Project Eternity Social Round-up

I'd like to apologize to our readers for taking such a long time to put together another social round-up for Obsidian Entertainment's upcoming Infinity Engine-inspired Project Eternity, but hopefully the wealth of information project director J.E. Sawyer and the rest of the crew have been releasing on the official forums and other social media will make up for it.

First of all, producer Brandon Adler and project director J.E. Sawyer have given some vague snippets of info on the belated fulfillment website:
It is close, but we aren't ready to talk about it. Once everything is together we will give a larger update about it.

I can't go into all of the details, but once everything is ready we will present to the public fairly quickly.


Darren Monahan (one of the owners) is developing it along with Brandon Adler. It's going to be more expansive than we originally envisioned.

Project director J.E. Sawyer also discussed update #61's screenshot at length:
How will the shadows/lighting work in the final product? The environment is gorgeous, but the conflict between the apparent lighting of the scene and the way the character models are illuminated makes the latter seem out of place.

The character models only have one shadow cast from the scene's directional light. Per-character point light shadows (especially multiple point light shadows) would get computationally expensive very fast. The characters are lit using dynamic lights placed in scene, but real-time lighting is never going to perfectly match the pre-rendered lighting of the scene.

Yeah... umm... I hate to do this, but you've asked us, so here I go: You've stayed very true to the concept art, and that I love. Still, in my probably not so humble opinion, the final in-game render is somehow lacking in the shadows department. It seems all tad to bright, and that takes a bit of mystery from that particular scene

The stronger/less diffuse our environment shadows are, the more difficult it becomes to place characters in the scene and not have them stick out. We tended toward relatively diffuse shadows in the IWD games for that reason as well.

Looks amazing! My only (minor) concern is that the individuals inside the room are visible, even though the party is outside of the room, with the door closed. I'm guessing that fog of war/mob detection is not in yet? :) 

The back end/data for fog of war is implemented, but not the rendering side. Also, if the room were obscured by fog of war, that would be a pretty crummy screenshot.

All of the cloth parts on the party members (and the enemies, actually) have tint maps applied so you can go buck wild with colors if you'd like. E.g. the character at the center of the party in red brigandine. We initially set him up with mustard yellow brigandine and purple pants. He certainly stood out, but looked pretty garish. 

Ultimately, those color choices are all up to you. Companions will come with defaults that we think look good, but if you want to alter their primary/secondary colors, you can. We're trying to always leave cloth areas on the arms and legs (at least) to use the primary/secondary color channels.

Oh, one other thing: everyone in that screenshot is a human or elf. We have orlans, aumaua, dwarves and (soon) godlike models as well. Since this was mostly about showing the environment and some of the new armor types, we didn't switch out the races. 

How are you guys handling cover? I don't mean tactical cover, I mean buildings and things that the characters walk behind. Do they become transparent when the party walks behind them, or does the player just lose sight of the party? The strategically-placed broken column made me wonder. 

Portions of the character become translucently rim-lit with their selection circle color when they move behind occluding geometry.


How does the door in the screenshot animate when it opens? Will it be like the IE games where the door just appears open on the other side or will it actually fling open?

Animated doors are 3D objects and will open "smoothly".

Ropekid, will there be capes/cloaks? What about various idle animations?

We have implemented cloaks but I think we're still debugging a few things with them. I don't want to commit to having them until we're sure they won't cause problems.

Fidgets (we call the looping stance the "idle" and the occasional gesture within an idle a "fidget") are scheduled but are lower priority than a lot of other animations. Seth got in our combat stances recently, which are pretty nice.


We use a 2D depth map render that is separate from the final/beauty render. The depth map has the same resolution as the final render and allows the in-game renderer to determine the depth each corresponding pixel has from the camera. Because it's an isometric render (no parallaxing), it simply does a per-pixel comparison to sort the scene.

Some information on class design, also from project director J.E. Sawyer:
Level is a big factor in your total defenses, but the character's class determines the starting point of each defense stat (which can be further modified by attributes, spells, abilities, talents, and equipment). For example, fighters start with the highest Deflection score and they maintain that advantage as they level up. If a fighter really wants to focus on holding a line in melee over doing damage, he or she can equip a shield and gain an even larger Deflection bonus. Unless you're higher level than the fighter, it's very unlikely that your Deflection-based attacks will come close to his or her Deflection defense, meaning you'll wind up missing a lot more than 5% of the time -- and it will probably be impossible to crit them. If you want to hurt fighters, use attacks that target Reflexes or Psyche, which are their weakest base defenses.

Most classes have at least one ability or spell that shifts the defense they are targeting with standard attacks. E.g. barbarians have a Brute Force ability that allows them to temporarily switch over to targeting Fortitude. Against our sample fighter, that would have the two advantages of ignoring the target's high base Deflection and ignoring the bonus provided by the shield. As another example, druids have a spell called Firebrand that creates a weapon made out of pure fire. In addition to doing only fire damage, it targets the Reflexes defense.

So, while it's true that an equal defense and accuracy will result in a 5% chance to miss, it's rare that defense and accuracy will actually be equal on any given attack.


I was wondering if you could, perchance, provide a little more detail on the new-character class bonuses, and how they'll potentially affect things as the character progresses throughout the game (in the grand scheme of things), as there was some concern that a simple permanent +3 Deflection for a Fighter (for example; don't know if that's accurate to the actual game's design) would be inconsequential when it comes to Attack-vs-Defense scales. It would be very much appreciated, if you have the time.

In D&D, +3 is still +3 even when other bonuses from other sources enter the mix that makes up the aggregate. Even if it comprises a smaller proportion of the total, it's still valuable. All other things being equal, a class' starting defense bonuses will always shift their overall balance. A 5th level barbarian may find a number of ways (items, spells, etc.) to get his or her Deflection on par with a 5th level fighter's, but given access to the same methods, the fighter would still maintain his or her class advantage.

In PE, for any standard attack, every 5 points of defense translates to a 5% shift to miss, graze, hit, and crit (sometimes negating the possibility entirely). In D&D terms, a 15 point defense advantage is similar to +3/-3. Shifting the odds always helps.

If you're comparing it to D&D, this is a similar progression except for two differences: 1) we maintain differences between classes (given equal level) as an integer rather than as a proportion and 2) because we use a 100 point base scale instead of a 20 point base scale, we have finer control over per-level advancement.

E.g. in Pathfinder, a fighter starts with +2 Fort, +0 Ref, +0 Will. They progress like this (Fort/Ref/Will)

5th +4 +1 +1
10th +7 +3 +3
15th +9 +5 +5
20th +12 +6 +6

A rogue starts with +0 Fort, +2 Ref, +0 Will. They progress like this (Fort/Ref/Will)

5th +1 +4 +1
10th +3 +7 +3
15th +5 +9 +5
20th +6 +12 +6

It's a regular progression: they start with an advantage in one (for some classes in D&D, two), and advance at regular intervals. Of course, at high levels this falls apart because the proportional gulf between good and bad saves becomes so wide that characters have to overcompensate or inure themselves to specific effects to avoid being sucker punched. At 1st level, the difference between the good and bad saves is (effectively) 10%. At 10th level, it's 20%. At 20th level, it's 30%.

To make matters worse, typically the obvious "good" stats for a given class reinforce the better saves and neglect the worse saves. Fighters often have a high Con, which means they are likely to have an even higher total Fort than normal. They might have a decent Dex which can bolster their Reflex, but it's rare that they have a high Wis (and consequently, Will save). Because save DCs are often balanced around the "hard" targets, it means that the weakest saves of a class combined with the weakest (or least important) ability scores for that class make them really, really vulnerable. This is why Pathfinder has a special Bravery feature for fighters at higher levels -- otherwise they'd run or freeze in terror more than half the time a comparable caster chucked a Will-based fear effect their way.

In PE, fighters start with the following defenses: 25 Deflect, 15 Fortitude, 10 Reflexes, 10 Psyche. Rogues start with the following defenses: 15 Deflect, 10 Fortitude, 30 Reflexes, 5 Psyche. Every level, every character gains +3 to all defenses. At 6th level, the fighter would have 40 Deflect, 30 Fortitude, 25 Reflexes, 25 Psyche. The rogue would have 30 Deflect, 25 Fortitude, 45 Reflexes, 20 Psyche. The fighter's worst defenses are still Reflexes and Psyche, but they're only "just as" bad (by the same margin) as they were at 1st level. The same applies to the rogue's Psyche and Fortitude. And while the rogue did "catch up" to where the fighter's Deflection was, the fighter maintains the same 10 point advantage over the rogue that he or she did at 1st level.

Of course, Attributes, gear, Talents, Abilities, etc. also can all feed into your defenses, but those are much easier to switch around than your class and level. I.e., if you find yourself particularly vulnerable to a particular type of attack (meaning, what defense it targets), the cause is likely easier to remedy in PE than it would be in D&D (because class is such a large component of that value as levels rise).


Classses are like Careers in real life. They're not supposed to be equal and balanced. They're supposed to vary wildly in terms of power.

In D&D, you have a class like Mage. Very weak early on, then they evolve and become the most powerful class. And you've got Warrior-types. Their progression is Slow and steady. They'll never be the weakest class, and they'll never be the most powerful class. They'll be something in between from beginning to end.

This is as it should be. I shudder at the thought of some boring, soulless, robotic system where everything is equal, thereby removing all motivation to replay the game (after all, why bother Playing a warrior for your new playthrough, when you know that the game's challenge will be exactly the same as it was when you played as a mage?)

If classes excel and suffer at different elements of gameplay but are still roughly as viable as each other throughout the game, the challenge isn't the same at all. "Balanced" does not mean "the same". Playing different classes should make challenges out of obstacles that were previously easy and turn cakewalks into struggles in equal measure. If you just want the game to be harder overall, that's what the level of difficulty is for.


When you say "high maintenance," I get the worrying idea that if I have wizards in my party I'm going to have to spend most of my time controlling them instead of my character or someone else "lower maintenance," unless I specifically spec them to have fewer active abilities. Can the AI handle running active abilities without my constant guidance so I can pay more attention to other classes if I want?

I don't particularly like playing wizards in DnD games because of their often cludgy rules and whatnot, but playing Icewind Dale I had to spend most of my fighting running the wizard, sorcerer and cleric instead of the other guys. I would like to avoid being obliged to "play" the wizard by default, regardless of my own character choice, so I'd really like it if the AI of allied NPCs can handle its own "maintenance."

I apologize, but that is the way that the game is likely to play. Part of the IE (specifically, BG and IWD) combat experience involved higher micromanagement of spellcasters and we believe it's something the majority of our backers enjoy. We want to put in automated controls for things like auto-pausing and similar high-level functions, but we don't want to transform tactical (i.e., in the moment) decision-making into strategic (i.e. pre-conditioned) routines.

That said, our lower levels of difficulty will not require as much micromanagement overall, and we have included the slow combat mode if you'd like more time to think without constant pausing/unpausing.

Well, can I instead get by without wizards or "high maintenance" classes? I mean we've got 11 classes to play with, so someone has to be left out, right?

Sure. You could double up on certain classes or ignore other ones (like casters) entirely. A fighter/paladin/monk/barb/chanter/rogue party would be a lot lower-maintenance than a caster-heavy party, all other things being equal. Of those, the paladin, monk, barbarian, and rogue are all moderate-maintenance classes by default.

Heavily-armoured wizards are definitely supposed to be viable, as are active-ability melee classes, but I don't think it's been mentioned as to whether it goes as far as wizards being able to wade into melee or fighters being able to sling fire.

Fighters can't sling fire as part of their normal class Abilities, but they will probably have some optional Talents that allow them to use magic.

Wizards can wear armor and hoof it into melee if you want to, but they aren't particularly well suited to it unless you're really prepping for it. If they're unprepared and are tangling with a "real" melee opponent they get badly beaten pretty quickly.

Don't worry; we will support gishes. Just don't expect them to beat fighters or rogues at their own game (i.e., absorbing attacks and doing a ton of single-target damage, respectively).

The goal of wizards in PE is, ultimately, flexibility. If you need/want someone to help the melee dudes, a properly-grimoire'd wizard can do it. If you want to cause a bunch of AoE damage to soften up a pack of scrubs, a properly-grimoire'd wizard can do it. If you want to target a specific defense or a specific damage resistance with a certain type of attack, a properly-grimoire'd wizard can do it. But if you want to go toe-to-toe with an equal-level fighter in melee, you're going to die. If you want to trade damage with an equal-level ranger, you're going to die. Wizards have more potential abilities (i.e., spells) available to them than any other class (casters included), but their access is always limited by their equipped grimoire. Switching grimoires disables spellcasting for a short while so it's really not something you want to do in combat unless you're desperate.

If hybrid classes cannot be as good at any single role as pure classes, even when specced to only do one job, there will be no point in using any of the hybrids at all. You might as well not include them in the game because no one will want to use them. I mean I'm just one guy on the internet, but I feel it would be a huge mistake to design hybrid classes this way.

I think the margins of efficacy and the mechanics of redundancy are important considerations. Combat is the sort of obstacle where every participant on a side generally contributes to their side's collective strength -- even if they're performing the same actions with lesser efficacy than their teammates. This is generally different from stand-alone skill checks where redundancy is pointless (e.g. opening a locked door).

A character who has the flexibility to perform their class' main combat "things" at a mildly diminished capacity in exchange for filling another role at 75%-90% capacity -- I don't think that's an unappealing or bad character at all, especially in the context of a party-based game. If a character just flat-out stinks at a given role, even given optional builds, then yeah, there's not much that's appealing about that.

Arcane Veil is every wizard's short-duration "holy shit" defensive ability that can be used if someone bum-rushes them. While it's true that fighters start with Defender, they get more localized area-denial abilities as levels rise. Defender also increases the reach of a fighter's Engagement, so they can effectively cover a choke point that's wider than their reach would indicate. You (or an enemy) can choose to break Engagement, but if you're hit with a Disengagement Attack, it does more damage and you stop to play a hit reaction (unless you're a barbarian using Wild Sprint).

Are wizards the only ones with spell levels or does that encompass other 'casters' as well (Druid, Cipher)? I think I vaguely recall that Priest's used Prayers and Cipher - Powers, but I am not sure whether they were spell-level based or not.

Wizards, druids, priests, ciphers, and even chanters all have levels of spells/prayers/powers/phrases/etc.

How much 'metamagic' will there be? Spell schools, spell sequencers, spell contingencies, spell turning, spell immunity, counter spells etc

We're still developing magic. I can say that wizard spells are the most expansive of any class' Abilities. Wizards currently have a 4th level spell called Minor Arcane Reflection that is effectively Spell Turning. They also have a 4th level spell called Minor Grimoire Imprint that allows them to temporarily "steal" spells from an enemy's grimoire and cast them for free. We'll probably be implementing more wacky wizard spells as time goes by. We started with the classics like Spell (Magic) Missile, Fireball, Crackling (Lightning) Bolt, etc.

Barbarians and rogues have the easiest "outs" for Engagement but there's a cost associated with each one. Wild Sprint still provokes Disengagement Attacks (it just ignores the stop effects) and it lowers the barbarian's already naturally bad Deflection. It's really easy for fighters to crit a barbarian who tries to blow on by them. Rogues can use Escape to break Engagement without penalty, but that's a per-encounter ability and they don't have any class Abilities that make them any faster than most other characters.

Is it possible to make a non-Monk pugilist?

Not really. You can punch dudes but unless you're a monk the damage never scales and the base damage stinks to high heaven. Transcendent Suffering is what automagically turns a monk's fists into SuperFists.

If I wanted to be the best arquebusier in the world, what class should I be? I'm thinking rogue or fighter, but who knows.

Ranger would be ideal but rogues can also be pretty brutal with them. Fighters are not great ranged characters since so many of their Abilities are focused on being in melee. However, if you want to use one of your weapon sets to open with a volley of gunfire and then switch to melee (Darklands-style), you can do that. It just leaves you without a lot of other options mid-fight (unless you increase your total number of weapon sets).

Hey Ropekid, I'm curious as to how you're thinking about handling equipment with the classes. What I mean is, for instance, will PE Rogues be assumed to be dual-wielding Daggers (classic Rogue) or could you equip a Greatsword and pump Str to dish out some sick sneak attacks? (a neat Pathfinder build).

Similarly with Fighters, say: are they optimized for sword and board or is two-handed/two-weapon/one-weapon Dualist-style viable?

I understand that not all equipment is equal and that, for running up and tanking the BBEG a sword and board is the best option, just wondering how much leeway there is.

BONUS QUESTION: What class would make the best Dualist/Swashbuckler and how viable would a lower damage, dodgy Dex melee fighter be?

PE rogues work best when they are using melee or missile weapons at close range, but they can use any "style". No matter what weapons they use, they do a lot of damage, crit more than other characters given the same Accuracy, and have Abilities to immediately spike damage if the circumstances are right. If you really want to do a ton of damage in one hit, a two-handed weapon like a pollaxe, estoc, or morning star is the way to go. If you want to do a bit more damage over time and have a more regular damage output, dual-wielding is the way to do it. If you find yourself in a situation where you're getting subjected to a lot of melee attacks, by all means, equip a shield.

Yes, for a fighter to be extra-fightery, a shield is great and can put his or her Deflection into the stratosphere. However, you may be in a situation where a) enemies aren't attacking your Deflection, but some other defense (e.g. druids can cast Flaming Brand and make melee attacks against Reflex) b) skyrocketing your Deflection isn't as important as doing more damage or c) enemies have shield-defeating attacks like flails. In the case of a) your shield gets you nothing and in the case of b) or c) it may make sense to equip a two-handed weapon or to dual-wield.

Specific weapon choice can also matter. E.g. if your enemies have shields, you might want to use a flail (or flails -- dual-wielding flails is one of my favorite things). If your rogue is getting squished in direct melee, use a pike (you can attack from farther away). If your enemies have annoying armor, use stilettos or an estoc (they negate a flat value of DT).

RE: Monks and fists: it's not really a matter of realism; I just don't want to make another class of weapons just to support what is ultimately a niche weapon type for one class. The idea behind monks is that they are the bare-fisted brawlers, so that's their "deal". Other characters can fight with fists, but they don't gain bonuses with them and they have a pretty huge array of other weapons to choose from.

Couldn't you just give the classes other than monk an optional talent that gives them the monk's scaling fist damage, so you could still make a punchy wizard gimmick build or whatever, but you wouldn't have to deal with a new weapon class?

We could. The Talents are the most loosely defined aspect of character advancement right now.

So if you're chasing someone who has Escape or similar... won't that lead to weird situations where you have to avoid attacking them? If you attack as soon as you're in range then they'll get away, so you have to use move command first to get as close as possible before clicking attack. In short, increased micromanagement.

Then again, is that level of micro a bad thing? I do remember doing a lot of fun path blocking to keep those snakemen off my casters in BG2.

Escape is an active-use ability, not something that's reflexively triggered by someone attacking. It also can only be used once per-encounter, so that limits how easily a rogue can shake a persistent attacker. The ideal use of Escape is against a fighter as long as the rogue has somewhere to go. A barbarian could (if he or she hasn't already used it) Wild Sprint to catch back up to the rogue, but that is a dangerous game to play for a barbarian. They have a lot of Stamina, but their Deflection is low and rogues crit easily. Barbarians also don't excel at single-target damage, so once the barbarian is on the rogue, math is often not in the barbarian's favor. Trying to Escape from a monk may be the most futile effort. Monks are inherently faster, their Deflection is average, they turn damage into Wounds to power active-use abilities, and many of those active-use abilities inflict status effects that target Fortitude or Psyche (rogues' worst defenses).

I don't know if this has been said anywhere, but will spells/abilities have different casting times like in BG/IWD or will everything take equally long to use? I kinda prefer the first since it feels like a good balancing mechanic for powerful effects.

We will probably have three casting/ability use durations: immediate, short, and long. Those will be set to specific intervals and will be consistent between characters. E.g. a fighter's "short" ability use is the same as a wizard's "short" casting time, etc.

I think Chanters can't do anything else while chanting and while the word combination mechanics seem nice, they still look pretty fire and forget. How do you make Chanter combat involving?

Chanters can absolutely do other things while chanting. They have good Stamina/Health, balanced defenses, and balanced melee/ranged Accuracy. Also, as their phrases tick over, they generate the ability to unleash Roars, which are powerful active-use (often offensive) abilities. Chanters are meant to mix it up in standard weapon combat and periodically unleash super slams.

If they were like bards in 2nd Ed. AD&D, they would wind up the same way: unplayed. Chanters don't play instruments. All of their power comes from using ancient literary phrases or idioms to awaken the memories of lost soul fragments around them. I'm trying to make them beefy enough to contribute to front-line combat even if they aren't doing anything dazzling outside of their Roars.


Only fighters continuously regenerate Stamina during combat. After combat, Stamina regenerates for everyone very rapidly.

We initially had constant Stamina recovery for all characters but it didn't make sense for everyone.


I think most people don't know enough about the mechanics to make any accurate predictions but each person sees one part of the elephant and they extrapolate from that.

I think a big part of the claim falls on the assumption that all the class mechanics allow the exact same options for talents and abilities (or effectively the same options, when you come down to brass-tacks and calculate each one out) across the board so that it really does play like a class-less system and that classes are just a hollow title to pacify those masses.

The other part of the claim lies on the fact that we've heard a lot about how classes aren't distinguished like they were in IE games (rogues are skill-buffed characters, mages are OP nuke throwers, fighters are good low-level fighters and meat shields, etc), but we don't really understand much about how they are currently distinguished in game-play and when you take away what distinguished them, but don't replace the descriptions with new ways that they are distinguished, it's hard for posters to understand what's what.

Classes don't currently share any Abilities at all. If you're not a fighter, you're never going to be able to take Defender. If you're not a monk, you're never going to be able to take Transcendent Suffering. Some Talents can be taken by any class (e.g. the weapon style Talents), but many of them are class-specific (e.g. Grimoire Slam).

I disagree with your description of what we've said about the classes. We've repeatedly stated that fighters are extremely durable, reliable, and excel at holding positions, that rogues are the best single-target, single-hit damage dealers of any class (yes, significantly better than fighters), that monks are high-mobility melee status-infliction machines that use Wounds as an expendable resource, that wizards have high flexibility and, in addition to their traditional area-nuking abilities, have a variety of personal and single-target buffs.

We've also said that if you try to play a class completely against role, you can run into trouble. There's an important distinction between what you can build and how you play. We don't allow characters to take Talents that are effectively dead-ends for their class. You also gain Talents at about 1/3 the rate that you gain Abilities, so they comprise much less of your character's makeup. In 3E/3.5, a fighter is practically made of feats and you can really botch a character even playing in the pool of combat feats.

For comparison, in PE you can buy light magic Talents for your fighter that give the character some neat flexibility, but you can't completely redefine what the fighter fundamentally is. And if you buy a set of Talents, we aren't setting up long Talent chains like the feat chains that exist in 3/3.5 -- e.g. taking Whirwind Attack requires Combat Mobility, Dodge, Spring Attack, Dex 13, Int 13, and a +4 BAB. Our Talents have a flat layout with simple prereqs and are designed to be valuable on their own for any class that is allowed to take them. In 3E/3.5, it's really easy to build a low-efficiency fighter who isn't good at, well, fighting. A PE fighter can diversify a bit, but at his or her heart, he or she will still be great at doing the job that all fighters' Abilities prepare them to do: absorbing damage, hitting reliably, and holding ground.

Similarly, if you want to gish it up with a wizard, there are spells and Talents that can lean you in that direction, but you can't outlast a fighter or hold ground like they can and you can't reliably spike damage in melee round after round like a rogue can. Now, there are things that you, the melee wizard, can do in melee that the fighter and rogue can't. You can surround yourself with a big fiery shield and make illusory duplicates of yourself. Those differences are cool and why you would want to play a gish wizard over a fighter or rogue even though you ultimately can't do their "jobs". But because grimoires are designed for flexibility (because wizards are designed for flexibility), if you get tired of being the glowing Daffy Duck gish hopping all over the place or if you're in a situation where you can't even stand next to the melee big kids, you can switch to an AoE damage grimoire and be a "traditional" wizard. There's no dead end in playing as a gish (even though you will be challenged in other ways) and more importantly, your gish-emphasizing build options don't dramatically impact your ability to do regular wizardly things. You shouldn't reach a point in the game where you go, "Wow, I regret taking these Talents because this character can't do anything well."

In 3E/3.5, class roles are less well-defined and it's easy to build a character that is bad at any job -- whether it's their class' job or otherwise. In 4E, class roles are very well-defined almost to the point of being straight-jacketed. In building PE's classes, I found that trying to draw strictly within the lines of a class' role was limiting in a way that wasn't enjoyable -- and I didn't believe that players would find it enjoyable either. That's why I've tried to use the approach of making classes "role-ready" instead of "role-constrained". PE's characters of any class are always ready to fill their class role regardless of the Talents you've taken because their per-level class Abilities have a much more dominant influence on their overall capabilities. There are always efficiency gains to be made in how you build, but compared to 3E/3.5, the number of viable builds should be much higher.

Play-wise, if you want to put a monk in a tanking position or run a wizard around in melee, the rules aren't structured around building restrictions to discourage you from doing that. In many fights, it will be totally viable even if it's inefficient. In some circumstances or at higher difficulty levels, it will be more difficult to play in this way, but if you find that you need to "fall back" to standard roles, you should be able to do so because your character can't be fundamentally built contrary to his or her class.


Abilities are always granted by gaining levels in a class. I.e., they are ALWAYS class-specific and they are gained per-level.

Talents are a more general pool of optional goodies (some of which are still class-restricted) that you gain (currently) every three levels.

Edit: To use a D&D example, Wild Shape would be an Ability, Dodge would be a Talent. Only druids* gain Wild Shape, but any character class can take Dodge.

* I'm sure there's some other class in a splat book that will prove this wrong.

There's nothing final about Ability progression/selection. Right now we're implementing them as a fixed order progression. That may change for one or more classes depending on how it feels. There's nothing that necessitates gaining them in a strict order.


Regarding the rogue's sneak attack; is it a damage range bonus, e.g. 3-5, or something else?

Right now it is a percentage bonus to damage.

No, please. There goes down the toilet the big incentive to play a rogue with fast/small weapons. :(

The damage output of small/fast weapons is terrific, especially if the target is in light armor. This isn't D&D, where weapons like daggers and shortswords are only situationally good but otherwise bad.

There's some confusion WRT the rogue's "spike damage." What does this mean exactly? Some think you mean burst damage as opposed to spike damage. Can you please clarify for the noobs?

I think in close to all cases where I've written "spike damage", I've used it as a verb. A rogue can spike his or her damage output, i.e. dramatically increase it for a short duration. Rogues have a few Abilities for doing this, the most notable of which does increasingly more damage based on how low the target's current Stamina is.

On attributes, equipment, encounters, stealth and skills:
No cover/melee engagement penalties like 3E/3.5/Pathfinder. Even in a turn-based tabletop game those penalties get really odious (cover + in melee = effectively +8 AC) and hard to avoid, so most ranged characters have to take Precise Shot (or get it as part of their class). I'd rather mitigate the per-shot damage done by ranged weapons and not use cover/melee engagement rules.

In the end, the result is effectively the same: less damage done over a given period of time. With unmodified accuracy and lower overall damage, it's more normalized/less spiky.

Probability wise, that's true. I just wonder what it will do for the verisimilitude to have every single ranged attack doing some damage, regardless of how difficult the shot? Arrows have effectively become shotgun blasts.

Every single ranged attack won't do damage. They can graze or miss just like standard melee attacks.

And if you want a shotgun blast, that's what blunderbusses are for.

Not a problem in Pathfinder or 3e. In those systems, ranged weapons deal less damage than other 2handed weapons, but can be used froma distance to make up for that shortcoming. I have little dount that PE will have viable ranged combat.

Our ranged weapons do good damage, but can't compete blow-for-blow with two-handed melee weapons. The exceptions to this are firearms, but they are relatively inaccurate and slow to reload.


I can't guarantee that you'll be able to "ghost" areas in PE, but if you build a party with the Stealth skill as a focus, you may be able to circumvent a lot of encounters if you so choose. If you dabble in Stealth, you will probably wind up using it more for combat positioning.

While rogues do have a bonus to Stealth, so do several other classes, and no class has an inherent penalty to sneaking. It will be quite possible for you to keep your entire party close in overall Stealth values if you choose to focus on that skill.

The weakness of the proposed system is that it sounds like the detection radius will be entirely based on the character doing the sneaking, say 20 yards while standing still, and 40 yards while running in full plate. This makes it sound like the perception of any antagonist plays no role, meaning that all enemies, be it an over sized bat or a half blind pig, will have the same ability to detect you. 

This is incorrect. Creatures have their own individual detection radii.


*We can't specialise in a single weapon. You need to specialise in an entire category of weapons. Ok. That's digestible.

*Attributes. Attributes will govern different, even vastly different, aspects of my character. How will having a universal "damage" and "accuracy" stat (whatever the final name is) help define my character from a rpg standpoint? It won't and it bothers me.

*Finally, something that hasn't been mentioned yet. Combat styles. I'm talking about weapon and shield, two-handed, ranged weapons, dual wielding. I'm afraid, considering their design philosophy of extremely flexible characters that constantly need to switch weapons and whatnot, that being able to specialise in combat styles is unlikely.

Gaining proficiency or specializing in categories of weapons has existed (at least as an option) in several editions of A/D&D, including Combat & Tactics (2nd Ed.), 4th Ed., Pathfinder, and 3.5 UE's Weapon Groups.

Structuring Attributes so there aren't dump stats promotes more role-playing options because non-viable builds should be much less common. Speaking as someone who has played a lot of gimmick builds and characters with sub-par ability arrays for a given class, while it can be very enjoyable to role-play a high-Cha fighter in 3.5/Pathfinder, those characters are typically (barring the use of a lot of special/house rules) pretty ineffective at doing the job their class is supposed to do. Choosing to play certain character concepts becomes an implicit difficulty slider and I don't think it's in the players' interest to link those two things.

We have combat style Talents and they should allow you to stick with a fighting style even if an enemy's specific armor strengths/weaknesses promote switching to a different damage type. You also typically have another option: switching a character's targets. We try to structure fights so you have a variety of enemies to contend with.

Overall, weapon/damage type switching should be occasional, not frequent. We currently have seven damage/resist types. In the vast majority of cases, on any given creature or type of armor, most damage types have the same DT as the base DT. One, two, or possibly three damage types will have a higher or lower DT. Those differences will also be consistent, so if you see someone in mail -- whether it's run-of-the-mill or some awesome magical variant -- their relative DT is always going to be worse for Crush than the base. If you see someone in plate, their relative DT for Shock is always going to be worse than the base. If there are two size variants of a monster, its relative damage type strengths/weaknesses will typically be maintained between the variants.

Will there be mono-like encounters? Like, a room full of monsters that are immune/highly-resistant to fire? Or will those not be present at the game at all?

I agree that most encounters should be varied so as not to make certain playstyles gimped, but if there's a single instance or two in the game where players who rely on certain strategies/builds need to do something radically different in a single room to move past it... that could be interesting. Though, maybe it's not worth it?

My flimsy question is: How true are you sticking to that design philosophy? Will it be permitted to have encounters ever designed that way?

We may have encounters like that. I think it's good to mix up encounters so they're not constantly balanced around maximal uses of resources or overly formulaic. Sometimes it's nice to just roll a bunch of scrubs in an otherwise difficult dungeon. Sometimes it's nice to see the balance of enemy types/resistances radically shifted. As long as it doesn't present an insurmountable roadblock, it becomes an opportunity to think/use your characters a little differently.

Hopefully not the other way around... "oh, the blue skin rats, they can only be hurt with electricity". Not with fire, not with swords, not with nuclear armageddon, only electricity, 14 amps will do nicely.

No, we really don't want to do that at all. I think it is better to occasionally say, "Of the ten ways you could attack this guy, these three are not good" than to say, "This is the one way out of ten that you can viably attack this guy." The former encourages you to examine other possibilities, but you can arrive at a number of solutions, some of which work better for your particular character/party. The latter can leave you in a situation where you're just SOL (or at a severe disadvantage).


That shield talk got me wondering; have you given any consideration to the historical development of shield sizes decreasing as body armor technology improves, i.e. some kind of mechanical limit in "defence stacking"? I remember in IWD2, i.e. 3rd edition D&D, better armor reduced the max AC bonus you could get from high Dex, but I've no idea if this also applied to shields and body armor in some form.

More generally do you try to stay within a certain historical period when it comes to weapon and armor design of the world, or is it more of the classical D&D mix and match fantasy?

Heavier armors do come with their own drawbacks (currently, attack/action speed). We're not being simulationist about armor/weapon statistics. Also, armor absorbs damage and that function is separate from the four defenses (Deflection, Fortitude, Reflexes, Psyche) which are the targets to hit.

The technology of the Dyrwood (and nearby Vailian Republics) is 16th century, with a few exceptions. Glanfathans are at Early Middle Ages tech but have no problem using the more advanced technology of their neighbors. Though really, our own history is really "mix and match". People continued using outclassed/old-fashioned weapons and armor long after new tech had been invented (though often, they got their asses kicked).

Would there be any reason to use a single one-handed weapon without a shield?

We're probably going to give you increased Accuracy if you fight with a single one-handed weapon sans shield.

Now that you mentioned it, what type of morningstar will your morningstar be?

Our flails are one-handed handle + ball and chain weapons. Our morning stars are two-handed long-shafted weapons with a fixed, spiked head.


Some people aren't fans of our overall Attribute design, which I've focused on always providing benefits to characters of any class even if that means the Attributes aren't very realistic. If you want to make a Strength-based rogue, that's viable. So is a Strength-based wizard, an Intellect-based barbarian, etc. As with taking Talents, you may be de-emphasizing some core aspect of your character by using a non-traditional Attribute array, but there aren't class/attribute combos that are fundamentally bad.


RE: Skills: PE's skill caps are more like Pathfinder's than 3E/3.5's. Regardless of how many skill points an individual rogue puts into Stealth or Mechanics, he or she will always have a bonus in those skills that other classes don't have (Pathfinder requires a rank in each class skill for the bonus, but it's similar otherwise). A rogue who neglects those skills may be running on par (or below, if neglected enough) with a fighter who specifically maxes them out, but a character who focuses on the skills their class gains bonuses to will always be ahead of a character of the same level from a class that does not gain bonuses in that skill.

Probably an easier way to explain it: all characters have the same number of skill points per level and the same access to skills. However, every class has a few skills in which they receive a constant bonus -- always and forever.

On usability features and difficulty:
Is there a sort of fallback AI for some classes? Ie; if my wizard isn't casting Fireball he won't stand around but will do...something? I'm not sure if casters have a default attack ability like using a wand or similar.

Characters can all auto-attack. Most players will probably equip wizards (specifically) with wands because wizards gain advantages with them and they're decent supplemental weapons with long range.


I'm going to tune the higher levels of difficulty to be like the hardest IWD2/BG2 fights all the time. Standard difficulty will be less difficult, but still challenging.

Changing the level of difficulty generally means changing what creatures (and how many) are present in encounters. Outside of Path of the Damned, we don't have any plans to scale health, level, accuracy, defenses, etc.

AI won't change based on level of difficulty.

Right, that's what I'm interested in. Does difficulty affect THA LEWTS or not?

It depends on the creatures. On average, you will probably get a small increase in loot, but it's possible to get less in certain circumstances. Level of difficulty both activates and deactivates creatures, effectively replacing weaker enemies with more powerful ones. Sometimes the total number of creatures remains the same (rarely, goes down) while the challenge goes up. E.g. if there's a fight that involves wurms (baby dragons) on the sidelines, the fight could go from having 3 wurms to 5 wurms to replacing all of those wurms with a single drake (adolescent dragon).


Auto-pause features/options are in btw.

Some of the classic auto-pause settings got new sub-options. E.g. "Enemy Spotted" now has a sub-option to halt the party when the pause is triggered. We also put in a few new ones like Extraordinary Defense which will come up when a PC's attack's pre-roll Accuracy is 50+ points below the enemy's corresponding defense. So if you take a bunch of 2nd level scrubs and start hurling standard melee attacks at a 7th level fighter with a large shield (i.e., super-high Deflection), it's probably going to pause right after your first attack and give you feedback indicating your chances of hitting the guy are really, really bad.

As an anecdote, I was recently doing tuning in a dungeon area where there were relatively narrow halls (not too bad, but not wide-open) with high-level fighters + large shields and casters behind them. I dogpiled the entire party with standard attacks on the enemy fighters and the dudes would not die. Even if I grazed them, their Constant Recovery would wipe out most of the pitiful damage pretty quickly. If I tried to walk away from them, they would stave my head in with a Disengagement Attack (fighters' DAs tend to be extraordinarily accurate/bad news). The caster behind them was chucking out a slow-moving line spell that fries enemies and heals allies, so that was sweet. I had no rogue (so no Escape), no monk (so no Rooting Pain), no barbarian (so no Wild Sprint), no flails to offset the shield's Deflection bonus. My wizard had Deflection-based spells or Reflex-based AoEs that would blast half the party. I had to use the cleric to light them up with a single-target Reflex-based spell that lowered their Deflection, then keep hammering them with the cipher's Psyche-based attacks.

We're still doing a looooooooooooooot of tuning (and will be for the rest of the project), but it's good to see the classes presenting these challenges and it's fun to use the given party's composition to come up with a solution using what we have.


We currently show the spell radius but do not yet highlight characters who will be affected. Many AoE damage spells are friend or foe. Fireball and Crackling Bolt, in particular, can blow up your party easily.


Our current plan is for save anywhere (except during combat).


We already have the slow combat implemented and it works pretty well. I'm sure we'll be adjusting it more as we fine-tune combat pacing and pathing, but it's an enjoyable alternative to full speed vs. full stop.

Some minor info on portraits and the interface:
Excellent update, one question however, will the characters responses in the dialogue UI be so stiff? The innkeeper description gives a definite sense of character, but the responses are somewhat formulaic. If these are just placeholder responses then never mind, I apologise for raising the issue, or is it I wonder a design choice so that responses appear as neutral as possible, and thus don't compromise the player characters own imaginings of personality? Be interested to find out.

It's the opening node of the dialogue so the choices are straightforward. Players will have more personality-laden replies when they make sense in the context and will go somewhere interesting.

Kerning on the text in the conversation UI is abysmal. This has bothered me about all the infinity engine games I'm a designer with a font fetish. Any other imperfections have been erased by time and nostalgia, but that one niggling flaw has never stopped bothering me.

NGUI only uses bitmap fonts, so we'll have to adjust the kerning and other errors by hand, but we're aware of it.

Hmm... I hope the inclusion of numbers in the dialogue means that dialogue will be keyboard navigable? (E.g. Pressing the "2" key will select the second dialogue option).

It is.

Is the way that Osmaer's dialogue is split up into four pieces indicative of the way multi-line dialogue will be designed throughout the game? My initial reaction is that it's unnecessary to have "Osmaer - " placed before each line as in this example as opposed to having a block with "Osmaer" at the top and the body indented. When there's no alternate character (NPC or PC) speaking between Osmaer's statements nor extra descriptive lines it may be cleaner to remove the extra identifiers. Would that work?

We'll probably reveal them as individual notes but keep them together in the log.

One thing to consider is if one is color blind, the text should still be at least readable. Not being color blind myself, I don't know what dark green on dark brown background would look like, if it would have enough contrast or not.

Tim and I both have forms of color-blindness (though his is worse than mine). We don't have any difficulty reading the text.


Kaz is working on the inventory UI. The icons below the portrait are "mine" (i.e. just placeholders) and represent (top) Stamina, Health, Main Hand Accuracy, Off-Hand Accuracy and (bottom) the four defenses: Deflection, Fortitude, Reflexes, and Psyche. Of course, it's still a work in progress.


Any word from the devs on whether custom portraits can be used and if so, what are the dimensions?

We want to support custom portraits and make it very easy for players to drop in whatever they want. We're still playing around with portrait sizes but right now they're around 660x850 on the character/inventory screen and get downsampled for use on other parts of the GUI.

Ouch. I was hoping you'd avoid the downscaling solution. Having to make thumbnails might be slightly less user friendly for people wanting to put their custom portraits in, but it allows for stuff like full body portraits that focus only on the face in the thumbnails, etc.

We may still have two separate assets, but for our portraits we will be going for a "head and shoulders" (BG1/BG2) approach rather than the "knees-up" (IWD1/IWD2) style. The full body portraits are dramatically more time-consuming and much more limiting when it comes to player concepts. Our "paper dolls" are pretty detailed because they're using the same 3D models of the characters that you see in the game. The knees-up portraits create more visual discontinuity when you see them next to the game models.

Yeah, we're doing head and shoulders portraits because it will allow us to do many more of them. Also, it's generally easier for players to match a portrait to a variety of character concepts if the body/equipment isn't visible. The IWD/2 "knees-up" portraits were cool but very time-consuming. On IWD, it took almost all of Jason Manley's time on the project just to generate the portraits.

As an aside, are the current companion portraits final, or will you swap them later for different pictures?

No, not final at all. Most of our current portraits in-game are using cropped images from non-portrait pieces of art (e.g. our wallpaper image from the Kickstarter drive). Even if we use some of those images as a starting point, the portrait artist(s) will be creating images specifically for portrait use in-game.

On the stronghold design:
Stronghold tax income is more for balancing maintenance costs (i.e., paying hirelings and repairing damaged upgrades) than a major source of income. Your major source of income will almost certainly be finding money and loot in the world.

Your player house is located within the boundaries of the stronghold, but it is its own building with its own upgrades/benefits.

I think physically splitting the house and the stronghold increases the likelihood that the player will choose to not visit one or the other (most likely, the house) over the course of the game.

It doesn't need to be a major source at all.[stronghold income] It becomes a problem if it's an endless source.

It isn't an endless source.

Good. Because in your other games with keeps/strategical resources, it was endless.

I didn't design the stronghold resources/systems in any of the previous games I've worked on. If a system gives out a resource periodically, it should either expire after a number of periods have elapsed or the periods should stop elapsing.

Some tidbits on lore and the city of Defiance Bay:
How do the Orlan's defend themselves against the aggressions of the larger races? As a largely disliked race they must have some way of defending themselves as a society despite their diminutive frame and physical deficiency.

For the most part, they've defended themselves by retreating and/or resorting to merciless guerrilla tactics. Ruthless orlan behavior is the source of much of the dislike of orlans as a group. In this part of the world, the only area where orlans are on equal social footing with other races is in Eir Glanfath. In Glanfathan legends, orlan and elven tribes both took the same vow to defend the Engwithan ruins within Eir Glanfah. Since then, the tribes have intermingled to such an extent that despite the physical differences between them, Glanfathan elves and orlans treat each other as equals.


So, people can be born without souls and no-one really notices, eh. What is the difference between a person with a soul and a person without a soul?

It's immediately noticeable. A person without a soul (born that way or made that way) is in a persistent vegetative state. They have no awareness or volition. It is ultrabad and most people in this condition die unless they are cared for constantly.

Yeah, I was pretty much imagining a kind of anti-hospital, wards full of soulless bodies being attended by "nurses" (possibly mentally deranged people who think of them as their actual children) at the command of some monstrous asshole who is preparing to convert them to his (or her!) slaves once they reach an appropriate age.

Most children who reach adolescence without souls do so because they are cared for by extremely patient parents who a) hope that someday it can be fixed or b) just can't bear to let one of their children die. Not all parents choose to do so.

Well magic and stuff is all soul power so at least some non-human stuff must have souls. It's probably necessary for consciousness, so I don't see how plants would have them though.

Animancers and ciphers don't know a ton about souls in plants, but at least some large, old trees in Eir Glanfath have them. Dyrwoodans found this out the hard way during the War of Black Trees.

We're still talking about the exact process/circumstances under which soulless children become wichts, but soulless children are absolutely born in a persistent vegetative state and the overwhelming majority of them are either killed by their parents (directly or just left to die) or they die before leaving infancy despite attempts to care for them.


Based on a lot of the questions people were asking here and elsewhere, we (the designers) got together and worked out the details of how wichts are created. I think the majority of you will be satisfied with how they (and their lore) are presented in the game.

The important thing to keep in mind here is that the implication of a universe with souls (especially ones that transmigrate) is that the soul is actually the person. A body is just a vehicle the person is driving around.

This is correct. Animancers and philosophers question to what extent the physical body plays a role in defining a person, but the soul is the essence of an individual's consciousness, memory, and personality.

Presumably, damage to the body can make certain capabilities of the soul inaccessible.

Yes, though a soul separated from a physical vessel but lingering in the mortal world (a "lost soul") has neither the limitations nor the advantages of being connected to a body. Lost souls, unlike "housed" souls, are extremely difficult to perceive and interact with.

Is there an afterlife in P:E, or somewhere for all these souls to go while they wait to be resurrected by a cleric?

When people die, their souls almost always move toward portals to (presumably) the realms of the gods. Some souls stick around, becoming "lost souls", though they can do very little in the mortal world other than postpone their own reincarnation.

Animancers will sometimes capture souls when a person dies, but placing a soul back into a body that has already died usually produces *~ bad things ~*. Engwithans placed warrior souls into bronze armor to create animats (a Vailian term, not theirs), but animats have limited memory and consciousness. Dyrwoodan and Vailian animancers have been creating their own constructs using contemporary mechanics and metallurgy. Making a construct takes a lot of time and money, but the results have exceeded what the Engwithans were capable of on an individual scale.

E: To clarify, animancers aren't magicians. They're scientists that specialize in understanding how souls work. Most of them also advocate using technology to manipulate souls with various goals. Naturally, this is an ethically contentious field.


In terms of layout, the closest comparison is probably Athkatla. Our cities have a number of distinct (large) districts with their own thematic style, sets of interiors, and unique quests/content. They are set up to be easy to navigate, both physically (pathing) and visually. Generally there's an exit from each district along each side of the map, so backtracking isn't necessary if you just want to move to another district. There are exceptions -- e.g. Ondra's Gift, which contains the docks, can't be exited via hopping over the levee and into the ocean -- but we try to make navigation and exploration enjoyable.

Finally, artist Kazunori Aruga replied to some questions on the Eternity forums:
Hello, my name is Kazunori "not a vertical slice video" Aruga and I'll be answering some of your questions!

I think I speak for everyone in that we're very happy to have you on the Project and I have loved every piece of your concept art I have seen so far.

The area/city art has been my favourite so far.

Hopefully the portraits are up there with IWD and fight the minimalists for a larger portrait size in the UI ;)

Thanks! I like to think portrait work as the ice cream waiting for me when I finish my UI vegetables. Not to say I dislike UI, but I'm fairly new to it and there is a learning curve I'm working to overcome. It doesn't mean portraits will be a cakewalk either, but it is one of my favorite things to do and I'm very much looking forward to getting started on those.

Anyway, Heyo Kaz; what areas are the images on the 2x2 split picture from?

Those were sketches I did to establish the general mood for the various districts in Defiance Bay. Someone had asked how long they took, and the answer to that is two days each, six days in total.

Also, as far as the paper doll inventory, could you guys try to create something like this only cropped in on the characters so it looks like they are standing on something and not arbitrarily floating in space?

Not going to promise anything, but if it fits with our layout / aesthetic I don't see why we couldn't include a simple dais or platform to help ground the character.

In the area concept art shown in this update, I'm guessing the top left is the rich part of Defiance Bay, the bottom left is the flooded ruins of the old city, and the bottom right appears to be a nearby quarry. What's the top right illustration? It reminds me of a park, yet it also appears...swampy?

I can't go into too much detail about specific districts. The top right image shows a district that is fairly preserved since colonial times, making it one of the older parts of the city. It boarders a river and is actually built below the water level, with a large levee protecting the district. The last description was added after the sketch, which is why you see the streets above water level in the picture.

That's all for now, though while you digest all this info, it's worth noting that project director J.E. Sawyer has pointed out that this Tuesday's update is going to be a "fairly lore-heavy" class update.