Jump to content

Average Raw DPS Done by Bosses


Recommended Posts

I'm curious as to what numbers people are seeing in the category of unmitigated DPS output from nightmare mode (and Denova) bosses. Combat logs are somewhat deceptive here. The numbers I see in torparse (in the category of Damage Taken) are in and around 1k (less for some, more for others). Kephess jumps to about 1.2k in his final phases, but 1k seems to be about the magic number. (note: this is for 8 man ops) I suspect though that the combat logs are hiding some of the mitigation that tanks have (e.g. armor or resistance).

 

What exactly is torparse showing here? Is this pure, unmitigated damage? Post defense+shield? If I knew exactly what the combat logs were measuring, I could work backwards through my expected mitigation to figure out the raw damage output of the boss, I just don't have that information right now. Having a raw DPS number would be massively helpful in prioritizing tank itemization, particularly for shadows/assassins which have non-mitigation forms of survivability.

Edited by KeyboardNinja
Link to comment
Share on other sites

Combat logs are somewhat deceptive here. I'm curious as to what numbers people are seeing in the category of unmitigated DPS output from nightmare mode (and Denova) bosses. The numbers I see in torparse are in and around 1k (less for some, more for others). Kephess jumps to about 1.2k in his final phases, but 1k seems to be about the magic number. (note: this is for 8 man ops)

 

What exactly is torparse showing here? Is this pure, unmitigated damage? Post defense+shield? If I knew exactly what the combat logs were measuring, I could work backwards through my expected mitigation to figure out the raw damage output of the boss, I just don't have that information right now. Having a raw DPS number would be massively helpful in prioritizing tank itemization, particularly for shadows/assassins which have non-mitigation forms of survivability.

 

Torparse is showing damage you actually took after Defense, Shield/Absorb Rating, and Armor are taken into account.

 

If you avoid a hit because of Defense, then that will show up as an attack that does 0 damage in Torparse (you can look at attacks broken down by time to see examples of this). It will be excluded from average damage per hit calculations.

 

If you Shield an attack, then Torparse will treat this as a reduced size hit, and it will be factored into average damage per hit calculations. If you want to understand what the original unabsorbed hit would have been, that is pretty difficult to determine (since the tank may be wearing trinkets that changed his Absorb rating dynamically for example). Shields (counter-intuitively) are not the same thing as Absorbed in the combat logs. Shielded attacks use the term "glancing." If you look at the log view in Torparse instead of the analytical view, you can spot these hits by looking for the word "glances".

 

Absorbed damage, e.g., via a sage bubble, is excluded from DPS and average damage per hit calculations, so the actual damage a tank takes is much smaller than what is reported by Torparse as a result. These show up as absorbs in the combat log (and should not be confused with tank shielding).

 

If hits are armor mitigated, then you won't know what the original unmitigated hit would have been without doing math, since that mitigation is already factored in to the damage shown in a combat log. Damage type is included, however, so you can tell whether or not an attack was mitigated by armor or not.

 

Finally, tanks may have popped cooldowns that affect the size of the hits, and those will of course reduce the DPS and average damage per hit of a boss as well.

 

So to sum up, in order to get all the way back to the original hits before any mitigation at all, you would want to look at the attacks a boss did broken down by time in torparse's analytical Damage Dealt view. Then you would need to determine which attacks were shielded by looking for glances in torparse's raw log view. You would then need to inflate the glancing damage numbers back to their original values based off the tank's Absorb Rating at the time the attacks landed.

 

You could then compute an average damage per hit for each unique ability. You would then need to look at how often the attack happens (e.g., the boss "swing timer" for that ability) and assume no avoidance. So for example if the average unshielded hit is 5k and it happens every second, then the boss worst case will do 5k DPS with that attack.

 

If the attack does damage that is mitigated by armor, you could then do the math to compute the original unmitigated value only for those attacks.

Edited by Kihra
Link to comment
Share on other sites

Torparse is showing damage you actually took after Defense, Shield/Absorb Rating, and Armor are taken into account.

 

Exactly what I needed to know. (thanks!) I can actually walk backwards from there fairly easily, since I have numbers handy representing my total (infinity-averaged) mitigation on various attack types. Cooldowns do throw a wrench in things, as do things like debuffs (e.g. armor stacks when fighting Fabricator), but it should still be fairly straightforward to just look at averages calculated over periods not including cooldowns.

 

The value of this sort of information becomes manifest in calculations like determining expected points of mitigation over time from the proc'ing absorb relic. These numbers in turn are necessary if you want to directly compare the survivability benefits of specific types of mitigation vs the survivability benefits of self-healing (and self-heal proc'ing relics).

Link to comment
Share on other sites

Good luck factoring in things like -5% damage from Slow Time or -4% damage from Static Field, etc. :)

 

That's a good point. For the moment, I'm going to just ignore those factors. The reason being that I'm considering my own damage mitigation numbers, and I can safely assume that I will always be maintaining the appropriate debuffs on my targets. Thus, Slow Time can be straight-up ignored. Force Breach is trickier, as it affects the value of defense rating, but that factor will be subtracted out when doing comparisons. Thus, numbers in isolation may appear slightly deflated, but comparative aggregates should be just about right.

Link to comment
Share on other sites

×
×
  • Create New...