a DM, not with what is literally written in the book.

[kazaryu]

So, I think that this doesn't actually lead to infinite HP, but not due to the semantic argument you present. The rules (for the most part) are written in common English, and in common English, the word "reduction" can refer to any time that a bigger number becomes a smaller number, period.
.

but if you apply that logic, then you're not looking at RaW you're looking at RAI. as common English is entirely dependent on context and interpretation. just because a word *can* mean a few different things, doesn't mean that it always does mean those things. context is whats important, and its fairly clear given the context of the rules that 'an effect that increases max HP ending' is not the 'reduction of max hp' that the necromancer feature refers to.

in short: if you're going to use a 'common english argument' then you have to interpret rules based on how you *actually* understand 'common english'. you can't apply an extra level of specificity to the rules that you wouldn't apply in a normal conversation.

[Greywander]

Exactly. If you earn a bonus one year, but not the next, you wouldn't call that a reduction in pay. Your wage/salary hasn't changed, you've just lost the extra bonus.

[LudicSavant]

Exactly. If you earn a bonus one year, but not the next, you wouldn't call that a reduction in pay. Your wage/salary hasn't changed, you've just lost the extra bonus.

It would be said that you got an increase to your annual earnings, then a reduction to your annual earnings.

The operative issue in your example isn't the words "increase" and "reduction," it's whether the bonus was considered part of your wage/salary to begin with.

[Chaos Jackal]

Divine Intervention text offers an appropriate use for the feature that the user can literally ask for. The DM should not ban it because by RAW, it's appropriate.

I don't know all the renditions of Pun-Pun, but Sharrukh is setting locked, so it's subject to DM's whim as well. I think your level of "prejudice" against DI is similar to this technicality.
I disagree on that premise; when theorycrafting, you use various sources, irrelevant to the setting at hand. A DM could ban kobolds as well. This falls into the general "everything is ultimately subject to the DM's whim". Your sarrukh example is similar to using Golgari Agent to get animate dead on a warlock; theoretically possible... as long as the DM doesn't leave Ravnica out.

But for Divine Intervention, even if an example might be given, you still have that pesky "the DM chooses the nature of the intervention"; that's not about what sources are allowed. It's more akin to using conjure woodland beings with the assumption that you'll always get pixies.

In the former case, the theory just stumbles on what the DM might allow in general for the game, which is ultimately the case for everything. In the second case, however, the theory requires DM adjudication on the spot; the very thing you're doing is literally "hope your DM gives you exactly what you want at the moment". It's not even bending the rules or going for a favorable interpretation or doing creative lawyering. It's flat out mother-may-I.

For the record, I see where you're coming from. Arcana cleric in particular being able to request a wizard spell through Divine Intervention is utilizing the wording of the rules to an unusual end. But even if you get that through, you still need to have the DM say yes every time you ask for it... and that is where I feel we're getting out of theorycrafting.

Point taken. You may even be right. I just view the fiat required differently. Call it "prejudice" if you want.


The rule specifically allows the player to choose a magical tiny object. The wish ring is even appropriate fluff wise, because it grants you a measure of your patron's power that is thematic. On what grounds exactly is the DM going to say no? On power level considerations, which is the reason why things get banned.

So theorize away with the knowledge that the DM is never going to let you do that.
Fair enough. Like I said, I can be swayed on that one. Consider me swayed.



Yep, infinite Create Magen too. Necromancer is fun.
Right, forgot about that. Thanks for the reminder. And yeah, necromancers are indeed pretty busted.



I disagree. I understand that it could be interpreted that way, but I feel like that's a bad faith interpretation looking to exploit the rules, not unlike the peasant rail gun. I've seen a number of badly written rules, and I'm not afraid to call those out and implement houserules to patch them up, but I don't believe this is one of those cases. It's not just a case of the rules saying or even implying one thing and a DM simply not allowing it, I don't think the rule actually says what you're saying it does. But perhaps this all comes down to an argument of semantics. My entire reason for disagreeing with you here is this:

That a reduction in max HP is not the same as the expiration of an increase in max HP.

A "reduction" is an effect, not the lack of an effect. If you don't agree with that statement, then you're probably not going to agree with my interpretation of Inured to Undeath. Inured to Undeath prevents effects that would reduce your max HP from being applied to you, but it has no interaction with effects that increase your max HP, like Aid. When Aid expires, that's not a reduction in max HP, it's simply the loss of an effect that increases max HP.

TL;DR, infinite magen is fine, but Aid stacking for infinite HP is not.
Well, an effect that increases your max HP expiring is gonna lead to your max HP being reduced. It is indeed an interpretation looking to exploit the rules, but the rules are vague enough to allow it; exploiting things is also part of theorycrafting. You won't be allowed to do it, but it's there... in theory.

It's also why Ludic's interpretation above might or might not be correct, depending on your ultimate reading; if you rule that you keep the HP from aid thanks to Inured to Undeath, do you still have the aid effect on you in some manner and therefore can't be affected by another aid (thus allowing you to stack different means of increasing max HP, such as aid and heroes' feast, but not the same thing infinite times) or is the aid still considered expired regardless of your HP staying increased, and can therefore be used on you again?

Inured to Undeath hasn't been very thoughtfully worded. I won't argue for any interpretation here, as it's not really within the scope of the thread; for what it's worth, I wouldn't allow aid to stack or magens to be created without reducing your HP, and neither would I ask for either as a player. The fact that an argument can exist and sides of the argument can lead to infinity or unintended boosts (be it infinite magens alone, infinite magens and increased HP, or infinite magens and infinite HP) is what puts Inured to Undeath within the theorycrafting brokenness scope that the OP asked about.

[Hael]

Its pretty hard to break 5e in tier1-2. Moon druids are the worst offenders in tier1, but they’re only maybe a factor of 60-70% better than the corresponding martial (say a barbarian). So its not like they are orders of magnitude better.

The most broken classes in the game are chronurgists, peace and twilight clerics. The former is like only a few percent better than other wizards until lvl 10 and wizards aren’t broken until tier3 in general. The latter 2 are roughtly twice as good as other clerics. So eg if you had 2 life clerics healing the party, you could recreate the effect of a twilight cleric in aggregate. Thats not quite right, but you get the picture.

So that’s broken, but its nowhere near what I could do in pathfinder. Where you could literally have two almost identical characters that do everything almost exactly the same, but one character could do ten or a hundred times more damage per round.

Of course in tier 3-4, spells break the game into Pun pun land, but thats always been the case in DnD, and no one ever plays that way..

[LudicSavant]

Its pretty hard to break 5e in tier1-2.

Hmm. Trying to think of some examples that fit into that category.

I guess I've seen some crazy TO proposals involving Enlarge/Reduce -- there's no limit to the original size of Objects that can be affected, so if you get a big enough object, you could potentially do some weird @#$% that wouldn't be permitted in any real game.

[Chaos Jackal]

Hmm. Trying to think of some examples that fit into that category.

I guess I've seen some crazy TO proposals involving Enlarge/Reduce -- there's no limit to the original size of Objects that can be affected, so if you get a big enough object, you could potentially do some weird @#$% that wouldn't be permitted in any real game.

What do you mean "you can't enlarge the mountain"?

[Unoriginal]

Hmm. Trying to think of some examples that fit into that category.

I guess I've seen some crazy TO nonsense involving Enlarge/Reduce -- there's no limit to the original size of Objects that can be effects, so if you get a big enough object, you can do some weird @#$%.

How could a big enough object break the game, at any tier?

[diplomancer]

you have to be at least lvl 3 warlock in order to fully coffeelock, since you need the aspect of the moon invocation, which requires the tome pact boon.

otherwise you *have* to long rest in order to not die. (that is, unless your DM handwaves you being able to sleep without long resting). this means that, for you, the earliest it can be a problem is lvl 8. (sorc5/warlock3).

but, that aside: infinite 1st level spells is still a problem, thats infinite healing words/cure wounds. infinite shield/absorb elements...there are probably others im not thinking of atm, but really i don't think i need them.

Or be a Warforged who also doesn't need to sleep in the first place.

[kazaryu]

How could a big enough object break the game, at any tier?

doubling the size of something thats feet across means nothing on a geographical scale. doubling the size of an object several miles across? thats a whole nother ball game.


Or be a Warforged who also doesn't need to sleep in the first place.

warforged still sleep, just like elves do. its just a different type. That being said your point is taken. 'sleep' for 4 hours then do something to interupt the long rest (which requires 8 hours).

[Unoriginal]

Or be a Warforged who also doesn't need to sleep in the first place.

Not needing to sleep does not inherently grant a character the ability to short rest every single time they don't act for one hour, unless the DM decides to rules that way.

So the key condition for coffeelocking isn't Aspect of the Moon or Warforged, it's "having a DM who allows it". Like every other supposedly OP shenanigan.

No wonder TO is losing steam.


doubling the size of something thats feet across means nothing on a geographical scale. doubling the size of an object several miles across? thats a whole nother ball game.

Do you have an example? I honestly don't see anything.

[diplomancer]

Not needing to sleep does not inherently grant a character the ability to short rest every single time they don't act for one hour, unless the DM decides to rules that way.

So the key condition for coffeelocking isn't Aspect of the Moon or Warforged, it's "having a DM who allows it". Like every other supposedly OP shenanigan.

No wonder TO is losing steam.

I agree. Coffelock is 100% DM dependant (though not for the reason you point out either, I believe. If a character fulfills the condition for a Short Rest, he gets a Short Rest, in the same way that if a character fulfills the conditions to cast Fireball, he casts Fireball). But also not because of the Xanathar long rest rule, as this rule, on its own, does not apply either to Warforged or to Warlocks with Aspect of the Moon.

[Arkhios]

No wonder TO is losing steam.

Good riddance, I'd say. TO is, imho, rather toxic approach to the game itself.

[Brookshw]

Good riddance, I'd say. TO is, imho, rather toxic approach to the game itself.

What this guy said.

[Eldariel]

Good riddance, I'd say. TO is, imho, rather toxic approach to the game itself.

How come? The whole point is to explore what the rules system does, not use those in a game. If anything, I think it could help avoid toxic gameplay: people can TO when they wanna break the game instead of doing so in play. As a by-product, the community knows what abuses exist and thus doesn't get blindsided by such in game.

[KorvinStarmast]

The worst offenders are Moon Druids at early levels, from there it mostly takes 9th level spells for real shenanigans. I don't think a moon druid is an offender, given how low the AC of most beasts is, but it sure has power spikes. Also, there are six saving throws in the game. Druid at Tier 1 and Tier 2 only has two proficiencies or three with the resilient feat.

Coffeelocks were the talk for a while- but it really takes some DM handwaving to be really possible. Sorcadin and Sorlock multiclasses tend to rule the meta for actual “Munchkin” Optimization. Yep.

Well, unlike 3.X, 5e does not even pretend that anything can be done without DM-buy-in. Which is nice, I think.

There's ways for a character to get multiple Concentration effects going at the same time (such as with glyph demiplanes or simulacrum loops). This is the kind of silly stuff that I would consider in the realms of TO, rather than things I would do in a real game. That's the key; TO is a game within a game that is fun as a mental exercise.

Gate has a pretty bonkers new use to it in 5e. The reason a lot of people don't notice it is because they keep thinking of Gate as a tool for summoning minions, and thus miss the forest for the trees. My use for gate is open it and then shove an enemy through it.

What use is that? Forget about calling allies, you set up a total deathtrap somewhere, speak the name of someone you don't like, and they get Gated right into it. No save. Can you send me a PM with a bit more detail on this, I appear to be missing the "at table application" angle on this. Something about "what's in a name" and "True Names" seems to be a limitation here.

Good riddance, I'd say. TO is, imho, rather toxic approach to the game itself. I agree. It can be fun as a mental exercise, though. Sometimes, cheese is fun.

[BoutsofInsanity]

It should be considered by the way, an achievement, that this late into the games life cycle, their are only four to five busted things in the game. And even those are mostly high level.

Even as much as powerful as caster's are, they aren't leagues above other characters outside of a few select spells. And that's only at high levels. Tiers 1-3 I'd argue are pretty balanced in terms of PVE, effectiveness and fun for all parties.

Even in Tier 4, if the DM has been handing out magic items, or as encouraged, using some items as templates to make others, most everything still functions as intended within a PVE setting. Looking back at how easy it was to bust things in 3.5, or instead, using a class to do things better than a classes intended feature (Druid to Fighter for example), the fact that a Rogue still very easily makes a fantastic Rogue, or a Fighter still outclasses nearly everything in Sustained Damage Output, the Wizard still has the flexibility, the Monk still shines as a Skirmisher/Support etc... That all of these classes and builds function without much investment, or that to do things better takes a sustained effort to convert a class via abilities and feats is fantastic.

In terms of T.O. there isn't many options there. Like others have said, it's mostly at high levels, requires DM buy in, and unlimited time. As far as practical optimization goes, even then, most of the stuff is relatively easy to find, use, and most of the time not outshine other party members.

5e, for all it's flaws has an incredible solid base, strong central mechanics and the flexibility to run nearly any kind of game. Be it high powered, or gritty and deadly.

[Willie the Duck]

Y'know, I think what has really changed is the culture. Sure, optimization is still a game many play. However, I think people are just a little less impressed with it than they once were. Maybe the same thing has happened with 3e since the early 2000s.


Probably because of the aforementioned white room theorycrafting. There is only so many way to say "you've cherry-picked the circumstances or the datas you used and excluded the rest to get those results, so it isn't as impressive as you're trying to convince us it is".


Part of the reason that 3.x had (and probably has, honestly) a thriving TO community is that 3.x had a TON of books with a TON of player and player-adjacent options, and it didn't rely heavily on DM adjudication (because there was a poorly-thought-through rule for everything in 3.x).

I think there are/were more reasons. Certainly the average gamer's relationship to the internet when 3e was released played a role. Having* to engage the optimization game somewhat to get classes of differing tiers to be readily playable in the same game after a certain level means it's kinda hard to argue against some optimization, and then it becomes harder to draw a 'too much' line. 3e's vague impression that it was a low-res reality emulator (or at least a relatively exhaustive rules set, as compared to AD&D and 5e, which hew much farther towards 'DM toolbox for as-needed') played a part. Overall though I think a lot of it was just 3e coming out as something wildly different from AD&D**, and a lot of people who had been optimizing or min-maxing with GURPS or Hero System or Shadowrun or whatever for years suddenly felt that D&D was allowing the same opportunity.
*bracing for deluge of exceptions.
**Player's Options series notwithstanding


I agree. It can be fun as a mental exercise, though. Sometimes, cheese is fun.
Cheese/Cheeze is fun, right up until it disrupts the normal function of the thing. If you're not hurting anyone else, there's no problem. I do think that Optimization talk often drowned out entirely too much on Wizards.com and the 3e forum here when I was a regular contributor, but that's the only transgression, if any. I will say that as time has gone on, I have found 'really good at TTRPG optimizing' to be significantly less impressive than I once did, but I think that mostly has to do with me and becoming a time-is-precious middle aged full-time professional and such, rather than any specific change in game or associated gaming culture.

[No brains]

I like TO for the same reason I like all absurdity. Is it useful? Probably not. Is it the best kind of correct? Technically yes.

Another piece of TO I've seen discussed is making a computer out of chains of Magic Mouths. https://forums.giantitp.com/showthread.php?539861-The-Arcane-Programmer-Guide-(-Official-Rules-Technique-)

[Dr.Samurai]

I think the disdain for TO is totally misguided. TO can't hurt you or your game. It's all for fun.

And yes, it requires DM buy in at times, that's what the "T" stands for, "theoretical". When we were tweaking Pun-Pun, it was under xyz assumptions.

Meanwhile, so-called "practical optimization" can indeed create imbalances because not all DMs and Players are prepared to handle all levels of optimization. Practical Optimization is seen as "okay" or can more easily be justified, and as such more likely to show up at a table, where the other players might not "buy in" to that level of optimization. If you can't handle X level of optimization, you're doing it wrong.

I can say, confidently, that my party would get easily overshadowed by some of the more basic optimization on these boards. So someone coming in with some tricked out "practical optimization" build will totally unbalance our game. Meanwhile, a TO build would simply not be allowed or even attempted and have 0 impact on our game.

Anyways, there's no reason to call it toxic or pretend like it's a problem. There will always be optimizers. The only reason TO is not around these days is because this edition has severely limited, as explained pretty well by Psyren, the potential for this type of optimization.

[Greywander]

Good riddance, I'd say. TO is, imho, rather toxic approach to the game itself.
It's only toxic when people think they can actually get away with doing it in a real game. As a mental exercise, it can be quite fun, and as others have said, it exposes flaws in the system and allows us to patch them with houserules rather than getting blindsided. In a way, it's not unlike the people whose job is to test security systems by attempting to break in; the goal isn't to break into the place, it's to find the weak points that might allow someone to break in, and then address those to prevent it from happening.

Case in point: coffeelocks. Now that we know these exist, a reasonable houserule can be implemented to "patch" them. E.g., you can only recover expended slots, you can't just generate new slots infinitely.

[Psyren]

There's ways for a character to get multiple Concentration effects going at the same time (such as with glyph demiplanes or simulacrum loops). This is the kind of silly stuff that I would consider in the realms of TO, rather than things I would do in a real game.



Gate has a pretty bonkers new use to it in 5e. The reason a lot of people don't notice it is because they keep thinking of Gate as a tool for summoning minions, and thus miss the forest for the trees.

What use is that? Forget about calling allies, you set up a total deathtrap somewhere, speak the name of someone you don't like, and they get Gated right into it. No save.

Fair points, those could both fall into the realm of a TO discussion.


Well, unlike 3.X, 5e does not even pretend that anything can be done without DM-buy-in.

Yes - I think this mindset shift for D&D is a huge deal. I'm a big fan of crunchy systems, but I think 5e has done a much better job than other systems of striking that balance between assisting the DM with tools and guidance but still leaving them in ultimate control of their campaign (or at least, more in the favorable position of saying "yes, you may" than "no, please don't" - and conditioning the players to think this way as well.)


I think the disdain for TO is totally misguided. TO can't hurt you or your game. It's all for fun.

And yes, it requires DM buy in at times, that's what the "T" stands for, "theoretical". When we were tweaking Pun-Pun, it was under xyz assumptions.

Meanwhile, so-called "practical optimization" can indeed create imbalances because not all DMs and Players are prepared to handle all levels of optimization. Practical Optimization is seen as "okay" or can more easily be justified, and as such more likely to show up at a table, where the other players might not "buy in" to that level of optimization. If you can't handle X level of optimization, you're doing it wrong.

I can say, confidently, that my party would get easily overshadowed by some of the more basic optimization on these boards. So someone coming in with some tricked out "practical optimization" build will totally unbalance our game. Meanwhile, a TO build would simply not be allowed or even attempted and have 0 impact on our game.

Anyways, there's no reason to call it toxic or pretend like it's a problem. There will always be optimizers. The only reason TO is not around these days is because this edition has severely limited, as explained pretty well by Psyren, the potential for this type of optimization.

Thanks - and I'll add, while I don't have anything against TO and know it can't hurt the games I play at, I also more greatly appreciate systems where it's harder for people to pull off simply because it tends to mean they are more elegantly / thoughtfully designed.

Like, do I actually expect someone to show up at my table with Pun-Pun or the Wish and the Word, of course not, but the fact that it's even possible (and remains possible) in 3.5 means the designers were at best lazy.

[Grod_The_Giant]

There's ways for a character to get multiple Concentration effects going at the same time (such as with glyph demiplanes or simulacrum loops).
I've heard of a few of the TO things you've mentioned, but not all of them. Do you mind elaborating on your list a bit more?


Good riddance, I'd say. TO is, imho, rather toxic approach to the game itself.

I think the disdain for TO is totally misguided. TO can't hurt you or your game....Meanwhile, so-called "practical optimization" can indeed create imbalances because not all DMs and Players are prepared to handle all levels of optimization.
Dr.Samurai has the right of it. Convoluted tricks to play as a sandwich, break people's minds by jumping stupidly high, or get arbitrarily high Strength scores aren't going to be usable at the table, aren't intended to be used at the table, and outside of a few outliers (anyone remember LordTippy?) aren't used at the table. They're experiments in applying formal logic to normal human language and seeing what breaks, not actual character concepts.

"Practical" optimizion is where you run into trouble with people pulling overpowered builds off the internet and overshadowing the rest of the party. It's an exercise in finding unambiguous synergy and the best of possible options. There's no easy point for a DM to say "nice try, but no"--shutting down a good PO build means adding new houserules, not just a sensible reading of the rules. I'd be willing to bet that 95% of complaints about optimizers ruining the game come from this sort of behavior.

And in terms of PO, I think 5e actually has the opposite of 3.5e's problem. Because balance is generally so good, builds that do manage to beat the curve are really noticeable.

[LudicSavant]

Fair points, those could both fall into the realm of a TO discussion.

https://forums.giantitp.com/images/sand/icons/icon_thumbsup.png


I've heard of a few of the TO things you've mentioned, but not all of them. Do you mind elaborating on your list a bit more?



Dr.Samurai has the right of it. Convoluted tricks to play as a sandwich, break people's minds by jumping stupidly high, or get arbitrarily high Strength scores aren't going to be usable at the table, aren't intended to be used at the table, and outside of a few outliers (anyone remember LordTippy?) aren't used at the table. They're experiments in applying formal logic to normal human language and seeing what breaks, not actual character concepts.

"Practical" optimizion is where you run into trouble with people pulling overpowered builds off the internet and overshadowing the rest of the party. It's an exercise in finding unambiguous synergy and the best of possible options. There's no easy point for a DM to say "nice try, but no"--shutting down a good PO build means adding new houserules, not just a sensible reading of the rules. I'd be willing to bet that 95% of complaints about optimizers ruining the game come from this sort of behavior.

And in terms of PO, I think 5e actually has the opposite of 3.5e's problem. Because balance is generally so good, builds that do manage to beat the curve are really noticeable.

I remember LordTippy.

As for the things you want me to elaborate on, which ones? Of the two mentioned in the quoted text:
- Simulacrum loops is basically having simulacrums summon more Simulacrums.
- Glyph demiplanes is basically making a Superman tollbooth out of Glyphs of Warding (which can bypass the Concentration limit).

Obviously these things all fall into the realm of TO discussion and are not things that should ever be used in a real game.

[Naanomi]

I had a great time on the old wizard optimization forums. It was rarely something I'd ever use at the table (excepting, perhaps, the 'this concept is terrible unless I optimize it) sense. It was a fun activity *seperate* from the game itself.

That being said, there is a... continuum... of optimization, and there is a space where what is 'practical optimization' and what is 'theoretical' can blur and have variability from table to table.

In 3.X you could sometimes accidentally stumble across that line without meaning to be malicious. I've seen a Hulking Hurler get preposterous from a well-intentioned player, and an uber-charger. I accidentally made a build that could take infinite turns (I just chose not to) because I wanted to make an Archery using Crusader.

In 5e the only thing I can think of along these lines that one could... accidentally stumble into without intentionally being exploitive is... I guess a really dedicated Necromancer going all in on minionmancy? Conceptually sounds, thematic, but often game breaking on multiple levels.

[kazaryu]

In 5e the only thing I can think of along these lines that one could... accidentally stumble into without intentionally being exploitive is... I guess a really dedicated Necromancer going all in on minionmancy? Conceptually sounds, thematic, but often game breaking on multiple levels.

similarly a summoner summoning 2 dozen creatures as an action

[Naanomi]

similarly a summoner summoning 2 dozen creatures as an action
To a degree, although at least that is temporary... An intelligent foe might flee such a display and return after the summons had disappeared. A Necro really working it has their army on hand 24-7 (or whatever the time frame and calendar of the local plane is).

Also, big summons are concentration bound... A skelemancer can have hundreds of minions around at high levels if that is the focus they want

[ATHATH]

This is sort of borderline, but a Necromancer with the feature that makes them immune to max HP loss can entirely avoid the permanent max HP loss from casting Create Magen, allowing them to convert money (which is nowhere near as valuable in 5e as it was in 3.5e) and time (or just time if you can cast Wish, IIRC) into decently strong permanent minions at a fixed, non-DM-dependent rate.

[No brains]

Question: can Necromancers abuse Homunculi with their un-lowerable hp, or does that just keep the homunculus from getting any hp that could be drained from the necromancer?

[Eldariel]

Question: can Necromancers abuse Homunculi with their un-lowerable hp, or does that just keep the homunculus from getting any hp that could be drained from the necromancer?

Works with Magens at least.

[Aquillion]

Malleable Illusions + Mirage Arcane is up there in the silliness department, although the DM does have some flexibility to downplay how real Mirage Arcane is (but Sage Advice implies that people can drown in a Mirage Arcane lake, which takes the combo to outright absurdity.)

Even just from the spell itself, though, it is clear that you can make terrain impassable and that it is actually impassable even to enemies who can see through the illusion, which means you can wall enemies off at will in a mile radius with a standard action for ten days after casting the spell, with no concentration required. That is a bit powerful. You just need to lean on the Sage Advice if you also want to be able to, say, drown them in lava or something.

[LudicSavant]

Question: can Necromancers abuse Homunculi with their un-lowerable hp, or does that just keep the homunculus from getting any hp that could be drained from the necromancer?

Sage advice says yes.