EOTS and Crit Diminishing Returns

Comments

• pingconchere Member Posts: 0 Arc User

  November 2013 edited November 2013

  I may not be one of the people you mentioned, but I do have insight on this. What I can say is this question is not as easy to answer mathematically as it may initially seem. I say this because I worked on a similar question regarding the new Ranger Archery capstone feat: what would the optimum crit percentage be to get the most benefit from the feat? While I haven't posted my results on this yet, I did eventually come to an answer and will make a post soon about it. That's beside the point. It took me a couple of hours of dedication to derive the proper equation to find the answer. Just to give an example of the ugliness of an equation needed to solve something like this...:
  
  

  

  
  Beautiful, isn't it?
  
  I will see if I can come up with for your specific question, but don't expect an answer anytime soon. Hopefully someone has done this already and will answer you promptly.
  
  EDIT: I forgot to mention that this is also subjective to what encounters you're running, as some make more use of EOTS than others. If you could give me the encounters you're using, that would be great. However, I will most likely just use autoattacks to make this easier.

  0
• j0shi82 Member Posts: 622 Arc User

  November 2013 edited November 2013

  I've done some testing a while ago with my Thaum CW if it's more valuable to stack Power over Crit, because most of the time you'll have a 100% critrate because of EotS anyway. My DPS was higher with stacking Crit anyway.
  
  That's not answering your question directly, but might help anyway. The diminishing return of Crit kicks in early (~1800), but most CWs will go as high as 3k.

  0
• stoxforum1 Banned Users Posts: 265 Bounty Hunter

  November 2013 edited November 2013

  tldr: power scales a lot better than people think because of debuffs.
  
  People always look at this wrong in my opinion. Most people take this to "soft caps" so that their stats are the most efficient. The problem with that is there may be other stats where you gain more damage before ever reaching that cap. What you really need to look at is simply where one stat is more effective than another for increasing your damage.
  
  Editing this to add some information and food for thought. Each point of power adds different amounts of damage to different skills. A couple examples from my last time running numbers (and these will change depending on how much power you have and maybe from your base weapon damage too?) but 25 power which is equivalent to 1 weapon damage added 11.2 chill strike damage, 4.8 shard damage, 2.2 CoI, 6.8 steal time, 2 magic missile, 13.8 OF. These were to the tooltip damages and all I did to calculate was to take the tooltip damage, remove a ring with 125 power (pyro band) which gives you 125/25 power = 5. So take damage from ability - damage from ability with ring off, / the number you found above. Now while crit is more effective if you just run numbers early on, it's only a theoretical comparison if you do it that way. Real world damage has debuffs, lots and lots of debuffs and buffs in a good party.
  
  So let's say for the sake of simplicity a skill has a base damage of 1000 and the person is getting 50% more damage from buffs (but this can definitely be higher.) This would be 1000*1.5 = 1500 damage from that skill instead or an increase of 500 damage.
  Now the damage multiplier from shield in my last test was about 10 (a little less) every 25 power so I'm going to use that as an example. If I added 1250 power this would say add 500 damage to the base of the skill. Now 1500 with 50% more damage = 1500*1.5 = 2250 or an increase of 750 instead of the 500 we saw previously. See how it is scaling better now?
  
  Now from where I am which is 0 crit slotted but about 1470 on gear, adding 1250 points into crit would add about 5.2-5.3% crit chance.
  
  If you want to know where it is more effective, factor in added power for the lower crit amount and revert back to the base damage for the higher crit amount but remember to add in the added crit chance diminished by about 20% since you'll have eye of the storm slotted which partially negates your crit because of it's 100% crit while activated. This may in fact be the answer the original poster was looking for. Ideally for calculations you will figure out the added damage from power for each skill and the actual damage done to see the debuffs. With all the damage and hit data from a parser you can extract how much additional damage each point of power does and that % increase. Then plug these into the statement at the start of this paragraph. I end up doing these calculations on a few good runs about once a month or after big updates (HV nerf.)

  0
• niadan Member Posts: 1,635 Arc User

  November 2013 edited November 2013

  Thank you for your insight and help. It became very evident to me while leveling up companions in preparation for the arival of MOD 2, that with EOTS and a lower crit rating (due to no augment pet slotted) I was stll critting most of the time. I do not have the math skills to devise the formulas necessary to model this problem, and respect those that can.

  0
• j0shi82 Member Posts: 622 Arc User

  November 2013 edited November 2013

  Approaching this issue from a mathematical side is a mess. The Ranger Passive adds another 100% Crit after a Crit so that's doable. But for this problem you would need to process the procrate for various skills and cooldowns and such. I would do a field test with your rotation instead, that should be way faster. Try 5/10/15/20% Crit from the stat alone and monitor your DPS.

  0
• niadan Member Posts: 1,635 Arc User

  November 2013 edited November 2013

  stoxforum1 wrote: »

  tldr: power scales a lot better than people think because of debuffs.
  
  People always look at this wrong in my opinion. Most people take this to "soft caps" so that their stats are the most efficient. The problem with that is there may be other stats where you gain more damage before ever reaching that cap. What you really need to look at is simply where one stat is more effective than another for increasing your damage.
  
  Editing this to add some information and food for thought. Each point of power adds different amounts of damage to different skills. A couple examples from my last time running numbers (and these will change depending on how much power you have and maybe from your base weapon damage too?) but 25 power which is equivalent to 1 weapon damage added 11.2 chill strike damage, 4.8 shard damage, 2.2 CoI, 6.8 steal time, 2 magic missile, 13.8 OF. These were to the tooltip damages and all I did to calculate was to take the tooltip damage, remove a ring with 125 power (pyro band) which gives you 125/25 power = 5. So take damage from ability - damage from ability with ring off, / the number you found above. Now while crit is more effective if you just run numbers early on, it's only a theoretical comparison if you do it that way. Real world damage has debuffs, lots and lots of debuffs and buffs in a good party.
  
  So let's say for the sake of simplicity a skill has a base damage of 1000 and the person is getting 50% more damage from buffs (but this can definitely be higher.) This would be 1000*1.5 = 1500 damage from that skill instead or an increase of 500 damage.
  Now the damage multiplier from shield in my last test was about 10 (a little less) every 25 power so I'm going to use that as an example. If I added 1250 power this would say add 500 damage to the base of the skill. Now 1500 with 50% more damage = 1500*1.5 = 2250 or an increase of 750 instead of the 500 we saw previously. See how it is scaling better now?
  
  Now from where I am which is 0 crit slotted but about 1470 on gear, adding 1250 points into crit would add about 5.2-5.3% crit chance.
  
  If you want to know where it is more effective, factor in added power for the lower crit amount and revert back to the base damage for the higher crit amount but remember to add in the added crit chance diminished by about 20% since you'll have eye of the storm slotted which partially negates your crit because of it's 100% crit while activated. This may in fact be the answer the original poster was looking for. Ideally for calculations you will figure out the added damage from power for each skill and the actual damage done to see the debuffs. With all the damage and hit data from a parser you can extract how much additional damage each point of power does and that % increase. Then plug these into the statement at the start of this paragraph. I end up doing these calculations on a few good runs about once a month or after big updates (HV nerf.)

  
  Stox, you are mapping out exactly where I am heading. Thanks!

  0
• borken69 Member, Neverwinter Beta Users, Neverwinter Guardian Users Posts: 56

  November 2013 edited November 2013

  Round numbers, EotS has a 30% uptime +/- 5%. Meaning that if you have a 30% crit chance, your effective crit chance is 51%.
  
  .3 + (.7*.3) = .51
  
  Obviously, this is a massive generalization instead of a complex formula but if you want quick and dirty math, your crit chance should be around 30%. I like to sit at just over 33% so I have a 1/3 shot at rolling a crit when EotS is down.

  0