List of PoE1 Inconsistent Stuff

[MaxQuest]

List of PoE1 stuff that in my opinion is inconsistent or potentially confusing. It would be nice to have it fixed, changed or improved in PoE2:

 

1. Modals

Problem:

- there are no clear rules which would state which modals are mutually exclusive

- it is unclear why Galant's Focus is mutually exclusive with any of ranger's modals

- it is unclear why Reckless Assault is not mutually exclusive with anything, why Savage Attack switches-off Cautious Attack, Defender and Guardian Stance

- modals exclusivity table for reference: link

Suggestion:

- create modal categories and two category groups:

.- Unrestricted (can have any number of modals of this group active at a time)

.- Restricted (can only 1 be active at a time), for example: Defensive Self, Offensive Self, Ranged Mastery, Supportive Aura

- this way it's clear, you can have one defensive modal, one offensive, one ranged (swift/vicious/twinned-arrows/powder-blast), one aura and any number of unrestricted modals enabled at the same time.

 

2. Lashes

Weapon damageRoll goes against enemy respective DR.

Weapon lash, is 25% of damageRoll that goes against 25% of enemy respective elemental DR.

One could expect that lash has similar behaviour to regular weapon damage, just is multiplied by 0.25. But it doesn't deal MIN damage, meaning that it can be completely eaten by DR.

 

3. Weapon UI speeds

Expected behaviour:

- fast weapons are faster than average

- average weapons are faster than slow

Encountered behaviour:

- slow 2h melee weapons have the same speed as average 1h melee weapons

- fast ranged weapons have the same speed as slow 2h and average 1h melee weapons

- weapons base speed values for reference: http://i.imgur.com/QqMIkII.png

Problem:

- creates confusion for the player, when he tries to evaluate relative efficacy of one weapon over the other. One will think that all fast weapons have the same attack rate.

Suggestions:

- rename fast melee weapons into "very-fast"

- rename average and slow melee weapons into "fast"

- this way all weapons of the same UI category will have the same speed.

 

4. +x Attack Speed tooltips

Problem:

- not a single talent, property or buff that states that it increases attack speed, actually affect it.

- instead they reduce recovery duration

Suggestions:

- update all related tooltips to reflect the true state of things "+x attack speed" => "reduces by x% recovery duration" / "reduces by x% base recovery duration"

- change recovery_factor from 1.2 => 1.0; this way x% at least will reflect a closer to reality reduction in action duration (provided there are no other similar effects active yet)

Notes:

- current attack speed system is actually great. It is deep, and in current "ecosystem" (items/talents/buffs/encounters at hand) quite is balanced. There is no need to revamp it, just to tweak a little; update tooltips and overall make it more transparent to the player (perhaps even showing system-calculated action duration)

- reference: link

 

5. Damage over Time

Problem:

- DoTs do no benefit from Elemental talents like Scion of Flame, Spirit of Decay and so on

- DoTs do no benefit from Creature Slayer talents like Beast Slayer, Ghost Hunter and so on

- The rule of thumb is that DoT DPS is not influenced by INT. Regular DoTs obey this principle, increases both their duration and damage, DPS remaining the same. But DoTs with fixed damage (like wounding shot) violate this idea. Duration gets increased by Int, total damage remains the same, and DPS is going down.

- There is a lack of unambiguous rules regarding stacking. Wounding does stack with itself. Deep Wounds doesn't. There is no way to tell if Dragon Thrashed or Shining Beacon stacks with itself, until we test it. Mouseover tooltips display only one appliance active at a time; but they do stack.

Suggestions:

- DoTs should benefit from Elemental and Creature Slaying talents. Or at least explicitly stated in their tooltips that they do not.

- DoTs with fixed damage, should not have their duration affected by Int.

- DoTs should have an extra field in their tooltips. Stacking:

.- Stacking: (unrestricted or) complete

.- Stacking: (restricted to) one per caster

.- Stacking: (restricted to) one appliance

Current DoT mechanics for reference: link

 

6. Shield Bash

Expected behaviour:

- in terms of dps [1h + bashing_shield] is between [1h + shield] and [1h + 1h]

Problem:

- provided enough recovery reduction, [1h + bashing_shield] falls behind of [1h + shield] in terms of auto-attack dps (provided there are no special proc effects)

 

7. Immunities

Problem:

- When trying to apply Envenomed Strike to a poison immune enemy you get "misses with addional effect (Fort:Imm. = Imm..)" message in combat log. That is not that descriptive

Suggestion:

- When an effect is "missed" due to an immunity, combat log should state just so.

- It would be nice to have all Ground, Poison, Diseases spells and abilities have it stated in their tooltips.

- (Optional) If the target is immune anyway, maybe it would make sense to make it impossible to try to apply related effect, which results in a waste of charge/spellusage/casttime.

 

8. Self buffs that last until hit or crit

Compare the following 3 wizard spells and their tooltips:

- Wizard's Double: +40 Deflection until hit or critically hit (no duration limit)

- Mirrored Image: +25 Deflection until hit or critically hit for 60s

- Ironskin: +8 Damage Reduction until hit or critically (no duration limit)

Problem:

- inconsistency

- wrong tooltips

Reality:

- Wizard's Double: +40 Deflection bonus; bonus disappears on first incoming hit/crit

- Mirrored Image: +25 Deflection bonus; bonus is reduced by 5 on each incoming hit/crit

- Ironskin: +8 Damage Reduction; bonus disappears on 10th incoming hit/crit

Suggestion:

- ... at least make tooltips reflect real spell effects and decay)

 

9. Elemental talents

Problem:

- as was already mentioned in #5, they do not increase damage of elemental Damage over Time effects

- the second problem is related to inconsistent interaction with weapons that deal elemental damage:

.- Matching elemental talent (e.g. Scion of Flame) increase damage dealt by pure elemental weapons (e.g. Firebrand)

.- Matching elemental talent (e.g. Spirit of Decay) increase damage dealt by elemental/physical weapons (e.g. Bitercut corrode/slash). And they even increase the physical damage. (i.e. in case of Bittercut, Spirit of Decay increases slash damage as well)

.- Elemental talents DO NOT increase damage dealt by physical/elemental weapons (yes, even if enemy has high physical, but low elemental DR, and elemental damage is dealt) (e.g. Durance's Staff, crush/burn)

 

10. AoE/FoE inconsistency

If you have high int, AoE spells get increased area of effect. This yellow zone is FoE meaning that friendly targets inside it should not get hit.

Problem:

- Powder Burns disregards it's FoE zone. Allies standing in yellow part of the cone are still being hit.

Note:

- When the system uses a check for party member, Animal Companion should also pass it. Iirc, depending on patch, pets had problems with FoE zone of some AoE abilities.

 

11. Weapon procs and status effects

Problem:

- there is no way for a player to know exactly how long is the proc duration (e.g: the stun of stunning weapons; or prone of overbearing weapons)

- game also doesn't communicate if duration is modified by hit quality or not

Effects for reference:

 

 

Desiccating (Captain Viccilos Anger)(on hit) : roll vs Fort | duration: 0 on miss; 3.0s on graze; 6.0s on hit; 9.0s on crit
Disorienting (The Vile Loners Lance)(on hit) : auto-applied | duration: constant 5.0s
Interfering (Lost Thayns Reach)(on hit)      : auto-applied | duration: 5.0s on hit; 7.5s on crit
Overbearing (We-Toki)(on crit)               : roll vs Fort | duration: 0 on miss; 1.5s on graze; 3.0s on hit; 4.5s on crit
Persecuting (St. Wygelts Cudgel)(on hit)     : auto-applied | duration: constant 6.0s
Stunning (Starcaller)(on crit)               : roll vs Fort | duration: 0 on miss; 1.0s on graze; 2.0s on hit; 3.0s on crit
Stunning (Godansthunyr)(on crit)             : roll vs Fort | duration: 0 on miss; 1.0s on graze; 2.0s on hit; 3.0s on crit
Wounding (Drawn in Spring)                   : auto-applied

Property effects:

- Desiccating: inflicts minor fatigue (-10% max. endurance, -10 accuracy)

- Disorienting: inflicts -5 all defenses

- Interfering: inflicts -5 accuracy

- Persecuting: drains 5 resolve from the target and you gain +5 Deflection

- Overbearing: chance to prone

- Stunning: chance to stun

- Wounding: +25% raw damage inflicted over time

 

 

Will probably add more stuff as I remember it.

P.S. Feel free to add to the list, I will append it.

Edited March 5, 2017 by MaxQuest

[Greensleeve]

Just posting to make sure people read this. It'd be lovely if this stuff was fixed in Deadfire. 

[George_Truman]

I have always thought Ironskin was god awful. Now it makes more sense.

[Madscientist]

Thanks a lot:

- about modals: Yes, that was very confusing. In the beginning I thought I could have only 1 modal at a time. So I asked in the forum if my char should take savage attack or vulnerable attack because both were offensive modals. The answer was: Take both.

 

- about attack speed: What does "Attack Variation" mean in the table?

 

- about dots: As written in the other thread, it should be describes as "Dot of Doom causes 10 points of fire damage every 3 seconds over 10.5 seconds (=45 damage total)" instead of "Dot of Doom causes 45 fire damage over 10.5 seconds." Since each tick is reduced by 25% of the DR, it would be nice to know how big each tick is and when they are applied. ( All ticks are applied at 0sec, then every 3 seconds and if the duration cannot be divided by 3 there is a smaller tick at the end. But I know this only because of MaxQuest, not because of any information in the game.)

The problem with stacking comes on top of that.

 

- about immuneties: What do you mean with the last sentence?

"- (Optional) If the target is immune anyway, maybe it would make sense to make it impossible to try to apply related effect, which results in a waste of charge/spellusage/casttime."

 

When it is immun, the effect is not applied and it cannot be applied, because this is the definition of immunity.

Maybe you mean that no roll is made in that case, it just shows that the target is immun.

 

examples (I do not know an example in PoE, so I create my own):

1.) There is a weapon that causes poison on hit and you attack a target that is immun to poison.

The combat log says: You caused 10 pierce damage, poison failed (target immun)

 

2.)There is an earthquake spell that causes crush damage and can knockdown the target.

You cast this spell on a bird. (Birds fly so they are not hit by the earthquake)

I have no Idea how to say this in the combat log. I had such a situation in some Final Fantasy games, but there is no combat log. There was only the word "missed" visible at the enemy instead of a damage number.

[Aotrs Commander]

I think the phrasing (and possibly even how the function works) of DoTs ought could be much improved, Don't tell me "does 100 damage over twenty seconds," tell me "does five damage per second for twenty seconds." Or, if "tick" is a unit, then tell me "does 3 damage per tick for 33 ticks" (where "tick" is a infotabbed doofer like all the other terms, which mousing over tells me "one tick is half a second" or something.

 

If you were really, really going to push the boat out, something like "does 17 damage per [unit] for 13 [units], (221 damage total)."

 

(Many - if not most, I would hazard - of us play tabletop RPGs, and I doubt many of us have problems grokking "does 5 damage per round for 10 rounds", whereas "deals 50 damage over ten rounds" is not something that is nearly as clear and not something you see in any set of RPG rules I have personally encountered.

 

It would way more transparent.

 

More wordly, yes, but one thing that has ben brought abundenly to my attention after I have spent much of the last two or three weeks revising, updating and indexing D&D 3.5, sometimes, there is no substitute for being wordier oif it makes it ultimately clearer. As there PoE is not tied to a page count, and can afford to have these little help box doofers, it seems to me that it could afford to be a bit more explicit and still not clutter up the interface or anything.

 

 

 

(Tangentially related, I have never quite understood why some devs don't come outright and tell you the nitty-gritty of the game mechanics (Paradox Development, lookin' at you in particular!) rather than leave the community to puzzle it out. Why not just tell everyone straight "this is how this works, chaps," from the get-go?)

Edited March 4, 2017 by Aotrs Commander

[Dr. Hieronymous Alloy]

List of PoE1 stuff that in my opinion is inconsistent or potentially confusing. It would be nice to have it fixed, changed or improved in PoE2:

 

 

4. +x Attack Speed tooltips

Problem:

- not a single talent, property or buff that states that it increases attack speed, actually affect it.

- instead they reduce recovery duration

Suggestions:

- update all related tooltips to reflect the true state of things "+x attack speed" => "reduces by x% recovery duration" / "reduces by x% base recovery duration"

- change recovery_factor from 1.2 => 1.0; this way x% at least will reflect a closer to reality reduction in action duration (provided there are no other similar effects active yet)

Notes:

- current attack speed system is actually great. It is deep, and in current "ecosystem" (items/talents/buffs/encounters at hand) quite is balanced. There is no need to revamp it, just to tweak a little; update tooltips and overall make it more transparent to the player (perhaps even showing system-calculated action duration)

- reference: link

 

 

 

I'll disagree to the extent that I think the current system is too confusing. Typical players don't even know what "recovery" even is (which is why they just called it "attack speed" -- it sounds simpler intuitively.) Just speaking as someone who writes a lot of guides -- the current system is too complicated to explain to most players.

 

I'd also add that the addition of a separate "reload speed" category adds a further layer of confusion and complication AND causes balance issues in the endgame for bows vs crossbows/guns (as "recovery" and "reload" are distinct, and it's much harder to reduce "reload.") 

 

The whole system needs to be simplified. Combine reload with "recovery" and rethink the terminology throughout. Importantly, have an IN GAME DISPLAY ON THE CHARACTER SHEET of your attack speed / time between attacks with each weapon, so the player can see without having to scribble a bunch of math.

 

 

2. Lashes

Weapon damageRoll goes against enemy respective DR.

Weapon lash, is 25% of damageRoll that goes against 25% of enemy respective elemental DR.

One could expect that lash has similar behaviour to regular weapon damage, just is multiplied by 0.25. But it doesn't deal MIN damage, meaning that it can be completely eaten by DR>

 

Correct me if i'm wrong here, but from what I remember, Might does not effect weapon lash damage, which leads to a lot of potential confusion (especially involving things like Soul Whip, which is technically a lash). Just another way the game doesn't follow player expectations.

 

 

 

 

5. Damage over Time

 

I would suggest standardizing this by having Int give extra damage ticks, while Might gave more damage per tick, across the board, for ALL DoT effects. It's intuitive, balanced, straightforward, and simple.

Edited March 4, 2017 by Dr. Hieronymous Alloy

[Hynkel]

Yup, any given DoT effect should also never stack with itself, to prevent multiple Dragon Thrashed silliness.

[Madscientist]

- about attack speed: I think tyranny showed the total recovery time for your current equipment (base+weapon+armor). Unfortuanatly it did not tell you how long the attack animation is. So you still do not know the time for a full attack cycle (attack+recovery)

 

- I think dots are not so bad the way they are now. Int adds more ticks and both mig and int increase the total number. The problem is to communicate it better.

[tinysalamander]

AFAIR in PoE1 charmed enemies where treated as allies and got buffs just like everyone else. That might have been fine (although I don’t really care about this part) they also could be healed whether I wanted or not (AoE heals, hello) and retained various buffs after charm expired. It’d be nice if PoE2 would do something about it if it wasn’t fixed.

[Dr. Hieronymous Alloy]

Yup, any given DoT effect should also never stack with itself, to prevent multiple Dragon Thrashed silliness.

 

Well, weapons that apply DoTs would be really nerfed by that. Maybe just make them always stack, and balance them all appropriately.

[JerekKruger]

Correct me if i'm wrong here, but from what I remember, Might does not effect weapon lash damage, which leads to a lot of potential confusion (especially involving things like Soul Whip, which is technically a lash). Just another way the game doesn't follow player expectations.

 

I'm fairly certain that's wrong. First, as I understand it, Soul Whip acts like other +X% damage bonuses in that it is an additive multiplier of weapon's base damage, so you're right that Soul Whip doesn't benefit from Might. However Lash's essentially act as a second multiplier, since they take the damage done by the weapon after all other modifiers have been added, multiply that by 25% and do that much extra damage of the appropriate elemental type. As such Lash's do benefit from Might, and in fact also benefit from the bonus damage of things like Soul Whip and Sneak Attack.

 

I don't know whether Soul Whip is technically a lash, or what that would mean if it was.

[Dr. Hieronymous Alloy]

ok, here's an additional weird inconsistency, copied from a post on SA:

 

In response to the question "how do elemental damage talents work":

 

It depends on the order that the damage is listed.

 

If it is element/physical, then the corresponding talent will work.

 

If it's physical/element, then the corresponding talent won't work.

 

So, the talent will work if it's Bittercut or Stormcaller, but if it's Durance's staff it won't.

 

Edit: Also, they don't raise base damage, they're additive and only increase the damage as a proportion of the base.

 

 

https://forums.somethingawful.com/showthread.php?threadid=3807509&userid=0&perpage=40&pagenumber=149#post469964944

[JerekKruger]

Yeah, I assume the way dual damage works is a quirk of how the game is programmed. There are also some situations where the weapon will use it's first damage type even when it would do more damage with it's second damage type (I can't remember the details, but it has to do with DR being fairly close).

 

As for not raising the base damage: that's how all + damage things work. It's why changing Sabres from 13-19 to 11-16 +20% was a nerf, even though (11-16)*1.2 = 13-19 (near enough).

[dukeisaac]

AFAIR in PoE1 charmed enemies where treated as allies and got buffs just like everyone else. That might have been fine (although I don’t really care about this part) they also could be healed whether I wanted or not (AoE heals, hello) and retained various buffs after charm expired. It’d be nice if PoE2 would do something about it if it wasn’t fixed.

I kinda liked the extra micro in this but I wouldn't be mad if they removed it

[Boeroer]

Great thing MaxQuest. Like Dr. Hieronymous Alloy said, you could add the problem that Firebrand, Cadebald's Blackbow, Bittercut and Stormcaller work with elemental boosters, but Durance's Staff doesn't - just because of the order of damage types. How could anybody who doesn't roam these forums day by day know this stuff?

 

And this thread should be pinned.

 

Just my opinnion.

 

Pin intended... 

[Madscientist]

 

ok, here's an additional weird inconsistency, copied from a post on SA:

 

In response to the question "how do elemental damage talents work":

 

It depends on the order that the damage is listed.

 

If it is element/physical, then the corresponding talent will work.

 

If it's physical/element, then the corresponding talent won't work.

 

So, the talent will work if it's Bittercut or Stormcaller, but if it's Durance's staff it won't.

 

Edit: Also, they don't raise base damage, they're additive and only increase the damage as a proportion of the base.

 

 

https://forums.somethingawful.com/showthread.php?threadid=3807509&userid=0&perpage=40&pagenumber=149#post469964944

 

 

One solution would be, that the damage+ talent (like scion of flames) is always applied when the elemental damage is used (including dots).

But in this case it would lead to a different problem. Lets assume an enemy has 10 crush DR and 11 fire DR and you attack with durance staff (when scion of flames works). The game would use crush damage because of the lower DR, but fire damage would be higher if the +20% produces more than 1 point of damage.

So at the moment I do not see a system that will produce consistent results every time.

 

Anyway, I think it is bad that the effect depends on the order in which the 2 damage types are listened.

 

I also like this topic to be pinned.

[JerekKruger]

One solution would be, that the damage+ talent (like scion of flames) is always applied when the elemental damage is used (including dots).

But in this case it would lead to a different problem. Lets assume an enemy has 10 crush DR and 11 fire DR and you attack with durance staff (when scion of flames works). The game would use crush damage because of the lower DR, but fire damage would be higher if the +20% produces more than 1 point of damage.

So at the moment I do not see a system that will produce consistent results every time.

 

This would be very easy to solve though: when you attack with a dual damage weapon the game should make its roll, then check how much damage this would result in for both damage types (factoring in all modifiers and the enemy's DR), then resolve at whichever came out higher.

[AndreaColombo]

Hope the devs have eyes on this.

[Madscientist]

 

One solution would be, that the damage+ talent (like scion of flames) is always applied when the elemental damage is used (including dots).

But in this case it would lead to a different problem. Lets assume an enemy has 10 crush DR and 11 fire DR and you attack with durance staff (when scion of flames works). The game would use crush damage because of the lower DR, but fire damage would be higher if the +20% produces more than 1 point of damage.

So at the moment I do not see a system that will produce consistent results every time.

 

This would be very easy to solve though: when you attack with a dual damage weapon the game should make its roll, then check how much damage this would result in for both damage types (factoring in all modifiers and the enemy's DR), then resolve at whichever came out higher.

 

 

In this case the game has to calculate the damage for both types, compare them and use the bigger one.

I assume ( I do not know ) that the game selects first what damage is used and then it calculates the damage. I assume the game wants to calculate as little as possible in order to save computer resources.

[JerekKruger]

I assume the game wants to calculate as little as possible in order to save computer resources.

 

That might be the case, but it really shouldn't be. Performing simple calculations like this takes a computer an incredibly small amount of time, and there aren't actually all that many to do in combat. Of course for each such roll the game has to look up various values in various locations (character stats, monster stats, ability modifiers etc.) and that might be a bottleneck, but I don't think it should be.

[Aotrs Commander]

 

I assume the game wants to calculate as little as possible in order to save computer resources.

 

That might be the case, but it really shouldn't be. Performing simple calculations like this takes a computer an incredibly small amount of time, and there aren't actually all that many to do in combat. Of course for each such roll the game has to look up various values in various locations (character stats, monster stats, ability modifiers etc.) and that might be a bottleneck, but I don't think it should be.

 

 

When you look at something like, say Crusader kings 2 that has to deal with the calculations and weighted decisions for hundreds of AI characters in real time (at its maximum speed), for AI War or something that is constantly deciding what sometimes thousands of unit are doing in real time, performing calculations for maybe no more than twenty or so should be no trouble, really. The slow down for PoE when you sometims have lots of enemies and such on the screen is, I suspect, far more graphics-related (not necessarily just graphics) than dealing with those sort of simple calculations; the sort of thing even the likes of I could program into a spreadsheet.

 

(I mean, it's a simple command, excuse my spreadsheet formula IF("damage 1" > "damage 2", "damage 1", "damage 2"). I can't imagine that would be difficult or resource consuming for the computer to run.)

[Madscientist]

I do not know much about how computers work and I have never played PnP RPG, but I think about how would it be if I have to sit at a table and roll some dice myself to get a result.

 

I did a run in the park and did some thinking:

- talents that give +20% fire damage give this bonus to all fire damage (same for all other elements)

- add talents that gives +20% slash/crush/pierce damage (for balancing. If you say it would be OP because you attack more often with a weapon than with an element, just reduce this number to e.g. 10%) (EDIT: I did some thinking if this amkes sense and I came to the conclusion that it does not make more or less sense than having a talent for elemental damage)

- If you make an attack that uses the best of 2 (or more) damage types:

1.) look at the target DR

1a) If the DR for both damage types is different, use the damage type with the lower DR

1b) If DR is the same, select randomly which damage type is used (to avoid a systematic bias towards one type, as we have it now with bittercut vs. durance staff)

2.) Calculate the damage with the selected damage type

 

This rule would be easy to understand. Under some conditions it might be possible that the other damage type would have caused a bit more damage. But if we have a system that checks tons of conditions in order to select something, most players will not understand it at all.

Edited March 5, 2017 by Madscientist

[LuccA]

This thread is very on-point and important. Nicely done!

 

Two things I am also not very sure/ maybe it could be explained better in the game:

- Is every percentage bonus in any stat applied to the base value of that stat?

- Most of status effects are applied to an attribute and something that is probably affected by that same attribute. Let's say "Dazed" causes a penalty to the Resolve and the Will defense (I'm not sure). The penalty to resolve will also generate a penalty to Will. Is that on top of the penalty directly to Will or is it already included in it?

Edited March 5, 2017 by LuccA

[Madscientist]

This thread is very on-point and important. Nicely done!

 

Two things I am also not very sure/ maybe it could be explained better in the game:

- Is every percentage bonus in any stat applied to the base value of that stat?

 

 

Damage is calculated like this:

damage done =  ( base damage * ( 1+ modyfier1 + modyfier2 + . . . ) - target DR )

 

example: base damage = 10, you get +50% from crit, +50% from sneak attack, + 20% from might and the target has 5DR

 

damage = ( 10 * ( 1 + 0,5 + 0,5 + 0,2 ) - 5 ) = ( 10 * 2,2 ) - 5 = 22 - 5 = 17

 

Edit: You will always do 20% min damage, which would be 3,4 in this case. So if the enemy has more than 14DR, you will always do 3 damage.

 

Sorry, I do not know the answer to your second question.

Edited March 5, 2017 by Madscientist

[JerekKruger]

- If you make an attack that uses the best of 2 (or more) damage types:

1.) look at the target DR

1a) If the DR for both damage types is different, use the damage type with the lower DR

1b) If DR is the same, select randomly which damage type is used (to avoid a systematic bias towards one type, as we have it now with bittercut vs. durance staff)

2.) Calculate the damage with the selected damage type

 

This rule would be easy to understand. Under some conditions it might be possible that the other damage type would have caused a bit more damage. But if we have a system that checks tons of conditions in order to select something, most players will not understand it at all.

 

Why not go with the simpler solution of calculating the potential damage for both sources then using the higher of the two? I can (almost*) guarantee it will have no noticeable affect on performance and it ensures that the weapon does what it's described as doing i.e. using whichever damage type is the best.

 

Assuming you intend your method to be used in combination with a change making talents like Scion of Flame apply only when the weapon uses it's burn damage, your solution will sometimes cause the weapon to use the damage type that is weaker.

 

- Is every percentage bonus in any stat applied to the base value of that stat?

 

For damage, the vast majority of +X% damage bonuses work as an additive multiplier of the base damage of the weapon/ability/spell you're using. By "additive multiplier" I mean that if you have several such bonuses, you first add them together and then multiply your base damage by the total (rather than multiplying by each bonus in turn). The only exception I can think of off the top of my head would be Lashes, don't add directly to the +X% damage modifier of the weapon, but instead take the final damage, multiply it by 0.25 (or 0.3 if you have the appropriate elemental talent) and do that much damage against the enemy (reduced by 0.25 of the enemy's DR). As such Lashes are essentially pure multipliers.

 

For other bonuses I am less certain. For example, I don't know whether a 15 Intellect Paladin wearing the Boots of Zealous Command would have a 1.35 (1 + 0.2 + 0.15) multiplier on their Zealous Auras or a 1.38 multiplier (1.2*1.15). I suspect that it is the former, but have never tested it directly.

 

*It definitely shouldn't, but depending on exactly how Obsidian have coded it I guess it could, though I really doubt it.