[X] My Assistant

It appears that Shortsword of Slaughter + Rapier of Slaughter was overall better than Rapier of Slaughter + Shortsword of Slaughter by 5072/4992 turns = 1.6% while your adb was only 13386/13267 = 0.9% more.

This is a slightly strange result to me, especially considering that at least for regular attacking, I expected Rapier of Slaughter + Shortsword of Slaughter to perform better. 13267*(1+0.96*0.5) = 19635 > 19343 = 13386*(1+0.89*0.5) so about 1.5% more damaging with the reverse setup.

My only explanation is that it is because you used DwD for testing, which is not a normal arena, and probably most important is how damaging your Frenzy Blows is.

Your two rapiers result could have been worse. Compared to Shortsword of Slaughter + Rapier of Balance there is 12695/12054 = 5.3% adb difference, yet the turns result is roughly comparable 5214/4871 turns = 7%. The bad thing about two rapiers is that you can only inflict one ailment, and if you get a crit then I think your second ailment is wasted. Or would a crit inflict two stacks of penetrated armor?

I think there might be very minor methodology issues regarding your use of 2 Holy + 1 Fire + 1 Elec but you did the best you could with the equipment you own. I think any errors in the results will be very, very tiny. There might also be some errors because only your Balance rapier has overpower, but again practically speaking it's not feasible to do better with limited resources.

Your two rapiers result could have been worse. Compared to Shortsword of Slaughter + Rapier of Balance there is 12695/12054 = 5.3% adb difference, yet the turns result is roughly comparable 5214/4871 turns = 7%. The bad thing about two rapiers is that you can only inflict one ailment, and if you get a crit then I think your second ailment is wasted. Or would a crit inflict two stacks of penetrated armor?

As far as I know, off-hand strike triggers weapon procs separately, so it should be possible to apply 2 stacks of penetrated armor at once... though I'm not sure about the exact implementation, so I'm not very certain about that either way.

...So I checked it, and it really blows big haha. A critical hit using two rapiers will only inflict 1 stack of Penetrated Armor. Only 1 message comes up in the log, and indeed only 1 stack is inflicted.

Even worse, it does not appear to be possible to inflict 2 stacks of Penetrated Armor at once, even after many tries, via luck and getting each rapier's PA chance stat to trigger instead of via a global crit.

I presume, but did not test, that the chances of inflicting 1 stack of Penetrated Armor should at least go up when wielding 2 rapiers, due to both rapiers having a PA chance stat.

Considering how terrible this is, it's remarkable that 2 rapiers performed as decently as they did, roughly comparable to their combined adb. It is a testament to PA being the a very powerful ailment, especially when targeting enemies individually. (Bleeding Wound can function almost as well as PA but only if you spread attack so everyone bleeds simultaneously, which DW usually won't do).

useless test:

CODE

mACC    |mEvaded |mHit   |mEvaded chance
--------|--------|-------|---------------
190.0%  |26      |1823   |1.43%
191.8%  |19      |1952   |0.97%
192.6%  |13      |2023   |0.64%
193.9%  |8       |1998   |0.40%
194.0%  |6       |2007   |0.30%
194.5%  |9       |2062   |0.44%
194.7%  |54      |56000  |0.15%
194.8%  |0       |62963  |0.00%
194.9%  |0       |61436  |0.00%
195.0%  |0       |53842  |0.00%

edit: add more test datas
test samples before 194.7% are small so the chance have no specific meanings, they just means: can be evaded
and the samples after 194.7% are large enough that I think monsters can no longer evade you spells if you have over 194.8% macc.
strange number. wiki says 200%, before the test I thought it would be 195% because of my battle data when I first became mage
maybe someone can test the physical evaded chance as well

useless test #2:
0% cast speed bonus, no haste, cast 36 x Arcane Focus (and other actions are all item use which cost 0 action time), cost total 64 tick time ( from the scroll expire time )
so the Arcane Focus action time should be 64/36 = 1.778 tick, strange number too, but anyway much bigger than the wiki said. The Heartseeker should have similar time cost maybe
edit: tested in Isekai, LV.385, 424 sup.prof. as Nezu said, there's a cast speed bonus from proficiency factor, the AF case speed should be 2x

This post has been edited by what_is_name: Jun 21 2021, 17:32

useless test #2:
so the Arcane Focus action time should be 64/36 = 1.778 tick.

In Isekai, I've noticed that Arcane Focus cost much tick. It's very dangerous in tower rush if I cast AF brainlessly. To make sure the safety, I only cast AF when most mobs have died.

useless test #2:
0% cast speed bonus, no haste, cast 36 x Arcane Focus (and other actions are all item use which cost 0 action time), cost total 64 tick time ( from the scroll expire time )
so the Arcane Focus action time should be 64/36 = 1.778 tick, strange number too, but anyway much bigger than the wiki said. The Heartseeker should have similar time cost maybe

Nah, it's just 2x, there's a cast speed bonus from proficiency factor. Do you have somewhere around 475 supportive prof?

Nah, it's just 2x, there's a cast speed bonus from proficiency factor. Do you have somewhere around 475 supportive prof?

oh that make sense. more or less, it's a Isekai test and I have about 1.1x level supportive prof there

oh that make sense. more or less, it's a Isekai test and I have about 1.1x level supportive prof there

The cast speed bonus formula uses a slightly different factor to the counter-mitigation & counter-resist formulas - instead of being relative to monster level, it's relative to each spell's proficiency floor and ceiling.

For example, Arcane Focus is castable with 0 prof, and the ceiling is thought to be 980 (uncertain) - so the formula is as follows:

CODE

spell_speed_bonus = min(0.25 * ( your_prof - 0 ) / (980 - 0), 0.25)

The cast speed bonus formula uses a slightly different factor to the counter-mitigation & counter-resist formulas - instead of being relative to monster level, it's relative to each spell's proficiency floor and ceiling.

For example, Arcane Focus is castable with 0 prof, and the ceiling is thought to be 980 (uncertain) - so the formula is as follows:

CODE

spell_speed_bonus = min(0.25 * ( your_prof - 0 ) / (980 - 0), 0.25)

good to know that. I'm 424 sup prof when doing the test, so the Arcane Focus action time should be

CODE

2 x (1-min(0.25 * ( 424 - 0 ) / (980 - 0), 0.25)) ≈ 1.784

as my test sample is small, that's close enough

This post has been edited by what_is_name: Jun 6 2021, 04:33

useless test:

CODE

mACC    |mEvaded |mHit   |mEvaded chance
--------|--------|-------|---------------
190.0%  |26      |1823   |1.43%
191.8%  |19      |1952   |0.97%
192.6%  |13      |2023   |0.64%
193.9%  |8       |1998   |0.40%
194.0%  |6       |2007   |0.30%
194.5%  |9       |2062   |0.44%
194.7%  |54      |56000  |0.15%
194.8%  |0       |62963  |0.00%
194.9%  |0       |61436  |0.00%
195.0%  |0       |53842  |0.00%

test samples before 194.7% are small so the chance have no specific meanings, they just means: can be evaded
and the samples after 194.7% are large enough that I think monsters can no longer evade you spells if you have over 194.8% macc.

edit to add more test datas. no any spells evaded by monster in my test when macc over 194.8%

further detailed test between 194.7% and 194.8% macc, calculating macc by

CODE

mAcc = 80 + WIS*0.04 + SUM(Equips_mAcc) + Prof_mAcc

here are the results in Persistent and Isekai:

CODE

Persistent, lv.419
WIS = 981
SUM(Equips_mAcc) = 53.17
ProfMAcc = Cloth.Prof*0.05 + Staff.Prof*0(unslotted)
------------------------------------------------------------------
Clo.Prof |mAcc     |mEvaded |mHit  | mAcc on Character Statistics
---------|---------|--------|------|------------------------------
446.032  |194.7116 |2       |6293  |194.7%
446.243  |194.7221 |3       |1865  |
446.327  |194.7263 |3       |3486  |
446.377  |194.7288 |3       |2677  |
446.431  |194.7315 |2       |1638  |
446.478  |194.7339 |1       |1590  |
446.509  |194.7354 |5       |1302  |
446.539  |194.7369 |0       |1257  |
446.566  |194.7383 |0       |1258  |
446.589  |194.7394 |0       |1232  |
446.606  |194.7403 |0       |1004  |
---------|---------|--------|------|------------------------------
446.656  |194.7428 |0       |6316  |194.7%→194.8%
---------|---------|--------|------|------------------------------
446.866  |194.7533 |0       |3534  |194.8%
446.927  |194.7563 |0       |2677  |
446.978  |194.7589 |0       |1923  |
447.031  |194.7615 |0       |1621  |
447.079  |194.7639 |0       |1704  |
447.124  |194.7662 |0       |1332  |
447.154  |194.7677 |0       |1295  |
447.185  |194.7692 |0       |1373  |
447.211  |194.7705 |0       |1256  |
447.234  |194.7717 |0       |1015  |

CODE

Isekai, lv.389
WIS = 917
SUM(Equips_mAcc) = 49.24
ProfMAcc = Cloth.Prof*0.05 + Staff.Prof*0.02 = C/S.Prof*0.07(Staff.Prof=Cloth.Prof)
------------------------------------------------------------------
C/S.Prof |mAcc     |mEvaded |mHit  | mAcc on Character Statistics
---------|---------|--------|------|------------------------------
411.685  |194.7380 |1       |998   |194.7%
411.693  |194.7385 |2       |1221  |
411.705  |194.7394 |5       |1323  |
411.717  |194.7402 |1       |1232  |
411.731  |194.7412 |6       |1288  |
411.745  |194.7422 |1       |1550  |
411.767  |194.7437 |4       |1604  |
411.789  |194.7452 |1       |1793  |
411.814  |194.7470 |1       |2026  |
411.838  |194.7487 |2       |846   |
411.844  |194.7491 |0       |3700  |
411.870  |194.7509 |0       |3147  |
411.890  |194.7523 |0       |3915  |
411.917  |194.7542 |0       |2391  |
411.939  |194.7557 |0       |4004  |
411.986  |194.7590 |0       |2584  |
412.005  |194.7604 |0       |1221  |
412.016  |194.7611 |0       |1315  |
---------|---------|--------|------|------------------------------
412.029  |194.7620 |0       |1268  |194.7%→194.8%
---------|---------|--------|------|------------------------------
412.042  |194.7629 |0       |1275  |194.8%
412.056  |194.7639 |0       |1602  |
412.078  |194.7655 |0       |1575  |
412.100  |194.7670 |0       |1805  |
412.124  |194.7687 |0       |1975  |
412.148  |194.7704 |0       |2427  |
412.170  |194.7719 |0       |4231  |

seems the calculated macc are not that exactly match the value on character stats
but at least both results show the critical value that monster can evade the spells is before 194.8%
and the result seems not related to other stats beside total macc, at least not that different between my Persistent and Isekai Persona when testing.

well but anyway still useless data, no mages care about macc as everyone have 200%+macc (IMG:[invalid] style_emoticons/default/laugh.gif)

@Basara Nekki, You da MVP (^o^)b
Btw, I use Shortsword+Rapier (Both Slaughter) and found it quicker and easier to clear arena when I use the Shortsword in my Main Hand than when I use Rapier in my Main Hand ((IMG:[invalid] style_emoticons/default/happy.gif)")

Btw, I use Shortsword+Rapier (Both Slaughter) and found it quicker and easier to clear arena when I use the Shortsword in my Main Hand than when I use Rapier in my Main Hand ((IMG:[invalid] style_emoticons/default/happy.gif)")

That's because of your level. When you'll be higher with higher defenses (say, above 350-400) the situation will reverse.

useless test:
...
test samples before 194.7% are small so the chance have no specific meanings, they just means: can be evaded
and the samples after 194.7% are large enough that I think monsters can no longer evade you spells if you have over 194.8% macc.
strange number. wiki says 200%, before the test I thought it would be 195% because of my battle data when I first became mage
maybe someone can test the physical evaded chance as well

since my DW stats happen to perfectly in the range, I test the physical evaded data too

CODE

pAcc    |hit     |evaded    
--------|--------|-------
194.7%  |2606    |2
194.8%  |10850   |0

not detial like the magical evaded test but seem enough to say that
the 194.8% magic number that monster can no longer evaded you attack is the same in physical and magical.

as the evaded chance in test result seems to "suddently" become 0, seems there is a stupid theory can explain it

the effective evade chance resolution is 0.1%, if the effective evade chance lower than 0.1% then it become 0.

threoretically the evade chance for monsters is:

CODE

effective_evade_chance = round(counting_evade_chance)
counting_evade_chance = raw_evade_chance * player_counter_evade = Evasion_level * 0.5% * player_counter_evade
player_counter_evade = max(200% - player_hit_chance, 0)

when the monster have max Evasion chaos forged it has 10% raw_evade_chance

if the hv system use the half rounding method, then to lower than 0.1% effective_evade_chance, the counting_evade_chance need to lower than 0.05%, equals to need 195%+ player_hit_chance

using the 194.7%~194.8% player_hit_chance in test result, the counting_evade_chance is 0.052%~0.053%

so the test result may because:
1) the test samples are too small, they can still evade until you have 195% acc
2) the rounding method used by hv system round 0.052x% to 0

since there is 0 evaded count in over 100k+ hit samples, I would assume the reason 2 is the truth



the counting stats above are incorrect, check stats by BlueWaterSplash here

I realized that you forgot about this monster stats formula:

Evade = 1 - (1 - min(10 , (AGI / 100) , (AGI - Level) / 75) / 100) * (1 - chaos_evasion_rank * 0.5%)

Nearly all monsters have a natural evade of 10%, this combines with the 10% from chaos, to a total of exactly 19% evade.

When the player has 194.7%~194.8% hit chance, this monster's evade is 19% * 0.052 = 0.988% ~ 1.007% so your test is basically correct. This means that when effective evade drops below exactly 1%, it suddenly becomes 0%.

effective_evade_chance = round_down(counting_evade_chance)

...

This post has been edited by what_is_name: Jul 5 2021, 03:49

Rather than it being your reason 2) I think a more likely explanation is that the game rounds twice somewhere, as many other tiny aspects of this game have a similar phenomena.

Anyway, I didn't realize that having 194.8% accuracy is enough to perfectly prevent evasion, thanks! So it's not 200% after all.

This also means that if your goal is to get just enough accuracy as needed and not waste credits on upgrades, then it will be "harder" to prevent yourself to rise above 200% accuracy (now 194.8% accuracy). My level 397 persistent 1H character already has 194.1% accuracy and I guess will probably reach over-accuracy soon.

I have wondered for a long time how the resist-lowering math of the shock shield works. Is it plainly subtractive, or instead working like anti-evade and anti-resist? Maybe you know?

I realized that you forgot about this monster stats formula:

Evade = 1 - (1 - min(10 , (AGI / 100) , (AGI - Level) / 75) / 100) * (1 - chaos_evasion_rank * 0.5%)

Nearly all monsters have a natural evade of 10%, this combines with the 10% from chaos, to a total of exactly 19% evade.

When the player has 194.7%~194.8% hit chance, this monster's evade is 19% * 0.052 = 0.988% ~ 1.007% so your test is basically correct. This means that when effective evade drops below exactly 1%, it suddenly becomes 0%.

effective_evade_chance = round_down(counting_evade_chance)

Note that in reality you experienced effective evade about 0.1% and not 1% around here. This is because most monsters have a natural evade a little below 10% ignoring chaos, because most monsters don't quite have enough AGI.

scaled_stats = int(0.01 * base_stat * monster_level + (monster_level ^ 1.076675) * 0.3325)

AGI is level scaled AGI. In order to reach (AGI / 100) = 10 the monster needs 1000 AGI. At level 500 this means 146.45 base AGI, which every PL 2250 monster will have. Some monsters at a lower level and PL may not reach 10% natural evade.

In order to also reach (AGI - Level) / 75 = 10 is more complicated but at level 500 requires 1250 AGI, or 196.45 base AGI, which a few monster species cannot reach even at PL 2250. But they can still get close to 10% natural evade.

As your player level increases to 500, and as the overall monster population rises in average PL, you will likely experience many more evades around 194.7% but it should still drop to no evades at 194.8%, nice find.

That's because of your level. When you'll be higher with higher defenses (say, above 350-400) the situation will reverse.

Not just reverse, now I do complete overhaul and change my style from DW to 1H (IMG:[invalid] style_emoticons/default/biggrin.gif)
Now I am using Rapier+Kite Shield (don't have any decent Buckler).....
Still wearing Light Armor instead of Heavy Armor, though

This post has been edited by Adhinferno Bloodmoon: Jul 5 2021, 03:08

I realized that you forgot about this monster stats formula:

Evade = 1 - (1 - min(10 , (AGI / 100) , (AGI - Level) / 75) / 100) * (1 - chaos_evasion_rank * 0.5%)

Nearly all monsters have a natural evade of 10%, this combines with the 10% from chaos, to a total of exactly 19% evade.

When the player has 194.7%~194.8% hit chance, this monster's evade is 19% * 0.052 = 0.988% ~ 1.007% so your test is basically correct. This means that when effective evade drops below exactly 1%, it suddenly becomes 0%.

effective_evade_chance = round_down(counting_evade_chance)

...

WoW nice catch, this perfactly explain it, and I wrong count the decimal point too (IMG:[invalid] style_emoticons/default/laugh.gif)

I have wondered for a long time how the resist-lowering math of the shock shield works. Is it plainly subtractive, or instead working like anti-evade and anti-resist? Maybe you know?

No, I have no idea how it work too.

Not just reverse, now I do complete overhaul and change my style from DW to 1H (IMG:[invalid] style_emoticons/default/biggrin.gif)
Now I am using Rapier+Kite Shield (don't have any decent Buckler).....
Still wearing Light Armor instead of Heavy Armor, though

It's Ok, if you don't have Power pieces it's probably better to use Light then Heavy; especially now that you still are below L.250.

At 250 you'll unlock the 1H-block ability, then going Heavy will probably be a better choice even if only some pieces will be Power. imho.

Forge/Upgrade Materials

While developing the script, I found something more concise about this.

[docs.google.com] https://docs.google.com/spreadsheets/d/1gpJ...g#gid=480170546

The material list consists of 13 groups
- 5x [6 Low-Grade Materials]
- 7x [5 Low-Grade Materials] and [1 Mid-Grade Material]
- 8x [4 Low-Grade Materials] and [2 Mid-Grade Materials]
- 7x [3 Low-Grade Materials] and [3 Mid-Grade Materials]
- 8x [2 Low-Grade Materials] and [4 Mid-Grade Materials]
- 7x [1 Low-Grade Material] and [5 Mid-Grade Materials]
- 13x [6 Mid-Grade Materials]
- 7x [5 Mid-Grade Materials] and [1 High-Grade Material]
- 8x [4 Mid-Grade Materials] and [2 High-Grade Materials]
- 7x [3 Mid-Grade Materials] and [3 High-Grade Materials]
- 8x [2 Mid-Grade Materials] and [4 High-Grade Materials]
- 7x [1 Mid-Grade Materials] and [5 High-Grade Materials]
- 58x [6 High-Grade Materials]

The amount required varies based on the equipment's pxp, not the quality in the equipment's name.
- Equipment with pxp 313 or less are from index 1
- Equipment with pxp between 313 and 335 are from index 16
- Equipment with pxp between 335 and 348 from index 31
- Equipment with pxp 348 or higher are from index 51

This post has been edited by sssss2: Jul 28 2021, 00:40

I ended up reforging my rapier and going for B5 O4. I'm pretty sure Overpower is worth considerably more than Fatality, even with Overwhelming Strikes.

Overpower

If average monster parry is 20% then Overpower 5 grants (1 - 0.2 * 0.8) / (1 - 0.2) = 5% more hits. In Research for 1H monsters were stunned 62% of the time, this changes the above calculation to (1 - 0.2 * 0.38 * 0.8) / (1 - 0.2 * 0.38) = 1.645% more hits.

The player also gets Overwhelming Strikes in 1H style, which I'll estimate to provide 20% counter-parry on average without haste, and 60% counter-parry with haste. Stacked additively with Overpower 5 the above calculation becomes (1 - 0.2 * 0.38 * 0.6) / (1 - 0.2 * 0.38 * 0.8) = 1.618% without haste and (1 - 0.2 * 0.38 * 0.2) / (1 - 0.2 * 0.38 * 0.4) = 1.568% with haste.

There are times when Overwhelming Strikes grants 100% counter-parry which is not uncommon with haste. In these instances overpower does nothing. Haste was used when performing Research for 1H which some might consider a flaw. The experimental result was that Overpower 5 increased the hit rate by 1.3075%.

The added overcharge granted by overpower is worth extra, however it's roughly already accounted for. Because even if 1H misses an attack, it will usually perform a counter attack and keep overcharge at a constant level for that turn.

For non-imperil rapier spread style only, added hits from overpower have the valuable effect of preventing previously stacked PA from wearing off. I estimate this effect is worth 0.68 * 0.68 * 0.66 = 0.3 hits, meaning overpower becomes 1.3x better with this viewpoint.

Overpower = Fatality (1H non-imperil rapier spread style only)

Let's suppose the average monster parry rate today is 22%, then Overpower 5 without haste improves hits by (1 - 0.22 * 0.38 * 0.6) / (1 - 0.22 * 0.38 * 0.8) = 1.792% and may improve turns by 1.792% * 1.3 = 2.330% which is equal to Fatality 5. Keep in mind the math is fuzzy and PA wearing off applies less in arenas with small mobs.

For other 1H styles the numbers suggest overpower is somewhere between 0.5x to 1x as beneficial as fatality. Even assuming 25% average monster parry, overpower can't give more than (1 - 0.25 * 0.36 * 0.6) / (1 - 0.25 * 0.36 * 0.8) = 1.940% more hits.

The percentage of time enemies are stunned affects overpower the most. When I wrote that, I assumed that haste usage would not affect how often enemies are stunned. That should theoretically be the case, but because of discretization issues I later discovered the following:

Once any enemy recovers from stun, it no longer matters whether you have haste or not in terms of the number of counter attacks you perform (except for increasing the stun duration from 4 to 6~7 turns, which is the biggest effect on loss of counters).

This is because ~95% of all enemies we encounter are chaosed enough to have over 25% attack speed bonus (more than half chaosed, so 10+ levels to attack speed). Combined with the 25% attack speed bonus of PFUDOR difficulty this means that whether we have haste or not, most monsters will attack 1H players twice every time they recover from Stun.

And if we approximate that 1H style deals 3 counter attacks and 2 stuns every turn, then this creates a perpetual loop. Because in many turns, 2 monsters will recover from stun, and those 2 monsters will attack 4 times, for an average of 3 counter attacks (75% chance) and once again 2 stuns (70%+ chance of stun with a counter attack).

New data on how often enemies are stunned is badly needed, as sssss2 used haste for his test many years ago. Without haste, enemies ought to be stunned much less often, and overpower would drastically improve.

Butcher

Rapier + 5 Power of Slaughter: 5 * 51 / (14400 - 5 * 51) = 1.803%

Shortsword + 5 Power of Slaughter: 5 * 1.2 * 51 / (15000 - 6 * 51) = 2.082%

Butcher is the only weapon potency that improves counter attacks, so calculated derating factors = 2.4 / (2.4 + 0.75 * counters) for Fatality and Overpower are:

CODE

stage         spell usage   difficulty   counters/attack   derating factor
--------------------------------------------------------------------------
4 mob arena    haste          pfudor         0.85              0.74
5 mob arena    haste          pfudor         0.95              0.72
4 mob arena    none           pfudor         1.17              0.67
5 mob arena    none           pfudor         1.30              0.65
item world     haste          pfudor         1.25              0.66
item world     none           pfudor         1.75              0.58
item world     imperil        pfudor         2.33              0.51

Fatality

I'll copy Research for 1H and assume +60% crit damage without fatality, presumably from a savage piece or two.

Fatality gets better at high level because crit chance increases. The peerless full slaughter warrior has 53.2% crit chance. With Heartseeker it becomes 57.88% crit chance to do +75% crit damage. The base multiplier of the main hit is 2.434 which becomes 2.49198 with Fatality 5, an improvement of 2.49198 / 2.434 - 1 = 2.382% (compare with 2.45373 / 2.4 - 1 = 2.239% in Research for 1H).

Taking ~1 counter/attack, the derating factor is 0.70 so Fatality 5 increases overall damage by 2.382% * 0.70 = 1.667%.

Butcher and Fatality have always been pretty close, so O5 B4 could be best. However arenas have bigger average mobs with the current game version, so Butcher is improved these days (a derating factor of 0.51~0.60 may be generally applicable now).