How to figure out damage per second?

[neltymind]

I started playing PoE yesterday and I am still unsure which kind of build I want so use to play trough the game. 

 

While comparing several builds I was trying to figure out the damage per second of builds and... came to a dead end. 

 

Every weapon has an attack speed. What exactly does that mean? I mean, in numbers. I read attacking gets faster if you use two weapons. Again, how much? I need numbers...

 

I build a barbarian, for example. He is solely focused on dealing damage. But I am completely unable to figure out if I am better off using a two-handed weapon, two one-handed weapons with average speed or two one-handed weapons with fast speed (but lower damage) because I have no idea how often my character strikes in each setting and thus I am unable to calculate the average damage per second. 

 

Can someone help me with that? I can't find any answers so far.

 

Any help is appreciated.

[Emptiness]

In order to get the numbers you are looking for it is necessary to record combat (using a 3rd-party program) and then manually count the frames between attacks.  This process would have to be repeated for each weapon configuration you wanted to compare.  I haven't done it myself, but I imagine it's a fair bit of work.

 

If you poke around the forum you may be able to find some people who have done this, and you may be able to read about their comparisons and make a decision based on that information.  Or, you may not find information that specifically applies to your situation, in which case that effort would be wasted.  For example: Dexterity and the armor you are wearing affect your dps, so even if you find a comparison of the weapon configurations you are interested in, that comparison may be taking place at a different Dexterity than you have or while wearing different armor.

 

Another complicating factor is that dps will be affected by the enemy's damage reduction, which is different for different enemies.  So, unless you're going to test (or calculate) your dps for every possible damage reduction value and then switch weapon sets to the optimum one for each specific foe, the effort you put into precise dps calculations won't necessarily give you the information you need to make the "right" choice of weapon configurations.

 

The alternative is the less rigorous method of trying out different things and picking the one(s) that seems best.  It may not get you hard numbers, but it will (probably) take less time and effort.  I imagine it is a question of whether you feel compelled to be "the best", or can cope with merely being "good enough".

[Epsilon Rose]

In order to get the numbers you are looking for it is necessary to record combat (using a 3rd-party program) and then manually count the frames between attacks. This process would have to be repeated for each weapon configuration you wanted to compare. I haven't done it myself, but I imagine it's a fair bit of work.

 

If you poke around the forum you may be able to find some people who have done this, and you may be able to read about their comparisons and make a decision based on that information. Or, you may not find information that specifically applies to your situation, in which case that effort would be wasted. For example: Dexterity and the armor you are wearing affect your dps, so even if you find a comparison of the weapon configurations you are interested in, that comparison may be taking place at a different Dexterity than you have or while wearing different armor.

 

Another complicating factor is that dps will be affected by the enemy's damage reduction, which is different for different enemies. So, unless you're going to test (or calculate) your dps for every possible damage reduction value and then switch weapon sets to the optimum one for each specific foe, the effort you put into precise dps calculations won't necessarily give you the information you need to make the "right" choice of weapon configurations.

 

The alternative is the less rigorous method of trying out different things and picking the one(s) that seems best. It may not get you hard numbers, but it will (probably) take less time and effort. I imagine it is a question of whether you feel compelled to be "the best", or can cope with merely being "good enough".

This is one of my problems with PoE. I have no real way to gauge how much a character is contributing or how good their build is, outside of things like spells, with any degree of precision. This makes it really hard to improve or plan out new builds. There really needs to be a base line dps calculator or a way to figure it out.

[Infares]

The equations and frame counts are known, you can probably take that information and create a simulator and parse what configurations in base stats, equipment, and class/ability/talent loadouts yield the highest DPS.

[Caerdon]

The equations and frame counts are known, you can probably take that information and create a simulator and parse what configurations in base stats, equipment, and class/ability/talent loadouts yield the highest DPS.

 

The last time I checked no one had yet figured out how exactly all the attack/recovery speed modifiers stack, and as a result, there was no established theory capable of explaining the timings from video analysis.

[aimlessgun]

Another rough method is to just do some fights with each build and compare character total damage to the Time In Combat stat. Due to time scaling this will yield very low numbers per second, so I just compared damage per minute between characters. For example 20+ damage per minute by game stats is "pretty good" as far as I can tell. 

[Emptiness]

This is one of my problems with PoE. I have no real way to gauge how much a character is contributing or how good their build is, outside of things like spells, with any degree of precision. This makes it really hard to improve or plan out new builds. There really needs to be a base line dps calculator or a way to figure it out.

 

This is only a serious impediment to gameplay if you insist that all your characters are perfect from the very first time you play.  Accept that perfection is impossible, make choices that are interesting to you, play the game, try different things, make mistakes, learn from them, make better choices, and so on.  If you calculate the optimum choice for every decision that the game presents to you before you play it then the game will hold no challenge for you and there will be no point to playing it.  The journey is more precious than the destination.

[Climhazzard]

Some one posted this recently... 

 

http://i.imgur.com/5LhXCjJ.png

 

And there is a version for sabers

 

https://docs.google.com/spreadsheets/d/1SHdaQ8yNSvJhxcqFeP9ykXdua_EsjSAb9ypzPA8Oink/edit#gid=1848365110

 

I don't know who did the work, therefore I don't know how accurate it is, nor can i credit them for it, it doesn't take into account talents or weapon specific mods, but it's a good base line.  Dual wielding sabers is a pretty safe bet for good damage, battle axes and maces probably work pretty good too, as well as two handing an estoc.  Some builds may use specific unique weapons, like a reach barbarian using the tall grass pike.

Edited April 12, 2015 by Climhazzard

[Epsilon Rose]

 

This is one of my problems with PoE. I have no real way to gauge how much a character is contributing or how good their build is, outside of things like spells, with any degree of precision. This makes it really hard to improve or plan out new builds. There really needs to be a base line dps calculator or a way to figure it out.

 

This is only a serious impediment to gameplay if you insist that all your characters are perfect from the very first time you play.  Accept that perfection is impossible, make choices that are interesting to you, play the game, try different things, make mistakes, learn from them, make better choices, and so on.  If you calculate the optimum choice for every decision that the game presents to you before you play it then the game will hold no challenge for you and there will be no point to playing it.  The journey is more precious than the destination.

 

It's not about being able to calculate everything perfectly in advance, it's about learning. Without proper information, how do I guage how well something is working? Did my tank die because my tacktics were bad and I didn't support it enough? Was it's defense not high enough? Was it simply not doing enough damage? Even if I win, could I have done better if I had a different weapon? I don't know and I don't have an easy or intuitive way to find out. Even the battle log and pause on inefective isn't a particularly good tool, because it pauses when the enemy makes an inefective attack (why would I care about that enough to want a pause?).

 

There are too many factors to simply try every permutation and see what works best in a feasible amount of time or at a reasonable skill level and to many degrees of success or failure for that to give you good informaiton anyways.