Go Back   Nodiatis Forums > Nodiatis Forums > General Discussion > Bug & Exploit Reports
Forgot?

Reply
 
Thread Tools Display Modes

Old 12-30-2012, 08:23 PM   #41
chrisdj
Boss Hunter
 
chrisdj is offline
Join Date: Apr 2009
Posts: 249
Default

my combined int/cnc is 930 currently int 413 cnc 517 hope that helps and really appricate all this testing on everyones part as frankly this skill has driven me nuts over the years! lol

shame we cant have this changed to a straight 25% max damage increase at 100 skill
  Reply With Quote

Old 01-01-2013, 03:45 PM   #42
Thomas Jordan
Seer's BFF
 
Thomas Jordan's Avatar
 
Thomas Jordan is offline
Join Date: Nov 2009
Location: Louisiana
Posts: 740
Default

Playing with the Chrisdj Norm03 Lake Goddess data set, the single largest I found.

Mean = 837.4214876

Standard Deviation = 198.2222134

With that mean and standard deviation, his expected damage range is 639 - 1,035. Only normal hits larger than 1,035 are going to be recognizable as being caused by "Exploit Weakness."

His mean is higher, as it should be, since he is using Ranger's Revenge Arrow and Demontendow Bow, both of which do more damage than my Demontooth Arrow and Soulthief Bow.

His combine INT/CNC is 930; mine is 856.

His standard deviation is 198; mine is 175.

It's hard to tell if his higher INT/CNC is making any difference at all. Since his is higher than mine, we'd expect him to have a tighter hit range, and thus a smaller standard deviation. This is probably being masked by his harder hitting equipment, which has a raw damage range of 64:458 compared to my raw damage range of 14:297. His spread is 394 damage; mine is 283.

Need to do some more playing with the data. I can increase my INT/CNC with Raiden legs, but that's a ways in the future. So I'm still not sure what all of this is telling us.
__________________
Quote:
Originally Posted by Glitchless View Post
Confirming TJ correct.
  Reply With Quote

Old 01-01-2013, 04:04 PM   #43
Glitchless
Administrator
 
Glitchless's Avatar
 
Glitchless is offline
Join Date: May 2008
Posts: 8,314
Default

If you want to see if INT/CNC (damage modifying stats for archers) is working you compute the ratio of your actual average damage to your (max + min damage)/2 and compare that figure to someone else or to yourself after changing your INT/CNC. Comparing the standard deviation of two data sets that have different spreads is not going to tell you much of anything.

Also note with the last patch we made all hits past the 25th end up in the 25th # file same file since they will all have the same exploit weakness factor.
__________________
Computing the probability that at least one of the following events will occur:
P(a or b ... or z) = 1 - P(!a and !b ... and !z)
Probability
  Reply With Quote

Old 01-01-2013, 04:16 PM   #44
Thomas Jordan
Seer's BFF
 
Thomas Jordan's Avatar
 
Thomas Jordan is offline
Join Date: Nov 2009
Location: Louisiana
Posts: 740
Default

Didn't figure it would Glitch, and unfortunately the four of us all have different equipment sets, so there is no easy way to directly compare us. There's a lot still to learn from this data, so I'm still playing with it.

I do fight some fights most days in Redwood Forest, and usually end up using a whip and a shield. I have less combined INT/CNC with that equipment set than I do with my normal equipment set. Unfortunately, those are long fights for me, so the data file isn't very large yet. Guess I'll try to get in more of them each day to build up the size of the data file. Then I can compare me to me with two different INT/CNC values.
__________________
Quote:
Originally Posted by Glitchless View Post
Confirming TJ correct.

Last edited by Thomas Jordan; 01-01-2013 at 04:55 PM..
  Reply With Quote

Old 01-20-2013, 02:42 PM   #45
Glitchless
Administrator
 
Glitchless's Avatar
 
Glitchless is offline
Join Date: May 2008
Posts: 8,314
Default

We've stopped recording data as the results are now conclusive. Taking a look at maximum normal hits numbers 0, 5, 10, 15, 20, and 25+ using equipment set 1830000000 (Files thomasjordan/Eqp1830000000Norm0/5/10/15/20/25Cluster of Animated Roots.csv) we have:
1st hit: 911 max
6th hit: 950 max
11th hit: 999 max
16th hit: 1034 max
21st hit: 1057 max
26th+ hit: 1133 max

This equals a % increase on those hits of
1st hit: 0%
6th hit: 4.8%
11th hit: 9.6%
16th hit: 13.5%
21st hit: 16.0%
26th+ hit: 24.3%

Analyzing a different creature and using crits instead of normal hits, files thomasjordan/Eqp1830000000Crit0/5/10/15/20/25Flying Squirrel.csv. This second set doesn't have quite as much data, the actual max doesn't appear to have been reached on the 21st hits that only had 13 data points:

1st crit: 1485 max
6th crit: 1553 max
11th crit: 1625 max
16th crit: 1693 max
21st crit: 1672 max
26th+ crit: 1827 max

This equals a % increase on those crits of
1st crit: 0%
6th crit: 4.7%
11th crit: 9.5%
16th crit: 14.1%
21st crit: 12.7%
26th+ crit: 23.1%

Clearly the damage trends upward as more hits are landed in a way exactly one would expect of Exploit Weakness. The files will remain for you to download and analyze as you wish.
__________________
Computing the probability that at least one of the following events will occur:
P(a or b ... or z) = 1 - P(!a and !b ... and !z)
Probability
  Reply With Quote

Old 01-29-2013, 07:11 PM   #46
Lurkee
Seer's BFF
 
Lurkee is offline
Join Date: Mar 2009
Posts: 707
Default

Quote:
Originally Posted by Thomas Jordan View Post
Glitchless:

Like usual, this is rapidly going nowhere.

I've been a dedicated bow fighter for about three years now. I've had an epic 99 bow set for about 18 months, and began upgrading to legendary 100 bow gear about six months ago.

My Master's thesis was about decision theory applied to management decisions. I understand probability. My doctoral dissertation is about decision theory applied to mass evacuation decisions. I understand probability. I ran combat simulation centers for the US Army for 20+ years. A few used deterministic models; most used stochastic models. I understand probability.

I can tell the difference between a hit for 800 damage and 1,000 damage, or, following Stormy's reasoning with low level bow gear, 100 and 125. When this topic first came up my exploit weakness skill was 28. Even then, I was able to convince Glitchless the skill was not working properly, and he corrected an error in the formula.

Well, my skill is 93, and that should represent a big difference in long fights. Thanks to the role back of the damage mitigation patch, I now routinely fight in Redwood Forest, Endless Canyon and sometime the Pit. Those are all 4+ minute fights, which gives the exploit weakness skill time to maximize. I am not seeing anywhere near a 25% increase in damage over these long fights, and neither is anyone else I have communicated with via PM. I'm not seeing much if any increased damage at all.
lolol
__________________
If you make a sad face when you sing I hate you, die.
  Reply With Quote
Reply

Bookmarks

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump



All times are GMT -5. The time now is 07:30 PM
Boards live since 05-21-2008


vBulletin skin developed by: eXtremepixels
Powered by vBulletin® Version 3.8.6
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.