Guild Wars 2

I’ve been trying to work out some of the more puzzling game-rules/coding that runs some of the secondary mechanics and attributes in GW2 and there’s very little info available.

So if anyone’s interested perhaps we can post and reply as a community to some of the more head-scratching examples in this thread.

If it gets a bit too confusing perhaps someone from the testing community can help clear things up.

Puzzle 1: Protection’s real value.

Ok so the tool-tip states Prot is worth 33% damage reduction. But depending at which point the mitigating clause enters the code it de-values.

An example:

1000 weapon damage x 2000 power x 1.0 (skill co-efficient) / 3000 armour = 666.6 true damage

(so in this example, 3000 armour gives a damage reduction of 33.3%, roughly the same as prot).

However, if the mitigating Prot clause is added post true damage it multiplies against itself:

666.6 × 0.33 = 219.9 damage. (i.e dealing 446.7 true damage overall).

Therefore, if the clause is 2 part, the armour mitigates for 33.3% whereas the Prot mitigates for 22% damage. Giving an overall damage reduction value of 55.3%

If the coding works together though (the mitigating factors are combined), then it works as the supposed 66.3%

Q: When/how does Protection enter the mitigating clause?

Puzzle 2: Anti-condi duration. Scaling or combined?

Similar to the Prot puzzle, the way condi duration is valued is v sketchy.

Lets take attributes vs food values for example:

An example:

You gain +40% condi-duration from attributes but face an enemy with -40%. (Using Warrior Sword Flurry as an example).

Full Flurry is worth 4s baseline. 4 × 1.4 = 5.6s adjusted.

5.6 × 0.6 = 3.36s adjusted.

Therefore, if condi-duration factors are isolated clauses (from both sources). Anti-duration is worth more than Plus-duration.

However, if there is a coding clause that compares the duration attributes and gives a resulting value overall, it comes out as equal (Negating).

Q: Does Condition Duration resolve in an independent clause or not?

Puzzle 3: Damage Scaling Abilities/Traits.

It is very unclear how percentage damage increase compare in real terms to flat increases.

Take the Warrior Trait ‘Destruction of the Empowered’. It increases outgoing damage by 3% for each boon on your target.

However, it isn’t clear when the multiplier is added. Is it only added to the ‘weapon damage’ quota of the DD formula? Is is multiplied after the entire DD damage formula but before mitigating factors? Is it multiplied post mitigating factors?

This is a difficult to account for because it obviously scales incredibly well depending on your power and might stacks.

It is also unclear how it interacts with Crit. Is it multiplied post Crit and therefore worth both 3% increase in DD and Crit %?

Q: At what point are scaling clauses added to the damage formula?

• suggestion – make trait sourced multipliers scale before mitigating factors but vulnerability scale post-mitigation, making vulnerability worth more – placing it higher in priority (or make it consider crit and be worth +1% on DD and +1% Crit).

I know this is a bit of a nerd-fest, but I thought it’d be useful to some and interesting to others to try and work it out.

I may be way off but I can’t be the only one who scratches their head with this stuff when trying to craft builds.

Please feel free to add your own mechanics ‘puzzles’ below.

Insight and discussion is welcome.

For puzzle 2, all + and – duration sources are added before being applied to the final result (cap of +100% duration). Neither is more valuable than the other from a purely mathematical perspective.

In other words, +40% duration and -40% duration cancel each other out perfectly.

This is not “my opinion” FYI, but how the game actually works.

I do think it’s logical to assume condition duration would resolve in an independent clause, do you have a source at all? Or is this from personal testing?

Its not that I wouldn’t believe you but is there something you can link to show-case the theory?

Personal testing. Also: http://wiki.guildwars2.com/wiki/Condition_duration

Specifically:
“Condition duration from various sources increases (and decreases) additively as a percentage increase on the base condition duration.”

Protection is simple. If you would average 1000 damage every 2 hits against a target without protection, you will average 1000 damage every 3 hits against the same target with protection. This is true regardless of the values of any other modifier present.

Ok, so the decreasing values must interact with the baseline value of a skill and disregard the incoming adjusted value.

Thank you for clearing it up

Can you help me with the others? :p

@Coldart

I follow the idea of protection as an over-time concept but what I’m trying to work out is how it relates to other mitigating factors, namely your armour.

If you looked at a singular value of damage. Would an Armour value end up being worth more than the prot (relatively) because they are measured independently?

They are multiplied together. Whatever damage you would normally take with your armor rating, Protection multiplies that by .67.

Just please, please don’t try to use the actual percentage reductions. 2,000 armor is a 99.95% damage reduction, for reference, so trying to work with percentages just is confusing.

Protection is calculated after armor reduction.
Condition duration is calculated before duration reduction. (No the don’t cancel each other out.)
+% damage modifiers are calculated after power/critical calculations and are calculated multiplicatively.

In general base calculations are gonna be done before modifiers and personal calculations are done before enemy calculations (conditions durations for example.)

Before last April’s patch we had ‘Steady State’ Weapons in Heart of the Mists that had specific damage values (without a range) that allowed for much more consistent data sets for testing various mechanics and for minmaxing of theoy-crafting in the Mists, but for some reason this where removed.

Regarding #3, the order of how multiplicative modifiers are applied on both the attacker and defender shouldn’t matter.

Say you attack something:
Start with the multiplicative identity 1
Your weapon has 1000 weapon strength (x1000)
Your attacking skill has coefficient 0.7 (x0.7)
You have 3000 power: 2300 base, 20 stacks of might, and +100 damage nourishment (x3000)
You critically hit with 1125 ferocity (x2.25)
You have a 2 traits that give you 10% bonus damage (x1.1^2)
Your target has 2000 armor (x0.0005)
Your target has 25 stacks of vulnerability (x1.25)
Your target has protection (x0.666)

You can apply these modifiers in any way you want, and the resulting number will be the same.

@Drarnor

Ok so what you’re saying is Protection is a flat-rate value added to the damage reduction you’ve already gained from Armor? It doesn’t interact independently as a mitigating factor? (i.e it’s an attribute gain that takes place before damage is adjusted?).

I don’t understand your point about 2000 Armour being 99.95% damage reduction however. I was considering real values of base-line gear armour + Toughness. Maybe I’m just being thick but can you explain a little?

@Tman

I have to admit, this is how I thought the engine worked. Although its logical to assume factors are considered in the same mitigating/modifying parts of the formula, I can’t imagine the servers would be up to it.

This is why I made the post because if the values are calculated in isolation it makes a significant difference.

@Altair

I see what you’re saying about it not mattering at what point its added into the formula. I don’t know why I was trying to isolate it.

Wow, that is pretty strong. Does anyone have any info on the comparable real value of scaling damage vs power increases? (When considering trait choices, for example)?

Thanks for the help though guys, much appreciated.

Byron.1902:

@Drarnor

Ok so what you’re saying is Protection is a flat-rate value added to the damage reduction you’ve already gained from Armor? It doesn’t interact independently as a mitigating factor? (i.e it’s an attribute gain that takes place before damage is adjusted?).

I don’t understand your point about 2000 Armour being 99.95% damage reduction however. I was considering real values of base-line gear armour + Toughness. Maybe I’m just being thick but can you explain a little?

Protection, being a multiplication on a final result, can be calculated before or after other factors with no difference in output. It is a separate factor that multiplies incoming damage by .67, with armor (and other damage reductions) factoring in separately.

@Tman: Sorry, but you are completely wrong on condition duration. Say you have someone wearing Runes of Melandru (-25% condition duration) and they get hit with a 6 second chill from a Necromancer with 5 points in Spite (absurdity of that statement aside, it’s +25% condition duration). The resulting chill lasts 6 seconds. On someone without the Melandru runes, the same Chill would last 7.5 seconds.

That is what the wiki means when it says “Condition duration from various sources increases (and decreases) additively as a percentage increase on the base condition duration.” Add and subtract the modifiers before applying them to the final result. It takes like 30 seconds to test this for yourself in a custom arena.

Byron.1902:

—

Does anyone have any info on the comparable real value of scaling damage vs power increases? (When considering trait choices, for example)?

—

Damage equation is skill coefficient * weapon strength * power * modifiers.

So for example 10% increased damage is same as getting 10% more power, 10% stronger skill or 10% higher weapon strength.

Since skill coefficients are stable and people often have same weapon strength what you really care about is power * modifiers.