Lamppost in Winter

Ah, you beat me to it. SomethingAwful has deigned to allow me access to the Joshposts again. Interesting that high level afflictions being "less gnarly" was a conscious design choice, but that feedback's been submitted by the team already.

I was thinking putting that incoming buff affliction duration modifier people were suggesting for Res to Con.

"Deadfire cancelled, Josh Sawyer's Bike Simulator 2018 greenlit"

Well, it's official. I really have to wonder what happened to the "Graze downgrades Affliction" concept? I still think that's a really good idea.

Would Intellect controlling only Will and Magic Damage not make it a dump stat for most non-caster builds?

I would've been down for removing character death in POE1 (except maybe for hirelings), because a companion dies and none of the others give a crap. The party banter doesn't recognise it either; Eder and Zahua joked about Durance as if he were still alive for me. Apparently PoE2 is incorporating the relationship system and character deaths, so I at least want to see how that plays out.

I guess it's unlikely we'll get some cool general talents as well then? Probably a bit more work than making new Fighter passives, after all. As much as I don't envy the work of coming up with new abilities for not just Fighter, but a lot of the passive-deficient classes, I can't imagine another solution that won't put out a large section of people.

My problem with this is that it would feel like another health bar to whack at, while the advantage of the current system is that it's very clear and deliberate.

Didn't see this in the compilation of known user-reported bugs, so thought I'd add my voice to the pile. Happens consistently.

Where are you guys getting that you can't stack Concentration? Unless you mean something else that I'm not getting?

I remember an interview that said Black Jacket had bonuses to Accuracy and Penetration when using the "correct" damage type against an enemy. I guess this was removed because it's basically the Devoted effect but for any weapon.

Agreed, these kinds of changes seem good in theory, but in actual practice the effect they have on game mechanics does not, in my opinion, warrant the added level of complexity that arises from their implementation. If anything I actually think the fact the Pillars uses a single accuracy stat for all attacks is one of it's better features, especially in Deadfire where multiclassing and hybrid builds are such a dominant feature - I want my character to be just as competent at swinging his sword as he is at slinging a fireball etc. I also like the more universal stats system, like Might for all damage (RIP), so I can't really place why Accuracy vs. All defences bothers me so. Something about the defence system is fairly complex while the Accuracy system being simple rubs me the wrong way. I think my preferred solution would be just a straight attribute contest, like Might vs. Fortitude (or even Might vs. Might), but that's another conversation for another game. This is more just brainstorming a way to graft that onto the existing systems.

I think it should stay Per because that makes sense, but Mechanics should give a bonus to detection.

You can shift a single-target spell much the same as the AoE retargeting. But I agree that it's quite annoying when you end up having to cancel your spell or reposition your character that you just lose the spell.

Yeah, I would hope this is just for testing purposes for a more elegant solution.

It is probably overcomplicated, yes (I didn't even realise how much until I put it in writing). I would hope the difference from attributes is small enough that overall Accuracy is still more important, and having a high Might or Resolve or whatever is just a little extra, but I suppose it could run into inflation problems. I was also thinking there could just be some mutually exclusive generic talents that give a flat bonus vs. a particular defence, kind of like a reverse version of the Superior Defence talents. Not exactly the same as what I'm suggesting in the OP, but adds a bit of variety.

Having 4 defences, all influenced by attributes, and only Accuracy targeting them has always bothered me, both mechanically and flavour-wise. I've often thought about what a system where different attributes also affect your ability to overcome different defences would look like. Completely overhauling the system wouldn't be reasonable, or even desirable, given how everything is tied to Accuracy, but I think there's a way to have a system where Accuracy is diversified a bit. I would suggest certain attributes give an Accuracy bonus, but only towards a certain defence. For example, Might give +accuracy against Fortitude, Resolve gives +accuracy against Will, etc. (using the old attributes system for simplicity's sake; the Strength and Resolve changes complicates this a bit). A rough proposal of what it might look like: Might: +Accuracy against Fortitude Dexterity: +Accuracy against Reflex Resolve or Intellect: +Accuracy against Will Left out Deflection in this one for now because I'm not really sure what to do with it. Could just not have any particular bonus. This, I think, would make attacking different defences a bit more interesting, and be a little more logical than the current system flavour-wise, e.g., a very intelligent or willful person is better at overcoming somebody's Will defence. Keeping everything tied to Accuracy would make this easier to implement in the current game (if it can be at all). Problems I can see with this: Overcomplicates things - not much we can do about that, really, apart from clear explanations and tooltips Not much really makes sense as a bonus vs. Reflex, though total realism isn't the goal in the first place Inflation of defence numbers - might not be too big of a problem since at the very highest base attributes go to about 20 Might being changed to Strength - with the new system, perhaps Resolve could instead target Fortitude for spells, while Strength remains targeting Fortitude for martial characters

To nick an idea from Fallout, what if we could tag a Skill at character creation that we want to focus on, and growth of that skill is doubled (maybe not doubled, but increased somehow)?

Anybody know if that's intentional? The simple additive percentages of the first game were rather more intuitive, by design if I recall correctly.

Right now, it mainly affects casters because they don't have enough Penetration to go through armour, but the patch might alleviate this.

I do like healing on Resolve. I feel it's a fairly simple way to make that stat more enticing. Maybe also have some effect on healing received?

Quoting what I previously said on the flavour subject I find it hard to explain, but I think of damage as being raw "force" from a spell, and Resolve seems to me like willpower and force of personality, which matches that more than Intellect. Duration of effects feels to me like a more "subtle" effect than damage, and therefore more fitting to Intellect. Bit of a sidenote, really, since the stats seem to have verisimilitude as a secondary concern. Your point about being able to have resolute Barbarians and intelligent Wizards being more the standard with your proposal is an interesting one. I think Intellect would have to be reworked more though, otherwise it's going to be even more of a dump stat for martial classes than Resolve was previously.

Josh speaks on the matter

Josh speaks

Assuming the changes go forward, I think Strength should stay for summoned weapons. Flavour-wise, they're supposed to be physical objects that are conjured into existence. Perhaps some exceptions for things like Kakaloth's Blights. Mechanics-wise, would it not stop melee wizards from being uber-tanks, since they have to split between weapon damage and the deflection boost/traditional spell damage? Though were I to make a melee summoned weapons wizard, I'd probably find the damage increase from Might to be more worth it than the bump in Deflection from Resolve. I wasn't around for the PoE beta/early patches, but from what I heard Deflection used to be governed by Per? Anybody remember why that was changed? I feel like moving Deflection to Per wouldn't make it too OP, especially since Per-based Interrupts are out.