MaxQuest

It may not be as easy. Let's imagine the following situation: your character is steadily attacked by an enemy. His accuracy equals your deflection, and his PEN equals your AR. He hits you for 10 damage on hit, and you have 100 hp. On average he has 35% to graze and 50% to hit, so ~6.75 damage per swing. Now imagine you can take 3 extra AR (and reduce all incoming damage furthermore by 75%) OR extra deflection. Question: how much deflection do you need to achieve the same reduction on average, as those extra 3AR gave you? Answer: in this situation you would need ~52 extra deflection. Yeap, this would solve this problem.

Valid point. What's the level of broodmother suggested by Joe? It's around 9? Had played it, yes. Generally each point of Armor was providing an extra 6% to EHP vs physical damage (which was changed to 5% when 7.07 kicked-in). The thing is, it won't work that well in Deadfire, because: - it would render penetration weapons as completely subpar. - we would be taking quite a lot of damage even if our AR is much higher than enemy PEN. For example at AR-PEN = 9: damage going through = 1 - (0.05 * 9)/(1 + 0.05 * 9) = 68.9%

^ 43% vs 57% you politician!

That would be... a gargantuan task. I understand that devs can code the declination rules into the game, but there is no way it would take into account all exceptions to those rules... like: - Санта Мария: Команда Санта Марии готовится к столкновению. - Дева Мария: Команда Девы Марии готовится к столкновению. "Санта" is coming from other language - and is not getting declined... Btw, here are the current voting results:

Gonna give a somewhat contradictory feedback. > From one point of view it's quite important. If you try to imagine how much it would be annoying, think about the inappropriate usage of "your vs you're". > From another pov... it's something I personally don't really care about, because am not going to play Deadfire in Russian. I've tried localized PoE1, and didn't like it, reasons being: Now a few thoughts about what to do with all this: - v1. leave as is - kinda poor choice, but oh well... there can be more important features that require attention - v2. use workarounds - you can avoid the declination of ship name, by prepending the word "ship" before it (and declining it instead); or by omitting ship name altogether (by saying "our ship", "enemy ship", stuff like that) - v3. proper localization - particularly ask the player to input all proper declinations, and also let him indicate the gender of <ship name>. EN Shipnames: - Black Pearl: The crew of Black Pearl braces for impact. - Sea Wolf: The crew of Sea Wolf braces for impact. - Santa Maria: The crew of Santa Maria braces for impact. - The Virgin Mary: The crew of The Virgin Mary braces for impact. ^ all names remain the same. RU (same shipnames as above): (v3 approach) - Черная Жемчужина: Команда Черной Жемчужины готовится к столкновению. - Морской Волк: Команда Морского Волка готовится к столкновению. - Санта Мария: Команда Санта Марии готовится к столкновению. - Дева Мария: Команда Девы Марии готовится к столкновению. ^ all names were declined, and use different suffixes RU (same shipnames as above): (v2 approach) - Черная Жемчужина: Команда корабля «Черная Жемчужина» готовится к столкновению. - Морской Волк: Команда корабля «Морской Волк» готовится к столкновению. - Санта Мария: Команда корабля «Санта Мария» готовится к столкновению. - Дева Мария: Команда корабля «Дева Мария» готовится к столкновению. ^ all names remain the same, but you have to add the declined form of word <ship> before them Btw, adjectives do change depending on the gender of the ship as well: P.S. If you go for proper localization, the best way imho is to ask the player to input the declined names himself. P.P.S. Aramintai you ninja

The thing is you have to double their AR in order to over-penetrate. There is a lagufaeth with 9 PEN. On crit it would be 13.5, and it's not that hard to achieve 16 AR, or lower their PEN. P.S. Btw what is the hardest encounter in current beta?

Weapon-based priests lost the healing efficiency (since RES is likely to be dumped in favor on STR), while Buffing-offtanks (that had low MIG, high RES) gained the healing efficiency. True that. In PoE we have to specialize into: - buffer + ~offtank (like 2/12/18/10/19/17) vs - damage_dealer + heals (like 18/9/17/12/19/3) In current beta we have to specialize into: - weapon_damage_dealer + [buffs if learnt] (like 18/12/17/10/18/3) vs - spell_damage_dealer + [heals if learnt] + [buffs if learnt] (like 3/10/18/10/19/18) vs - healer + ~[spell_damage if learnt] + buffs if learnt] (like 3/14/18/6/19/18) vs - buffer + ~{[offtanking] or [weapon_damage] or [spell_damage and/or heals]} (like 3/14/18/6/19/18) In my current oppinion the main limiting factor to versatility is the amount of spells we can learn. I am not sure how I feel about two different attributes being responsible for damage though... Yeap, personally I haven't spent much time healing as well. Most of healing was coming passively from Shod-in-Faith boots, Beloved Spirits and Silver Tide racial (had chanter and priest moon godlikes with maxed MIG/INT). Ofc had some Moonwell scrolls and Healing potions for emergency, but almost never used them. Fire priest gets buffed though. Because he completely dumps STR. And: - if he remains ranged - he remains as before but with extra 15 deflection - if he goes melee - he can also passively offtank (provided a bit of PER will be traded for CON) And just in case: no I am not going to be biased because I've published that build )) I was giving it to my primary dotter... since MIG coefficient was multiplied by INT coefficient for the total damage calculation. 20 -> 24 MIG, or ~ +9% in healing (x1.42/x1.30) would not make much difference. But yes in Deadfire it would be wasted on healer, and on dotter. It would go to someone with high INT AND accuracy, in order to make use of the aoe stun.

I have a feeling that not only resolve became less useful in Deadfire compared to PoE1, but also deflection itself. In PoE1 rising defenses in general and deflection in particular was an important part of downgrading the quality of incoming damage (crit->hit->graze->miss). With DR being additive it allowed for a larger amount of damage being prevented by armor. And viceversa, if you had low defenses - your armor was less effective: Having to endure an average crit compared to average hit was a big deal. While at the moment it isn't. If you character is a melee phys-dps'er and your AR > PEN + 2, from defensive point of view you can easily offset 15 deflection from resolve, by just 5 extra CON, and put the now free points elsewhere, specifically in MIG or DEX in order to end fights faster / offset the actions slow-down. Thoughts?

Limiting buffs to be cast during combat only, allows making them more powerful, since there is also opportunity cost at hand. On the other hand though... spells in Deadfire have a noise parameter. I wouldn't be surprised if at some point, we would be able to start buffing before actual combat start, rather when we spot the enemy. And if we use something loud - they will come to investigate. Makes sense. You either: - disable them, or - gather enough concentration, or - bring enough casters, such that they would find it hard to interrupt them all Btw, there have been mentioned 'ritual' spells in this thread. Just a weird thought or rather association, as it makes me think of several robed figures drawing pentagrams together: a ritual could be an assisted spellcasting. E.g. a priest starts casting a very slow spell (let's say Storm of Holy Fire); and another priest (if he has access to the very same spell) can assist him, which results in twice faster casting speed, but at the cost: it's enough to interrupt only one of them - in order to interrupt the spell/ritual.

Roger that) Agreed for druids. But I've already included a point about hybrids. As for priests: Maybe they will. But so far we can theorycraft further. It's entertaining) Another approach could be: think of possible playstyles and character roles inside of party. And adjust stats in such a way, that would empower roles that are lacking something. Let's come to a common definition of caster. For example by casters I meant: wizards, druids, priests, ciphers and chanters. - So with the new system: RES is useful for casters that cast spells; STR is useful for casters that deal damage with weapons. - With the old system: MIG was useful for all of them; while RES was useful only for melee ones (and even than resolve impact was quite marginal, because 15 deflection doesn't really make a lot of difference, unless you have a lot of it) Or you use another definition for 'casters'? Like any character, who is focused exclusively on actual casting? (including support paladin and scroll-rogue?) I did. I wanted to accentuate that Josh has emphasized RES as not being useful for ranged casters, when in fact, it wasn't useful for any ranged character. But now that you mentioned it, I see that it was confusing. Is it better this way?: before / after Nope. That's why I was asking if I've missed something. Although I have a feeling that 1st (now 2nd) point is related to RP perspective. I will edit it to make more clear.

That's not fair) Agreed. High defenses (especially if coupled with high outcoming damage) do indeed add to challenge. So do pre-set enemy immunities as this stresses your usual routine and forces to adapt and be ready for plan B. Although to be fair, casting times and enemy interrupt abilities are also part of challenge. For a solo'ing priest the hardest part was to self-buff in peace, before he can unleash all his damage potential. Also remember those abbey monks and their prones and interrupts. Well... PoE1 casting durations were mostly enjoyable. In current beta though not that much, as most spells/powers do not feel like they are worth the spent time. I doubt the changes in casting durations had disruption as their goal. The changes are likely being dictated by the new interruption system, and the fact that now vancian casters can throw 2 of their most powerful spells per rank in every encounter without having to worry about spell conservation. Extra: - Casters have fewer spells to cast now. Once they throw them all, they are dry. - Interruption is easier and clearer to see if there is a decent time window during which you can interrupt. - Also random incoming interruptions were making our characters effectively cast slower. Now that they are gone, cast durations were made longer to match that effective average. But on the other hand there were players who were using preemptive hard cc, and thus avoiding all incoming interruptions and thus having no cast slow-downs. So imho, the spells were over-slowed, and especially cipher powers, since ciphers didn't get the ability to cast 2 powers per rank per encounter for free. Well, the main goals is indeed enjoyability, which comes from fun + lack of frustration. And frustration comes from stepping into a trap option, i.e. making a choice that after some time proves to be unviable, or that makes your character a liability to party. A little reasonability and consistency helps when it comes to estimation, spell comparisson and thus avoiding such situations. Also it reminds me of my last party: I had a priest with 18 dex, and wizard with 17. And when issuing Painful Interdiction with Shadowflame at the same time during pause at the start of combat, I could be sure that enemy defenses would get debuffed before the check for paralyze. I mean durations weren't jumping around, and I could get a consistent result. Hehe, it happens)

I hear you, and been thinking to note down all pros and cons. At the moment it's something like that: Do you think I have missed something? Also, I have noted quite a few "the old might was good; just give resolve something useful for casters"; and have briefly brain-stormed a few possible things that could be influenced by RES: Concentration related: - generate 1 concentration every 100/RES seconds. - 1 concentration is generated every 8s. 1 point of RES increases the speed of this generation by 5%. Penetration related: - you act with higher determination. Each point in RES increases your PEN by 3%. - 1 determination point is generated every 5s. 1 point of RES increases the speed of this generation by 5%. Determination points are automatically consumed on next offensive action, adding 2 PEN. Power level related: - you act with higher determination. Each point in RES increases your effective power level by 3%. Affliction/Inspiration related: > each incoming hit or crit, drives you closer to getting resolve (rank 1) affliction (shaken) > each party kill, drives you closer to getting resolve (rank 1) inspiration > resolve is used slow down the first, and speed up the second Defense related: > high resolve converts part of incoming hits to grazes > low resolve convers part of incoming hits to crits Defense and offense related: - each point of resolve converts a part of incoming attacks one category down (crit->hit->graze->miss); and a part of outcoming attacks one category up. Also, we could technically separate spells in: - damage: uses MIG (RES), PER, INT (sometimes), DEX - heals: uses MIG (RES), INT, DEX - cc: uses PER, INT, DEX - buffs: uses INT, DEX and think if it's feasible to make resolve increase the effect of buffs.

Yeap, as offensive side: we don't want to get interrupted. But if that spell is sooo good we can still try and cast it. And in well-balanced party there should be someone who will help with concentration, or will assist with an AoE hard disable. As defensive side: we want to interrupt enemy spellcasters. If they would be able to unleash all their potential instantly... it would hurt. In PoE1 we had many spells at our disposal, and it was a great feeling that I have 5 different general strategies, and that in this given situation I can use this approach or this one. Versatility. In Deadfire though... dunno, we'll have to see. But I can imagine a playstyle around: - send a tank to gather enemies - start casting a slow heavy hitting aoe spell that would seal the deal in one crit - toss a mass disable, and relocate the tank right before impact My bad. Think of something between OOMPF and KABOOM! Kidding I wrote BAMF just to denote a heavy hitting spell, as that was the first association I had when thinking about the sound it makes on impact.

^ Not gonna bet) but I would expect him getting close with Xoti, if the player helps them synchronize their religious beliefs.

If I understood right, Josh wants some spells to be 6s because it increases the time window when you can interrupt them. And because of this drawback these few spells can be buffed for the BAMF effect, making interruption a really important aspect of the game. At the same time, imho there are no spells that worth of 6s cast duration yet. But imagine if there was something like... AoE Disintegration or Overwhelming Wave with the radius of Interdiction. - if all spells are instant - spells would become OP also it will be hard to follow what's happening on the field. - if spells have instant cast duration and long recovery - this would severely disfavor casters from wearing armor - if spells have long cast duration and instant recovery - there would be no reason for casters to not wear the heaviest of armors - if spells have long both cast and recovery durations - chances are that fights will be over before the casters will be able to adequately contribute to the fight Tbh I like them in the same order, and for the same reason. In PoE1 I avoided casting slow spells, because 4s was literally too slow. On the other hand, since in Deadfire there is no Slow mode at the moment, plus we have a new interruption system which is reliable instead of being chance-based, having 6s cast duration for a few aoe damaging spells (provided that their damage is proportionally buffed) can be quite ok, especially if they deal direct damage / have instant effects, like hilfazer wrote above. So very-slow (6s) is being reserved for them, just like Andrea noted: I would gladly have something like: - Fireball: 3s cast duration + 3s recovery - Let's say Meteor (think of fireball with x2 damage): 6s cast duration + 3s recovery - Mental Binding: 3s cast duration + 1.5s recovery - Whisper of Treason: 1.5s cast duration + 1.5s recovery - Summoned Weapons: 0.5 or 1.5s cast duration + no recovery - Thrust of Tattered Veils: 1.5s cast duration + 3s recovery

Do I understand it right?: in order to match final result indicated in tooltips: speed coefficients need to be added up, their sum converted to duration_coefficient and after that base recovery value is divided by it?: A = 0 + 0.5 + 0.2 = 0.7 B = 0 a = B - A = -0.7 => coef = [0.7 / (1 - 0.7)] = 2.333 => 2.0s / 2.333 = 0.857s ~= 0.9s (which matches SaruNi's experience) If so, there must be something off,.. because if those B and A are equal (e.g. situation when there are no buffs) coef would be zero instead of 1, leading to infinite recovery when in fact it is 2s. Not sure yet, but have a feeling that tooltip result does not reflect real value of current recovery. The source code for recovery calculation is still the same in Deadfire as it was in PoE1. And if I do the computation in the same manner as before, I get the value of 0.6s:

I agree that 6s is horrendously slow.But am still not sure how I do feel about it taking just 0s. I was thinking about it taking either: - 0.5s or - 1.5s, provided there is also some new talent. Let's say Battlemaster Conjurer: summoned weapons take 1s less to conjure and refund the spellusage when they expire. (or even make them not use spellusages at all) Reason for it being 0.5s instead of 0: - it allows for showing animation - it helps against accidental double-click - it adds some opportunity cost, even if minor Similar thing about wizard self-buffs, like: Mirrored Image, Wizard's Double, Llengrath's Displaced Image, Llengrath's Safeguard, Arcane Veil, etc. If it was 0, chances are we would just go for subclassless wizards, and set AI Behaviour to use them all at once at the start of every fight. I was thinking about having a few consistent "casting" categories (like instant, fast, average, slow) with recovery duration matching the casting durations. Examples: And then having recovery modifiers indicated in spells tooltip, that would proporti8onally increase or decrease the final value of recovery time. For example I think that interruption spells should be really fast, but since they have an utility component, means their dps should be lowered down a bit to compensate, either via base numbers or via longer recovery. At the same time, the recovery of invocations and cipher powers could be lower than of equal spell, because they have a cost attached related to chants/focus build-up. Also chanting is stopped mid invocation recovery. I haven't thought these deep yet. But the idea is that: - something of weight 0 - can be very-fast cast - something of weight 1 - can be just fast - something of weight 2 - can be average Generally the higher the impact a spell has - the slower it should be. But need to take also cost and ranking into account...

I wonder who's gonna be gay this time. Not gonna be surprised if it's the most popular and funny characters as always, just to screw straight players. Heh, as a straight male player romances often screw me because my mains are mostly female characters. That's also why I prefer female companions to not be straight, but bisexual. Am pretty sure am not alone in this. And here is my logic around sex selection: On a mildly related note:

Just want to mention that actually the former was presented by Josh as main motivation. In his tumblr post he mentions: And in his twitter post he makes emphasis on ranged casters: Also, resolve is still a poor stat for ranged rogues and rangers (not to mention ranged fighters, monks and barbarians; although these don't really apply).

Yeap That example is from Beta 1. I don't know if soul and bitting whips stack now, but they 100% did back then. Example: image 16.2 * (1 + 0.5 + 0.2 + 0.4 + 0.2 + 0.15) * 1.15 = 45.64 ~= 45.7 I was thinking about additive all the way 19.6 * (1 + 0.4 + 0.2 + 0.15 + 0.18) * 0.5 = 18.9... but that's still off from 18.5 That's pretty accurate. And yes, graze does not lower PEN.

Yeap. Crit damage bonus is additive. As for graze, it looks like a multiplicative malus: example. But if might is multiplicative (see below) something doesn't add up. There is no easy answer to that, because we have several sources that seem to contradict each other. > According to UI tooltip steps: all damage coefficients are additive. > According to Source Code (assuming I understood it right) all damage coefficients are additive. > According to UI tooltip result (if we try to figure out how the end result was achieved): Might is applied multiplicatively, while all other damage coefficients are applied additively. And BMac confirms that this indeed is the case here, and tooltip steps - being just an UI problem. Funny enough, none of the above would explain the graze above. In this case looks like all damage coefficients (including Might) are additive, while Graze itself is a multiplication by x0.5 Agreed that both graze and crit should be applied multiplicatively. This is also reflected in suggestion v4. As for perception - it was a flat stat for dps'ers in PoE1, and a bump stat for debuffers because it made cc reliable, and helped a lot in bringing defenses of high-def bosses down. And in Beta 2 there are no bosses, no Broodmothers/Battery Sirens, nor cc is worthy the cast time yet.

Since majority in this thread agree that spellcasting should be faster, there is also another question: - should there be a few consistent spell categories, or it's ok to have a ton of them? And in order to start from something, here are the values for PoE1, and Deadfire Beta 2: Notes: - as you can see, spells in PoE1 did not always obey their displayed speed category, yet they still were more consistent than in Deadfire Beta 2 - unlike in PoE1, in Deadfire some instant spells can be used during pause (for example Disciplined Strikes and Second Wind, but not Swift Flurry)

You are using two fast 1h weapons with base values of 0.5s attack / 2.0s recovery, right? P.S. Looking at AttackBase.NotifyAttackComplete() I see that nothing has changed in recovery calculation logic in comparisson to PoE1. Well except for recoveryFactor which is a constant, and allows for a different ratio between attack and recovery durations. But still something is odd, specifically base values and rounding: - fast melee weapons have base 0.5s att and 2.0s rec - slow melee weapons have base 0.7s att and 3.0s rec which is technically impossible, since they both go through the same function and thus should be proportional. My current guess is that fast weapons actually have 0.46(6)s attack duration (or exactly 14 frames). So it's likely is: - fast melee weapons have base 14 frames att and 60 frames rec - slow melee weapons have base 21 frames att and 90 frames rec Will be able to tell precisely when will find the value of new recoveryFactor. SaruNi is speaking of recovery duration) Dexterity bonus is applied multiplicatively. Both in PoE1 and in Deadfire. Also DEX does not allow you to reach 0 recovery... but huge numbers can bring you close to it. Specifically if you are at 0.0499 recovery it could be rounded to 0 by UI. So let's compute how much DEX would we need for UI to indeed display 0 recovery duration for a fast weapon (provided there are no other bonuses/maluses): 2s / dexBonus = 0.0499s 40 = dexBonus 40 = 1 + (dex - 10) / 33.3 39 * 33.33 = dex - 10 1298.7 + 10 = dex result: 1309 dex or above

Are you sure that's a good example? I am interested in the arguments presented in that video. Can you share the link? This does not reflect the current state of the things. Not to mention that youtuber is speaking in absolutes. Hmm, it sounds a bit black & white: - community feedback being impure - and devs initial vision being pure If Obsidian is adamant about any system/change they implement, they would not need our feedback, nor this beta thing. It would be obvious that they would like to not "taint" the upcoming masterpiece, and just present it when it's completely ready. Forced?

I suppose that it's not that testers really hate the beta (if you saw any hate)... rather there could be subtle feeling of potential disappointment. How to put it better... Look Deadfire has a bigger potential than PoE1, because: - it is bigger (sequel adds many new features) - devs are more experienced now, and are already familiar with many possible pitfalls So it is natural to expect for Deadfire >> PoE1. And having such expectations is ok. While the beta itself atm is: Beta << PoE1. Which results in that little bitterness. But is actually also ok, since Deadfire didn't come through a comparable number of iterations yet. Good question, and you are on point with the split feedback. As I see it there are two general approaches to balancing: #1. You create a detailed mathematical model from the very ground. You start with a set of axioms that list what you want to achieve and build a pyramid from bottom to top. It's balanced, it's stable on every stage of development. It maybe scallable, but you are limited that you can't change the grand blueprint. #2. You have several nice ideas, that are cool but not necessary balanced or thought to their depths. You translate each into a corresponding system and throw it into the common cauldron. After that you try to figure out what's ok and what is requiring tuning. You test->estimate->tune->and repeat. In his presentation here Josh is talking about iterative approach. So it's closer to #2. They test, estimate, compare their own conclusions with collected feedback and implement changes. With each iteration getting closer to an etalon state, just like getting closer to solution in bisection method. Experience, vision and access to more info, really help in sorting the feedback, and estimate which one was off, which one was on point, which one was short-sighted (for example because the player didn't take X or Y in account). We all do)