thelee

there appears to be a lot of places in game where the designers forgot that over-time effects also apply upon initial application, and not just every tick thereafter. This ends up messing up the calculus for all ApplyOverTime effects (of which I assume Unbending uses). (In particular it means that Disintegrate and Cleansing Flame in particular do way more damage than stated in tooltip.) edit 2 - fyi in case anyone isn't aware about applyovertime i added a new section summarizing some stuff https://gamefaqs.gamespot.com/pc/227477-pillars-of-eternity-ii-deadfire/faqs/76599/applyontick-vs-applyovertime edit - i just did a fight the other night in SSS (the "challenge" before you go find artifacts) and the fighter there had unbending. I had to get my characters to stop attacking it because they had accumulated enough stacks of unbending that they were healing literally like ~100+ health per tick (dude surged from near death to healthy while konstanten was trying to gear up his instakill invocation). I didn't think it was possible in typical situations for unbending to heal > 100% of damage dealt like that, but there you go. i suspect something funky might be happening when combining unbending stacks because that in particular seemed outrageous. on a related note (while we're talking about random mechanics) - refreshing defense and concussive tranquilizier interact in a way that favors refreshing defense. in that same fight, i could not clear the fighter's refreshing defense. concussive tranquilizer nominally clears 30s away, but with maia using a blunderbuss each hit with concussive tranquilizizer added a gigantic amount of time to the refreshing defense duration. mathematically, even if each bullet added 5s to refreshing defense and only the initial bullet cleared 30s of duration away, it still doesn't work out to how it ended up being (should've been a net loss of -10s to the buff). I have no idea how the mechanics of that actually played out. I ended up with like 500s of refreshing defense on that fighter despite all my attempts at trying to use concussive tranquilizer on him. if that's something you can dig into, that would be great

also an important thing to remember is that the cipher can't self-target with pain block. if you're running a vela challenge, you also can't target vela with pain block.

what choices did you make at Motare O Kozi? "Unfinishable" quests don't exist - if you made it impossible, it should become a failed quest. If it truly is unfinishable, it's a bug.

have you played much potd before? IMO you really need a healer class of some sort, so tekehu single druid would be a good pickup for fifth. Without it, all you've got left is some weak mid-late game minimum sustain from troubadour and mid-late game robust from ydwin, which combined imo is nowhere near enough, even with potions and scrolls. (especially because if you rely on a cipher for healing and it's not a psion you are utterly screwed when you don't have good PEN)

true - i did carve out important on-hit effects as exceptions where stacking isn't relevant (i cited toxic strike and marux amanth as big examples). i think this is a YMMV - over time the type of gear i use doesn't generally rely on that many important on-hit/crit effects so it's less of a huge advantage for dual wielding. it's also important to note that while two handed weapons have +1 PEN over equivalent one handed weapons in general, many two-handed weapons have no way of further increasing their PEN, only estoc. So in practice, one-handed weapons might do better in low-PEN situations than two-handed weapons. (though estoc is pen king). again, these days my parties tend to be so planned ahead in advance that i have ez solutions to PEN/AR problems instead of it being an ad-hoc problem to solve (e.g. some source of tenacious inspiration, potions, etc.) this is very true, but for me mostly means i don't use two-handed weapons (or atleast avoid using them) on casters or caster hybrids, unless i have a lot of action speed bonuses. i don't think this is that big of a deal for more pure martial characters, at least once you have a few levels (and therefore more health [edit: to withstand some sluggishness when you need to heal]) under your belt.

Two handed weapons have +1 over "equivalent" slow one hand weapons. Great sword does two damage types and does high damage, which puts its equivalent as the sword, which has 6 PEN.

was gonna say, streetfighter + beast of winter set sounds very effective with perception afflictions. i tried it myself, but my streetfighter playstyle was heavily based on flanked and <bloodied health (instead of perception afflictions per se) that it didn't work out as well i'd like (though it was pretty decent). blunderbuss modal would be pretty useful for this, but you'd really have to make sure that it expires before you refresh it if you want to maximize the beast of winter gear payoff (which is actually semi sub-optimal for a streetfighter due to how recovery bonuses have a delayed effect)

Ooo that's a clever use of that. I should do that, to get those annoying back-line enemies within easy reach.

precisely i've found that it seems like a lot of players don't bother with explicit interrupts of spells, at least based on some of the posts i've seen.

you don't need one tank to do all the engagement in the world, though it can be nice. I'm a little less sanguine about using terrify (or other effects) in place of actual engagement, but there are plenty of easy ways to get +1 or +2 depending on what you need based on how many semi-tanky melee-ers you have. my main concern though is that any tank multiclass that relies on a caster for survivability sounds extremely flaky to me against arcane suppression. it's do-able but i feel like players need to be aware of this going into it and that it requires a little bit more skill and metagaming to not just be utterly crushed in fights where a caster has access to arcane suppression (and very rarely, arcane cleanse) edit - though i'm not sure ironskin is suppressable? it doesn't have a duration.

Depends on your definition of "strong" I suppose. The metagaming possibilities of great swords are limited enough (e.g. no real degenerate combos) that even a single-class streetfighter would make for a powerful great sword build.

nope. there isn't an ultimate steam achievement. it's a whole other thing. people complained a lot in poe1 days about obsidian making nigh-impossible challenges steam achievements (some people just like to get to 100% achievements - full disclosure that is literally the only reason why I did the ultimate in poe1), so obsidian didn't put any of the really hard stuff as steam achievements for deadfire.

In the SSS DLC you can redo fights that include spirits.

that's a very good question and unfortunately i think needs to be tested (am unable to do so atm)

It's a very specific issue to the cavern of xaur-tuk-tuk and also one level of the oathbinder's sanctum. I have no idea why it's just those places. But basically if you happen to do a save/load cycle within those areas, it screws up the hostility of some of the enemies there. for the xaur-tuk-tuk case, when you try to leave the cave after having killed all you could, the frightened xaurip will run out of the cage and come talk to you, so everything can proceed normally, so nothing is really broken (you just avoid fighting a couple of random enemies). this is probably why it never got fixed - it's so low priority because hardly anything is broken because of it, that no one could be bothered in the little time they had left.

as an additional option, priest of berath gets a great sword spiritual weapon. It's potentially a +31% corrode lash which is probably the best brute-force damage great sword in the game. But no synergy with Ring of Focused Flames, which probably in practice makes Firebrand better (which has a +25% fire lash).

As the person who wrote that the key word is "generally". There are exceptions, annoyingly (chanter chants is another big one). In general if the gamefaqs guide mentions an alternate scaling, it was tested and explicitly different from normal PL scaling, such as the self-heal from forbidden fist. (Carnage is another one that doesnt follow normal scaling)

1. i think there are plenty of wonderful great swords. Twin Eels for example is excellent sustain in many fights. The real problem with great swords is the poor modal (as you mention) and their poor PEN - they have the worst of any 2h weapon, and unlike 1h weapons you have no modal to boost its PEN. On PotD this is a real big problem. I had great success with a devoted, the bonus PEN really helps. Outside of that, you should look towards parties and builds that help you mitigate the PEN and accuracy problem (so you can keep the damage modal on). Supporting/playing as a priest (champion's boon), chanter (shield cracks) are good, playing as a monk (thunderous blows), cipher (hammering thoughts, body attunement), berserker (frenzy) are all ways of helping with PEN, for example. (Edit: Oh, and boeroer's build. Like others mention it's really not that hard to get a huge deflection bonus temporarily with a modest deflection bonus more persistently. It's only hard/tedious to get that huge bonus persistently)

the damage penalty is only for dual-wield weapons, so two handed weapons are just as good as in poe1. Honestly, PoE1 should've also had some sort of balancing because dual-wield was just so good for full attacks. mathematically, a full attack with a dual-wield is almost equivalent to a two handed weapon; the nature of the damage bonuses and inversion of penalties are such that you pretty much have a similar effect whether or not you use 2w or 2h. It doesn't matter that rogues get so many additive damage bonuses because the penalty is essentially multiplicative. The main difference is that the two handed option will generally have +1 PEN over "equivalent" dual wielded options, and you have two chances to hit with a dual-wielded option. That may certainly matter for proccing something very important like Gambit, Toxic Strike, or Marux Amanth, but in general I don't think is a big deal otherwise. Math follows in spoiler tag.

You should read this and each megaboss subsection: https://gamefaqs.gamespot.com/pc/227477-pillars-of-eternity-ii-deadfire/faqs/76599/megabosses A scroll or two of maelstrom ain't gonna get you far I'm afraid.

I ran a wael+devoted+monastic unarmed training build and it was fun and pretty simple. It's kind of like a glass cannon build, but you trade off some of the high-end damage for easier survivability. Arcane Veil+cloth/leather+fighter regen is pretty good even on potd. Tactical barrage works great to boost arcane veil. Yes, champion's boon and mob stance work great in general and for this specific setup as well. Also you get +2 PEN which is nice as well. I think most martial classes will combine pretty well with wael. Only ones I'd shrug at are mage Slayer and Shifter. If you're doing a whispers setup, you could also do non-streetfighter-rogue with wael for extra riposte; the riposte chances stack multiplicatively.

B doesn't follow from A. Unity probably handles worker threads very badly, this is probably the most certain statement one could make. YMMV though. I tried special K/whatever on my AMD 2700x and didn't get much of an improvement back in the day. Most people are very very bad at properly measuring performance ("most people" also includes some hardware review outfits), so anecdotal stories of getting +10-15 fps are overblown to me since a bunch of other things might be different with your tests. I had gotten some initial gains, but with repeated testing it mostly seemed to be an artifact of having improperly measure cold-start-deadfire performance to non-cold-start-deadfire performance (and also confounding factors of things happening randomly in Nekataka). The more I tested it, the less it seemed like I got any gain whatsoever. (Though now that this thread has been necro-ed, I will probably give it a shot on my newer 3700x build just to see if it helps.) However, this has almost nothing to do with CPU overheating. Any competent PC build should be capable of dissipating the heat from 100% cpu usage, and the cpu should also be capable of thermal throttling (same goes for GPU). There's nothing special about Deadfire or Unity that would cause this. It's hardware problems, pure and simple. Saying that normal cpu usage causes cpu overheating is misinformation. The game won't run like ass on all high end systems. It's more complicated than "higher core count = bad". In terms of bad performance, I've seen reports from people with virtually identical systems as something I use, and I don't have the same problems. Saying "it doesn't happen to me" is extremely relevant in such situations, because it rules out software, or at the very least points to an underlying issue that is nonobvious and more difficult than "use this mod and it fixes everything." I've had Deadfire running on a haswell + 1060 pc and had no stutter issues (aside from the ones that always happen when you cast mirror image), so if other people with similar or better setups (including one person who had an ostensibly identical setup) are having stutter issues it rules out (obvious) software issues because the software is literally the same within a platform (e.g. steam literally verifies with checksums). That being said, this is not a defense of Deadfire or Unity - for one Deadfire has serious cpu+memory leak issues. I think the severity is actually a relatively newer issue than from release - I don't know when it got this bad. It was extremely noticeable to me during my Ultimate run - I had to explicitly save/quit before and after my Belranga attempt because the sheer amount of spider corpses in my fight would cripple my otherwise pretty beefy PC, and if I continued on without quitting the game after beating Belranga I'd be stuck with 10-15 fps in fights, even if I quit to menu and long after I had left the Belranga map. In addition, every time you do a save/load cycle, something isn't being cleared properly (this was extremely obvious back in the day when hsip to ship combat battles or audio effects would have weird post-load effects). On both my newer build (a watercooled setup from a month ago) and a slightly older build (2700x w/ vega 64), I get 100% gpu usage and >60 fps (hard to say by how much because my main display is an hdtv with a max of 60). However, the more I save/load, my gpu usage starts dropping and my fps also starts dropping - basically my performance is becoming increasingly CPU-bound and is throttling my PCs ability to display frames faster. It is definitely not temp throttling - especially in my watercooled setup, my CPU and GPU temps stay cool the entire time. All I have to do is exit the game and come back and everything is fine again. This doesn't happen in other Unity games I've played, there is definitely something spaggheti-code going on with Deadfire (as if patches 4.0-5.0 weren't clue enough about the spaggheti code). edit: tl;dr - instead of doing full VO they should have hired more programmers and QA testers. OR cut some features and have their existing programmers and QA testers have better coverage on the remaining features.

you could do it but i'm not sure what a multiclass monk gets you over most other multiclass options. for example, in my tactician/skaen build, i picked up monastic unarmed training anyway. For multiclassing, that's very close to identical to a multiclass monk's fists. i think the real wins come from single-classing monk and getting really powerful fists and also whispers of the wind (but i'm not a monk expert). if you're just interested in doing something different, though, go for it, like Raven Darkholme says priest is the most important part of the "standard" ultimate meta and everything else can be fudged. (unless you're super hardcore like the one guy who did it as a single class wizard)

1. Can you take a screenshot or a picture? I don't know what pop up you could be talking about. 2. There's no conspiracy to exclude anyone from credits ("purposefully omitted"). It's way too late to do anything about it now. You should have received a survey well before release to fill in what name you want to appear in the credits and, failing that, you should have emailed support (again: before release) if there were issues with that. Back in PoE1 in the "developer's commentary" they noted that the first tavern was a kickstarter backer reward but the backer went AWOL and they did it themselves, so the lore in that tavern talks about an owner that left and never came back (and that's how you can get that dog). I always wondered how a player could miss out on a reward they paid out for, but well I guess...

for extra credit, a lot of the builds that aren't explicitly touched recently or good for 5.0+ are probably still viable (only really gimmicky/metagamey builds or builds made for 1.0ish will likely have fallen out of viability), but it'd be a lot of work to go through everything to see what still works or not.