by Thott
This comparison is a series of simulations, comparing the three priests
in different activities. The purpose of these simulations is
to compare priests in experience group situations, not on
raids. Taking values from these simulations and applying them to
raids is like taking 100 meter dash numbers and applying them to a space
walk. Not that people won't do it anyway, but it will be silly.
All of these simulations share certain base assumptions:
- All the priests are the same level, either 60 or 65.
- All the priests have useful buffs active before the simulation starts (typically mana/hp regen buffs).
- All the priests have any AA's that will be useful to them, including:
- Spell Casting Mastery
- Cannibalization
- Mental Clarity (or equivalent FT)
- Spell Casting Fury
- Spell Casting Reinforcement (Mastery)
- Healing Adept
- Healing Gift
All priests are assumed to have the same total number of AA points.
- All the priests have FT+MC 8, meaning 8 points of mana regeneration from AA and items.
- All spells are assumed to cost 40% less due to specialization (+SCM3).
Some tests say the mana reduction is higher, others say it is
lower. One theory is that mana reduction from specialization
varies by spell level. In any event, the same value is used
for all priests.
- Mana reduction from focus items ignored, or if you prefer, included in the 40% used above.
- +20 mana/tick when sitting.
- +1 mana/tick base when standing.
- All the priests have 4000hp and 4000mana.
- Cleric CH heals 4000hp every time, shaman/druid mini-CH's heal their max value.
There are other assumptions specific to individual simulations, click on a graph to see them.
How the Simulations Work
Each simulation is done similarly. The simulated priests have a
spell script that they follow. Each spell takes its normal cast
time, and once it is complete, each spell has a period of time after it
where all spells are greyed out. Spells are only cast if there is
enough mana to cast them. Each spell of course takes mana.
The end result is a complete simulation of the priests in question,
just as if they were in game doing it.
For example, the cleric healing script checks to see if there is enough
mana to cast CH, and if there is, it starts casting it. 10 seconds
later, the CH finishes, and 4000hp is added to the total number of hit
points healed. If there isn't enough mana to cast CH, the cleric
sits down until there is. Every tick, mana from buffs and
meditate (if the cleric is sitting) is added to its mana pool.
Every second, the total hit points healed is divided by the number of
seconds since the simulation started, and that value (hit points per
second) is plotted on the graph.
Thus the graphs are showing the damage per second a tank can take without
dying, thanks to the priest. A mob will have to have a higher dps
value than what is shown to have any hope of killing the tank.
|
The simulated shaman does not cannidance, meaning the shaman does
not sit down between casts of cannibalize IV, when mana is low.
I considered adding this, but shamans don't enjoy it, and I don't think
shamans should be balanced assuming they use it. If a shaman wants to
go through that trouble, more power to them.
Shaman cannibalize was incorrectly giving 1200 mana instead of 1066 -
fixed now in all simulations.
I removed fungus breastplate from all priests, although this only affects
shamans. This really isn't a correct choice. Yes, it's
difficult to get a fungus breastplate - but it's difficult to get to 60,
and it's difficult to get AA points, and I assume both of those.
Shaman gain more from fungus breastplate than any other priest, and should
be balanced in the post-60 world assuming they have it, or something
like it (I suspect more regen breastplates will come in the future,
since innate regen is now capped), but because many shamans objected,
I removed it.
The graphs didn't change all that much, but now, with accuracy improved,
shamans can be happy and carefree again, instead of whiney and angry.
Fleshed out the AA's used list above.
|
Intro | Mitigation | Healing only | Damage | Errors | Next >>
|