Originally posted by Tempest_Stormwind:
Thank you for your reply. I'll try to address everything.
Marandahir wrote:Why don't you want to go the way of the 5e Bard with full-casting and limit the magic in other ways?
Because I figured that having a delayed spell-level access would be an important clamp on their power, especially when combined with a spellbook mechanic. It turned out to be the key to quite a bit of how this class actually performs, and you'll see delayed spell access showing up again and again in my explanation for how this works.
The old artificer had access to every spell list and could craft items of any spell on them as if he was a member of the same class of the same level. (Actually, the same level +2, but that's a different factor.) That is, a 3rd level artificer could craft spell items, but only those that a 3rd level cleric or wizard could cast... but without any of the limitations present in either class. This was a huge part of the problem.
If the artificer were a few levels behind the others, then the others still bring out their signature spells "on time", with the artificer picking up
both class's slack... in the much-weaker-spell department. While a wizard's throwing around Dimension Doors and the cleric calling down Flame Strikes, my artificer is building universal lockpicks (Knock) or lie detectors (Zone of Truth), spells the wizard or cleric were using earlier but now might crowd out of their spell preparation loadouts, knowing they can trust the arty to cover the "unexpected lower-level utility magic" niche.
If they had full spellcasting progression, all of that goes out the window and I'd have to design an entire new spell list from scratch. All the testing I've done so far suggests the delayed spell level access is
immensely important to keeping the artificer reined in - even moreso than the spellbook, in some cases. (All of Rampant's complaints about oversized spellbooks led me to run a few stress-test conditions where the artificer's book of schema was full of every spell ever. Curiously,
it didn't break (though it did get hard to manage), largely because of the delayed spell-level access.
Don't just look at the final level of spells. Look at
when each spell level comes online.
Also, consider craft reserve in this. A delayed spell-level access allows for craft reserve to supply a few "extra" spells (through devices), but because the
actual spells are a few levels behind, the
total number of spells you toss around is reasonably good (though quite a ways behind a full caster; you average about 9 fewer spell levels, all told. If you ignored craft reserve, you'd average over
20 spell levels behind a full caster.)
It seems to me the "racing ahead of everyone else" happens regardless of the max-out level because of having access to everyone's spells. This is the fundamental problem, in my opinion. If you had a select Artificer list, like everyone else has for their classes, you could get 9th-level spells just fine. You could even have a Magical Secrets-like feature if you want more diversity, but here's the rub: Artificers have rarely, if ever, been about Illusion, Enchantment, or Divination. Sure, they'll make items related to these schools, but they really are more the mad scientists, tinkerers, spell blasters and supporters. Bard gets access mainlyto healing conjurations/evocations, some supportive conjurations, abjurations, and necromancies akin to the Cleric, and then sound-based evocations (for the music stuff) and Illusion, Enchantment, and Divination spells akin to the Wizards of those schools. Artificers have a clear spell-list that goes in an almost exact opposite direction: healing conjurations/evocations, ome supportive conjurations, abjurations, and necromancies akin to the Cleric, and then Trasmutations, Abjurations, Conjurations, and Evocations akin to Wizarding schools. It wouldn't be that hard to develop a list of spells that are scream "Artificer."
Actually, artificers only had a list like that in 4e, where they played
very differently from the original, and it's the original I'm looking at. The original artificer's "spell list" was full of infusions. It couldn't heal at all, for instance, unless they were repairing constructs. This artificer's ability to heal comes from its access to every spell (with a delayed spell level access and a spellbook "gate"), and the ability to create items based on those. (Or, more likely, the healing potions; you can make the basic healing potion at level 1, and it's only slightly less efficient than a Cure Wounds device. The higher-level healing potions cost more reserve, but are more efficient than a similarly-levelled Cure Wounds device and don't require you to learn a schema. Same argument - the artificer isn't healing, he's creating devices that heal for him. His own spells don't do healing the way a bard's do.)
If I create a specific bardlike spell list, then I'd probably have to restrict the artificer's ability to create magic based on other lists, and at that point I'm not working as a magical engineer (capable of making the perfect tool for the job). I'm just a bardlike spellcaster with special descriptions of what it means to "cast" a spell. Compare to how this artificer works, with spellcasting from a narrow object/construct-focused list, but the ability to create simple potions (not tied to any list) or arcane devices that function like spell scrolls for spells from other lists.
I think ultimately that the complications and weirdness of the class come from your desire to make the class the ultimate generalist caster, and yet, it seems odd. If Artificers can learn any spell because they could find scrolls of any spell and know how to read and translate those into magic, why can't the Wizard do the same? Why can't Wizards scribe cure light wounds into their spellbooks? The logic that the Artificer should be able to learn any spell suddenly makes the Wizard feel impotent, which, while some of us would love seeing them knocked down a peg, doesn't make sense from an in-world perspective. It makes a lot more sense to limit the Artificer's allowable spell-pool. And when you do that, you could then make it a full caster, since it wouldn't be overwhelmingly better than every other caster. The nigh-unlimited versatility was what killed the Generalist wizard, after all. There are other ways of making a class versatile.
I'm
not copying Cure Wounds (note the name!) into a spellbook and learning how to cast it. I'm reverse-engineering the basic magical principles of how it works, and learning how to recreate those principles through objects. A wizard learns their own magical traditions, complete with specific trappings of focus items that show a specific style of working with magic; it's different for warlocks, clerics, bards, and every other spellcaster. They learn to solve their version of the problem. The artificer does that too - except his version of the problem can't be directly solved, and instead has to be jury-rigged together from first principles without use of any of the specialist shortcuts a wizard or other caster learns.
Mechanically, the reason this works is because of the limits placed on arcane devices and prototypes. An arcane device of Cure Wounds is
costly in terms of craft reserve (one point of reserve
per level of the spell, and you're still limited by your (delayed!) spell level progression; for instance, a 7th level artificer can only create a 2nd level Cure Wounds device, and doing so would cost 2 points of reserve; assuming 15 starting Int, she'll only have 10 points
total to spread around across all of her devices, potions, and guild abilities).
The delayed spell access,
again, helps a lot here. This is a level where wizards are learning Evard's Black Tentacles, Conjure Minor Elementals, Greater Invisibility, Polymorph, and Stoneskin. The artificer has to wait until
next level to pick up Haste or Fireball (and has to pay 3 points out of ~12 for each of them!), and the level after that the wizard walks away with Dominate Person, Hold Monster, Scrying, and Telekinesis. Which one's feeling impotent here?
You seem to think that this should be a full caster, then point out issues with making it a full caster. Those issues were legitimate in the 3e artificer (whose item creation was completely disconnected from its delayed spell level access, you'll note; mine isn't), and I took great steps to try to learn from them here.
Also, the way you're doing it makes the Bard feel impotent, which is really odd. The Bard, the Wizard, and the Sorcerer all got versatility as their schtick this edition, but in completely different ways (Wizards can cast more often and with more reliable choices each day, Bards have astounding accesss to various spell lists through their dabbling with magical secrets, Sorcerers can convert spell slots into spell enhancers or spell enhancers into spell slots to negotiate around specific issues). This Artificer basically takes the Bard type of versatility and completely outshines the Bard in it, despite the Bard starting from 1st level with magic, and your Artificer only getting 7 levels of magic. The math might work in terms of not overpowering other classes, but what you've presented outshines both the Wizard and the Bard, a feat that for an arcane class that is almost incredible, especially when Artificers have always been the "working class" Arcane class – less sparkly but just as effective. This guy plays Rogue, Bard, Wizard, and while he doesn't do their jobs better than them, he does them in grander ways. That's a problem.
Um, no, he doesn't.
- Spell level delays, again, are a huge part of this. Seriously, look at it level-by-level; build some sample characters, and you'll see. Don't just look at level 20 and see "7thh level spells!!!!". Look at how long it took to get there and how diverse the higher-end spells of, say, Valor bards (not evel Lore bards) have gotten. (Note: At every level, a non-Lore bard, a sorcerer, a wizard, and a Moon druid will all have more spells ready to go than an artificer will, assuming equal casting ability scores. Those are all classes that don't get bonus spells ready.)
- Off-list access is regulated through a spellbook, which can (barring DM intervention) only be expanded through levelling (one spell per level) or finding Spell Scrolls (note that 5e is quite stingy with these; I did a specific analysis for scrolls(x) upthread). Both of these are also limited by your delayed spell level access. You'll have a good assortment of devices to build, but you won't have the entire book on demand as you seem to think they have. (Seriously, Rampant's helpful criticism on this has had me test this rather extensively; you're versatile but nowhere near as diverse as you seem to think.)
- Craft reserve is required for every off-list spell. These cost one point per spell level, you have to pay for each individual casting, and in most cases you have to choose them in advance since it takes a short rest to build one such device or replenish one you used earlier. Furthermore, these devices do not make use of your proficiency bonus nor your Intelligence score - they have their own DC/attack mod, since they follow the rules for spell scrolls. (If you try to conserve craft reserve by focusing on lots of 1st level spells, even at 20th level, they'll still only be DC 13/+5; a wizard doing the same would see DC 19/+11 with that.) Even then, again, you're on a delayed spell level, so you can't get the really good stuff that everone else is bringing fresh to the table. Every bit of reserve you spend also can't be spent on your potion bandolier, your Personal Weapon Augmentations, Magecraft, or your guild abilities. (Note that because of the delayed spell progression, you'll be using that weapon augmentation as your primary offense.)
- If you're not using craft reserve to access those unconventional spells, you're using Prototype. Prototypes have a failure chance, which is enough to discourage many people from relying on them. (Remember how people would get bent out of shape over a 10% arcane spell failure chance in 3e? You don't begin to see failure chances that low until around level 11 or 12, depending on your guild, and even then that's limited to very low-level spells. i.e. you have a 9/10 chance of using a 1st-level Heroism spell 12 levels after bards had a 10/10 chance of doing the same, and at the same level that a bard can reliably toss out Irresistible Dances. You do have a perfect chance of using Elemental Weapon, Energy Ward, or Jumpstart, though, since your thing is to magically modify and enhance objects.) ANd this doesn't get into what happens on a mishap.
On a related note, the delayed spell level access means you
cannot rely on an artificer for party-required spells unless they're already stale. If I design an adventure where the climactic fight involves an enemy spellcaster who has very powerful magic, but no way of dealing with Counterspell, for instance, the fight might go from "rather hard" to "rather easy" if the party has Counterspell, which several casters get at level 5 and can cast, as needed, until their slots run dry. The artificer doesn't get Counterspell until level 8, if he chooses to use his one schema for that level on it, and has to build the right number of them in advance to be able to do the same, and each of those eats up craft reserve that could have supplied a better tool, an augmented weapon or prototype, a potent guild ability (like a few strong alchemist bombs), or the right potion for the job.
I really don't see how all of this - especially the delayed spell progression, which you continually ignore - amounts to outshining bards and wizards. Would you care to provide an example?
As for rogues, the one area they can outdo rogues in would be magical traps, and even then it costs craft reserve to do it. Rogues can apply Expertise to
skills, (and their skill list is broader - especially including Persuasion, Perception, and Stealth) as well as to their thieves' tools (which is a consistent bonus applied everywhere, any time); artificers can only apply it to a tool they're proficient in, and doing so requires them to keep some craft reserve free between short rests (see above on how this resource matters). Rogues also have Fast Hands (Thief) or Mage Hand Ledgermain (Arcane Trickster), allowing them much greater speed or versatility in tackling mundane traps, even if a Magecraft-boosted artificer has a similar bonus (probably a
lower bonus, actually, since due to Personal Weapon Augmentation, artificers have less of a reason to boost Dexterity, and disarming mundane traps is still a Dexterity (Thieves' Tools) check).
Magical traps require Intelligence (Arcana) to disarm normally, so
wizards are actually better at that than rogues already, and
bards (with Expertise in Arcana and possibly a decent Int score anyway) can trump them all. The artificer is
supposed to be especially good with magic devices, of which magical traps are one possible form, so I don't mind them having a way of dealing with these in an especially good way, particularly given the limited nature of the artificer's skills outside of this situation (i.e. look at
anything the rogue does
except disarming traps, and they equal or exceed what the artificer's capable of doing most of the time. That's because the artificer is
related to that niche, but only where the devices are concerned.)