D&D 5E Artificer Class, Revised: Rip Me A New One

[RealAlHazred]

Originally posted by Xanchairothoss:

My issues with the Artificer in previous editions were that it pitted the DM against the party. Artificers were great for the party, but a head-ache for the DM. Treasure leveling is a pretty important aspect to maintaining party effectiveness and ensuring you don't end up in a game where every encounter is a cake-walk. Artificers threw CR out the window by making it exceedingly difficult to determine party effectiveness against a challenge.

From a DM’s perspective it wasn’t the Artificer, specifically, that one needed to be worried about. It was the creation of super optimized Magic Items that limited the DM’s ability to challenge the party effectively -- potentially causing a DM to fumble in either over-estimating, or under-estimating. Most multi-use magic items often trumped spells and offered unprecedented combos. Spellcasters had limitations in spell selection, however rapid magic-item creation restrained by only amount of gold eliminated those limitations by placing control in the hands of a PC. When ability selection for said magic items was open to creative fiddling by anyone but the DM, nearly anyone in a party could eventually acquire the abilities of a high level caster due to the Artificer constructing any item without a real leash.
 
Secondly, with specific classes I knew what level to scale a challenge at. Magic items gaining a reduction in cost to produce really called all of that into question. I could limit the amount of treasure a party gained, but doing so nerfed the Artificer. Not doing that made it totally unbalanced. It was a scaling issue. I could plan for 'X' magic item, but what happens at high levels when I've effectively got a party of 5 Gestalt characters who can individually hold their own against a Tarrasque, due to full optimization via magic items through the Artificer?

Artificers as PCs were either game breaking or super-lagging. They weren’t limited by spells the same way spellcasters were. If a player told me they wanted 'X' item, I could make said item, tweak as necessary and surprise them a bit too. That kind of discovery was no longer present with the Artificer. There was no DM to player relationship.

The model, conceptually, for an Artificer PC very much suited a DM vs. the PC style game. I felt that there either needed to be a lot of communication and trust, or a pre-established antagonism couched in a mutual agreement of fair-play and RAW between an Artificer and DM.
 
In 5E, the Artificer, conceptually, is anathema to the themes and focus of the game, ergo you're working against the fundamental backdrop of the game's design when building this class. For the Artificer to play like it did in previous editions requires extensive mechanics for magic item construction that aren't present in 5E. Where does that leave us? DM fiat -- which actually resolves the main problems with the class in previous editions, but creates new ones.
 

Tempest_Stormwind wrote: 5e, by default, leaves quite a lot up to the DM. That's not an assertion I can change, but it's one I can work with.

 
This is very true, yet the following points are relevant to constructing this class and I feel that you keep skirting past their relevance by mistaking thorough justification mechanically for what is inherently a series of psychological problems that need to be resolved.
 

rampant wrote: You're trying to make the artificer run on setting rather than it's own power. The class is unbalanceable as long as it's powers remain undefined. 
[…]What are you actually trying to keep/preserve/accomplish with such a bloody awful and UNBALANCEABLE power acquisition system. Because yes it is unbalanceable, because it's either random or DM controlled, neither of which is a balanced class feature. 
 
[…]you should TRY to design classes so that they don't care about the economy, because every game world is gonna have a different one.

 
So we’re faced with this class, much like some issues with particular Ranger abilities, only being relevant by virtue of the DM ensuring they are. Why is this an issue? Because it places the onus for relevance as wholly the responsibility of the DM. It’s not that that is entirely a bad thing, after all the DM is responsible for that in regards to the other players. It’s the degree though, that’s the issue. The DM is forced to do extra work in order to ensure that particular members of a party are getting a chance to shine. It eliminates the ability for the player to determine that for themselves and make that happen on their own. All other classes can be placed in any adventure and shine, but Rangers and Artificers need special attention to make sure they do.
 
Should I, as a DM, be doing this for everyone regardless? Absolutely, but doing so with other classes doesn’t pigeonhole me to specific niche situations to the degree necessary for a Ranger. Most importantly, doing so doesn’t necessitate that a fundamental aspect of the game be considered constantly. Namely, the repercussions for every conceivable combination the introduction of a spell or magic item will have upon my game due to a player (the Artificer) having the ability to modularly re-purpose it. This forces me to constantly babysit the Artificer and their unlimited spell selection capability. It makes it entirely my fault if the player does something game-breaking, because I allowed it. It kills my fun, by making every random treasure roll a situation I need to consider from all possible angles of abuse. It forces me to know every spell and intuit the repercussions of it being used in an unorthodox manner by a player who can munchkin it, and is indeed encouraged to, by the very nature of its class. Finally, it forces me to play the game against the game, rather than run the game.
 

rampant wrote: The reason I suggested a custom magic item creation system in the first place was because the DMG magic items are inherently outside the realm of what a class should messing with. 

 
I think THIS is the direction to go in. What I’m thinking is we eliminate spells from the equation entirely by transforming the class into a pure magic-item builder whose selection of abilities allow for the imbuing of items that duplicate spell properties. Schema would be more like Warlock Invocations, and secondly, their scaling advancement would work similar to how Warlocks cast spells. Warlock should be the template for this class. Multiclassing into Artificer should be akin to multiclassing into Warlock, i.e. it doesn’t, in any way, improve spell-slot selection.
 

Tempest_Stormwind wrote: Also, I can tell you didn't read my last reply. You didn't say the magic word.

 
This is a problem. You can’t obligate people to read thousands upon thousands of word walls of text in every single reply you make. You justify doing this as “being thorough”, and yes that is the case, however when the end result potentially serves to inevitably exhaust or bore those who disagree with you, or prevent anyone from wanting to post due to the the strong likelihood of a wall of text reply, often people give up and might not respond. It’s something to consider.
 
Long-windedness and filibustering by stating an argument in so much detail and rigour that it is almost impossible to understand is not the way to go. It shows a lack of consideration for social etiquette. You don’t need to cram every one of your cogent points into a single comprehensive response that is roughly the length of a short novel. It doesn’t matter how nuanced and thoughtful your walls of text are, the fact that near every reply you make is a wall of text is in itself disruptive to the process. Just remember: the longer it is, the less of it people will read.
 
Defending or fostering arguments with a giant chunk of text that contains so many diffs, assertions, examples, and allegations as to be virtually unanswerable due to overwhelming your audience, or obligating them to read every detail you write in order to respond is self-defeating if your goal is collaboration. Bombarding people with so much information is a form of hostility, rendering it impossible to keep up without the other person replying with a wall of text of their own.

If you want people to contribute, use brevity and answer directly, not thoroughly.

 

[RealAlHazred]

Originally posted by Tempest_Stormwind:

Xanchairothoss wrote:My issues with the Artificer in previous editions were that it pitted the DM against the party. Artificers were great for the party, but a head-ache for the DM. Treasure leveling is a pretty important aspect to maintaining party effectiveness and ensuring you don't end up in a game where every encounter is a cake-walk. Artificers threw CR out the window by making it exceedingly difficult to determine party effectiveness against a challenge.

From a DM’s perspective it wasn’t the Artificer, specifically, that one needed to be worried about. It was the creation of super optimized Magic Items that limited the DM’s ability to challenge the party effectively -- potentially causing a DM to fumble in either over-estimating, or under-estimating. Most multi-use magic items often trumped spells and offered unprecedented combos. Spellcasters had limitations in spell selection, however rapid magic-item creation restrained by only amount of gold eliminated those limitations by placing control in the hands of a PC. When ability selection for said magic items was open to creative fiddling by anyone but the DM, nearly anyone in a party could eventually acquire the abilities of a high level caster due to the Artificer constructing any item without a real leash.
 
Secondly, with specific classes I knew what level to scale a challenge at. Magic items gaining a reduction in cost to produce really called all of that into question. I could limit the amount of treasure a party gained, but doing so nerfed the Artificer. Not doing that made it totally unbalanced. It was a scaling issue. I could plan for 'X' magic item, but what happens at high levels when I've effectively got a party of 5 Gestalt characters who can individually hold their own against a Tarrasque, due to full optimization via magic items through the Artificer?

Artificers as PCs were either game breaking or super-lagging. They weren’t limited by spells the same way spellcasters were. If a player told me they wanted 'X' item, I could make said item, tweak as necessary and surprise them a bit too. That kind of discovery was no longer present with the Artificer. There was no DM to player relationship.

The model, conceptually, for an Artificer PC very much suited a DM vs. the PC style game. I felt that there either needed to be a lot of communication and trust, or a pre-established antagonism couched in a mutual agreement of fair-play and RAW between an Artificer and DM.

I've seen issues like this first-hand. The way we resolved them was gentleman's agreements: no custom magic items (though upgrading and merging printed items was fine), no infinite loops, and so on. Even then, all the games I've had artificers in also had immense time-pressure elements to them; we eventually rigged up a portable base that allowed us to create items on the road and multi-threaded crafting through the use of Dedicated Wrights. That was the only way we could keep our gear up to par.
 
And yet, functionally, this was akin to managing downtime days, just with far more elaborate math involved in the relevant activity. This is assumed by default in 5e - you get N downtime days between adventures, decide how to allocate them. You also have a system where magic item treadmills are gone - you don't need to spend every waking moment slaving over a forge to have math that works. Add in the requirement for formulas (which lets DMs set which items they determine are appropriate for their game in a way that's far less negotiable than simply allowing or disallowing them, like in previous editions) and it becomes rather hard to view this artificer as a magic-item factory.
 
This is deliberate: the game doesn't assume magic items are freely available, so I carefully designed the artificer to avoid creating a magic-item-market-style-economy. This includes the short-lived items (arcane devices and potions; neither of these are usable by enough people over a long enough time to make them worth anything on the market, and since vital craft reserve doesn't return until the item's used, artificers are not incentivized to give control over their use to anyone else) as well as the more dramatic permanent items. He can duplicate items you find, or create new items based on old ones (depending on how you rule on formulas; for instance, does dismantling +1 Scale Mail give you a formula for any +1 armor or specifically for +1 scale mail?) but due to the intense costs involved (including the sacrifice of the original item and the prolonged number of downtime days) and the lack of existing infrastructure to support buying these items, this won't translate into introducing economies where they don't belong.
 
This, in turn, combined with the need for formulas, prevents players from kitting themselves out in Iron Man suits of perfectly customized equipment. DMs remain in full control of what items are handed out, as always in 5e. Players can still create items, if they have the resources (including downtime days) to spend, without developing the expectation that they can "win" D&D if they pore through enough splatbooks and assemble the perfect Christmas wishlist of magical tools to build. This puts a rather strong leash on the adversarial attitude that the old artificer encouraged.
 

In 5E, the Artificer, conceptually, is anathema to the themes and focus of the game, ergo you're working against the fundamental backdrop if the game's design when building this class. For the Artificer to play like it did in previous editions requires extensive mechanics for magic item construction that aren't present in 5E. Where does that leave us? DM fiat -- which actually resolves the main problems with the class in previous editions, but creates new ones.

I don't see how it must necessarily fall into this adversarial attitude. That attitude came out of assumptions in the game and how the mechanics worked - two things which dramatically changed between editions. If you look at how Keith sees the artificer - which is almost identical to my vision for it, conceptually - and breaks it down to core abilities, the only one which potentially causes tension with 5e is the proficiency with creating permanent magic items, since 5e doesn't assume that such items are widely available. (My solution was to link that proficiency with discovering formulas - based on items the DM consented to accept by giving out in the first place. If the player dismantles an item that the DM doesn't want them duplicating, the rules already support adding in troublesome quest requirements (i.e. it isn't really fiat) - one of the only example formulas in the entire game suggests having a flame tongue require being forged in lava, which isn't exactly common in your own forge.).  
 
In terms of the need for item creation mechanics, I find the actual item creation rules present in the DMG are sufficient, assuming you don't believe that PCs should be able to invent customized magic items out of nowhere to introduce into the game. The way to handle that does require some fiat, namely working with your DM to devise a new item and place its formula somewhere in the world for the PC to discover (either through adventuring or through invention). Considering how this requires a dramatic shift in assumptions about the game and it isn't absolutely central to the artificer concept I'm using, I'm fine with this particular aspect being handled by fiat.
 

This is very true, yet the following points are relevant to constructing this class and I feel that you keep skirting past their relevance by mistaking thorough justification mechanically for what is inherently a series of psychological problems that need to be resolved. 
 

rampant wrote:You're trying to make the artificer run on setting rather than it's own power. The class is unbalanceable as long as it's powers remain undefined. 
[…]What are you actually trying to keep/preserve/accomplish with such a bloody awful and UNBALANCEABLE power acquisition system. Because yes it is unbalanceable, because it's either random or DM controlled, neither of which is a balanced class feature. 
 
[…]you should TRY to design classes so that they don't care about the economy, because every game world is gonna have a different one.

 
So we’re faced with this class, much like some issues with particular Ranger abilities, only being relevant by virtue of the DM ensuring they are. Why is this an issue? Because it places the onus for relevance as wholly the responsibility of the DM. It’s not that that is entirely a bad thing, after all the DM is responsible for that in regards to the other players. It’s the degree though, that’s the issue. The DM is forced to do extra work in order to ensure that particular members of a party are getting a chance to shine. It eliminates the ability for the player to determine that for themselves and make that happen on their own. All other classes can be placed in any adventure and shine, but Rangers and Artificers need special attention to make sure they do.

I disagree slightly with this, but that slight disagreement makes all the difference.
 
Ranger abilities are extremely dependent upon adventuring environment and foe composition, true - but the DM needs to create environments and foe distributions already, as part of the process of creating the adventure in the first place. If he doesn't decide, for instance, to use Forsts and Goblins, then a ranger who picked those environments and foes is SOL. There's no way to adapt said ranger to the world the DM made, so such a character creates more work for the DM, as you say.
 
However, it's also true that the DM must decide which treasures to place in the world as part of creating the adventure. He's doing this anyway, regardless of who's in the game. The artificer only introduces a handful of basic potions to the game (and even then, only in the adventure itself; he can't sell them and they're not any more available at merchants); his schema were already present (through other spellcasters), he gets "enough" schema on his own to not rely on scrolls (an artificer character doesn't force a DM to change his mind on placing scrolls any more than a wizard character does), and the way Salvage Essence works, it adapts the artificer to the world, based on choices the DM is already making. 
 
That's the difference:  a ranger is fixed in specific choices, and will not adjust to a world that doesn't conform to those choices, which creates more work for the DM. An artificer adapts automatically to the world the DM is making, without any special effort on the DM's part.
 
There are only three corner cases where the artificer might require more work from the DM.

• The first of these is in how much downtime the DM hands out. Giving out too few days is akin to saying that magic item creation is unavailable - which is fine, this artificer doesn't actually require magic item creation to function. Giving out too many gives the artificer plenty of time to work on items, if he has a formula and the means to do so. While the DM doesn't need to modify his downtime schedule to account for this, he does need to be aware of what signals that sends. This is similar to setting short rest frequency between long rests: plentiful short rests favor warriors (and warlocks), few short rests weakens them. However, downtime day determination only happens once per adventure, instead of every few hours of the potentially-multi-day adventure.
• The second is if the DM decides to hand out powerful magic items that he doesn't want the artificer to duplicate. There are lots of ways of regulating this - a ban works just fine, but in-world, you can use a formula that can't be fulfilled, you can give the item plot significance, or you can control the downtime days. These do require extra work (and they also punish players, because a salvaged item is not recoverable), but the situation where they occur is sufficiently niche that I'm not worried about it. It's enough that the class has a way to control this without it spiralling into chaos.
• The presence of construct-only spells implies the existence or use of constructs in the game world in a way similar to the ranger's favored enemy or the cleric's turn undead. I got around this by having enough of those spells function on objects in the game world (Inflict Damage, for instance, works just fine for demolition as well as fighting constructs), and due to the cleric-style casting, there's no opportunity cost to this ability in a world without constructs. Therefore, the artificer only introduces extra work for the DM if the DM already wants to use constructs in his story.

I've seen how much work the artificer used to force on the DM by its mere presence. I tried my best to make sure this artificer doesn't impose on the DM in a significant way. If you think I've failed in that goal, I'd very much like to know how to address it.

Should I, as a DM be doing this for everyone regardless? Absolutely, but doing so with other classes doesn’t pigeonhole me to specific niche situations to the degree necessary for a Ranger. Most importantly, doing so doesn’t necessitate that a fundamental aspect of the game be considered constantly. Namely, the repercussions for every conceivable combination the introduction of a spell or magic item will have upon my game due to a player (the Artificer) having the ability to modularly re-purpose it. This forces me to constantly babysit the Artificer, and their unlimited spell selection capability. It makes it entirely my fault if the player does something game-breaking, because I allowed it. It kills my fun, by making every random treasure roll a situation I need to consider from all possible angles of abuse. It forces me to know every spell and intuit the repercussions of it being used in an unorthodox manner by a player who can munchkin it, and is indeed encouraged to, by the very nature of its class. Finally, it forces me to play the game against the game, rather than run the game.

That was certainly true in 3.5. I don't think it's true here. To the extent where it's relevant, I'm working as a designer to minimize it, and am still open to suggestions.
 
Let's consider your concerns.
 

• Repurcussions of introducing spells to the game. I don't see how this argument applies any more to the artificer than it does to any other spellcaster. Clerics and druids already bring their entire spell list to the game, and wizards learn about twice as many spells from their list as artificers get total schema. The artificer also must split his limited schema choice across multiple spell lists, and can't pull a new spell out of the ether (since actually using those spells always requires giving you fair warning - usually in the form of "I'm building this during my short rest" alerts long before the next encounter starts). The only exception is a magitechnician's Prototype, but even then, you still have the 
• Related, but worth pointing out in a separate point: The Concentration mechanic goes a long way to keeping this from going out of control. Yes, an artificer can create devices that mimic spells on different spell lists. He still only has one concentration slot. He can't Hex and Hunter's Mark a foe, nor can he Hold Person someone in a Wall of Fire. The Concentration mechanic allows me, as a designer, to look at the spell combinations out there among spells that do not require concentration, and see if anything breaks if list barriers are broken down. If I do my job right, a DM doesn't need to worry about that, because the system (via concentration) and the designer (via how these rules work) do it for them.
• Introducing magic items to the game. As I said above, the only items the artificer always introduces to the game are the simple potions on the Infuse Potions list, and scrolls of the spells in their book of schema; these are introduced in a way that they can't change the economy. Since the artificer cannot introduce new items into the game other than this, the DM doesn't need to worry about the player pulling a killer interaction out of his hat. There are some potential interaction effects here when paired with some of the new spells - notably Synchronize, Power Surge, and Jumpstart - but that's the point of this exercise: finding out how to design the class so these interaction effects aren't worth worrying about. 
• Unlimited spell selection capacity: I go over this again and again with Rampant. The artificer doesn't have unlimited spell selection unless you specifically give it to them, which occurs in exactly the same circumstance that a wizard has unlimited spell selection capacity. People are familiar with wizards and know what "unlimited scrolls" means, and won't do it (nor does the system do it by default; it's actually quite stingy on that front). This is amplified by the artificer's slower spell level progression, since a level-appropriate scroll for a wizard is one he won't be able to copy. A DM using the time-saving tables will find the artificer still has a very limited spell selection. There is the possiblity of pulling a rabbit out of your hat by picking spells from multiple lists that break the game when combined in the same caster but don't break the game if combined across two casters (after all, if Wizard Spell A and Druid Spell B put together didn't break the game with two casters, why would they break the game on one, who has to spend extra time (and possibly a single concentration slot) to bring both of them to bear?). That's why I'm testing the bajeezus out of this, trying to find any such interaction.
• Incidentally, any implementation of the artificer that uses "any spell list" (including yours) will have that exact same problem, and will need to engineer its own solutions. Mine include the limitations of craft reserve, the spellbook, the delayed spell level access, the spell scroll's low DCs, the extra setup time (especially for arcane devices), and the mishap chance on prototypes.
• Playing the game against the game: Again, if I do my job right, this isn't going to happen, because the artificer, while exploitable like any class, won't introduce any completely adventure-wrecking interactions.

Many of these problems seem to stem from viewing this as a finished project, instead of a testing project attempting to catch these very issues as a designer and minimizing or removing their impact on DMs running with it. A designer and developer doing their job will nip these in the bud, whether your players are immature munchkins or gentlemen. If we don't do our job, then you rely on gentlemens' agreements in places where there should be rules, and any munchkin will turn it into a wrecking ball. That's incentive enough to do my job.
 

I think THIS is the direction to go in. What I’m thinking is we eliminate spells from the equation entirely by transforming the class into a pure magic-item builder whose selection of abilities allow for the imbuing of items that duplicate spell properties. Schema would be more like Warlock Invocations, and secondly, their scaling advancement would work similar to how Warlocks cast spells. Warlock should be the template for this class. Multiclassing into Artificer should be akin to multiclassing into Warlock, i.e. it doesn’t in any way improve spell-slot selection.

Part of the problem with this is it inches too close to Pathfinder, where every class had their own resource management minigame powering their abilities. 5e generally minimizes this; apart from spell slots, resource management is generally tied to single uses of specific abilities. There are a few exceptions (such as supremacy dice and sorcery points) but these are kept to a minimum. That's why I use craft reserve the way I do - it becomes a currency to work on spells, similar to sorcery points (except only really managed during rests instead of on a round-by-round basis), instead of an entirely separate point-based invention system that either only sees use during downtime or slows the game to a halt when it's used during an adventure.
 
I like the warlock design quite a lot, so I'm sympathetic to your fix (though I wish it was more specific). However, your specific concern seems to misread the artificer.
 
First, the artificer doesn't improve spell-slot selection if one multiclasses into it. You do get access to the book of schema, but those are fuelled off of [UNKNOWN=del]: Pact Magic slots
craft reserve. Your schema are totally different from your spells known, and your craft reserve is totally different from your spell slots.
 
Second, the warlock does improve spell slot variability, since you can cast warlock spells through your Spellcasting slots and your spellcaster spells through your Pact Magic slots accordingly. The artificer doesn't do this (beyond what other caster/caster MCs do, using the artificer's highly-focused normal spell list), since you can't cast schema nor can you build devices of spells you know.
 
If this wasn't clear, how can I make it clearer?
 

Tempest_Stormwind wrote:Also, I can tell you didn't read my last reply. You didn't say the magic word.

 
This is a problem. You can’t obligate people to read thousands upon thousands of word walls of text in every single reply you make. You justify doing this as “being thorough”, and yes that is the case, however when the end result potentially serves to inevitably exhaust or bore those who disagree with you, or prevent anyone from wanting to post due to the the strong likelihood of a wall of text reply, often people give up and might not respond. It’s something to consider.
 
*snip*

Apologies. If you look through Rampant's replies that triggered that, you'll notice that even when I was brief or clear on specific points, he didn't acknolwedge any of it. Length didn't matter at all - he refused to accept that a spellbook wasn't infinite in this edition, no matter how I phrased it. The only conclusion I could have was that he wasn't reading what I wrote, so I added sentences asking him to mention bananas if he read them. Multiple sentences, actually. He still didn't mention it. (And, later, when I wrote what you quoted, he didn't even acknowledge that there was a magic word, and that specific quote showed up in a very brief reply. The length didn't matter.)
 
Considering the alternative, on my side, would be to similarly ignore concerns that others are raising, what solution would you suggest?

 

[RealAlHazred]

Originally posted by Xanchairothoss:

Tempest_Stormwind wrote: Considering the alternative, on my side, would be to similarly ignore concerns that others are raising, what solution would you suggest? […] If this wasn't clear, how can I make it clearer?

I didn't read most of your comment can you guess why? I as well, already answered both of those questions in my last post and also indicated the likely reason he isn't answering your magic word. The answer isn’t to supply more information, the answer is to succinctly, and directly respond. Not write essays for every question or concern. Regurgitating ad-nausea in a variety of different ways with supplemented analysis, examples, commentary, or formula is the opposite of what you want to do.  

Tempest_Stormwind wrote: Part of the problem with this is it inches too close to Pathfinder, where every class had their own resource management minigame powering their abilities. 5e generally minimizes this; apart from spell slots, resource management is generally tied to single uses of specific abilities.

 
That’s not a reason not to do it. Form should follow function. You’re making this class far too nuanced, optimal, and intricate. It’s way more complex than it needs to be.

If we simply offer a choice of abilities to infuse objects with, that mirror how Invocations work i.e. a Schema[Invocation] that allows me to (for example) “spend the equivalent of a spell slot (or sorcery/Artificer points) to add a weapon enhancement of +1. It requires ‘x’ Artificer level to select this schema. The enhancement lasts for ‘x’ turns and scales up by +1 every ‘x’ Artificer levels I gain.”
Another Schema might allow me to infuse an object with Invisibility, Resistances, or grant the user Detect Magic, offer a delayed healing spell, give ‘x’ to a skill check or saving throw, allow me to fly for ‘x’ turns, breath water, cast Disguise Self etc.

The point being, the effects are considered beforehand. The infusions are limited to choices offered. At later levels you could potentially even make them permanent. And multiclassing wouldn’t improve spell slots for infusions (like Warlock spell slots don’t enhance others).

I wouldn’t even make this guy a caster. I’d give him the equivalent of sorcery points, and include costs to activate an infusion based on the Schema (Invocation) he/she selects.

 

[RealAlHazred]

Originally posted by Tempest_Stormwind:

Xanchairothoss wrote:I didn't read most of your comment can you guess why?

The amount of energy needed to address a point is an order of magnitude higher than the amount of energy needed to make that point. Naturally, some replies will be longer.

That’s not a reason not to do it. Form should follow function. You’re making this class far too nuanced, optimal, and intricate. It’s way more complex than it needs to be.

That's also true of 5e in general, though: the classes are designed remarkably well, but the complexity in their design is hidden by simplicity in how they're presented and how they play.
 
The tl;dr of most of the parts of my post you didn't read is "There is a difference between design and DMing. This is design." 
 
How can I simplify the class (Design Goal 3) while still meeting the requirements set out in Design Goals 1 and 2?

If we simply offer a choice of abilities to infuse objects with, that mirror how Invocations work i.e. a Schema[Invocation] that allows me to (for example) “spend the equivalent of a spell slot (or sorcery/Artificer points) to add a weapon enhancement of +1. It requires ‘x’ Artificer level to select this schema. The enhancement lasts for ‘x’ turns and scales up by +1 every ‘x’ Artificer levels I gain.”
Another Schema might allow me to infuse an object with Invisibility, Resistances, or grant the user Detect Magic, offer a delayed healing spell, give ‘x’ to a skill check or saving throw, allow me to fly for ‘x’ turns, breath water, cast Disguise Self etc.

You showed why I didn't do this in your own thread on this topic: enabling players the fun of building tactics creatively in play is the intent.  (See also Design Goal 2; this is how the original artificer felt, when it wasn't bulldozing entire games.)
 
Basically, a teensy list of self-contained warlock-invocation-style abilities is necessarily going to be less creatively-supporting than tapping into the diversity of the existing spell system, unless you're willing to design something better than that spell system on your own.
 
Also, Design Goal 3 - I don't want the class to take up an entire book on its own. The class itself takes up about 5 pages, including the class table and sidebar but not including art. A huge part of this is that I didn't invent a new system for it - just a way of managing the existing system.

And multiclassing wouldn’t improve spell slots for infusions (like Warlock spell slots don’t enhance others).

The tl;dr of another part of the post you didn't read is "The artificer, as written, no more enhances multiclass spell options than the warlock does."
 
In fact, since you can cast Spellcasting spells from Pact Magic slots and vice versa, the warlock multiclass is nowhere near as independent as you play it up to be.
 
A multiclass caster/artificer does pool spell slots together, just like any other caster/caster multiclass, but your schema and craft reserve are completely independent from your casting, unlike your pact magic spells (in which only the slots are independent).
 

EDIT: I know you dislike additional information, but there is a corner case I need to address.
[sblock] Namely, that's Prototype. Consider a Wizard 18 / Artificer 2. 
A multiclass spellcaster / artificer will have one list of spells known which he can use through his spell slots, and one list of schema he can build using his craft reserve. Our Wizard 18/Artificer 2 would be hypothetically able to learn the schema for a 9th level spell if he took the classes in that order (he's fixed to 1st level schema at level 1, but at level 2 he's open to any spell level for which he has slots), but he would only have at most 6 craft reserve, so he would never be able to build an arcane device to actually use that schema.
 
Prototype, however, is a spell that lets you create an item based on a schema. So, hypothetically, our Wizard 18 / Artificer 2 could prepare Prototype as a spell known and cast it using his 9th level wizard spell slot, creating an item of, say, a cleric spell.
 
However, Prototype is limited by your book of schema as well. You start with only 1st level schema, and cannot expand it without possessing scrolls or finding spells in the environment. You can't copy from your spellbook or anything, so chances are very good you'll only be able to replicate a very small number of 1st level effects upcast to 9th level. (A 9th level spell slot on a 10d8+5 Cure Wounds is much less impressive than a native 9th level spell like Miracle.) You may have a single 9th level schema if you took the classes in that order, but this would be completely unusable except through this method, which isn't as useful as a standard 9th level spell.
 
And after all of that, the prototype is still unreliable (requires an Arcana check to function properly). For 9th level spells, that's an incredible 80% failure rate. If you're willing to spend the increased casting time (1 minute, i.e. not in combat and you can't be concentrating on anything else) for Prototype and risk only a 20% chance of success on one spell per day at the absolute highest character levels (and trade out the high-end wizard abilities on the way, including an ability score increase / feat), I think being able to pull off one out-of-class spell is an appropriate payoff. 
 
It's still in his best interests to pick more manageable schema, though.
 
 
 
Suggested fix: If you think this is a problem, I can delete or adjust the text regarding multiclassing and schema levels, so the only schema our Wiz 18 / Arty 2 can learn are those a 2nd level artificer could learn. I'd prefer it to still allow scrolls, but I'm totally okay with blocking off innate learning like this.
[/sblock]

I wouldn’t even make this guy a caster. I’d give him the equivalent of sorcery points, and include costs to activate an infusion based on the Schema (Invocation) he/she selects.

That's... actually very nearly already what it is. Craft Reserve is the equivalent of sorcery points, using the pre-developed spells instead of an independently-developed invocation-style schema system, and the costs are set by the spell level.  
How can I make it clearer in the actual text what's going on? I thought I did a good job, as did others, but either you didn't read the class itself (which I don't believe; it wouldn't be fair to judge the wizard having not read the wizard, for instance) or you've spotted something that we missed (in which case I'd love to hear how to improve). 

 

[RealAlHazred]

Originally posted by Xanchairothoss:

 
Vis Craft Reserve. I meant Spell Points (p. 288 DMG), not Sorcery points. Have it scale like that. i.e. increase the amount of Craft Reserve.

Gut spellcasting entirely, instead dig through all magic items, evaluate their core elements and turn those into Schema that may be activated via Infusions of Craft Reserve points. This still permits the creative versatility and "fun" element that I sought and we thought was important. Your "pick any spell" ability for the Artificer totally encroaches on Magical Secrets (which is limited for a reason). You've turned this class into the most optimal and utilitarian class in the game. Sure, it's balanced mechanically. However, what are the drawbacks to playing this class? I'm not seeing many, if any. It's the perfect generalist and does the Bard's job better than the Bard. We already have a fully optimized Bard class (which is arguably the best class in the game). Having your Artificer be on par with the Bard is fine, but this guy outshines everyone, and that's a problem. This is a super class, and it shouldn't be. What it should do is allow creative tactical strategy for infusions, but it shouldn't be able to do EVERYTHING any class could conievably do, and then some, with spellcasting on top of that.

I worry that you're emotionally biased by your own creation and what I recommend is counter to what you've already built, and thus is not at all the direction you wish to go in. A well reasoned arguement does not mean the direction you're proposing is correct. I challenge you to take on my point of view and argue for it. I further challenge you to argue against what you've built.

 

[RealAlHazred]

Originally posted by Tempest_Stormwind:

Xanchairothoss wrote:Gut spellcasting entirely, instead dig through all magic items, evaluate their core elements and turn those into Schema that may be activated via Infusions of Craft Reserve points. This still permits the creative versatility and "fun" element that I sought and we thought was important.

That's possible, but a hearty endeavor, and it treads over design space that already exists (i.e. if spells already do that style of work, why re-invent spells?).
 
It also makes magic item treasure less relevant, since you can more-or-less-freely cook up optimal gear even in games where magic items are scarce. This hits the problem with the ranger you raised upthread (and I addressed in a post you said you didn't read), and it conflicts with Design Goal 1.
 

Your "pick any spell" ability for the Artificer totally encroaches on Magical Secrets (which is limited for a reason).

Mine's limited too, in different ways. Once a bard has a spell from another class, it's his to use as he sees fit. Once an artificer learns a schema, he still has to build the device to use it, and those carry greater limitations.

You've turned this class into the most optimal and utilitarian class in the game. Sure, it's balanced mechanically. However, what are the drawbacks to playing this class? I'm not seeing many, if any.

• Delayed spell access (there's a reason you don't rely on Eldritch Knights as primary casters)
• Reliance on time to build your devices (one device or potion per short rest, or a solid minute to cook up an unreliable prototype)
• Super costly prices (compare how costly an arcane device of level X is to an Elements monk's spells; ki points and craft reserve are pretty much on the same scale. They're comparable - and elemental monks aren't known for bringing much magic to the table!)
• Narrow spell selection (both in normal spells (which only really work on objects) and in schema (since you learn so few of them to begin with)
• Inability to make use of bonus-action or reaction spells except the rare few on the artificer list itself
• Very limited skill list (notably with zero social skills and very few exploration skills).

That's just off the top of my head. There are more.

It's the perfect generalist and does the Bard's job better than the Bard. We already have a fully optimized Bard class (which is arguably the best class in the game). Having your Artificer be on par with the Bard is fine, but this guy outshines everyone, and that's a problem. This is a super class, and it shouldn't be. What it should do is allow creative tactical strategy for infusions, but it shouldn't be able to do EVERYTHING any class could conievably do, and then some, with spellcasting.

It cannot do everything anyone else can do. In your own specific case, try convincing a guard to let you past. Bards are made for this with a few smooth words and a winning smile (and magic if all else fails); artificers have to build a mind-control device to have a chance of success here, which takes time, eats up resources, and has a chance of failure (low arcane device DC and Prototype failure rates), and after all that it produces an effect that the bard outgrew several levels ago.
 
I've been running this through various gauntlets for the past several months. If anything, test characters often came up too narrow in focus. Sure, the class as a whole has a lot of flexibility, but any individual artificer does not. 
 
Remember the 3e sorcerer? The class, as a whole, had a huge range of flexibility through access to the largest spell list ever printed. Despite this, no one remembers the sorcerer as being a paragon of versatility.

I worry that you're emotionally biased by your own creation and what I recommend is counter to what you've already built, and thus is not at all the direction you wish to go in. A well reasoned arguement does not mean the direction you're proposing is correct. I challenge you to take on my point of view and argue for it. I further challenge you to argue against what you've built.

Fair enough, but I also challenge you to look through the actual class and build an artificer of your own design, at a level of your choosing, that can rival the versatility of an equivalently-levelled Valor bard (not even a Lore bard, a Valor bard) that I build. You decided that the bard is your hallmark of versatility, which is why I'm selecting it here.
 

This challenge might require a few words, though.
[sblock] 
In favor of your design:

• Assuming an intricately developed modular invention system, the PCs will be able to create tools of their own design using simple, non-spell-based building blocks and exploiting the interactions between them. 
• With enough building blocks, the system could accommodate virtually unlimited potential devices, which could be concentrated in one truly epic Rube Goldberg device or spread out across multiple party members, potentially using the same resource to manage it.
• In theory, this system would be translatable to multiple classes, in effect turning anyone into whatever flavor of artificer you desire. 
• If an actual artificer class, subclass, or feat is designed, it can dovetail with this system (in a manner similar to how the spell-less ranger or the Martial Adept feat dovetail with the Battle Master fighter maneuver list). 

However, against this design:

• Spells already provide building blocks; this is reinventing the wheel.
• The game already uses spells as basic components common to all classes - for instance, there's no arcane/divine distinction, and what used to be spell-like abilities are now just "can cast X spell". Even sophisticated variants are just modifications to this base system - for instance, psionics, so far, is just innate spellcasting without components.
• Blocks based on the DMG magic items make the assumption that such items are appropriate for any world that would potentially allow this creation system.
• Furthermore, this system assumes that it is appropriate to give players control over which magic items are present in the world, something that no other class even comes close to doing (except possibly the ranger in regards to foes and environments, which you identified upthread as a problem).
• This system would still hit some of the same limits mine does, notably in requiring a solution to preventing it from introducing a magic item economy in any world that doesn't already have one. Including, notably, 5e's "default" world.
• Such a system is an immense undertaking. Assume a simple system: devices consist of a Red block, a Green block, and a Blue block, and each color of block has six possible states to choose from (note: there are 16 battle master maneuvers and 17 elemental disciplines; I did not make this number up out of thin air), you have to consider 216 possible three-block devices (assuming you can't duplicate a color) when evaluating their impact on the game. For comparison, the wizard has 230 spells (excluding cantrips but including the spells in Elemental Evil), and the wizard was closely scrutinized by the entire Wizards D&D staff and an Internet full of playtesters for years.

Note that none of these are against that specific implementation of the artificer, particularly as a DM or player - it's against the scope of this undertaking as a designer. 
 
 
Against my own artificer:

• The class involves tracking two sets of spells - your schema and whichever spells you've prepared.
• You need to be creative in describing the results of your devices to really capture the "inventor" feel - an "arcane device of Tongues" makes you sound like every boring spellcaster, but a helmet that speaks its wearer's words in a different language, made from a discarded hobgoblin skull and some spare silver dust, feels more like the mad inventor.
• Choosing one spell from any list in the book takes a surprisingly long time, so levelling up at the playing table takes a while.
• Craft reserve recharges at two separate rates, requiring a bit more bookkeeping than normal (I use two piles of stones (available, and coming back next short rest) and a storage bag (coming back next long rest).)
• Delayed spell access requires a teensy bit more arithmetic than usual, if you aren't just writing down the numbers on your sheet. 
• Delayed spell access also makes you feel like you're playing second fiddle if you learn schema from spell lists that other party members choose from (although this isn't true if you're focusing on other lists, since there aren't PCs of that class at your table to compare yourself to)
• Magecraft puts it on par with rogues on handling traps for one hour per craft reserve spent, meaning rogues are slightly less unique specifically in regards to thieves' tools
• Access to any spell list lets it crib unique spells (i.e. Hex, Hunter's Mark, or Bless), further reducing the uniqueness of those other classes*
• Extremely reliant on rests to re-tool its equipment; if you don't have the perfect tool already built and don't want to risk a prototype failing, it takes a short rest to pull a rabbit out of your hat
• For all of the hoops it needs to jump through, individual artificers still only have a small list of unconventional effects to draw upon; I may have been overly conservative in limiting flexibility
• It can't really "invent" new magic items, even if the world would normally allow for this, without the same DM fiat that any other class would need
• The construct-specific elements of it, notably the Golemists' Guild, implies a higher level of Constructs than most campaign settings employ
• If multiclassing is allowed, Expertise (Arcana) makes prototypes much more reliable at low levels, cribbing the 11th level magitechnician ability possibly much earlier; meanwhile, MCing into artificer might let you cast one more 5th or 6th level spell than you normally would (from any list).**
• All the testing I've done apparently hasn't translated into a simple class - there's still apparently plenty of complexity that I haven't successfully hidden.
• ...I'm running short on time and honestly starting to scrape the bottom of the barrel here.

 
*I do share this concern, and have been debating whether or not to require a spell to appear on at least two spell lists before the artificer can learn it, with the fluff justification that unique spells are too intrinsic to the training style of individual classes to duplicate from first principles without actually being in that class yourself. My testing has shown that this isn't necessary, but it does remain on the table.
 
**I addressed this point in the spoiler block in my previous reply.
 
 

[/sblock]Does that suffice?
 
 
When you mount criticism of my artificer, I ask too that you consider the three listed design goals in your evaluation:

• It has to be playable with standard 5e assumptions about magic items as well as in worlds like Eberron (including not forcing an economy where one doesn't belong)
• It has to feel like the original artificer; I concur with Keith's summation here
• It has to be at least as simple to play as 5e's spellcasters, and presentable without an incredible amount of text. (The math used to develop it can be longer, as long as that complexity is hidden in the final result.)

Your proposal here dramatically challenges Goal 1, compromises on Goal 2, and hits the second half of Goal 3 with a depth charge. This isn't to say that it's a bad implementation, it just means that our design goals don't overlap. I'd like to know which of those goals you disagree with or consider unreasonable, and why, and what your goals would be in a similar project.

 

[RealAlHazred]

Originally posted by Xanchairothoss:

"Note that none of these are against that specific implementation of the artificer, particularly as a DM or player - it's against the scope of this undertaking as a designer. "

Want to give it a shot? If not, if you're set on your design, it looks good to go from what I poked through (I think it might be too good, but I don't think it's broken). I think it could be more streamlined via what I propose, but you seem to believe that the building block method would be overly complex to test. I really don't think it would be, we'd be pulling core concepts from the baseline magic item list, which have been tested. As well, how it scales will make it fairly evident if something is busted.

I'm not going to justify my recipe via running it through your design goals, mainly because I take many of those as a given. You can do the mental gymnastic to twist logic to suit whatever argument you want when running your assumptions on my proposed build through parameters of your own design. I'm not interested in that intellectual pissing contest (my tone is neutral, not harsh -- internet does not convey tone well; don't take me the wrong way) of arguing and justifying. I'd rather spend the time and energy on building something cool.

 

[RealAlHazred]

Originally posted by Tempest_Stormwind:

I'd also like to apologize for my tone today - I've been having a really, really rough week, and upon a re-read, it seems I was being more than a little dickish, which you don't deserve.

Xanchairothoss wrote:"Note that none of these are against that specific implementation of the artificer, particularly as a DM or player - it's against the scope of this undertaking as a designer. "

Want to give it a shot? If not, if you're set on your design, it looks good to go from what I poked through (I think it might be too good, but I don't think it's broken). I think it could be more streamlined via what I propose, but you seem to believe that the building block method would be overly complex to test. I really don't think it would be, we'd be pulling core concepts from the baseline magic item list, which have been tested. As well, how it scales will make it fairly evident if something is busted.

I'm not going to justify my recipe via running it through your design goals, mainly because I take many of those as a given. You can do the mental gymnastic to twist logic to suit whatever argument you want when running your assumptions on my proposed build through parameters of your own design. I'm not interested in that intellectual pissing contest of arguing and justifying. I'd rather build something cool.

I'm not sure I have the time to dedicate to this that it would deserve, but I'd love to contribute. A standalone Masters of Artifice module could serve so many potential purposes, from Eberron artificers to Iron Kingdoms clockwork wonders, and could be implemented with or without a full artificer class. My concern is I will not be able to take on a freelance game designer gig until I'm finished my research; I do stuff like this in my spare time, what little I have. (The long posts written today were written while I was waiting for AI simulations to finish running.)
 
Perhaps due to the immense limits on my time, when I get an idea, my first thought is always "Is this idea necessary?". If something is close enough that a bit of reflavoring handles it, then I don't need to redesign it, or can re-imagine my idea as a tweak to an existing rule.
 
(It would have been buried in this extended thread, but I did respond positively to this idea from Rampant as well, though - with a potential interface sidebar (i.e. a way of allocating artificer craft reserve to a hypothetical invention module, or some similar adapter), but making sure that either works on its own without the other.)
 
That said, two key points (point 2 is technically 1a, but I can't do nested lists) that need to be addressed:

• Testing X and Y independently and being assured that they don't break is mum on whether XY, their interaction, is breakable. For instance, the function of the Scimitar of Speed was balanced against it being a scimitar and requiring attunement, and the function of Vorpal is balanced against it requiring attunement. A Vorpal Greataxe of Speed, especially in the hands of a barbarian, Vengeance paladin, or similar advantage-on-attack-rolls character gets so many opportunities to roll 20s and behead their targets that I'm positive such options were deliberately designed out of the game, and being a single item it still only takes one attunement "slot". This was an example I found with thirty seconds of looking at the book, not something I've paid much attention to in the past. (There are ways to adjust this, I know - an obvious one is saying such an item won't attune if there's another attuned item present - but identifying hurdles is separate from jumping them.)
• Introducing more than two items together makes things get even more complicated, since now instead of just X, Y, and Z, we have to check XY, YZ, XZ, and XYZ, which means more work than the entire magic item chapter itself (which was tested by WotC's team of professional developers and an Internet full of volunteers). 
• I think this works better as a completely standalone module rather than a subsystem in a single class. It certainly isn't "streamlined" when a huge block of text defining new mechanics and the options to work with them springs up out of nowhere (compare the Way of the Elements to the other two monk Ways, or the Battle Master fighter to the Eldritch Knight (which makes ample reference to an existing system), but a standalone block usable by multiple classes is held to a different standard and serves more people's purposes.

Don't get me wrong - by no means do I think this is unworthy. I just think that if the project can't solve problems like this early, it's going to have bigger problems later on. That, combined with the same philosophy behind Rip Me A New One, should explain why I sound pessimistic.
 
For what it's worth, I have done stuff like this before - I managed to engineer about 70% of an entirely new system together for handling divine magic / divine intervention in 3e without using the "spellcasting" framework, falling into the CoDzilla trap, or even being pinned to a single class. I thought it was better in many ways too - for instance, why should a priest of a war god be a tin-can-wearing band-aid box*, instead of a more typical barbarian shouting "IN KORD'S NAME!" while gloriously charging into the fray? The 3e approach was to use domains to turn the cleric into a more appropriate form; my idea was to use an analogous system to domains to attach appropriate divine influence to thematic characters, skipping the cleric entirely - but the sheer amount of testing this required to make sure it wasn't inadvertedly worse than before proved far too much, and in the end I abandoned the project. The alternative would be putting out material without proper testing, and, as I said, I'm a perfectionist. All my design work since has been smaller-scale, and much better-received.
 
*Hyperbole
 
Still, that doesn't mean a smaller start wouldn't be appropriate, to be expanded in time.
 
 
 
 
Also, yes, my design goals with my artificer are fixed, unless they can be shown to be poor choices. How that's implemented? Still open to suggestion. You seem to think it's fine, possibly "too good"; I'm curious where, specifically - especially in the context of single characters. (For instance, there are 16 battle master maneuvers, but any individual battle master only knows nine; similarly, although there are hundreds of spells, you can never combine more than one spell requiring Concentration on the same character at once.)

 

[RealAlHazred]

Originally posted by Xanchairothoss:

"A standalone Masters of Artifice module could serve so many potential purposes"

Not interested. Personally.

"I'd also like to apologize for my tone today - I've been having a really, really rough week, and upon a re-read, it seems I was being more than a little dickish, which you don't deserve."

FWIW I didn't get that read from you at all (dickishness).

"Perhaps due to the immense limits on my time, when I get an idea, my first thought is always "Is this idea necessary?". If something is close enough that a bit of reflavoring handles it, then I don't need to redesign it, or can re-imagine my idea as a tweak to an existing rule."

That pretty much sums up part of my inherent philosophy per supplemental material, post core books. So thumbs up!

"Don't get me wrong - by no means do I think this is unworthy. I just think that if the project can't solve problems like this early, it's going to have bigger problems later on. That, combined with the same philosophy behind Rip Me A New One, should explain why I sound pessimistic."

It's apparent that it's more of a time thing, than anything else. If you're busy, rest assured that I did scrutinize the hell out of your build. There's not much wrong with it, at all, mechanically speaking. As an aside, I'm not a fan of the Artificer class as a concept to begin with (I thought I'd dabble in a rough for a cognitive exercise, but even my proposed rough I wouldn't introduce to any sort of game, outside of the intent to test it; unnecessary now, as you already did the leg-work on many of the issues that would've come up).

1. It has to be playable with standard 5e assumptions about magic items (the rest is unnecessary addendum as all future settings are dependent upon this supposition; if they aren't that will be clearly stated, unless the designers are sloppy, in which case they won't sell books).

2. It has to feel like the original artificer. <--- ie. it has to fulfill the inherent trope, which in this case is ability to rapidly craft magic items. Anything alse can be erased, short of particular edition aspects that were popular (hard metric to incorporate, but easy if you have a forum with threads dedicated to such to pull your market research from).

3. It has to be at least as simple to play as 5e's spellcasters, and presentable without an incredible amount of text. (I would propably include "it has to be fun, accessible, and open to creative flexing".)

Have you achieved these 3? Yes. Move on. Congratulations

"my idea was to use an analogous system to domains to attach appropriate divine influence to thematic characters"

That sounds brilliant.