Originally posted by Tempest_Stormwind:
rampant wrote:Tomes? Do you mean like the wizard ability from 4e? Guess what, some major differences between the 4e books and the 3e/5e ones. 1st. Of all you could only have one extra spell per qualifying spell known (dailies and utilities if I recall correctly), although some feats would allow you to multiply this effect (however as you were paying with a feat that constitutes an opportunity cost). There were hard limits in place, and you were still picking off a pre-defined list of powers. Yes the 4e wizard learned more powers than any other class, that's completley ok, being the class with the most known powers is not in and of itself a problem. The problem is when you can buy permanent powers known via a non-character building resource without swtiching out old ones, and the other classes can't. Or in the case of clerics when you just automatically get the new powers for doing jack all. 2.) A 4e wizard had one book, he might make copies, but he couldn't have multiple books with different spells in them his known spells were his known spells, even if he managed to gank another wizard and decode the book he couldn't add them to his spells known until he opened up some spells known slots, either by leveling up or swapping out old ones (also at level up technically).
No, I mean the tomes introduced in Arcane Power, which really were the equivalent of runestaffs with a use limiter on them. The Tome of Replenishing Flame, for instance, contained two daily wizard fire powers of up to its level, and by expending your own unusued daily attack power of equal or higher level as a free action, you could use either one of them instead (though you lost your chance at the end of the encounter). These spells
could be copied into your spellbook, but they didn't need to be prepared or anything in advance the way a normal spellbook spell would - just having the tome on hand was enough. The way the level progression and price worked was basically standard for 4e (i.e. it's a money-sinking treadmill to keep the numbers going; like you I believe that removing the assumption of magic items from the game's core math was one of the best things about 5th), but it's still, functionally, "buying" two extra possible (prepared!) spells beyond your normal limits. Or, in your terms, buying (semi)permanent powers known via a non-character-building resource without switching out old ones. (I say "semi" only because the game expects you to rotate out lower-level tomes, but by then the a higher-level one with better powers in it will be available.) The only limit here is the descriptor.
Also, I'm much too simulationist for 4e, so I read point 2 as if it's something I'd say to attack 4e. In 5th, the schema are basically blueprints with personalized notation; if I'm in my fabrication lab in a big university with its library behind me, I don't see any reason why I couldn't construct a massive array of arcane devices (but not at once, since craft reserve won't sustain that many), even as a low-level artificer. In fact, that's kind of an apt description of House Cannith's Tinker's Guild.
One thing I toyed around with, but removed because there's no precedent, was using the space in a spellbook as a resource. (I originally said that schema take up twice as much space as a spell of the same level, and formulas from Salvage Essence took up 2^(R+1) pages with R being rarity (Common is 1), but with cleaner language. I removed it when I noticed there's no mention of the length of a typical spell nor the size of a standard spellbook in the game.) That would at least make schema clunky to transport without proper tools, since there's no Blessed Books and bags of holding are hard to come by. However, it turned out to add a lot of text, which violated Goal 3.
5e makes a big deal about how you don't have to have scrolls at every shop and such like, and frankly remocving magic items form the game's assumed mathematical progression is probably a good idea. The flat plus aspects of magic items were always the least interesting bits anyway, more of a chore to keep up with than anythign else.
I'm the same way. This is also the same reason why the Augmentation spells unique to artificers play with damage types instead - that way, including artificers doesn't bake +X weapons into the system any more than, say, adding a character who can cast Magic Weapon does. (There's Magic Armor, which is a unique artificer spell as well, but looking just at the artificer's spell list, it's a natural fit as a unique spell, and it competes for concentration with the others.)
However, the fact of the matter is that in many worlds it makes sense for there to be scrolls ont he shelves and in those worlds the wizard becomes a lot more versatile, just like your artificer does in a schema rich world, what you call 'automatically adjusting' is basically the problem because the class becomes vastly more or less versatile, and/or powerful based on the game world which is amajor problem becaus eit makes the class basically unevaluatable, you have to evaluate the game world for each campaing before deciding if the class is balanced, and whats more the class becomes more unbalanced the more appropriate it would be to the setting. You put the work of making the class balanced on the DM instead of the designer. Which means the designer isn't doign their job.
Let me get this straight: If the DM deviates from the rules, the game breaks. I'm supposed to anticipate every possible deviation and test for that, because this is the job for a designer.
No, that's wrong. Testing every possible deviation is impossible. What you do is provide a single system with a set of assumptions,
and then test within those assumptions to make sure it works. You include guidelines for deviations, and clearly mark those as "variants" and make sure people going off-road know what they're getting into, because that's the only way to account for the massive range of human creativity.
I'm doing the latter - I'm testing the artificer with assumptions for 5e standard. I'm fairly confident in how some aspects, notably Salvage Essence, translate with different sorts of item availability policies, but without having a set of concrete rules to follow for different policies, I don't have a way of knowing how the rest of the game will behave under those policies. As such, I don't have any benchmark to test against, let alone an environment to test in.
All of what I've just said, by the way, applies equally well to the fighter as it does to the wizard. The fighter in an Epic Heroism game (using the variant in the DMG; the nutshell summary is that it's the same, but short rests are 5 minutes and long rests are an hour, and high-level spells (6th+) still need an 8-hour rest to recover) is virtually unkillable and has an impossibly high damage output, so much so that I'm certain it wasn't tested under that paradigm. The wizard, being less dependent on short rests and more dependent on long ones, doesn't gain as much from such a paradigm (especially where the high-level slots are concerned), so it's clear that the variant influences different classes unevenly even without deploying the testing metrics. There are certainly other areas of the game that break open if you do this as well. However, it's a variant. It's not intended to be as robust as the main system, as it's only going to be used by people who are comfortable going off the beaten track, and those who do are warned of the risks when they do it.
What part of this is hard ot understand? the fact that learning new powers via the scroll system is 'harder' in 5e is irrelevant, the class needs to give a balanced number of splles/powers known, and if you want to exceed that number you should pay an opportunity cost. You yourself are fond of pointing out that it's not garanteed that you'll be able to add schema right? So if you have that sort of situation and you don't give them enough powers then the class is underpowered, if you do give them enough then when someone plays the class in a game where they do get the chance to expand they become overpowered.
Power's a spectrum, not a binary "under/over" thing.
Furthermore, there's still not one but
three throttles in place for how many of these can be deployed at once, even if you ignore the spellbook and the fact that artificers have slow spell level progression (so by the time they're building 3rd level devices, wizards are picking up 5th level spells; if there's a critical 4th level spell that you designed your adventure around, a PC wizard may have it, but a PC artificer won't). Let's assume that someone's actually got the infinite spellbook and the effect you want is of a level you can cast (this is actually a stronger set of assumptions than the artificer will ever encounter, so the clamping we get from these rules will actually be stronger in real play).
1) There's craft reserve, most obviously, which is a daily resource with multiple features demanding it (and since devices cost one point of that
per level of the spell, you wind up getting fewer and fewer of the strong ones; the fact that they're Spell Scrolls, which don't scale with your proficiency bonus or Intelligence, mean you have incentive to use your biggest ones).
2) There's also construction time (you can only build one of them per rest, and only if you have unspent reserve; if you walk into a dungeon with all your reserve in pre-built items, you can't replenish them until tomorrow, so your "versatility" is set entirely by what you prepared before the adventure began (just like an AD&D magic-user in this regard, since you need to build multiple copies of a device to use its effect more than once). If you walk in with no devices, keeping your craft reserve free to handle whatever problems you encounter with maximum versatility, you only get a chance to build one per rest. And every position in between on that scale.)
3) There's activation time. Using a spell scroll (and thus, an arcane device) always requires it to be in your hand (so you need to Use an Object to retrieve it, since they're not handled like ammunition; this isn't too severe but it does eat up your interaction options taken during movement most of the time), and being an object that requires your action to use (DMG p.139 on Spell Scrolls), you also need to Use an Object to activate them (Basic Rules p72 on Use an Object), meaning they take up your action
even if the spell inside is a bonus-action spell, and short of using Contingency there's no way to make use of a reaction-spell-based arcane device. (The flipside is that it takes only one action even if the spell inside is a long-casting-time spell, but none of the long-casting-time spells are combat spells, and many of them are already rituals, which you can cast from scratch faster than you can build the device and paying 0 craft reserve. You, or the tomelock that's already shown this isn't too much of a problem.)
Now, you can get around these restrictions by using Prototype, which also uses your Intelligence and proficiency directly, but that carries its own set of restrictions:
1) It consistently costs a spell slot, even if there's a mishap. (Magitechnicians can cast it from craft reserve, but there, it eats up the craft reserve in the same way as an arcane device does - long rest recharge.) You have fewer of these slots to waste.
2) Casting time - it's a 1-minute cast time, preventing its use in combat. (Magitechnicians, again, can get around this, which is one of the big selling points of the guild. However, it costs a Hit Die to do this, so it's at most once per level per day, and long rests only recover up to one-half of your level in spent Hit Dice at most, so this is
not sustainable
and it eats into your ability to bounce back during rests.)
3) It's risky; I posted the success rates above. Recall how people would get bent out of shape with even a 10% Arcane Spell Failure chance in 3.5? You have much more than that if you're trying a high-level spell, meaning you stick to a lower level one than you need unless it's an emergency. This exposes it to similar weaknesses of using lower-level spell scrolls. Also, that 3.5 bent-out-of-shapeness was in a paradigm where a failed spell was just a wasted action - here, it's also a mishap, which is not insubstantial damage against you to begin with (which I assume during balancing), and possibly much more if the DM is getting creative. (Again, the magitechnicians get around this to an extent, but only after level 11, and the fail chance on your high level slots is still substantial.)
4) It actually shares the same activation time limit as above (need to be in hand, need to use your action to set them off, and as such make reaction spells impossible and bonus-action spells less useful), augmented by the shorter duration on prototypes and without the exception for long-casting-time spells. (Magitechnicians can't get around this, and in fact feel it stronger, since they rely on Prototype so much. Even
they need to spend a turn (and a Hit Die!) setting it up, so you actually get the spell one round after you wanted it - and if circumstances have changed by then such that you no longer want it, tough, the slot is spent and the prototype's duration is ticking.)
All of that is independent of the size of your spellbook and the availability of spells to fuel it; it's also (more or less) independent of the delayed spell level acquisition that being a 2/3 caster brings with it.
These are not hurdles that can be easily ignored.
As for the spell scrolls I can sort of see the issue when you put it that way, warrior types get the weapon augment and the mages get scrolls, However weapon augmentation is a little sup-par compared to being able to hand off a scroll of just about anything. The artificer get's a nifty damage boost when using it but recall most of the other guys don't .
The reason you'd use Weapon Augmentation specifically isn't to increase damage (except for you - and that's because artificers lack any offensive cantrips or other weapon support, but are expected to fight with magically-boosted weapons). The reason you use Augmentation specifically is to exploit vulnerabilities or avoid resistances. (A low-level fighter against a mummy is in for a challenge, especially with a nonmagical sword. If that sword gets a fire augmentation, though, expect him to chew through the mummy in a round or two. Switching the damage type here
quadrupled the fighter's damage - but it's situational, and relies on the artificer making smart choices instead of simply locking on and deploying a strong buff by default. Augmentation is also a multitarget buff, which is pretty rare among the weapon buffs and changes in utility based on the number of cantrip-users vs weapon-users in the party.)
However, look through the
rest of the artificer's spell list, paying special attention to those that are buffs or those that create opportunities. None of them support caster playstyles except, possibly, the chess spells at high levels (i.e. Wall of Stone) and its "summon" spells (Animate Objects, Mordenkainen's Sword), none of which are online at any reasonable level. The closest it gets is Glyph of Warding, which is a noncombat support spell (and the way the artificer works, without using prepared spells, he can
only pick Explosive Runes with it, which doesn't favor any class in particular). The only wizards that could benefit from his buffs are warforged wizards, if you assume they're constructs (the rules on warforged are ambiguous as to whether they're constructs or humanoids), but the nature of the construct-specific buffs generally favor creatures exposed to direct combat (except for Energy Ward, which also favors nothing in particular). In fact, the defensive buffs that aren't construct-enabled require a suit of armor to work (Magic Armor, Armor Augmentation), and sorcerers and wizards don't wear armor.
The spellcasters that benefit most from artificer support - either on the team or through multiclassing - are actually the 1/3 casters, the arcane trickster and the eldritch knight (both of whom are Int casters with access to the wizard spell list, meaning they work well with both arcane devices and prototypes, yet they're primarily considered weapon-users instead of spellslingers). The arcane trickster can do magical things with Mage Hand, including using alchemists' bombs, and can benefit from Sneak Attack more easily with the aid of a homunculus (it's an extra body on the field, even if it isn't commanded). The eldritch knight has blast and defense spells of its own, and wears good armor and carries good weapons to enhance. (In fact, an EK can strongly consider a 3-level artificer dip to get short-rest-recharge, int-based attacks of a tunable damage type, two types of resistance (armor and shield), and a couple out-of-combat spells; everyone already expects him to fight with magic and swordplay, and this lets him interface the two perfectly. Plus, against foes with a basic energy vulnerability, an elemental cantrip and a matching-element-augmented weapon make a good War Magic combo. Meanwhile, an AT/artificer 3 multiclass has Expertise in Arcana, meaning his prototypes are much more reliable, even though he still can't use them in combat even if he's a magitechnician, and he can be in control of his own homunculus if he wants.)
Also even if the party does all pick up a scroll from the artificer and use them, that's not going nova, not at all. Yes you're chewing up a lot of artificer resources and the artificer is involved in a lot of stuff that round, but you're not actually being all that dramatic. See it's still spaced out over the same number of turns, and eats the same number of actions. The party is spending their actions to cast your spells, which can be incredibly useful I'd be the first to admit, but going nova generally means you're doing something unusually impressive for a single turn. Scrolls don't do that, they let you trade your turn for the spell in the scroll, yes it all draws from the scroll writer's resources so they probably feel like they went nova. However, the impressiveness over time equation for an artificer scroll alpha strike is not that high, because they're eating a lot of time.
A nova requires two things: A massive expenditure of resources in a short amount of time (such that you're "burned out" later on), and a stronger-than-usual output during that time. It's not related specifically to a single character's turn - it's a small amount of time and a larger amount of resources than would be sustainable or advised.
You're burning through the artificer's craft reserve in a single round (ROUND, not TURN, is the important part here) and hurling far more spells than the game would assume except in the 4-wizard-party case (a party the game doesn't suggest, largely because that's a glass cannon - here it's all of the cannon (for that round), none of the glass).
Look at the numbers for the very basic case about an 8th level party unloading four Fireballs in one round. Look at the foes you'll fight around CR 8. The actual CR 8 foes in the MM average 118 hit points with 13 Dexterity and thus a +4 Dexterity save unless they're explicitly proficient in it. Four Fireballs averages nearly as much damage as their HP, and with a Dex save that low, it's unlikely that more than one will pass, which is still 90 damage in one turn. And in this case - with a spell that probably isn't all that hot since it's a full spell level behind the wizard's range at this level - that can be done against an entire field of CR 8 foes with pretty decent reliability, without any special support beyond the devices, and without picking a better spell. The question to ask, then, is what the average
party would output, and how it compares to ~90 damage to huge clusters of CR 8 foes over a turn. And that requires far more work on my part than balancing the probabilities on prototype took - so it's a lot of work for a potentially exploitable nova enabler that the current rules prevent from deploying, and I'd need to do it against every possible spell at every possible level to be sure it'd work.
It would be cool, but it'd be
impossible to police at a rules-design level. It'd be
entirely up to the DM to prevent that from going wild - and wasn't it you earlier who said that if the rules force the DM to balance things, the rules aren't doing their job?
I'm fine leaving this, with limits, to the Spell Flasks on the alchemist. That shows up late enough that I'm fine with it (by 17th level, 6th level spells just aren't cutting it, and they cost a lot of craft reserve), and it has a couple other limitations too (notably, offensive targeted spells don't work - but that'll need testing, and I haven't hit that far with the alchemist yet.) You haven't so much as mentioned spell flasks, even though they do exactly what you want them to do.
I get not wanting to fix 5e with a single class, it's beyond the scope of the project I hear ya. The catch is you can't ignore the problems 5e has, you have to design around them.
You also have to design within the constraints of 5e. One of those constraints is that scrolls aren't freely purchasable. That's what I'm testing within. I can't foresee how available they'd become in games with more liberal economies, at least not without a point of reference (i.e. a 5e Eberron Campaign Setting Guide). But I do know that if I set it to be robust against all possible modifications (say, by adopting a spells-known model for schema), then the artificer ceases to feel like an artificer to me even in the assumed and shared 5e standard. I'd rather risk a potential imbalance in the face of a DM's houserule than ultimately producing a castrated version of the class - seeing as that's exactly the direction that WotC's Unearthed Arcana article tried to take, and if I thought
that worked, then I wouldn't have done this project at all.