Evade chance and defense penetration bugs

Started 8 Dec 2018
by Dimir
in Resolved Issues
I tested evade rate today on Pendragon and then on Phoenix and I found a weird mix of results. I can share all the Pendragon combat logs (just tell me where to send em), and I will post the summery below but here are the main points:
1. Evade rate is too low on phoenix vs low weaponskill and too high vs higher weaponskill.
2. A weapon's primary damage stat makes almost no difference to defense penetration when it should make a big one.

In case anyone reads this and wonders about the miss rate on Pendragon, I used the starting weapon to do low damage. That should make no difference as misses happen after evade chances and weapon bonus should only affect miss chance, just as a style bonus to hit only affects miss chance:
https://darkageofcamelot.com/content/grab-bag-april-25-2008

// PHOENIX
//===========================================
Cleric vs Infiltrator:
Infil base: 210 dex, 158 qui
Cleric base: 150 str
41.3% evade

Infil buffed: 348 dex, 236 qui
Cleric base
47.14% evade

Infil buffed: 348 dex, 236 qui
Cleric buffed: 283 str
47.12% evade

Cleric vs Armsman:
Infil base: 210 dex, 158 qui
Arms base: 220 str, 35+15 weapon spec
37.91% evade

Infil buffed: 348 dex, 236 qui
Arms base: 220 str, 35+15 weapon spec
42.82%

Infil buffed: 348 dex, 236 qui
Arms buffed: 353, 35+15 weapon spec
42.79%
//===========================================

// PENDRAGON:
//===========================================

Note for Cleric vs Infil: I should have logged more swings, because you can see the Infil is probably just hitting the 50% evade in rvr cap and the swing count is low enough that there is some varience.
Log 1:
"cleric 152 str, infil has 212 dex and 158 qui"

Evades: 399
Miss: 158
Hit: 196
Fumble: 5
Total attacks without fumble: 753
Evade chance 339/753: 52.98%

Log 2:
"cleric 152 str, infil has 353 dex and 239 qui (buffed)"

Evades: 331
Miss: 154
Hit: 178
Fumble: 3
Total attacks without fumble: 663
Evade chance 331/663: 49.92%

Log 3:
"cleric 288 str (buffed), infil has 353 dex and 239 qui (buffed)"

Evades: 394
Miss: 183
Hit: 186
Fumble: 2
Total attacks without fumble: 763
Evade chance 394/763: 51.63%
//=================
Log 4:
"armsman vs infil for evade test"
"armsman has 39+11 slash, 222 str, infil has 212 dex and 158 qui"
Evades: 399
Miss: 430
Hit: 484
Fumble: 12
Total attacks without fumble: 1313
Evade chance 399/1313: 30.38%

Log 5:
"armsman vs infil for evade test"
"armsman has 39+11 slash, 222 str, infil has 342 dex and 243 qui"
Evades: 626
Miss: 343
Hit: 432
Fumble: 13
Total attacks without fumble:
Evade chance 626/1401: 44.68%

Log 6:
"armsman vs infil for evade test"
"armsman has 39+11 slash, 362 str, infil has 342 dex and 243 qui"
Evades: 661
Miss: 647
Hit: 779
Fumble: 18
Total attacks without fumble:
Evade chance 661/2087: 31.67%
//===========================================

Issue Timeline

8 Dec 2018 Dimir created this issue
9 Dec 2018 gruenesschaf triaged this issue
9 Dec 2018 gruenesschaf changed the status of this issue from new to open
9 Dec 2018 gruenesschaf assigned this issue from Nobody to gruenesschaf
9 Dec 2018 gruenesschaf changed the status of this issue from in_progress to resolved
Sat 8 Dec 2018 9:08 PM by Dimir
Can't edit, but the "Cleric vs Armsman" line in Phoenix is suppose to be Infil vs Armsman.
Sat 8 Dec 2018 10:57 PM by gruenesschaf
You chose a pretty bad day to test it here, the update included a bug that made stats only count 100th (basically a *0.01 was done twice) and is already fixed but will only be there on the next update
Sat 8 Dec 2018 11:00 PM by Dimir
Well even so, it's very quick to test here (that is nice!) so I can re-do that test tomorrow. Being able to duel log would make it so much easier to test though

Return to Resolved Issues or the latest topics