Jump to content

Holy Flame damage bonus ...


  • Please log in to reply
50 replies to this topic

#41 grimnok

grimnok

    Veteran

  • Members
  • PipPipPip
  • 917 posts

Posted 18 February 2010 - 22:36

I've decided to do a field test myself to be sure.

The creature I'm killing is Banshee of Misery, which has 5875 average armor. I've done 200 hits in each of the 5 variants to get a semi-decent sample for analysis.

The damage on my character screen is 12832. I don't have a "Holy" enhancement.

#1 - No DD, no CA, no HF: Average damage 6990 (min 6115, max 7898)

#2 - No DD, no CA, HF 175: Average damage 9296 (min 8194, max 10452)
As you can see, this is ~33%, so HF applies -after- armor. If it applied to damage in character screen, I'd be doing 11.5k on average.

#3 - DD 20%, no CA, no HF: Average damage 9551 (min 8632, max 10491)
This is just DD, so 20% bonus is applied to damage; reflected in test results here.

#4 - DD 20%, no CA, HF 175: Average damage 12845 (min 11426, max 14189)
The DD bonus is applied to damage in character screen, 12832 * 1.2 = 15398. If HF was applied on top of that, ignoring armor, I'd be doing 14.9k damage, but I'm doing 12.8k. That confirms #2, HF being applied after armor.

#5 - DD 20%, CA 50, HF 175: Average damage 15432 (min 14194, max 16645)
Here are the results for both stacking and non-stacking variants:
a) 12832 * 1.2 * 1.125 = 17323 (stacking) | 17323 - 5875 = 11448 after armor | with HF, 15454 - pretty much my result here
B) 12832 * 1.2 + 12832 * 0.125 = 17002 (non-stacking) | 17002 - 5875 = 11127 | with HF, 15021 - not close enough to test results, so a) seems to be right



#42 fs_robotussin

fs_robotussin
  • Guests

Posted 19 February 2010 - 02:16

I've decided to do a field test myself to be sure.

The creature I'm killing is Banshee of Misery, which has 5875 average armor. I've done 200 hits in each of the 5 variants to get a semi-decent sample for analysis.

The damage on my character screen is 12832. I don't have a "Holy" enhancement.

#1 - No DD, no CA, no HF: Average damage 6990 (min 6115, max 7898)

#2 - No DD, no CA, HF 175: Average damage 9296 (min 8194, max 10452)
As you can see, this is ~33%, so HF applies -after- armor. If it applied to damage in character screen, I'd be doing 11.5k on average.

#3 - DD 20%, no CA, no HF: Average damage 9551 (min 8632, max 10491)
This is just DD, so 20% bonus is applied to damage; reflected in test results here.

#4 - DD 20%, no CA, HF 175: Average damage 12845 (min 11426, max 14189)
The DD bonus is applied to damage in character screen, 12832 * 1.2 = 15398. If HF was applied on top of that, ignoring armor, I'd be doing 14.9k damage, but I'm doing 12.8k. That confirms #2, HF being applied after armor.

#5 - DD 20%, CA 50, HF 175: Average damage 15432 (min 14194, max 16645)
Here are the results for both stacking and non-stacking variants:
a) 12832 * 1.2 * 1.125 = 17323 (stacking) | 17323 - 5875 = 11448 after armor | with HF, 15454 - pretty much my result here
B) 12832 * 1.2 + 12832 * 0.125 = 17002 (non-stacking) | 17002 - 5875 = 11127 | with HF, 15021 - not close enough to test results, so a) seems to be right



this test is clearly better than mine for 3 MAJOR REASONS:
higher armor on creature
no enhancements active
larger sample


clearly my results were not conclusive (like i said), but these results ARE conclusive. i think tang has his answer now...

thanks for your time and effort, mistle!

#43 BalianRW

BalianRW

    Veteran

  • New Members
  • PipPipPip
  • 755 posts

Posted 19 February 2010 - 04:18

It is also worth noting that since it seems apparent that HF is applied AFTER armor is subtracted, all the formulas are damage directly against the creatures HP and does not have to also pass through the armor (since we already subtracted the armor before we applied HF).

I had actually over looked this fact, until now, and wanted to make sure no one else had done the same thing.

#44 BalianRW

BalianRW

    Veteran

  • New Members
  • PipPipPip
  • 755 posts

Posted 01 March 2010 - 19:17

Ok, so based on both my analysis of robotussin's data as well as the analysis from Mistle/grimnok I believe that the formula for damage inflicted on the creature (after armor) is as follows.

(DisplayedDmg*(1+Holy)*(1+DD)*(1+CA) - MobArmor) * (1+HF) * f

Where Holy is the bonus from the Holy enhancement. Could be replaced with greenskin, aquatic, etc. . . Enhancement advantage.

The "f" is a factor I apply to cover the +/- 5% variance to ensure that I can kill the majority of the creatures I encounter.

Tangtop, my suggestion would be to switch the FSH to this formula, but keep the three options in the FSH settings. Those options could use factors (f) of 1.0 (very risky), 0.975, and 0.95 (very safe).

#45 fs_coyotik

fs_coyotik
  • Guests

Posted 02 March 2010 - 19:12

Ok, so based on both my analysis of robotussin's data as well as the analysis from Mistle/grimnok I believe that the formula for damage inflicted on the creature (after armor) is as follows.

(DisplayedDmg*(1+Holy)*(1+DD)*(1+CA) - MobArmor) * (1+HF) * f

Where Holy is the bonus from the Holy enhancement. Could be replaced with greenskin, aquatic, etc. . . Enhancement advantage.


What makes you so sure that the Enhancement damage is applied upfront? First of all, it would be extremely illogical from the programmer's point of view, because he would have to introduce creature type check in two places.

(DisplayedDmg*(1+DD)*(1+CA) - MobArmor) * (1+HF+Holy) * f (or +/- f?)

I would expect the programmer to treat all "damage against particular type" the same way, apply it after armor has been crushed out of the way, by a simple 1 + $enh1_level*$enh1_effect*is_creep_type1().

I haven't tried doing any major testing of the random factor (to see whether it's dependant on player level, creep level or total damage).

#46 sirwinkey

sirwinkey

    New Member

  • Members
  • Pip
  • 36 posts

Posted 02 March 2010 - 22:15

I am not quite to that point yet.

#47 fs_robotussin

fs_robotussin
  • Guests

Posted 02 March 2010 - 22:54

I am not quite to that point yet.



are you sure that you don't have any insights on the mathematical nuances of the game engine? it seems like you have a lot to say on every other topic

#48 sirwinkey

sirwinkey

    New Member

  • Members
  • Pip
  • 36 posts

Posted 02 March 2010 - 22:56

I am not quite to that point yet.



are you sure that you don't have any insights on the mathematical nuances of the game engine? it seems like you have a lot to say on every other topic

Just the newbie stuff thats all. I haven't gotten very far into it yet.

#49 Mister Doom

Mister Doom

    Veteran

  • Members
  • PipPipPip
  • 6,518 posts
  • United Kingdom

Posted 02 March 2010 - 22:56

I'm guessing multi. Someone got bored and didn't want to jeopardise their main account..

EnhancedShardoom1-1.gif


#50 sirwinkey

sirwinkey

    New Member

  • Members
  • Pip
  • 36 posts

Posted 02 March 2010 - 23:18

I'm guessing multi. Someone got bored and didn't want to jeopardise their main account..


Seriously? dude stop trolling me, not cool.

#51 BalianRW

BalianRW

    Veteran

  • New Members
  • PipPipPip
  • 755 posts

Posted 03 March 2010 - 02:15

Ok, so based on both my analysis of robotussin's data as well as the analysis from Mistle/grimnok I believe that the formula for damage inflicted on the creature (after armor) is as follows.

(DisplayedDmg*(1+Holy)*(1+DD)*(1+CA) - MobArmor) * (1+HF) * f

Where Holy is the bonus from the Holy enhancement. Could be replaced with greenskin, aquatic, etc. . . Enhancement advantage.


What makes you so sure that the Enhancement damage is applied upfront? First of all, it would be extremely illogical from the programmer's point of view, because he would have to introduce creature type check in two places.

(DisplayedDmg*(1+DD)*(1+CA) - MobArmor) * (1+HF+Holy) * f (or +/- f?)

I would expect the programmer to treat all "damage against particular type" the same way, apply it after armor has been crushed out of the way, by a simple 1 + $enh1_level*$enh1_effect*is_creep_type1().

I haven't tried doing any major testing of the random factor (to see whether it's dependant on player level, creep level or total damage).


Hmm, you may be right about this. Running the data I have through both calculations shows that it is a little more accurate (when compared to the control) if it is applied afterwards (i.e. with HF) like you suggest, however the difference is so small that i need more data to make the determination with any level of certainty.


0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

Font:
Arial | Calibri | Lucida Console | Verdana
 
Font Size:
9px | 10px | 11px | 12px | 10pt | 12pt
 
Color: