Jump to content

Welcome to Obsidian Forum Community
Register now to gain access to all of our features. Once registered and logged in, you will be able to create topics, post replies to existing threads, give reputation to your fellow members, get your own private messenger, post status updates, manage your profile and so much more. If you already have an account, login here - otherwise create an account for free today!
Photo

Proposed method for semi-accurate DPS measurement/comparison for melee/ranged weapon combos

game mechanics

  • Please log in to reply
1 reply to this topic

#1
peddroelm

peddroelm

    (6) Magician

  • Members
  • 609 posts
  • Deadfire Backer
 
 
We need to count the number of swings made the test character attacking un-interrupted for a long time interval (10 / 20 minutes).. (game log doesn't store 10 minutes worth of slow attacks so we'll also need to record the test and read the number of hits from the video after the test) ..
Hit resolution (mis..crit //damage rolls irrelevant at this stage) 
 
Simplest way to achieve this that I could think of, is attacking party member that doesn't retaliate and using console cheat - "god" to make sure he/she can take 10/20/30 minutes worth of beating ..
 
 
Manual Test 
 
-prepare test char with chosen weapons/armor/DEX/ other attack speed modifiers for test
-prepare outside chronometer (most phones have chronometer app this days) 
-iroll20s god
-move test chars in weapon range
-pause game 
-give attack order
-start recording video
-start chronometer & unpause game
..10 /20 /30 minutes later
-pause game when time reaches threshold 
-stop recording 
 
Review video and get number of swings in time interval (for each weapon if testing DW with fast/slow weapons, or they do different amount of damage) 
With this knowledge can calculate maximal damage inflicted in time interval (assume all attacks crit, weapon damage RNG roll is always average of min-max). 
 
 X_NR_swings * (((wpnmax+wpnmin)/2) * (1 + sum_damage_mods) + Transcenent_suffering_bonus/OSA) = total_maximal_"measured"_damage - fair for comparing set-ups  
 
Not 100% accurate since not all tested combos will finish an attack cycle right when test time runs out  (error will be slightly bigger for slow hard hitting weapons) but we can reduce this error's influence by increasing test time..
 
 
Semi-Automated test
 
Same steps but have some Windows macro software start and pause video recording and test .. (increased accuracy on measured interval) .. Will still have to manually count the number of swings from video later ..  
 
 
Fully automated test could set a break point on address value of health variable for test dummy and increase hit counter each time it takes damage and gets healed by "god mode" (no longer need to record and count swings after, but I don't know how to do that for pillars of eternity) ..So it will be semi-automated method for me .. 
 
I will update this thread with test results  (might take a day or two for first test results since I'll be traveling soon) .. You're welcome to do the same if this sounds like a reasonable testing method ..
 
 
 
 
 
   
 


#2
Daemonjax

Daemonjax

    (4) Theurgist

  • Members
  • 344 posts

You could automate it full using autohotkey and tapping the screen image to look specifically for hits or misses from your character to get a count.

 

I would manually edit the video to cut off the beginning and end before analysis.

 

 

I don't think what you're proposing is worth it, though.

 

Alternatively, you can do what I do and use virtualdub to count the number of frames between a reference point -- like when the recovery bar is completely depleted... just a handful of attacks is good enough.


Edited by Daemonjax, 12 April 2015 - 09:25 PM.






Also tagged with one or more of these keywords: game mechanics

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users