[X] My Assistant

I tested how attack speed affects debuff durations at the beginning of this same thread. This new explanation is good. It's simple and I believe explains the random variance in debuff duration when you have small amounts of attack speed.

It may also be that when you kill the last monster, the round immediately ends and the next one begins. So rounds don't start at 0 ticks, they might start at fractional tick. This is the inherent timing jitter I described previously which allows ailment durations to vary randomly (which is proper and a good thing).

This isn't so much research as rediscovery at this point, but I have a few sources for some time-related functions of the game I'll be updating the wiki with later.

This quote is from the change notes for the update that reworked action speed, it has not been noted to have changed since:

- Effect duration as well as heal- and damage-over-time ticks have been partially decoupled from the round counter, and now tick every 100 time units, independent of speed-affecting effects and actual action time.

From my own testing, I am quite confident new rounds start from 0 - however, there is some very slight imprecision with my numbers (to the tune of around 5 time units) so I will not post about this until I've discovered exactly what factor(s) I'm missing... I suspect it has something to do with rounding, but no approach I've attempted so far has fixed it.

This post has been edited by lestion: Jan 29 2020, 13:18

However you decide to formulate it, all I know is that when you have some amount of attack speed, the effective durations of all your status ailments becomes slightly random, such that your PA duration might be 7.01 turns instead of 7 turns. And that's a good thing because it helps attack speed do what it is supposed to.

When you apply a stack of PA on an Imperiled 80% PMit monster, the PMit falls from 30% 40% to 22.5% 30%, which represents an increase in damage of about 7.5% 10% of the initial, raw damage output.

It's not much of a difference but that's how I calculate it, not sure which is right.

Ah I see where the difference originates from. Are you sure Imperil reduces an original PMit of 80% down to 30%? I believe either Imperil or PA2 would reduce the PMit to 40%.

Imperil is an additive -50% PMit...
https://ehwiki.org/wiki/Spells#Deprecating_Spells
https://ehwiki.org/wiki/Equipment_Procs

You have literally not read the wiki we've just updated, you know... (IMG:[invalid] style_emoticons/default/dry.gif)

This post has been edited by decondelite: Jan 29 2020, 22:07

In that case Imperil is better than I previously thought. However I went to check my old initial test data on this and (?)I feel it matches the "old" Imperil better(?) I may eventually try to test this particular question more directly sometime. Comparing void damage ranges for plain PA2 versus plain Imperil on a school girl would be best.

edit: After further thought I realized I made a logical error in my original analysis of my test data. So I think decondelite is right now.

edit2: I'm confused again. Let's step through this. At my lower level a school girl has about 60% Pmit instead of 66%. With Penetrated 3 that would be 15% Pmit remaining. With decondelite's way, PA3 + Imperil would be 2.5% Pmit remaining.

So by decondelite's way, every stack of PA increases damage by about 15% of the full unmitigated damage. Throwing on Imperil at the very end like I did in my test would increase damage by 12.5% almost the same.

Maybe I was right to begin with. I don't know. I should test at some point.

This post has been edited by BlueWaterSplash: Jan 30 2020, 10:19

Monsters Statistics

- 2,513 monsters / 167,318 times for 20 x PFUDOR grindfest

CODE

==================================================================
    PL |  Count |  Average |  Encounter |    Ratio |  Accumulated
------------------------------------------------------------------
  2250 |     46 |      479 |     22,049 |  13.18 % |      13.18 %
  2150 |      2 |      479 |        957 |   0.57 % |      13.75 %
  2100 |      3 |      459 |      1,378 |   0.82 % |      14.57 %
  2050 |      1 |      410 |        410 |   0.25 % |      14.82 %
  2000 |      2 |      401 |        802 |   0.48 % |      15.30 %
  1950 |      5 |      361 |      1,806 |   1.08 % |      16.38 %
  1900 |      1 |      309 |        309 |   0.18 % |      16.56 %
  1850 |      4 |      302 |      1,209 |   0.72 % |      17.28 %
  1800 |     89 |      232 |     20,685 |  12.36 % |      29.65 %
  1750 |      5 |      261 |      1,304 |   0.78 % |      30.43 %
  1700 |     29 |      162 |      4,699 |   2.81 % |      33.23 %
  1650 |    221 |      129 |     28,558 |  17.07 % |      50.30 %
  1600 |    156 |      100 |     15,675 |   9.37 % |      59.67 %
  1550 |    323 |       70 |     22,611 |  13.51 % |      73.19 %
  1500 |    564 |       50 |     28,480 |  17.02 % |      90.21 %
  1450 |    327 |       24 |      8,010 |   4.79 % |      94.99 %
  1400 |    301 |       16 |      4,939 |   2.95 % |      97.95 %
  1350 |    117 |       11 |      1,329 |   0.79 % |      98.74 %
  1300 |    291 |        4 |      1,152 |   0.69 % |      99.43 %
  1250 |      6 |        1 |          6 |   0.00 % |      99.43 %
------------------------------------------------------------------
  1200 |      5 |       59 |        295 |   0.18 % |      99.61 % # MONSTER DATA IS OUT OF DATE
  1150 |      1 |       14 |         14 |   0.01 % |      99.62 % # MONSTER DATA IS OUT OF DATE
  1100 |      1 |      112 |        112 |   0.07 % |      99.68 % # MONSTER DATA IS OUT OF DATE
  1050 |      2 |       33 |         65 |   0.04 % |      99.72 % # MONSTER DATA IS OUT OF DATE
  1000 |      2 |       61 |        121 |   0.07 % |      99.80 % # MONSTER DATA IS OUT OF DATE
   950 |      1 |      100 |        100 |   0.06 % |      99.85 % # MONSTER DATA IS OUT OF DATE
   900 |      4 |       52 |        209 |   0.12 % |      99.98 % # MONSTER DATA IS OUT OF DATE
------------------------------------------------------------------
     0 |      4 |        9 |         34 |   0.02 % |     100.00 % # BOSS
==================================================================

================================================
  Rank |          Trainer |  Count |  Encounter
------------------------------------------------
     1 |         Nero-Arc |    200 |     16,332
     2 |          serorin |    127 |     14,135
     3 |         morineko |    182 |     13,296
     4 |           sssss2 |    200 |     12,230
     5 |          gc00018 |     97 |     11,336
     6 |      FreeloaderV |    101 |      9,213
     7 |     MidNightPass |     78 |      8,951
     8 |         RoadShoe |    200 |      6,409
     9 |         NerfThis |    200 |      6,384
    10 |            Koaen |      7 |      3,298
    11 |          danixxx |     17 |      2,953
    12 |        qdjseh001 |     66 |      2,874
    13 |         Mantra64 |     27 |      2,250
    14 |     Sushilicious |    116 |      1,980
    15 |       Hina Amano |    175 |      1,969
    16 |           DJNoni |     33 |      1,943
    17 |          Godde?? |     32 |      1,901
    18 |       Petal_Kiss |      7 |      1,744
    19 |        .@_@.@_@. |     49 |      1,663
    20 |         cylinnia |     28 |      1,490
    21 |         tempasdf |     14 |      1,428
    22 |         StonyCat |      4 |      1,423
    23 |         Honeycat |      5 |      1,345
    24 |           ddwiki |      4 |      1,255
    25 |   Kira.Yoshikage |     22 |      1,180
    26 |          lestion |     10 |      1,133
    27 |      Void Domain |      6 |      1,030
    28 |        sabregimp |     12 |      1,025
================================================

100 days after...

- 2,672 monsters / 167,241 spawned for 20 x PFUDOR grindfest

CODE

==================================================================
    PL |  Count |  Average |  Encounter |    Ratio |  Accumulated
------------------------------------------------------------------
  2250 |     54 |      472 |     25,485 |  15.24 % |      15.24 %
  2150 |      3 |      438 |      1,315 |   0.79 % |      16.02 %
  2100 |      8 |      434 |      3,470 |   2.07 % |      18.10 %
  2050 |      1 |      369 |        369 |   0.22 % |      18.32 %
  2000 |      3 |      390 |      1,171 |   0.70 % |      19.02 %
  1950 |      6 |      343 |      2,059 |   1.23 % |      20.25 %
  1900 |      1 |      283 |        283 |   0.17 % |      20.42 %
  1850 |     11 |      290 |      3,192 |   1.91 % |      22.33 %
  1800 |     90 |      208 |     18,704 |  11.18 % |      33.51 %
  1750 |     25 |      209 |      5,225 |   3.12 % |      36.64 %
  1700 |     31 |      152 |      4,726 |   2.83 % |      39.46 %
  1650 |    226 |      118 |     26,604 |  15.91 % |      55.37 %
  1600 |    220 |       85 |     18,798 |  11.24 % |      66.61 %
  1550 |    488 |       58 |     28,503 |  17.04 % |      83.65 %
  1500 |    825 |       28 |     22,882 |  13.68 % |      97.34 %
  1450 |    371 |        8 |      2,869 |   1.72 % |      99.05 %
  1400 |    205 |        4 |        766 |   0.46 % |      99.51 %
  1350 |     33 |        5 |        158 |   0.09 % |      99.60 %
  1300 |     56 |        8 |        421 |   0.25 % |      99.86 %
  1250 |      4 |       15 |         61 |   0.04 % |      99.89 %
  1200 |      3 |        1 |          4 |   0.00 % |      99.89 %
  1150 |      1 |       24 |         24 |   0.01 % |      99.91 %
  1100 |      1 |       39 |         39 |   0.02 % |      99.93 %
  1000 |      1 |       16 |         16 |   0.01 % |      99.94 %
   900 |      1 |       67 |         67 |   0.04 % |      99.98 %
     0 |      4 |        8 |         30 |   0.02 % |     100.00 %
==================================================================

- The number of pl 2250 monsters increased from 46 to 54.
- The average pl of monsters increased from 1712 to 1753.

CODE

================================================||======================================================
  Rank |          Trainer |  Count |  Encounter ||  Rank |     |          Trainer |  Count |  Encounter  
------------------------------------------------||------------------------------------------------------
     1 |         Nero-Arc |    200 |     16,332 ||     1 |  -  |         Nero-Arc |    200 |     15,111  
     2 |          serorin |    127 |     14,135 ||     2 | ▲ 1 |         morineko |    200 |     13,372  
     3 |         morineko |    182 |     13,296 ||     3 | ▽ 1 |          serorin |    113 |     12,802  
     4 |           sssss2 |    200 |     12,230 ||     4 | ▲10 |     Sushilicious |    153 |     11,147  
     5 |          gc00018 |     97 |     11,336 ||     5 |  -  |          gc00018 |     96 |     10,059  
     6 |      FreeloaderV |    101 |      9,213 ||     6 | ▽ 2 |           sssss2 |    200 |      9,659  
     7 |     MidNightPass |     78 |      8,951 ||     7 |  -  |     MidNightPass |     87 |      7,871  
     8 |         RoadShoe |    200 |      6,409 ||     8 | ▲10 |       Petal_Kiss |    200 |      7,853  
     9 |         NerfThis |    200 |      6,384 ||     9 | ▽ 3 |      FreeloaderV |    101 |      7,759  
    10 |            Koaen |      7 |      3,298 ||    10 | ▽ 1 |         NerfThis |    198 |      5,474  
    11 |          danixxx |     17 |      2,953 ||    11 | ▲10 |         tempasdf |     20 |      4,096  
    12 |        qdjseh001 |     66 |      2,874 ||    12 | ▽ 4 |         RoadShoe |    197 |      3,576  
    13 |         Mantra64 |     27 |      2,250 ||    13 | ▽ 3 |            Koaen |      7 |      3,099  
    14 |     Sushilicious |    116 |      1,980 ||    14 | ▽ 3 |          danixxx |     17 |      2,489  
    15 |       Hina Amano |    175 |      1,969 ||    15 | ▽ 3 |        qdjseh001 |     67 |      2,094  
    16 |           DJNoni |     33 |      1,943 ||    16 | ▲10 |          lestion |     18 |      2,037  
    17 |          Godde?? |     32 |      1,901 ||    17 | ▽ 4 |         Mantra64 |     17 |      1,851  
    18 |       Petal_Kiss |      7 |      1,744 ||    18 | ▲24 |           moanim |     27 |      1,839  
    19 |        .@_@.@_@. |     49 |      1,663 ||    19 | ▽ 3 |           DJNoni |     34 |      1,592  
    20 |         cylinnia |     28 |      1,490 ||    20 | ▽ 1 |        .@_@.@_@. |     19 |      1,518  
    21 |         tempasdf |     14 |      1,428 ||    21 | ▽ 4 |          Godde?? |     45 |      1,444  
    22 |         StonyCat |      4 |      1,423 ||    22 | ▲ 1 |         Honeycat |      5 |      1,342  
    23 |         Honeycat |      5 |      1,345 ||    23 | ▽ 1 |         StonyCat |      4 |      1,322  
    24 |           ddwiki |      4 |      1,255 ||    24 |  -  |           ddwiki |      4 |      1,141  
    25 |   Kira.Yoshikage |     22 |      1,180 ||    25 |  ?  |             ???? |      7 |      1,069  
    26 |          lestion |     10 |      1,133 ||    26 | ▲ 1 |      Void Domain |      4 |      1,018  
    27 |      Void Domain |      6 |      1,030 ||    27 | ▲ 6 |        in memory |     30 |      1,016  
    28 |        sabregimp |     12 |      1,025 ||    28 | ▲ 1 |   another planet |      2 |      1,006  
================================================||======================================================

- Some players are training their monsters.
- It is harder to meet my monsters than before.

Does Godslayer give magic damage boost?
Conclusion (preliminary): No, looks like it doesn't


How this was tested
- test at low difficulty, to see less effect of the 3% evade that godslayer brings
- Hell 2 runs IW
- Nintendo 3 runs IW
- Comparing damage per turn, corrected for number of cures at Nintendo

CODE

Hell - 81 round IW (avg 2 run s)
Godslayer                       Dohavin                    delta %
turns  damage/turn               turns  damage/turn
335,5  327129                   348,5  314586          + 4% for godslayer


Nintendo - 81 round IW (avg 3 runs)
Godslayer                       Dohavin                                  delta %
turns   damage   damage/turn*    turns  damage      damage/turn*
401    129362867 338349            385   128600593  348543           -3%, godslayer does less damage

*corrected for number of cures and health potions

This post has been edited by DJNoni: Feb 10 2020, 21:24

[attachmentid=138869]

Grindefest within 5989t, 13 horses, as charged cold mage with DDXI on cold attribute day, no flower vase and bubble gem. (IMG:[invalid] style_emoticons/default/tongue.gif)

I tied to full imperil on all monster, by using both down wheel and keyboadrs.
I splitted my T3 from cast to better use it. I didn't use gem, draught, mp and later sp.
I used hp, he before full cure and cure.

My basic prof was 552.6, and it took me 3529 turns to get monster full imperilled. I think less than 30 turns could be optimized through my action and strategy. Under ideal conditions(full prof), a charged cold mage could possiblel reach 56XXt or even less.

At the worst imp case I used 8 times to full imperil on 8 monster, 7 times on 7 monster. The full imp strategy had led to 138 full cures and 26 cures. --- I think I should full imperil all monster, I have tried to ingore some monster with poor(i.e 21, 28) cold migitation. But I finally found these monster might have too much magic migitate, involved by their tokens of chaos.
( I wonder know if imp could influence monster's CR )

I could recalled over 10 times 90% resists after I casted my Tier 3. (In old vesion of mb, it might be normal that your T3's icon showed prepared after use items, which results in 6 attacks form my data,)

I was very close to fail this time as previous sereval tries. Only one health at a moment where both of my LE, HE ,HP, FC and cure were in cd. The only thing I could try is to cast my T3 on 9 imperilled monsters. Luckly, I was alive by evade 3 monsters attack.

...

Did you mean to post the screenshot of your stats, or was it meant to be the end of battle monsterbation readout??

Impressive numbers. I was not expecting fests to be able to go that low.

Returning to the question we asked a month ago, I repeated my method of beating down School Girls, this time comparing the damages with Penetrated Armor 2 versus Imperil directly.

CODE

1H level 383
overwhelming strikes and spirit stance
8569 attack, Friday

12074~17738 main attack  21538~32306 crit
6112~8944 void strike  6261~8198 crit
6019~8803 dark strike  6095~9029 crit

Penetrated Armor 2

21537~31137 main attack  38899~55771 crit
10898~15569 void strike  10379~14791 crit
10478~15717 dark strike  10609~15717 crit

Imperil

21278~30359 main attack  37692~53428 crit
10380~15569 void strike  10509~15439 crit
11133~15455 dark strike  10478~15717 crit

A few more trials would make this data perfectly match but it's enough for our purposes. I believe this proves that Penetrated Armor 2 and Imperil reduce PMit by the exact same amount, consistent with my earlier damage tests.

Congratulations. You have managed to prove that 50% is equal to 50%. (IMG:[invalid] style_emoticons/default/unsure.gif)
More importantly. I thought that Imperil was additive, and given the stats of a SG, that doesn't appear to be the case.

How many attacks have you recorded?

This post has been edited by decondelite: Feb 29 2020, 01:57

I do these by going through one of the SG arenas manually, without hover. Depending on what data I'm going for, one arena would be enough. I did Arena 110 for this test and it wasn't quite enough to get perfect data. Most of the data has to be thrown out, because I only want to accept data with the correct amount of PA or Imperil. Each SG gets poked a few times then she gets too messed up and I kill her with Vital Strike.

However you don't need many hits if you study damages in this way, because the randomness is highly discrete: there are only 10~20 possible different random damage values. Luck eventually reveals the exact endpoints.

The proper way to do this is to reduce stamina first, because otherwise damages creep up due to proficiency increasing within the same arena. This is noticeable because it eventually increases the damages by 1, not the discretized amount. I don't actually do this though because my tests don't need to be ultra perfect in most cases.

I do these by going through one of the SG arenas manually, without hover. Depending on what data I'm going for, one arena would be enough. I did Arena 110 for this test and it wasn't quite enough to get perfect data. Most of the data has to be thrown out, because I only want to accept data with the correct amount of PA or Imperil. Each SG gets poked a few times then she gets too messed up and I kill her with Vital Strike.

However you don't need many hits if you study damages in this way, because the randomness is highly discrete: there are only 10~20 possible different random damage values. Luck eventually reveals the exact endpoints.

The proper way to do this is to reduce stamina first, because otherwise damages creep up due to proficiency increasing within the same arena. This is noticeable because it eventually increases the damages by 1, not the discretized amount. I don't actually do this though because my tests don't need to be ultra perfect in most cases.

Did you remember to exclude Konata from these tests (she has different stats to the others)? Did you make sure you only took values from turns after applying PA2 (as the damage occurs before the debuff, on the turn it is applied on)? Why are there such significant differences in the top-end comparing PA to imperil?

Not that I think any of these things are large enough to challenge the conclusion (that imperil is multiplicative, rather than additive) but I would like to ensure the test is as thorough as possible before making any edits to the wiki. (I think this was probably known at some point in the past and has just been misinterpreted when working on page updates!)

This is notably odd in one fashion, though: imperil is observably additive for specific mitigations (as can be seen with simple scanning before and after). I wonder why pmit (and presumably mmit) are applied differently, despite the same wording in the spell description?

This post has been edited by lestion: Feb 29 2020, 04:24

I excluded Konata. I took values after applying PA2. I've been doing this style of test for the last couple years so I have the method decently refined. Those differences are just due to randomness; I didn't do quite enough samples to ensure luck rolled each endpoint.

Note that you can actually fuse the crit and non-crit data for void strike and dark strike; the damage ranges do become perfect that way, as there are enough samples.

I had remarked on those oddities of Imperil in the past as well, but that's how it is.

copied from sssss2: research on effect of DD on holy mage (also, nice insights into the effect of infusionn and of gum+vase
___________________________________________________________________________________________
DD's effect on PFUDORfest

- Setup: holy mage (LHOH, 3 Radiant Phases, 2 Cottons)

- Type A: Day Bonus (holy mitigation -10% on sunday) + Infusion + Bubble-Gum + Flower Vase
- Type B: Day Bonus + Infusion
- Type C: no Day Bonus + no Infusion

CODE

Dæmon Duality           ||   VII  (+40%)         ||  VIII  (+45%)         ||    IX  (+50%)         |
prof_factor             || 0.946 | 0.996 ~ 0.998 ||   1.0                 ||   1.0                 |
Type                    ||     A |     B |     C ||     A |     B |     C ||     A |     B |     C |
----------------------------------------------------------------------------------------------------
Turn                    ||  2370 |  5924 |  6166 ||  2337 |  5736 |  6013 ||  2333 |  5634 |  5896 |
----------------------------------------------------------------------------------------------------
Infusion of Divinity    ||    25 |    52 |       ||    24 |    50 |       ||    24 |    49 |       |
Bubble-Gum              ||    25 |       |       ||    24 |       |       ||    24 |       |       |
Flower Vase             ||    25 |       |       ||    24 |       |       ||    24 |       |       |
----------------------------------------------------------------------------------------------------
Paradise Lost           ||   431 |  1024 |  1075 ||   426 |   993 |  1049 ||   424 |   978 |  1030 |
Banishment              ||   647 |  1484 |  1558 ||   636 |  1440 |  1520 ||   634 |  1414 |  1488 |
Smite                   ||  1054 |  1984 |  2057 ||  1036 |  1946 |  2019 ||  1032 |  1922 |  1989 |
----------------------------------------------------------------------------------------------------
Cure                    ||    31 |   677 |   733 ||    33 |   637 |   706 ||    31 |   622 |   687 |
Full-Cure               ||       |   116 |   128 ||     1 |   108 |   122 ||     2 |   101 |   117 |
Regen                   ||     6 |    13 |    13 ||     6 |    12 |    13 ||     6 |    12 |    13 |
Arcane Focus            ||     2 |     5 |     5 ||     2 |     5 |     5 ||     2 |     4 |     5 |
----------------------------------------------------------------------------------------------------
Health Draught          ||    25 |    36 |    36 ||    24 |    36 |    36 ||    24 |    35 |    35 |
Health Potion           ||       |    48 |    52 ||       |    43 |    49 ||       |    42 |    48 |
Health Elixir           ||       |    17 |    19 ||       |    15 |    17 ||       |    15 |    17 |
Mana Draught            ||    25 |    51 |    53 ||    24 |    49 |    52 ||    24 |    49 |    51 |
Mana Potion             ||    51 |    84 |    87 ||    50 |    79 |    84 ||    49 |    79 |    82 |
Mana Elixir             ||       |       |       ||       |       |       ||       |       |       |
Spirit Draught          ||     7 |    34 |    38 ||     7 |    35 |    38 ||     8 |    34 |    37 |
Spirit Potion           ||       |    37 |    41 ||       |    34 |    38 ||       |    33 |    36 |
Spirit Elixir           ||       |     4 |     6 ||       |     4 |     5 ||       |     3 |     5 |
Last Elixir             ||       |     1 |     2 ||       |     1 |     1 ||       |     1 |     2 |
----------------------------------------------------------------------------------------------------
Health Gem              ||     1 |    51 |    52 ||     1 |    51 |    52 ||     1 |    49 |    51 |
Mana Gem                ||     1 |    49 |    53 ||     1 |    51 |    53 ||     2 |    48 |    53 |
Spirit Gem              ||       |    54 |    51 ||       |    48 |    53 ||     5 |    46 |    51 |
Mystic Gem              ||       |    51 |    53 ||     2 |    50 |    51 ||     2 |    50 |    51 |
----------------------------------------------------------------------------------------------------
Scroll of Life          ||       |    27 |    28 ||       |    25 |    27 ||       |    22 |    26 |
Scroll of the Avatar    ||     7 |    13 |    13 ||     7 |    13 |    12 ||     6 |    13 |    12 |
Scroll of the Gods      ||     7 |    13 |    13 ||     7 |    13 |    12 ||     6 |    13 |    12 |
----------------------------------------------------------------------------------------------------

The worst equip now I am using is a radiant pants.

If I use Peerless Radiant instead...

CODE

           |  price           |  damage                   |  detail      
-------------------------------------------------------------------------
PRPPH      |  (at least) 100m |  +1.764% magic score      |  32651/32085
DD8        |  (40k Hath) 164m |  +3.571% effective damage |   1.45x/1.4x
DD9        |  (50k Hath) 205m |  +3.448% effective damage |   1.5x/1.45x

DD8 and DD9 seems to be expensive, but it is better than Peerless Radiants as there is no bidwar.

So, you guys should buy Hath, unlock DD series and let me buy Peerless Radiant Phase x of Heimdall cheap (IMG:[invalid] style_emoticons/default/smile.gif)

would've liked to have seen one with no day bonus & infusion, or day bonus & no infusion - 50 turns spent on infusions looks like it could be barely a turn gain there, depending on how strong day bonus is by itself

copied from sssss2: research on effect of DD on holy mage (also, nice insights into the effect of infusionn and of gum+vase
___________________________________________________________________________________________
DD's effect on PFUDORfest

- Setup: holy mage (LHOH, 3 Radiant Phases, 2 Cottons)

- Type A: Day Bonus (holy mitigation -10% on sunday) + Infusion + Bubble-Gum + Flower Vase

CODE

Dæmon Duality           ||   VII  (+40%)         ||  VIII  (+45%)         ||    IX  (+50%)         |
prof_factor             || 0.946 | 0.996 ~ 0.998 ||   1.0                 ||   1.0                 |
Type                    ||     A |     B |     C ||     A |     B |     C ||     A |     B |     C |
----------------------------------------------------------------------------------------------------
Turn                    ||  2370 |  5924 |  6166 ||  2337 |  5736 |  6013 ||  2333 |  5634 |  5896 |
----------------------------------------------------------------------------------------------------

Thanks for sssss2's research. (There should not be cure in Type A test, which would occupy half set of Bubble-Gum + Flower Vase.)

I think for elemential mage, the improve of DD would be more effective in Type A test but less obvious among Type B and C tests.

This research coincide with my unserstand of elemential mage and holy mage. Holy mages would be more competitive among type B and C while less competitive in Type A tests.
(My best Type A test record was 1966T as a lv453 40 forged Mystic cold mage with DD V. It happened on a Friday --- 5% bouns day 10 months ago, )

The data for Type B is awesome, I think sssss2 could easily finish Type B PFUDORfest within 5000 turns if he will. (The limit of cold mage might be around 5500 turns?)

I have benchmarked Dark spells to check their base "multiplicator":
https://ehwiki.org/wiki/Spell_Damage

Now we do have solid (and wiki-tracked) proof that... dark just utterly sucks. Basically, the spells have a higher MP cost, for the same damage output than elemental. Except we also need more prof to reduce to 0% dark mitigation, which then implies we have less EDB available to deal the actual damage.

The Excel sheet also includes further data that shows that Godslayer doesn't have any effect on magic damage, and that Imperil is a multiplicative layer, not additive.

Long story short: why the fuck is dark still more expensive than elements? Only possible answer: because it's cool nonetheless.

This post has been edited by decondelite: Apr 8 2020, 11:32

I have benchmarked Dark spells to check their base "multiplicator":
https://ehwiki.org/wiki/Spell_Damage

Now we do have solid (and wiki-tracked) proof that... dark just utterly sucks. Basically, the spells have a higher MP cost, for the same damage output than elemental. Except we also need more prof to reduce to 0% dark mitigation, which then implies we have less EDB available to deal the actual damage.

The Excel sheet also includes further data that shows that Godslayer doesn't have any effect on magic damage, and that Imperil is a multiplicative layer, not additive.

Long story short: why the fuck is dark still more expensive than elements? Only possible answer: because it's cool nonetheless.

so now those 'in memory of dark mages' monsters make sense, at least

I thought about this a little after reading what was said: I guess the amount of prof factor needed is sort of fuzzy. According to wiki advice, elemental mage needs between 0.7 ~ 0.79 (optimum to reach 0 elemental mitigation across all species) while dark does well at 0.8 (the optimum is still 1.0 to reach 0 elemental mitigation with imperil but there aren't nearly as many species that require it).

If we can fuzzily take 0.79 and 0.8 to be the same, then I guess there is not much prof requirement difference.

The magical spells themselves do equal or negligibly different damage as pointed out, especially considering the higher mana cost for dark/holy. I had sort of asked about this earlier here, but DJNoni explained that dark still does more raw damage than elements. I think(?) the "best staffs" for dark/holy are a bit better than the "best staffs" for elements: that can be the only source for it.

I don't follow mage gear so I didn't know or expect that dark is higher priced than electric, wind, etc. If it is, I suppose the reason could be that it's still considered slightly higher damage with imperil, with the possibility to play non-imperil style. Even then, the higher mana cost might make it less attractive.

If players could fight each other in 1v1 battles, probably dark mages would be the weakest, and holy mages the second weakest, all losing to elemental mages after a battle of mana attrition. Though all mages would be horribly decimated by melee warriors of any type.

If players could fight each other in group battles, mages or certain types of area-of-effect melee (dual wield's frenzied blows, etc) could be useful though.

I thought about this a little after reading what was said: I guess the amount of prof factor needed is sort of fuzzy. According to wiki advice, elemental mage needs between 0.7 ~ 0.79 (optimum to reach 0 elemental mitigation across all species) while dark does well at 0.8 (the optimum is still 1.0 to reach 0 elemental mitigation with imperil but there aren't nearly as many species that require it).

If we can fuzzily take 0.79 and 0.8 to be the same, then I guess there is not much prof requirement difference.

The magical spells themselves do equal or negligibly different damage as pointed out, especially considering the higher mana cost for dark/holy. I had sort of asked about this earlier here, but DJNoni explained that dark still does more raw damage than elements. I think(?) the "best staffs" for dark/holy are a bit better than the "best staffs" for elements: that can be the only source for it.

I don't follow mage gear so I didn't know or expect that dark is higher priced than electric, wind, etc. If it is, I suppose the reason could be that it's still considered slightly higher damage with imperil, with the possibility to play non-imperil style. Even then, the higher mana cost might make it less attractive.

If players could fight each other in 1v1 battles, probably dark mages would be the weakest, and holy mages the second weakest, all losing to elemental mages after a battle of mana attrition. Though all mages would be horribly decimated by melee warriors of any type.

So 'need' is questionable but certainly counter-mitigation makes a huge difference. In fact 3+2 holy/dark are significantly better than 4+1 for average damage all the way up until somewhere around 590 base prof with peerless robe & staff rolls, where there's only 2-3 counter-mitigation difference between the two, despite the EDB gap.

For mages, you must always consider that the speed is dictated by the single hardest monster in the round, and basically nothing else matters, because of the way mage damage is delivered in large areas of effect. This means it's usually worth expecting the worst case scenario - maxed out mitigation and full chaos tokens - almost reliably once per round.

Dark and holy have slightly stronger T2 and T1 spells, while T3 is equal. Decon has correctly noted (on Discord) that 3+2 dark loses more damage to the EDB difference than is gained from these modifiers, however, 4+1 dark does actually end up doing slightly more damage than elemental builds. It just requires absurd proficiency and equipment to do so. Part of the reason for this is, as you noted, Demonic Willow of Destruction being exceptionally good & having very high base dark damage, roughly equivalent to a prefix.

Comparing all peerlesses together, dark users gain about 6 EDB over elec/wind despite using robe instead of shoes as the prof slot, because of this. (Holy is harder to demonstrate because of the different suffix, but their average damage with HOH is actually higher than any other staff in the game even disregarding the holy debuff, albeit not by much.)

However, if we're comparing imperil to imperil, fire and cold should still be using willow staffs for grindfest/IW (works out slightly better on turn counts due to reduced imperil resists). In this case, 3+2 dark is actually roughly comparable to fire/cold 4+1 (both ~386% EDB), except it has marginally higher dark counter-resist than they get with their respective elements due to the prof factor (for what little it's worth - counter-resist does not actually make a huge difference to average damage, at least compared to the community perception of it). Cold debuff IS very, very good though, and may add up to slightly more turns saved in a fest.

You might be surprised about mage mana costs. They are not as high as you might think.