thelee

absolutely not, because this is a party-based game, and classes by design fulfill different roles. if a given class can't solo very well, that says nothing about it's quality/strength because solo play is essentially a whole other type of game. you can certainly try to rank classes by their strength in their ability to solo, but it's not a general ranking. example 1: in poe1, i completed the ultimate with a high-resolve paladin. i would never ever rate that as a very strong party-based build, because its sole strength was simply the ability to grind out fights via high defenses and sustain; it would basically be useless in any reasonable party because everyone else would DPS the enemy down way before the paladin's tankiness would ever become needed. example 2: tactician/skaen is probably the second best ultimate build, but i would consider it a supremely parasitic build for party-based build, because so much of what makes the class powerful relies solely on the mechanics of solo play and related challenges, and they literally don't work at all in a party-based build (unless everyone else is a */skaen or /*rogue of some sort). combos can certainly elevate the quality of a party-based class, but the more a class relies on that combo to be playable, the worse the class actually is (because maximum viability is a stated goal of the game design). If this were magic: the gathering with perpetual support and more aggressive balancing, wizards of the coast would step in and ban combos that are too good. from a party-based perspective, i think how you assess a class's strength falls into these questions: what is the class's nominal role(s)? (primary weight) how well does the class fulfill this/these role(s)? how necessary are these roles in a party? how well does the class interact with the metagame of equipment and other such (secondary weight)? for combos (and secondarily itemization), how reliant on these combos is a class? how viable are they without? (tertiary weight) for all of the above, is it weighted towards specific levels? (for example, if a class is only good at a role and can metagame well for a specific role by like level 19, this might make the class pretty bad for the role because much of hte game is weighted towards earlier levels and multiclasses may not ever reach the necessary ability level; tertiary weight) in my gamefaqs guide i tried to do #1 with some of the casters. e.g. from the perspective of healing, i'd go lifegiver->kind wayfarer->shieldbearer->other druid->other paladin->priest->troubadour->other chanter->(way way at the bottom)cipher. for resurrection, i'd go priest->troubadour->other chanter->paladin->druid. for party-buffing, i'd go priest->troubadour->shieldbearer->other paladin/chanter->druid. In other words, there's no catchall "best ranking" for classes because needs are different, and there's not even a "best ranking for defense" because there are different aspects to that that classes have varying strengths and weaknesses as part of balancing and diversity. so if you're thinking purely of a 'what is the best support class?' it's not a good enough question. you can probably assign some weights to how much you care about the various aspects of being a support class, but you need to split out their roles. and it becomes more of a question 'for this party, i'm missing healing and don't care about revive capabilities. which class should i put in?' At the same time, the answer to some of these actually change on difficulty. For normal difficulty, for example, i would rate "party buffing" as a very weak and unnecessary part of a party, and consequently any class that relies on fulfilling this role and any slot in your party for this role becomes worse and rankings shift in favor of just brute-force damage. On the other hand, on PotD party buffing and enemy debuffing I find to be much more valuable; consequently such spots in a party become more valuable and classes that are good at buffing or debuffing also become comparatively better. edit: for a combo like brilliant + duration extension, because it is so accessible to any class (all you need is a priest, not even a cipher) it's not actually worth considering as a factor in determining class strength anyway, unless the party role you are trying to fulfill is "i want to extend powerful buffs" at which point the ranking is sc priest->mc priest->(way way way down)non-evocation/transmuter wizard.

it also kinda helps the ghost heart along that the sharpshooter subclass is decidedly niche due to how strong of a downside a categorically universal recovery time penalty is, and the stalker which--while can be strong--arguably requires a lot more micromanagement and really leans further into bonded grief as a mechanic.

just to add, the confusing part about the initial roll-to-hit-an-inanimate-corpse doesn't show up anywhere in the combat log, it's completely invisible. with expert mode off (obviously not a choice in the ultimate challenge), you do actually see the to-graze/hit odds when hovering over a corpse, which is extremely helpful (because sometimes one corpse will be a better choice than another simply because of their defenses... again, don't ask me how an inanimate corpse can dodge a spell)

for casual players - not having to worry about bonded grief i bet. it's a really rough penalty and very few other (sub)classes have something like that where it actively punishes the player for poor micromanagement or metagaming. but that nonengagement is sweet. maia has like the best ranger subclass and pet combo, and ghost heart gets you some of the way there. an itinerant was my second choice for a run-through (after a bunch of nerfing happened) and between speeding my AC past the front line towards squishes and potions of impediment and concussive tranquilizer i was a better mage slayer than the mage slayer subclass.

this got me real down at first (i don't pay attention to backer updates much, so i wasn't aware of all the changes coming down the line) - so the backer beta was a real rough experience for me as a priest-lover (not to mention the loss of interdiction as a global priest talent instead of a generic spell). fortunately the changes they made to the priest subclasses helped placate some of my concerns on spell diversity/selection constraints (for the unaware, priests [and druids] used to not get bonus spells, and priests were even more restricted based on keywords related to their diety... think wizard subclasses but worse, you could basically have like one or two valid selections for a given ability level depending on your diety).

this is sort of my fault. there was inconsistent targeting between the base white wurms, and the upgrade. the base worked like it does now, and the upgrade worked like in poe1 (making it an amazing upgrade). i raised the inconsistent targeting, and they "fixed the glitch" and removed the poe1-style targeting from the upgrade. sorry about the nerf on the flip side, like @Boeroer said, they compensate for it by making the base damage and aoe much, much larger than in poe1.

Is this related to just an urban vs non-urban (suburban, rural) preference? I'll say that I'm unabashedly into urban environments, which colors the fact of why I appreciate some RPG cities so much (esp Fallout 3's DC). If they eliminated a bunch of the quest density, I would still enjoy it on its own merits I think. To reiterate, Athkatla was great because there was just so much discoverability and interactivity. It was just an added layer was that it was a richly realized city environment (bridge district was hella cool when I was a kid, and it blew my mind to learn later that there are actually cities outside the US that have big thick bridges with markets on them). But if they could do the same with a rural environment, it would still be great in its own respect. In that respect, I would agree that Dyrford is pretty great - as far as quest hubs go it was deep, there were nice little narrative easter eggs (such as the tanner opening right into the skaenite dungeon and them having a hidden book on skaen). But cities are still better

Early on I preferred Baldur's Gate (B1) to Athkatla, because it was "seamless", like it was literally one large fort town divided into nine maps, so the city felt "truly" big. Over time, Athkatla has sort of become my gold standard because of how dense each area is - I remember going into random houses and being surprised at interactivity I could find (like that whole Katana sword thing in the Temple District, or the items in one of the random nondescript-yet-heavily-trapped houses in the Bridge District). I don't think any game has met that high bar since; PoE1 and Deadfire were kinda there, but I basically have no incentive to go into random houses in these games, just the ones attached to a task/quest or ones that initiate a task/quest. Though I think Deadfire did the best job (of any of the mentioned games so far) of giving me an incentive to go to random districts, because they didn't concentrate all the best magic items on one vendor. Can be a little annoying, but I like the wandering. Other contenders I'd like to include: Imperial City (Oblivion), really felt legit huge, even though I mostly just spend time where the stores are DC, all of it (Fallout 3). Not just the small pockets of settlers and merchants, but the entire ruins. I thought this was and is an absolute blast to experience. And there's a real "they made art through adversity" feel here, because the game engine couldn't handle large outdoor areas, they had to break up DC by sections of metro tunnels that connected them, but I ended up loving this aspect of it, made it feel like a real city with real transport routes; whereas Boston in FO4's subway stations went nowhere and were just self-contained mini-dungeons, you could actually study the metro map in FO3 and get out at certain points to different parts of DC (or the DC suburbs).

Angband,

literally when I was doing the similar-but-easier Ultimate challenge for PoE1 it was so tedious I started a completely different and parallel game of PoE1 just so I could play the game "for fun" instead of staring at my paladin very slowly grind out enemies for hours at a time (it also proved to be useful in doing advance research of areas, so there was that). Deadfire Ultimate wasn't as tedious, but only because once I got scordeo's edge fights were pretty fast and dynamic... also this time around I had a Netflix subscription so I could watch tv shows while my character was buffing for minutes at a time I actually made it through six seasons of Frasier just from watching it during buffs and rehearsals.

whew! just wanted to confirm that my run finally got validated. i don't know what i'm going to do with the patch... frame it? it's funny, i think obsidian managed to both overestimate the difficulty and underestimate the difficulty at the same time (e.g. staff writing notes to je sawyer that this challenge was fundamentally impossible vs the fact that there are so many patches and entries on their plaque, but i think i'll be merely the fourth to have done it, after six months of the challenge being out).

The thing about comparing games in the past to games now in terms of sales trajectory is that the past is an extremely different market. You don't actually have to go further than Hollywood to see how the entertainment landscape has changed. Some of the major movies of the 90s made their money over huge timelines, and doing such a slow burn was not that bad of an idea. (Titanic notably didn't sell out at first release, and only after the initial weekend did theaters start selling out, and it actually became more popular over time; its highest grossing week was two months after release. By contrast, the similarly big-revenue movie Avengers: Endgame made 1.2 billion in its opening weekend alone and then trickled in additional sales to slowly hit its record; they even did a marketing blitz just to shove a few more butts into seats near the end of its theatrical run so that it could legitimately claim to break sales records.) I don't doubt that smaller indie productions might have to rely more on the long-tail of sales, especially because they don't have the ability to make as big of a splash as AAA studios. But that doesn't change the basic economics that are at play in the 21st century with multiple million dollar budgets that need to be recouped pretty quickly before moving onto the next big project. I would say small budget PoE3 is the least-risky way to test the waters - try a new setting, keep in TB mode from the start, etc. If that also stays small, then the writing is on the wall for this kind of a niche game and we'll see the IP explore different genres. I like Boroerer's idea of basically a small-scale hackathon using the Deadfire engine, or JE Sawyer's own idea of a Pillars Tactics game.

it's actually extremely weird. The game still tracks the per-second duration under the hood, but then rounds at the very end. all durations less than 6s in RTwP become 1 round all other durations are rounded down to the nearest 6s increment chanter chants are specifically adjusted to have more meaningful durations for TB mode. This means there's a huge range where effects just last 1 round, from 0.1s to 11.9 s. It also means that intellect has non-linear returns. For example, with no PL Scaling, the difference in duration on "Eldritch Aim" between 13 intellect (11.5s innate duration that rounds down to 1 round) and 14 intellect (12s innate duration rounds to 2 rounds) is massive, but then you need a gigantic amount of additional intellect to make a further difference in duration (you would need 26 intellect to get up to 3 rounds, which is possible but not exactly a number a causal build can hit; though with PL scaling a three-round Eldritch Aim is pretty easy to hit, and a four-round Eldritch Aim becomes feasible). I think @Raven Darkholme did some initial numbers in a video somewhere and concluded that 15 intellect is a good "all purpose" target for durations, due to how base durations tend to cluster around multiples of 5. But for great success, you should actually find the abilities and spells you care about the most and do the numbers to see what you need to get an extra round or two of duration (keeping in mind that PL Scaling gives you multiplicative returns). PS. I actually don't know how effects like Wall of Draining or Salvation of Time or Ooblit work in TB mode (I have only slightly played around with TB mode). The game might literally track all effects internally as seconds even in TB mode and so those effects (which give +1s/tick, +10s, and +3s to effects, respectively) might give you partial round durations that can be combined with other partial rounds to give you additional round of duration on things. At least, that would be much better than the possibility that Wall of Draining and Ooblit do nothing whatsoever or that Salvation of Time has "wasted" effect beyond 1 round.

😮 is this tested? I always assumed it was just going to cast the Arcane Veil effect, not be a separately implemented passive item buff. sweeet jebus. could a wizard/priest of wael do arcane veil as well then (they don't clash?) for a stonking +100 deflection? in such a case a conjurer should flee by casting arkemyr's brilliant departure-... oh wait, oh no

don't ask me how a corpse has a functional reflex defense

an alternative to trying to keep your pet alive is to lean into it a bit by using Vengeful Grief (Nimble and Tenacious upon pet knockout). It created some funny incentives in some fights, sorry Ishiza. (Also, if you're keen to micromanage, Ishiza is likely the most tankiest pet, because they are immune to engagement and so therefore can just constantly run out of melee range as the enemy winds up an attack.)

a long long time ago i reproted a bug about white worms not always exploding a corpse, and i finally figured it out while playing with my bellower what exactly was happening: you have to roll to hit the corpse. i don't know when this was added in, but i'm pretty sure it's just a bug we are stuck with. fortunately since 5.0 you can actually see your chance to hit hte corpse (whereas when i first reported it in like 3.0 there was no indication you had to roll to hit). so white worms has a huge aoe, huge base damage, and a hugely long sickened duration (if upgraded)--not to mention on a bellower you get a crapton of PL scaling. but you have to discount it by whatever odds you see over the corpse (so 72% chance to hit is effectively a -27% multiplicative penalty over the long run). it is also secretly a poison/disease attack, so enemies can be immune to it. because of that, revenge can be a much better choice at times during combat.

i would kindly disagree here. It is true that conjurer doesn't get as much of a benefit from the +2 PL as a few other subclasses, there are still a few spells that get decent benefit (e.g. corrosive siphon), but more importantly their unique summon that gives you a guaranteed, stacking +1 PL bonus, along with some random other bonuses. you do miss out on illusion magic, but you still get arcane veil; but most importantly the survivability of a conjurer comes from reach (qstaffs and pikes) or range (kalakoth, black bow), not from evasion. i think the bigger deal is that a single-class wizard isn't going to help out too much for a dedicated spellsword; my argument that the conjurer doesn't suck is simply because you get a bunch of other stuff at +1 PL as well, which may not be "on theme" for a spellsword (though i think slickens and frozen pillars and even binding web are decent complement). and citzal's enchanted armory at AL9 isn't worth the payoff for a spellsword, except on lower difficulties, simply because the survivability of standing behind a tank while using a reach/ranged weapon is almost infinitely greater than relying on an autoscaling breastplate that won't help significantly defend against the boosted PEN of potd and maybe even veteran difficulty. that being said, if you're set on a single-class wizard, an enchanter might be a better choice. you get access to illusion magic as well, and the spells you miss out on (transmutation, evocation) are more likely the ones less "on-theme" for a spellsword. The +10% recovery penalty doesn't even hurt you too much because summoned weapons don't have any recovery. the class bonus is a little less good than the conjurer, but might be worth the tradeoff to pick up more "on-theme" magic from the illusion school. 1. intellect does influence how long things are summoned for 2. intellect will boost the AoE of kalakoth's minor blights and citzal's spirit lance, so can be relevant. it will also influence how long the debuff of black bow will last for. you can forget about resolve, unless you are going to go with boroer's suggestion with a shield. the way defenses work (especially on PotD) you have to invest in a lot to get good returns from them, which sorta necessitates a shield (which itself obviates the ability to use most of the summoned weapons). arguably you can worry less about constitution because you have Vital Essence at AL2, which is an easy way to get Fit (+5 con) for a real long time. you can also pick up Tough earlier as a single-class, which gives you a massive health boost. perception can be important, but a wizard can just cast Aware if they're having trouble landing some abilities. Has really short duration early on, but scales generously all the way to level 20, and can be extended with a Wall of Draining to last an entire fight. that sorta leaves dex, might, and intellect as important stats. Personally I would never dump intellect below like 15 for most caster-y classes and only go down that low if it can be justified (a conjurer can sorta justify it because many of their spells last so long innately). Dex is the king stat for offense - it makes your spell casting faster (still matters for a spellblade/sword-type because a faster cast time means less chance of being interrupted) and increases your net damage close to 3% per point (whereas might--because it's an additive bonus with so many other bonuses--is closer to 2% on average); so even though you can boost dex easily with Fleet Feet or Deleterious Alacrity of Motion it's still worth investing in a lot (versus perception or con). that being said, if you do want to use something like concelhaut's corrosive siphon or a draining spell like ninagauth's missiles, might is more important because it's more important to get more out of those spell casts instead of doing them slightly faster. if you put a gun to my head to decide right now, i would probably do a stat spread like 15m 10c 18d 9p 15i 8r, with any racial bonuses added in on top of that, and then choose whatever +1 stat from background you want that gives you might, dex, or intellect. yeah, this is a real fun way to play. if you don't want to do 2h sword, there are shields that do some retaliation attacks (not as much damage though) for an even more defensive build. just be careful about arcane dampeners

oh wow i didn't know that, thanks for that pointer

(sorry for triple posting). Yeah, players don't play separate tutorials. They also don't read manuals. (As someone who loved the thick Ultima manuals with their art and bestiary, the decline of rich RPG manuals makes me sad and is one reason why I backed so I could get a cloth map.) There's a reason why so much info gets stuffed into loading screens. Codexes are alright, but none of this is really relevant because at this point we're just talking about people who have already bought the game and are playing it. I'll admit I'm the worst judge for how well mechanics are explained, since I would be happy if the game just shows you all the equations it uses, so I don't know what necessarily puts people off or not. Though frankly if millions of people could figure out THAC0 in the infinity engine games, I don't see why they couldn't also get anything about Deadfire (except maybe inversions).

yeah, a lot of people say what Hulk-a-saurus say, but it's not true. The vast, vast majority of sales happen up front, which is why landing a game launch well is so important. It is true that RPGs and similar genres might have better long-term performance, but that long-term performance is a very slow trickle and will likely happen at lower average sales prices.

the more accurate numbers have been pulled from the fig financial reports and fig dividends, which are legally bound by a specific formula. it's only updated every six months, but that's where the sub-"500k-1million" numbers come from. steamspy is quite inaccurate and would also count copies that don't actually count towards revenue, nor account for sales from other channels (GoG mainly).

i always considered that ability pretty underwhelming, but on my current run with a bellower it's actually pretty nice. you don't get extra bolts or jumps (from upgrade) from the +bellower PL but getting the bonus accuracy, penetration, and damage is very good, and you could trigger it pretty frequently (up until my evoker picked up some level 6 spells the bellower was actually outpacing everyone else's total damage done). with sasha's upgraded to give you full empower points, it works really well on a bellower to go from a super-charged empowered revenge and then into a 6-phrase or 7-phrase offensive chant right after. Constentine seemed to clue into this a bit, but having used Revenge extensively now it appears to be implemented very weirdly. I think it's less like a minoletta's which is one spell with multiple projectiles, and more like one spell repeated multiple times as part of one "master" spell (which is why nowhere in the tool-tip ability summary dose it mention there are three projectiles, you have to infer it from the description). This is most obvious with the jump upgrade - the jumps will still work even if there's only a single target in the fight, you'll get 6 zaps on a single target; literally no other jump spell works like that. At this point it seems like one spell repeated six times (half of them slightly different). I think this is why it might be interacting the Least Unstable Coil in such a bizarre fashion.

There are certainly "interesting" tactical decisions to make. I have definitely (and this is probably not very common for players) used AoE to deliberately knock out my own party members because they had something like a druid regeneration active and I wanted to pre-empt them from hitting the "permadeath" range (e.g. eder is at 20/80 endurance but 40 health and if i wait too long the regen will have their endurance catch up to their current health). That is certainly an emergent tactical decision to make. But I think it says more about the relative solidness of the rest of poe1's system rather than to the actual merits of the endurance/health system; i don't think comparisons to real-world help much in eliciting the merits of a game mechanic. It was certainly interesting and good that I could do the above given the other parts of the system (e.g. I am personally a little saddened by the greater prevalence of foe-only aoe in deadfire), but the fact that I felt the need to is not.

I like Deadfire's approach in broad strokes because it "rationalizes" the system a bit more. I think the main problem was the instances where it was poorly balanced: for martial classes, most abilities should only be 1 resource with uncommon 2 and a rare more than that, which is generally true, but violations of that make the class seem overly constrained (corpse-eater barbarian and rogue mostly, paladin to a certain extent for all the high-level stuff but they could also regen their resource if single-classed). Also Brilliant was way better for casters than martial for a related reason. I think a way to fix the "ability economy" you mention would've been to let picking up a new martial ability grant you an increase to your power pool by +(ability_cost - 1). That way you're less incentivized to stick to a few cheap low level martial abilities all the time (especially for rogue where their abilities escalate in cost very very quickly), and each ability you pick up never gives you fewer uses of abilities (whereas in the rogue case, you could go from 5 encounter uses of crippling strike and/or escape to picking up shadowing beyond and suddenly if you use this shiny new ability you only have 2 crippling strke/escape left, for a net loss of two abilities per encounter just from trying to use a shiny new ability). From a "rationalization" perspective I also like Deadfire because sometimes the level-up choices in PoE1 seemed arbitrary in whether they were per rest or per encounter. Of course, Deadfire also threw that out the window a bit by having item effects be mostly per-rest but also sometimes per-encounter for Balancing Reasons(tm) *shrug*.