It's a hard thing to determine for some classes. For others like mages, they know it's just one ability that they spam. For warlocks however, you start to wonder. Does Siphon Life actually do enough damage on a boss to make it worth using? Should I get Unstable Affliction or Ruin? How much of my DPS is in Shadowbolts?
I found out last night that a lot of a DOT lock's DPS is in Shadowbolts, which totally changes my perspective on how to itemize them. It means that +crit is more important to them than I originally thought.
These are the things that using a tool like Wow Web Stats can do for you. Want to know which spec does more DPS? Find out! What to know which of your abilities are doing the most DPS? Find out! What to see if you need more +hit? Find out! Go to www.lossendil.com and see for yourself how you can determine your DPS and your abilities' effectiveness. Seeing your stats can truly open your eyes.
Subscribe to:
Post Comments (Atom)
Posts
1 comment:
I'm getting into the 'study' of DPS more and more.
On the topic of +hit, +crit, and +damage... a typical raid boss is max level +3 levels, so at level 70 you will be looking at a base chance to hit of around 83% due to level only and not factoring in abilities or resistances. The maximum hit chance is 99%. So between talents and equipment, you want to get around 16% bonus hit (83+16=99).
Consider the following:
http://forums.worldofwarcraft.com/thread.html?topicId=36573662
Spell Hit Rating: 12.6 rating grants 1% spell hit chance
Spell Critical Strike Rating: 22.1 rating grants 1% spell critical strike chance
What is the spell coefficient?
If you have 1000 +damage to your spell, not all spells will gain 1000 damage. That 1000 is multiplied by a number known as the spell's damage coefficient.
Typically the coefficient of spell damage is equal to the spell's cast time divided by 3.5 multiplied by the amount of spell damage.
Many Warlock DoTs break this rule by having their coefficients artificially raised.
Fireball and frostbolt break this rule by having their coefficients artificially lowered when the improved fireball/frostbolt talent is taken (by 2% per talent point in the 'improved' talents). Frostbolt also suffers from a 5% penalty imposed on it as an arbitrary estimation of the value of the snare effect.
For more on coefficients for specific spells, see here:
http://www.wowwiki.com/Spell_Damage_Coefficients
and here:
http://www.wowwiki.com/Formulas:Plus_damage_and_Plus_healing
So what is more valuable?
Let's make some simplifying assumptions:
1) The critical strike damage bonus of a spell is 100% (it's actually 50% in most cases)
2) The direct damage spell being cast does 1000 damage every time, unmodified with a 3.5 second cast time and 100% coefficient.
3) The relative level of caster and target plus yada yada yields an 80% hit rate.
4) 100% hit rate is possible.
5) Probability enforced
Now, our caster will cast 10 shots with 3 configurations...
Scenario 1: 50% crit rate (0 +hit, 0 +damage):
4 shots do 1000 damage, 4 shots do 2000 damage, 2 shots miss.
Total damage: 4,000+8,000+0=12,000
50% crit equals about 1105 crit rating. We'll assume a base crit of some amount and trim that to 1000 +damage. Again, 10 shots.
Scenario 2: 1,000 +damage (0 +crit, 0+hit)
8 shots hit causing 2,000 damage each. 2 shots miss.
Total damage: 8 * 2,000 = 16,000
Note that we assumed a base crit of some amount, so there is a possibility of one crit. This case assumes there was no crit for comparison purposes.
Now, we could go 1000 +hit, but this would not show the value of +hit, since +hit has 0 value past 100% (99% in the 'real' world). To get the full 16% +hit from gear alone and not talents is about 202 +hit rating. We'll call that 300 and spend the remaining 700 points on damage.
Scenario 3: (100% hit, 0 crit, 700 damage):
10 shots hit causing 1,700 damage apiece.
Total Damage: 10*1,700 = 17,000
Based on this, +hit is better than + damage and both are better than +crit.
What did our assumptions skew?
Assumption 1 over-inflates the value of +crit. There are a few talents that improve the value of crits to this level (IE: ice shards, spell power and ignite for mages) but not all builds will include such a talent.
Assumption 2 the 1000 damage part doesn't affect the value of any of the stats, it just makes for cleaner math. The 100% coefficient assumption inflates the value of +damage for spells with less than 100% coefficient (frostbolt for example) and underestimates for spells with over 100% (a fully talented fireball).
Assumption 3 overestimates the value of +hit on mobs of equal level with no abilities that reduce a player's chance to hit. It underestimates +hit vs mobs greater than 3 levels above the player's.
Assumption 4 overestimates the value of +hit VERY slightly (and this is compensated for by overcompensating on the amount of +hit we put in to reach 100%).
Assumption 5... well... reality is bizarre. Sometimes you can beat the math and crit more than your crit rating suggests, for example. This could oversate or understate the value of +hit or +crit, which in turn could make either compare more favorably or less favorably against +damage, which is held constant.
I didn't do DoTs, but my guess is, +crit has no value and +hit is more valuable than +damage (vs mobs higher level than the player up to 99% chance to hit).
~Yim
Post a Comment