What attribute bonuses should have been

[PsychoBlonde]

@PsychoBlonde:

 

In all fairness, there is no stat choice that lets you land proper hits instead of grazes more often. So, while I get what you're saying about build options beyond "DAMAGE!", one of those is not "ACCURACY!" where stats are concerned.

 

I didn't say there was.  I'm saying that "MOAR DAMAGEZ!!!!" is not the "ONLY" build option.  Perception adds to your interrupt.  More interrupt = enemies don't do so much.

 

My rogue with stacked perception and interrupting blows TRASHES casters--they literally won't get off a single spell.  It was really kind of weird to watch.

[WebShaman]

I wholeheartedly agree with the OP - but sadly (as a few of us found out in the beta testing) Obsidian is not interested in changing the stats. We found this out early - we were to beta test for bugs, and not to critique and offer changes to the actual mechanics (that is the impression that I received). There were really some hot debates about this and we did fight the good fight...but it was hopeless.

 

As it is now. Like it or not, this is what PoE is stuck with. When I think of what it COULD have been *sigh*...

[VioNectro]

At the end of the day when it comes to attributes the question is, does it result in diversification through min/maxing or not. In DND it does because your class dictates what is going to get used for defense(armor or dex) and what attribute is going to get used for damage or number of spells. Currently in pillars this is not so much the case class matters very little in stat selection, rather it comes down to this: does he stand in front to soak damage or swing hard, and if he isn't in front then its might and dex for attack speed/damage. Things like interrupt/concentration/AOE/Duration become "support" attributes. The one thing that grits me personally about the stat system is there's no way to modify "To Hit" other than gear, your class, and your level. So there is no way to increase the amount of hits a character lands, so the only option left is DEX for more hits to be thrown out. From the reverse side deflection can get pumped up by two attributes making you harder to hit etc.

 

TLDR; this stats system needs some serious work in my opinion, currently I agree with the OP.

[Pixies]

The idea that there is any particular stat that is necessary for any particular class to be effective is just silly, IMHO. It all depends on how you build and how you play; what role you are assigning that character to.

 

I don't see how it's silly; it's the case with most professions in life and actually makes a lot of sense. Would Usain Bolt be helped by a 180 IQ? Quite likely - but his capacity for running fast is without a doubt the defining characteristic that makes him what he is - and you can be sure that's also why he devotes all his training towards that goal, rather than to be able to, say, perform surgery, calculate mechanics of materials or wrestle someone into submission. So not only are certain attributes necessary for success in a certain field; development within that field also tends to lead to fulfilling as much of the potential of those attributes as possible, quite likely at the cost of others. Complaining that most D&D warriors have high str and con seems like complaining that most basketball players are tall, explosive and have good ball handling skills.

[nipsen]

Really not sure what you're going on about here.  At first it sounds like you're saying it can't be done, then move on to say it'll be unbalanced?  The joke of it is, these stats are currently NOT widely stretched out into lots of systems, that's the problem!  They each modify like 3 variables and that's it. 

 

We already have complete imbalance in builds, so I'm not too worried about what the modders will come up with.  Or at least, I'll judge it when I see it.  I don't consider Obsidian to be some kind of prescient god-king of balance and any changes from their divine vision will ruin the game. 

 

Right. That's what I said as well. But the solution they've chosen is to remove certain variables from the attributes, make other player-controllable variables less significant. While making abilities, feats, spells, and so on comparatively more significant per cast, regardless of stats. And they're then going to balance those abilities to make them seem reasonable against player expectation.

 

Which is a great plan if you want to engage a Q&A team for at least the entire duration of the contract, or, more likely, for as long as the game sells enough to justify paying them. Because obviously you're not going to actually see a situation where two different adventure parties are not going to be able to deal out and take different amounts of damage.

 

And you get in a situation where not just the attribute system is cosmetic, but also where you will create strategies based on those expectations that simply won't work. And there'd be further tweaks, endlessly.

 

What I'm saying is that if you mod this now, you will also have to go through and normalize all critters, all items and all spells, abilities and hamsters, as well as the class variable tables down to what they originally might have been, if you don't also want weird unaccounted for buffs and nerfs.

 

Because half of the game now is put into specific implementations and tweaks for spells, feats, weapons and items. State of the art game-design in 2015, ladies and gentlemen.

 

 

To think a tactical RPG works with such streamlining is really limiting its potential - from my estimation.

 

You don't need to moderate yourself, you know. It's only on the internet that you will ever see anyone insist it's common to hear: "You know what? I don't care about that wizard - I think we should talk more about variables now!" in a role-play gaming setting.

 

And it's only on this forum that I've ever heard anyone create a philosophical justification for variable twerking. In the sense that people seem to genuinely argue that twiddling with variables is a transcendent act of Game. That if you twiddle enough, that in itself entails immersion. Which then in the last parting of the waves signifies nirvana, and perfect Gameplay.

 

Elsewhere, narrative is everything, and bending the rules to make them narratively consistent makes up about 90% of the game. But not here. And, sadly, neither at Q&A at Obsidian HQ.

 

But of course you're right. Designing an rpg around a cosmetic attribute system is an incredibly curious choice. And it's insulting to role-playing gamers who know what this means.

 

And it's insulting to people who used to play the old IE games as well. (Well, with the exception of the five people on this forum who got all their wishes fulfilled, of course. And who insisted anyone disagreeing with them must have been, simply, insane. 70 thousand backers thank you, I'm sure).

[loewenheim]

Now I won't deny that the stats could use some re-jiggering... back during the early Beta I put together a fairly extensive paper with Sensuki that made a number of suggestions to rebalance the attribute bonuses from stats. Some were adopted, many weren't. In any case, I'd agree that a few of the stats are a little weak at the moment - but they're still perfecting the system.

 

Interesting, is this available to read somewhere?

[loewenheim]

I wouldn't want requirement for equipment. At all. I very much enjoy the current system where I can have my mage in full plate armor wielding a battle axe or a barbarian with a backup blunderbuss if I want to. It's not very viable, sure, but I can do it and it pleases me.

 

Besides that, I think Obsidian did a decent job for attributes overall, but balanced them wrong. Perception should affect more than tanking ability (accuracy? critical chance/damage?). Constitution should be more potent overall, with boosts to defensive saves and perhaps lowering armor penalty to provide small dps boost? Might seems fine to me. Dex might be a bit too useful for everyone. Resolve is pretty much a tank stat, I'd maybe add interrupt resistance to aid casters a bit. Intelect is kind a niche stat for AoE damage dealers, it seems OK that way to me that way too.

 

Really, Perception is the only one that needs an overhaul. Resolve and Constitution are already tank stats, we don't need a third one. I'd be comfortable shifting some of Per's defensive bonuses to Con, and add an offensive component to Per.

I pretty much agree with you. Apart from Perception, what the stats do seems alright to me, it's just that the numbers are out of whack; e.g. Con doesn't have enough of an effect. I don't know what you mean by "add interrupt resistance" to Resolve, since it already does that. Do you mean add more of it?

[loewenheim]

I think the current stat system is fine from a gameplay perspective, although I agree it leads to some strange situations. However, I'm not a huge fan of the OP's system. It doesn't seem any better and just leads to the same old (this class needs this stat) results. It does, however, make more sense with regard to strength.

 

IMO the attribute system should look something like this:

 

• Strength: +Melee Damage, +Bow Damage (not crossbows), +Fortitude, -Armor Penalty*
• Constitution: +Health, +Endurance, +Fortitude, +Deflection, +Duration of Abilities, +Heal
• Dexterity: +Melee Accuracy, +Non-Spell Action Speed, +Reflexes, +Deflection, +Disengagement Defense
• Perception: +Ranged Accuracy, +Spell Accuracy, +Reflexes, +Buff AoE, +Max Ranged Attack Range
• Intelligence: +Spell Speed, +Offensive AoE, +Duration of Abilities, +Will, +Interrupt, +Crit Damage**
• Resolve: +Spell Damage, +Wand Damage, +Concentration, +Will, +Endurance as a percentage of Health, +Max Ranged Spell Range

*Armor Penalty: -Non-Spell Action Speed, -Reflexes based on the heaviness of the armor

**Crit Damage is a flat bonus to damage rather than a percentage

 

With this system you can have a

• Str+Con frontline fighter
• Str+Dex big weapon melee DPSer who relies on armor for some protection
• Str+Per bow/melee versatile DPSer
• Str+Int melee crowd controller
• Str+Res nothing
• Con+Dex meatsack nothing
• Con+Per healer
• Con+Int crowd-control tank
• Con+Res frontline caster
• Dex+Per gun/crossbow DPSer
• Dex+Int small-weapon crit DPSer who uses disables and disengagement for protection
• Dex+Res melee offensive caster (weak)
• Per+Int AoE magic debuffer/buffer
• Per+Res single-target magic DPSer
• Int+Res AoE magic DPSer

That seems pretty good to me (I'm biased) just a few combos don't work that well and with the right items/tweaks, that might be fixed too. Obviously, you can try and add a third or forth stat for added strengths and variety. Str+Con+Dex go well together, as do Per+Int+Res, not surprisingly.

At first glance, this actually doesn't seem bad at all. More complicated, but way better than OP's system. With this, I can see e.g. a wizard wanting each stat except Str. But one stat not being that useful for a given class is acceptable, I'd say.

 

If someone based a mod on this, I'd try it.

[loewenheim]

 

 

To note "Might" does denote physical strength in-game.

 

Yes, and the flaw there is that Might is used for those skill checks rather than Athletics which makes more sense.  Still a flaw, just a flaw in a few places in the game rather than a fundamental flaw in the design of character stats.

 

Yeah, this. Using Might for physical feats of strength in interactions is clearly a mistake, just not one in the stat system.

[nipsen]

^still - that was the elegance of the original system, that you only really needed the 6 primary abilities to describe any feat. Endurance would be used for feats that needed sustained effort. Might would be used for displays of raw strength. Intellect for complex reasoning and ability to quickly categorize data. Perception for spotting unusual details. Dexterity for delicate and accurate movement. Resolve for resisting the influence of others, or holding on to your convictions in the face of any danger, etc.

 

All of these would be displays of a type of strength, all situational, and all potentially as important as the other. Raw strength here meaning the ability to channel force, or as explained in the lore, channel the force of your soul power. How you would channel that force -- up to you.

 

Meanwhile, there's a duality to each pair of stats that would be reflected in the derived abilities. Might and dexterity would create incredibly skillful strikes, for example (and be analogous to a d&d "strength" maxed character). And if you focused exclusively on those abilities, you would have these elegant synergies with the other unfavored abilities.

 

Might and dexterity could go together to create a swordsman capable of placing very strong and accurate critical strikes. But it would go at the cost of situational awareness, endurance, resolve and intelligence (on the other hand, they would still strike really hard if they would hit). And so the uniqueness of it is that this was actually reflected in the game's mechanics as well, with the derived stats. For the sword saint in the example, he would have low interrupt chance, be easy to interrupt, and easy to hex.

 

Lack of endurance would also mean lower resistance to wounds, perhaps critically. So you would have to have this character use heavy armor and shields perhaps, that would protect them but slow them down. Always, these intuitive synergies. But still, the character would strike really hard when they'd be protected and buffed and could approach the target to connect a hit. This character would also naturally use his strengths for the role-playing parts - he would impress with raw physical strength, and be able to stab a fly stuck to the wall with a dagger. But they would be somewhat clueless when it comes to processing data, having any situational awareness, and so on. Any amount of variations on that character with a third stat would also work. And you'd get out anything from a swordsman fresh out of a dojo, to a brutish pirate, and all the way to a foil-wielding nobleman.

 

Of course, this would also open up for the option to create a supremely mighty wizard, capable of casting monstrously powerful spells. But at the cost of resistance, physical dexterity, resolve, perception, and so on.

 

In the same way, with for example Perception and Dexterity, you would have a character able to interrupt any ability trigger, any target change, any status change. So that say a dimwitted but industrious priest would be able to use his contact spells (against deflection, bypassing armor) extremely well. For a short time, when calling upon his god. And thanks to high intellect, he would always be guaranteed high lower output on his divine spells. But he would not be able to withstand divine influence from other deities in a critical moment, or deal out any amount of damage physically, or ever maximize his spells.

 

So again, you have this unique synergy with the role-playing aspect and the mechanics in the game.

 

And all the pairs of combinations like that would be viable in the game. Because if you used the strengths of the character, you could always find a situation where they'd be useful - you just had to use your wit to place your characters well and use their abilities to your advantage. This system was pure genius, and transformed the usual "see who runs out of potions first", or "wait until we see who is the lucky dps-er in this situation", into a game of strategic placement and finding a way to actually using your abilities to your advantage in a party.

 

The dojo trained swordsman would be invincible in a one on one fight, for example, and even moreso with the priest healing his wounds. But he would be helpless against an ambush, or if he was ever disabled (or crippled from a long fight).

 

In other words, the massive success of that system was in how it would translate the simple high-level narrative explanations into relatively simple gameplay-mechanics, with the kind of accuracy that a game-master (like me) dreams of. The only sacrifice you would have to make to bask in the magnificence of this system was simply to allow that neither single stat alone represents supreme and unparalleled skill in a specific domain, as we are used to from D&D.

 

Unfortunately, this system was deemed too complex to understand for mere mortal men, and we ended up with the current system instead. Enjoy. Praise and bless.

 

 

Now I won't deny that the stats could use some re-jiggering... back during the early Beta I put together a fairly extensive paper with Sensuki that made a number of suggestions to rebalance the attribute bonuses from stats. Some were adopted, many weren't. In any case, I'd agree that a few of the stats are a little weak at the moment - but they're still perfecting the system.

 

Interesting, is this available to read somewhere?

 

 

http://forums.obsidian.net/topic/68526-how-to-fix-the-attribute-design-in-pillars-of-eternity/

 

..As you can see from the title, this wasn't really a suggestion for a more holistic replacement of the other system, but a "critique" of the existing system at the time. The pdf link is presumably still dead. But highlights include, but are not limited to, how a damage distribution model that doesn't approach a 50% hit ratio, with an average damage output of 50% of the weapon stat, is a failed design.

 

This argument is sustained in the following way: assume golden standard. Demonstrate how the current system does not meet said standard. Insist that the failure of original system is now "mathematically proven". Which then naturally would allow any amount of unargued and utterly unrelated suggestions to be adopted. Because anything would of course be better than a "failed system". We are not allowed to question the "golden standard" distribution, of course.

 

Finally, dismiss anyone pointing out weaknesses with the approach here as making irrelevant objections. Since the circular argument above, and more comically the superior quality of all incidental and unargued suggestions, is now "mathematically proven". If you thought this was the silliest sophistry imaginable, you were, quote "insane".

 

And we are apparently still so convinced of the argument made here that Matt actually brags about how Obsidian adopted some of the suggestions. Obsidian on their end keeps bragging about how responsive they are to the "community's needs". But as you all know, very few of us actually wrote 50-page letters to the devs, insisting that Obsidian should placate us personally, at the cost of the other 70.000 backers. In fact, it's assumed that because Obsidian does not receive 70.000 letters, they all agree with Matt and Sensuki.

 

So now you know how the current stat system came into being. Eothas be praised.. and things.

[Arhu]

 

IMO the attribute system should look something like this:

 

• Strength: +Melee Damage, +Bow Damage (not crossbows), +Fortitude, -Armor Penalty*
• Constitution: +Health, +Endurance, +Fortitude, +Deflection, +Duration of Abilities, +Heal
• Dexterity: +Melee Accuracy, +Non-Spell Action Speed, +Reflexes, +Deflection, +Disengagement Defense
• Perception: +Ranged Accuracy, +Spell Accuracy, +Reflexes, +Buff AoE, +Max Ranged Attack Range
• Intelligence: +Spell Speed, +Offensive AoE, +Duration of Abilities, +Will, +Interrupt, +Crit Damage**
• Resolve: +Spell Damage, +Wand Damage, +Concentration, +Will, +Endurance as a percentage of Health, +Max Ranged Spell Range

*Armor Penalty: -Non-Spell Action Speed, -Reflexes based on the heaviness of the armor

**Crit Damage is a flat bonus to damage rather than a percentage

(..)

At first glance, this actually doesn't seem bad at all. More complicated, but way better than OP's system. With this, I can see e.g. a wizard wanting each stat except Str. But one stat not being that useful for a given class is acceptable, I'd say.

How about moving  "+Offensive AoE" to Strength and for Intelligence use "-friendly fire AoE".

 

In other words, STR would increase AoE size, at the cost of hitting friends. INT would not increase AoE size, but make a portion of AoE damage not hit friends, allowing for a more tactical application of AoE abilities -- similar to how it is currently done in the game.