Guild Wars 2

With a 50% crit chance, what will do more damage

10% crit dmg
or
100 Power

Using the manual stat adjustments in gw2buildcraft.com, at 50% chance power is better.

maybe this will address your question

http://wiki.guildwars2.com/wiki/Critical_hit

Acotje.5689:

With a 50% crit chance, what will do more damage

10% crit dmg
or
100 Power

That depends on how much Power and crit damage you already have. There’s no general answer to this.

Depends, here the forumla.

Your expected damage is: Expected damage = Base damage * (((0.50 + Critical Damage / 100) * (Critical Chance / 100)) + 1)

We know you have 50% chance. So…

Expected damage = Base damage * (((0.50 + Critical Damage / 100) * 0.5) + 1)

Now your question is +100 base damage or +10% crit damage. That separates the question to two formulas and which is greater?

Base damage * (((0.50 + (Critical Damage 10) / 100) * .5) + 1)
(Base damage100) * (((0.50 + Critical Damage / 100) * .5) + 1)

If your base damage was 2500, and had 0 crit damage increase previously.

2500 * (((0.50 + 10 / 100) * .5) + 1) = 3250 (crit)
(2500+100) * (((0.50 + 0 / 100) * .5) + 1) = 3250 (power)

It’s same! So, this is your root at 0 original crit.

or if power was a very high 3500
4550 (crit)
4500 (power)

or if power was a very low 1500
2000 (crit)
1950 (power)

Just fill in your stats and see which is higher expected damage.

Grumpy.8365:

Depends, here the forumla.

Your expected damage is: Expected damage = Base damage * (((0.50 + Critical Damage / 100) * (Critical Chance / 100)) + 1)

We know you have 50% chance. So…

Expected damage = Base damage * (((0.50 + Critical Damage / 100) * 0.5) + 1)

Now your question is +100 base damage or +10% crit damage. That separates the question to two formulas and which is greater?

Base damage * (((0.50 + (Critical Damage 10) / 100) * .5) + 1)
(Base damage100) * (((0.50 + Critical Damage / 100) * .5) + 1)

If your base damage was 2500, and had 0 crit damage increase previously.

2500 * (((0.50 + 10 / 100) * .5) + 1) = 3250 (crit)
(2500+100) * (((0.50 + 0 / 100) * .5) + 1) = 3250 (power)

It’s same! So, this is your root at 0 original crit.

or if power was a very high 3500
kitten (crit)
4500 (power)

or if power was a very low 1500
2000 (crit)
1950 (power)

Just fill in your stats and see which is higher expected damage.

Sorry for bumping an old thread but I am trying to understand the numbers here.

Its seems to me that no matter what number your crit damage is, replacing 10% crit damage with 100 power will always be more effective. Is that correct or am I doing something wrong?

OK wait, I see now. 100 power is better than 10% crit DMG, but there are no stat combos that allow you to replace crit DMG with power…

Grumpy used an old formula (well, old thread), but he also calculated it wrongly. He added 100 power to the base damage … let me redo the formulas …

The currently assumed formula for most skills is:

• baseDmg = wpnAtk * statPower * skillCoeff / statDef
  Ofc, weaponatk, statpower and skillcoefficient are from the attacker and statDef from the target.

Critting equals (50% + crtDmg) bonus damage (on crit) as Grumpy explained. The formula for avg damage stays:

• targetDmg = baseDmg * (((50% + crtDmg) / 100% * (crtCha / 100%)) + 1)
  Where critdamage and crtchance are taken from the attacker and are values in percent.

Now I have to compare the power addition to the crtDmg addition to check which is higher. Assuming you use the same skill with the same weapon at the same target the formula can be shorted to:

• dmgCoeff = (statPower + addPower) * (((50% + crtDmg + addCrtDmg) / 100% * (crtCha / 100%)) + 1)

So you need the “old Power”, “old crtDmg”, “crtChance” and the adders to compare the values.
The formula looks is similar to grumpy’s, but you don’t use the attack value, but wpnAtk * statPower .

I assume the statcombos were runes, not prefixes.

Nretep.2564:

Grumpy used an old formula (well, old thread), but he also calculated it wrongly. He added 100 power to the base damage … let me redo the formulas …

The currently assumed formula for most skills is:

• baseDmg = wpnAtk * statPower * skillCoeff / statDef
  Ofc, weaponatk, statpower and skillcoefficient are from the attacker and statDef from the target.

Critting equals (50% + crtDmg) bonus damage (on crit) as Grumpy explained. The formula for avg damage stays:

• targetDmg = baseDmg * (((50% + crtDmg) / 100% * (crtCha / 100%)) + 1)
  Where critdamage and crtchance are taken from the attacker and are values in percent.

Now I have to compare the power addition to the crtDmg addition to check which is higher. Assuming you use the same skill with the same weapon at the same target the formula can be shorted to:

• dmgCoeff = (statPower + addPower) * (((50% + crtDmg + addCrtDmg) / 100% * (crtCha / 100%)) + 1)

So you need the “old Power”, “old crtDmg”, “crtChance” and the adders to compare the values.
The formula looks is similar to grumpy’s, but you don’t use the attack value, but wpnAtk * statPower .

I assume the statcombos were runes, not prefixes.

That doesn’t make a difference does it? Power is still worth more?

The answer is:
“there is no simple answer such as ‘power is better’”.

If you have 100 power and want to compare + 10% vs + 100 power, the + 100 power is obviously better since it basically doubles your damage. But if you already have 1,000,000 power and compare the same, th e+ 10% crtDmg will obviously better since the + 100 power will barely have any effect.
Or … what’s the point of adding + 1,000,000% critdamage with 0% critical chance?

You need to insert the original values into the function
Or
For some players the + 100 power will be better and for some the + 10% crtDmg
Or
It depends on your stats

It’s more about the scaling of the skills than dependend on your stats, I’d say. And most skills scale quite well.

Of course, you can always calculate extreme cases, but the game has limits. A character with full power equipment and traits has something of about 2000 power, maybe add a few hundret for special effects. Add even more for buffs like might if you really want to. However, you don’t get to values like 4000, even 3000 seems way to high for proper calculations.

So I took 2000 power and 0 bonus crit damage as a starting point and looked at which crit chance value (so basically precision) changes would equal out (so 2100 vs +10%). It’s when the crit chance is a bit higher than 65%, which is also a very high value. It’s actually so high you don’t hit it with equipment only (if you want to keep your maxed power). You need skill effects or alike. Until that point, power is better off, given a power scaling of 1 (which is common, or at least I think it is).

If you take more natural values though, so your crit chance is a little lower and you don’t compare 0 vs 10 crit damage but rather 40 vs 50 or something, power is the better choice.

According to my calculations though, you won’t feel THE difference if you take one or the other. It’s like a 1-3% difference (depending on your stats :P ).

sorry im bad at maths but can you tell me which one is stronger
2337 power with 65% crit dmg and 39% crit chance
or 1977 power with 84% crit dmg and 41% crit chance

The first one.

I use the formula
(Power / 916) * (1 + (CritChance * (BonusCritDmg + 0.5) ) )
Note that CritChance and BonusCritDmg are percentages and thus values like 0.65 and 0.39 for the first example.

I get ~3.696 for the first and ~3.344 for the second set of data. The first coefficient is quite a lot higher, or the be precise, 3.696/3.344 ? 1.105 = 10.5% higher.

Keep in mind though that I assumed power to scale at a factor of 1, which might be wrong in some cases. Consequently, the formula is not correct with certainty.