Intellect & hit quality adversely affect total damage inflicted by Wounding Shot, Enduring Flames and "Wounding" unique weapons DOTs

[peddroelm]

WD = base_weapon_roll * (1 + sum weapon damage multipliers) 

 

Wounding shot will apply 100% of WD in DOT ticks..

Each DOT tick is again multiplied by might (YAY !!) and gets 0.25 * appropriate target's DR substracted ..

 

Higher INT , better hit quality - dilute the DOT ticks - more ticks will less damage per tick (more time needed to apply less damage ) .. Damage lost per DR is a fixed amount.. More ticks - more damage lost due to DR ..

 

Result - vs high DR targets - higher INT and better hit quality can end up REDUCING total damage inflicted on target  ..

 

Oh - even stranger things when you dump INT (3) .. Wounding shot does more damage than it should (even vs targets with 0 DR).

 

test data

http://forums.obsidian.net/topic/78410-proper-debugger-breakpoints-dot-hot-tests/?p=1693464

Edited May 20, 2015 by peddroelm

[peddroelm]

Perhaps a numeric example/test is needed to clarify 

 

Test subject  25 Pierce DR ( plate_armor_exceptional + blunting belt)

 

Ranger 20 Might 3 Intellect   Weapon 22-22 Arbalest

 

+0 ms |Health Damage 5.721313 // hit 28.6 - 25 = 5.7 Min  [22 * (1 + 0.3) = 28.6]

+53 ms |Health Damage 8.056580 // ?

+2994 ms |Health Damage 8.056580

+3032 ms |Health Damage 8.056580 

Total Wounding shot damage 8.056 * 3 = 24.168 

 

 

Ranger 20 Might 10 Intellect   Weapon 22-22 Arbalest

 

+0 ms |Health Damage 5.721313 //  hit 28.6 - 25 = 5.7 Min       [22 * (1 + 0.3) = 28.6]

+60 ms |Health Damage 3.049316  // (28.6 / 4) * 1.3 - 25 * 0.25

+3008 ms |Health Damage 3.049316

+3004 ms |Health Damage 3.049316

+3045 ms |Health Damage 3.049316 

Total Wounding shot damage 3.049 * 4 = 12.196

 

Ranger 20 Might 10 Intellect   Weapon 22-22 Arbalest

 

+0 ms |Health Damage 5.721313 //  hit 28.6 - 25 = 5.7 Min       [22 * (1 + 0.3) = 28.6]

+89 ms |Health Damage 1.239929 //  (28.6 /6) * 1.3 * 0.2 = 1.2393

+3003 ms |Health Damage 1.239929

+3026 ms |Health Damage 1.239929

+3001 ms |Health Damage 1.239929

+3094 ms |Health Damage 1.239929

+3001 ms |Health Damage 1.239929

Total Wounding shot damage 1.239 * 6 = 7.434

 

Ranger 20 Might 20 Intellect   Weapon 22-22 Arbalest

 

+0 ms |Health Damage 9.106628  // CRIT 34.1 - 25 = 9.1        [22 * (1 + 0.3 + 0.25) = 34.1 ]

+77 ms |Health Damage 0.985535 // (34.1 / 9) * 1.3 * 0.2 = 0.985.. 

+2978 ms |Health Damage 0.985535

+3022 ms |Health Damage 0.985535

+3011 ms |Health Damage 0.985535

+3021 ms |Health Damage 0.985535

+3005 ms |Health Damage 0.985535

+3022 ms |Health Damage 0.985535

+3029 ms |Health Damage 0.985535

+3002 ms |Health Damage 0.985535

Total Wounding shot damage 0.985 * 9 = 8.865

 

 

////////////////////////////////

////////////////////////////////

So in the above test 20 might Ranger with 22-22 arbalest vs 25 Pierce DR

 

Total wounding shot DOT

 

intellect 3 hit  24.168  

intellect 10 hit  12.196  

intellect 20 hit  7.434  

intellect 20 crit  8.865  

 

More intellect less damage, yes ? Better hit quality less damage, yes ? Working as intended, yes ?  

Edited May 20, 2015 by peddroelm

[peddroelm]

Numeric example/test Flame of Devotion // Enduring Flames

 

Test subject burn proofed  plate_armor_exceptional 20 pierce DR 19 burn DR

 

Paladin 20 Might 3 Intellect   Weapon 22-22 Arbalest

 

+0 ms |Health Damage 18.159912 // FoD hit 28.6 - 20 DR = 8.6 pierce + 9.6 Burn

[ 22 * ( 1 + 0.3) = 28.6    28.6 * 0.5 - 19 * 0.25 = 9.55 ]

+63 ms |Health Damage 3.833984

+3003 ms |Health Damage 3.833984

+3031 ms |Health Damage 3.833984

Total enduring flames damage 3.833 * 3 = 11.499

 

Paladin 20 Might 10 Intellect   Weapon 22-22 Arbalest

 

+0 ms |Health Damage 18.159912 // FoD hit 28.6 - 20 DR = 8.6 pierce + 9.6 Burn

[ 22 * ( 1 + 0.3) = 28.6    28.6 * 0.5 - 19 * 0.25 = 9.55 ]

+57 ms |Health Damage 1.115906

+3015 ms |Health Damage 1.115906

+3099 ms |Health Damage 1.115906

+3029 ms |Health Damage 1.115906

Total enduring flames damage 1.116 * 4 = 4.464

 

Paladin 20 Might 10 Intellect   Weapon 22-22 Arbalest (merciless hand & rabbit fur)

 

+0 ms |Health Damage 37.959900 // FoD crit 41.8 - DR 20 = 21.8 Pierce + 16.2 Burn

[ 22 * ( 1 + 0.3 + 0.25 + 0.1 + 0.25) = 41.8    41.8 * 0.5 - 19 * 0.25 = 16.15 ]

+78 ms |Health Damage 1.087219

+3002 ms |Health Damage 1.087219

+3032 ms |Health Damage 1.087219

+3031 ms |Health Damage 1.087219

+3128 ms |Health Damage 1.087219

Total enduring flames damage 1.087 * 5 = 5.435

 

Paladin 20 Might 20 Intellect   Weapon 22-22 Arbalest

 

+0 ms |Health Damage 18.159912

[ 22 * ( 1 + 0.3) = 28.6    28.6 * 0.5 - 19 * 0.25 = 9.55 ]

+57 ms |Health Damage 0.743958

+3019 ms |Health Damage 0.743958

+3013 ms |Health Damage 0.743958

+3026 ms |Health Damage 0.743958

+3027 ms |Health Damage 0.743958

Total enduring flames damage 0.744 * 5 = 3.72

 

Paladin 20 Might 20 Intellect   Weapon 22-22 Arbalest (merciless hand & rabbit fur)

 

+0 ms |Health Damage 37.959900 // FoD crit 

[ 22 * ( 1 + 0.3 + 0.25 + 0.1 + 0.25) = 41.8    41.8 * 0.5 - 19 * 0.25 = 16.15 ]

+80 ms |Health Damage 0.724792

+2989 ms |Health Damage 0.724792

+3016 ms |Health Damage 0.724792

+3031 ms |Health Damage 0.724792

+3008 ms |Health Damage 0.724792

+3092 ms |Health Damage 0.724792

+3007 ms |Health Damage 0.724792

+3030 ms |Health Damage 0.724792

Total enduring flames damage 0.725 * 8 = 5.8

 

 

So in the above test 20 might Paladin with 22-22 arbalest vs 19 Burn DR

 

Total Enduring Flames DOT

 

intellect 3 hit  11.499

intellect 10 crit 5.435 

intellect 10 hit  4.464

intellect 20 crit 5.8 

intellect 20 hit  3.72 

 

More intellect less damage, yes ? Better hit quality less damage, yes ? Working as intended, yes ?  

[peddroelm]

testing "wounding" weapon propriety

 

ranger 

20-20 tidefall

15% 2h weapon style

45% superb

 

10 might 3 int vs 0 DR

+0 ms |Health Damage 32.000000 // hit 20 * (1 + 0.45 + 0.15 ) = 32

+62 ms |Health Damage 5.307678 

+3004 ms |Health Damage 5.307678

32 * 0.25 - 5.3 * 2 = -2.6 

 

+0 ms |Health Damage 32.000000 //hit

+65 ms |Health Damage 5.307678 // hit dot

+2975 ms |Health Damage 42.000000 //crit

+68 ms |Health Damage 5.307678 // hit dot

+1 ms |Health Damage 5.076904 //crit dot  lower? 

+2972 ms |Health Damage 5.076904

 

+0 ms |Health Damage 22.000000 // graze

+58 ms |Health Damage 6.000000

22 * 0.25 - 6  = - 0.5

 

+0 ms |Health Damage 42.000000 // Crit

+63 ms |Health Damage 5.076904

+3007 ms |Health Damage 5.076904

 

Crit dot damage lower than hit dot for no reason (0 DR same number of ticks) ..

 

////////////////////////

might 20

 

+0 ms |Health Damage 48.005981 // crit 20 * (1 + 0.45 + 0.15 + 0.3 + 0.5)

+63 ms |Health Damage 7.802979 // 6 * 1.3

+2998 ms |Health Damage 7.802979

48 * 0.25 - 6 * 2 =  0

 

+0 ms |Health Damage 38.005981 // hit

+60 ms |Health Damage 8.703796

+2993 ms |Health Damage 8.703796 

again more DOT damage on hit vs crit

 

+0 ms |Health Damage 28.005981 // graze

+53 ms |Health Damage 11.406250 // 11.4/(1.3) = 8.77

28 * 0.25 - 7 = 0

 

//////////////////////////////

int 10 might 10

 

+0 ms |Health Damage 32.000000 // hit 

+62 ms |Health Damage 3.450012 // more int less damage

+2997 ms |Health Damage 3.450012

 

+0 ms |Health Damage 42.000000 //crit

+68 ms |Health Damage 3.299988

+3070 ms |Health Damage 3.299988

+3031 ms |Health Damage 3.299988

 

int 10 20 might

+0 ms |Health Damage 38.005981 // hit

+68 ms |Health Damage 5.657471

+2978 ms |Health Damage 5.657471

(38 * 0.25 / 2) - 5.65 / (1.3) = 0.4 ??

 

////////////////////////////////

int 20 might 20 0DR

 

+6766 ms |Health Damage 38.005981 // hit

+62 ms |Health Damage 3.771667

+2978 ms |Health Damage 3.771667

+3003 ms |Health Damage 3.771667

 

3.77 * 3 - 5.66 *2 = 0.01

///////////////////////////////

////////////////////////////////

 

tests vs 25 DR 

3 INT 20 Might

 

+0 ms |Health Damage 5.601196 // graze

+43 ms |Health Damage 11.406250

 

+0 ms |Health Damage 13.005981 / /hit 

+53 ms |Health Damage 8.703796

+3001 ms |Health Damage 8.703796

 

+0 ms |Health Damage 23.005981 // crit

+60 ms |Health Damage 7.802979

+2981 ms |Health Damage 7.802979

 

10 int 20 might

 

+0 ms |Health Damage 5.601196 // graze

+48 ms |Health Damage 7.414063

 

+0 ms |Health Damage 13.005981 // hit

+54 ms |Health Damage 5.657471

+2985 ms |Health Damage 5.657471

 

+0 ms |Health Damage 23.005981 // crit

+59 ms |Health Damage 5.071960

+3002 ms |Health Damage 5.071960

+3006 ms |Health Damage 5.071960

 

20 int 20 might

 

+0 ms |Health Damage 5.601196 // graze  28-25 = 5.6 min

+49 ms |Health Damage 4.942688

+3022 ms |Health Damage 4.942688

 

+0 ms |Health Damage 13.005981 // hit

+53 ms |Health Damage 3.771667

+3012 ms |Health Damage 3.771667

+3024 ms |Health Damage 3.771667

 

+0 ms |Health Damage 23.005981 // crit 

+65 ms |Health Damage 3.381287

+2990 ms |Health Damage 3.381287

+3111 ms |Health Damage 3.381287

+3006 ms |Health Damage 3.381287

 

 

 wounding DOT  3 INT >> 10 INT > 20 INT

[peddroelm]

bump

[peddroelm]

bump

[El Zoido]

Interesting, how did you acquire those numbers?

Did you run actual in-game tests and if so, how do you read-out such accurate values?

[durbal]

Interesting, how did you acquire those numbers?

Did you run actual in-game tests and if so, how do you read-out such accurate values?

 

Basic math and using eyes.

[El Zoido]

Well, my point is, I just want to be sure that this is what actually happens in-game, NOT just what you arrive at from calculating it according to the tooltips, which are wrong even according to the devs in those cases.

[peddroelm]

testing methodology https://www.youtube.com/watch?v=Yq59wttu2Po

 

test threads 

http://forums.obsidian.net/topic/78239-abilities-various-other-damage-related-mechanics-revealed/

https://forums.obsidian.net/topic/78410-proper-debugger-breakpoints-dot-hot-tests/

[peddroelm]

bump

[peddroelm]

bump

[celliott]

Hello all.  peddroelm, unfortunately I am unable to test these things in this way.  If you can add an item/talent/ability to a character without using some sort of external editor and showing the values we can go from there.  You can feel free to use cheats to add items that you can legitimately get in game, modify talents, etc, but it has to be done in game, with values that are possible to attain.  Otherwise I have no way of verifying the data, as using an external editor ruins the 'sterile' environment we test in.  Sorry for the inconvenience, and thank you for your ongoing support and work on these issues.

[peddroelm]

I think you can simply confirm those issues without the need for third party debugger.  (but I am surprised you are required to do your job without proper tools - like ability to "see" damage inflicted by damage/heal over time effects )

 

Much easier for the skills that work out of combat.  Notice test PC char health number (displayed in interface) before and after test hit (+ DOT time) . This would give you total damage inflicted .. You can alter  test parameters via console (better armor for DR on receiving char, altering INT/MIGHT attribute on the PC char doing the hits ) .. 

 

For the combat only skills - you'll need to test them during combat .. You can not toggle combat state via console (would help testing) as far as I'm aware of but you can achieve long enough (permanent) combat state multiple ways ..

 

One way would be to have a single low damaging/low accuracy no special attacks hostile permanently engaged to a high DR & deflection fighter (press X for the fighter to stand and not attack back) ..

 

After measuring the total test damage inflicted on the test PC (one test round/hit) you can console "heal party" to reset the state for another test .. (heals test PC to be able to receive another hit and heals the health & endurance of the fighter so the combat state will maintain for as long as it is needed) ..

 

Anyone can do this - its just going to be tedious work ..

Edited June 30, 2015 by peddroelm

[celliott]

Hello again.  Do you happen to have a saved game you can upload to Dropbox so I can take a look at your characters' stats.  I'm having a hard time coming up with the attribute scores on your characters in a legitimate way.  Not saying they aren't legit, just that I'm having a hard time hitting the same values.  Thanks!

[peddroelm]

the int /might score on my test characters aren't "legit" since I've use the a console commands to change them to the amount required for each test .. It would take extreme amounts of real time to test those bugs properly without at least some console help .. I did not use third party application to change char stats for tests - only the in-game console ..

Edited June 30, 2015 by peddroelm

[celliott]

Hello again.  I understand the stats were changed with the console.  I'll pass this information along to the programming team.  I will let you know however that they are going to ask me for a saved game and output log and probably a screen shot for the bug in our database.  It is policy to have those things on every bug if possible.  Thanks!

[peddroelm]

Hello again.  I understand the stats were changed with the console.  I'll pass this information along to the programming team.  I will let you know however that they are going to ask me for a saved game and output log and probably a screen shot for the bug in our database.  It is policy to have those things on every bug if possible.  Thanks!

Game log does not (currently) list  DOT damage ticks .. (which makes testing those kind of bugs difficult without third party application)

 

You are free to include the values picked up using the third party application or link to the OP. I do think they would find them useful .

[AndreaColombo]

Those scores are very easy to obtain legitimately in-game.

 

Here's a saved game for peddroelm's first test case (post n. 2): link

 

This includes:

 

• Protagonist wearing Exceptional Plate Armor (bought from Dunstan) and Blunting Belt (found adventuring; it is also sold in Gilded Vale.)*
• A ranger with 20 MIG, 3 INT
• A ranger with 20 MIG, 10 INT
• A ranger with 20 MIG, 20 INT

All rangers hired at the Goose and Fox inn in Copperlane; you can easily use console commands to level them up as appropriate/required (they're level 1.)

 

* note that in this game I have completed the Crucible Knights quest and therefore have the Second Skin talent that grants me bonus DR. Hence, the protagonist's DR is higher than what peddroelm mentions in his post. If you want the same DR, you can hire a Fighter at an inn and equip them with those items.

 

Because you'll be using "real world" weapons instead of weapons whose minimum damage is the same as the maximum damage, testing this is going to take a very long time as you'll need to get the exact same damage on hit from all three rangers in order for data to be comparable.

 

If you want to enter combat state, use a console command to create a hostile creature (example: BSC cre_blight_earth 1) and make sure the protagonist engages it. You may remove the protagonist's weapon to make sure the blight isn't killed too quickly (perhaps also give a shield to the protagonist; there is one in the stash.)

Edited June 30, 2015 by AndreaColombo

[celliott]

Hello again.  Thanks for the saved game, AndreaColombo.  

 

Peddroelm, yes, they will find all this information useful.  I will and always do link the thread in question to the bug database.  I understand that no information can really be gained from the output log in this case, it is just policy to ask for it and try and include it with the bug for all bugs.  That way we don't have to wonder which bugs need it and which don't.  I know this is sometimes an inconvenience and apologize for that.

 

As always, thanks for your support and information regarding this issue.  I will dig in and add all relevant information to our database.

[AndreaColombo]

My output log for your convenience: link

 

It is my opinion that fixing the issue for one ability (e.g. Wounding Shot for rangers in the case of my saved game above) would fix it for all other abilities as well.

 

However, if your programmers require saved games for peddroelm's other test cases as well, please ask here and I'll try to provide them.

[celliott]

Thank you very much, AndreaColombo.  I will add everything I can to the database and update this thread if needed.  

 

More than likely, yes, fixing it for this will fix it globally.  Just to be safe though, and to follow protocol, it is nice to have the files.

[lameover]

Hello!

Is there any news?

[celliott]

Hello.  It does not look like this is fixed yet, but it is marked as high priority and is hotlisted.  Thank you for your ongoing support.