[X] My Assistant

CODE

===============
monster_pl    
---------------
min     |  750
max     | 2250
average | 1328

It's another part of the analysis result.

more than a plain average i guess an average weighted on the number of mobs may be more useful. is it possible to have it?

well , already try 1H without haste & my result is cleartime in PF SG is slower than usual since my style using imperil alot (IMG:[invalid] style_emoticons/default/smile.gif)

in normal arena ( PF T&T as my test ) , I don't feel much different in cleartime speed ~ nearly same for me (IMG:[invalid] style_emoticons/default/huh.gif)


minus without haste is I need to cast regen/heartseeker more often than usual since it running out faster (IMG:[invalid] style_emoticons/default/heh.gif)

Well, I analyzed battle logs and proved from them, but you don't agree and still believe your assumption is right.

I think he has a point, did you analyze turn count also?
Stable +damage could decrease the average turn to kill more than the sudden -20 resist. Because over damage and whatnot. But thats just all talk without data.

nvm

This post has been edited by sssss2: Dec 31 2015, 03:30

It occurred to me that Butcher vs Fatality does not take into account the elemental reduction for the elemental strike. Of course the element and monster will vary, but on average it will be 50% reduction. Therefore normal attack becomes 2.15, Butcher attack becomes 2.20, Fatality attack becomes 2.20373, now Fatality is stronger.

But the elemental strike does not crit

where does that 50% reduction come from, btw?

Wait a sec...this is a 2 year old thread!

This post has been edited by LOL50015: Sep 10 2017, 22:23

Wait a sec...this is a 2 year old thread!

it is. actually i was going to tell him that these results were found by a high-level two patches ago and *may* not apply anymore in general (and to his level in particular), but since we're not sure we'll first discuss about his results. then about necroing (IMG:[invalid] style_emoticons/default/ph34r.gif)

It occurred to me that Butcher vs Fatality does not take into account the elemental reduction for the elemental strike.

What?
He explained in the first post why he weighted both the same.

- An average damage of normal hits is 15,828
- A sum of 3 strikes is 15,237 : it is slightly lower than normal hits' avg, but I'll consider they are almost same.

Ah I see. He assumed the rapier is infused with an additional elemental strike. Makes sense then. It might still be useful to consider the comparison without infusion.

It's true these results aren't hard, they are based on real world data for a particular player and level. I've done some rough estimates for myself, but think his calculations are an excellent reference.

Rather than start a new topic, I felt like putting some more info into this one. I realize most of the focus from sssss2 dealt with picking the element of your rapier and Butcher/Fatality IW choice. It also focused on more "end-game" considerations, and bases most of the research on Pfudorfest.

I'd like to instead introduce a concept that it seems others aren't really paying much heed to: kill speed. Most people are concerned with raising damage, but not much thought is put into why and how damage boosts impact how fast you clear something.

You can read up in the Wiki on how damage is calculated, how action speed lets you get more attacks relative to your opponents, and so on. I'm going to assume you know that stuff, can look it up yourself, or know it better than I do.

Here is some info copied from the start of a round in DWD Hell difficulty.

Spawned Monster G: MID=119799 (Al Kaphrahs) LV=316 HP=37506
Spawned Monster F: MID=128843 (In Memorabilia Of Darksage Ff8) LV=316 HP=50510
Spawned Monster E: MID=163011 (Giant Striped Savannah Slug) LV=316 HP=42676
Spawned Monster D: MID=4905 (Crosis) LV=316 HP=51356
Spawned Monster C: MID=117680 (Chance) LV=316 HP=50510
Spawned Monster B: MID=142393 (Goldnight Castigator) LV=316 HP=20189
Spawned Monster A: MID=23 (Yuki Nagato) LV=316 HP=543341

Same, only I'd leveled up and took the time to note the species of each non-SG monster.

Spawned Monster I: MID=117681 (Luck) LV=317 HP=50894 DGN
Spawned Monster H: MID=20 (Konata) LV=317 HP=321591
Spawned Monster G: MID=86622 (Kimi No Shiranai Monogatari) LV=317 HP=43756 MEC
Spawned Monster F: MID=147618 (Angel Of Deliverance) LV=317 HP=20632 CEL
Spawned Monster E: MID=117158 (Gwyn-lord Of Cinder) LV=317 HP=40110 CEL
Spawned Monster D: MID=23 (Yuki Nagato) LV=317 HP=547831
Spawned Monster C: MID=139694 (Kinoshitamikoto E-mag) LV=317 HP=54752 UND
Spawned Monster B: MID=21 (Mikuru Asahina) LV=317 HP=547831
Spawned Monster A: MID=140524 (In Memorabilia Of Darksage H3) LV=317 HP=56666 UND

Of interest, Yuki, Mikuru, and Ryouko have the same HP. At my level and difficulty, that works out to around 550k
Konata is much frailer at 320k, but still far tankier than player-built monsters which tend to top out at around 55k for the more durable giants, dragons, and undead. More glass cannon monsters like Celestials, Sprites, and humans may have around 20k hp.

Because we're relying almost entirely on void-type damage, there is very little concern for their resistances; it's all 0% resistance to us. We can knock physical mitigation down through penetrated armor (from Rapier's piercing proc) and/or Imperil.

I collected some damage values of some strikes. I'm not trying to give precise numbers here, so I didn't bother to gather hundreds or thousands of data points. We already know from lots of accumulated data roughly how the damage formulas work, so this is more just a snapshot example than an attempt for statistical analysis:

Spirit Stance, No Heartseeker, no penetrated armor:
Void Strike hits Yuki Nagato for 2637 void damage.
You crit Yuki Nagato for 8337 void damage.

No Spirit Stance, Heartseeker, No Penetrated Armor:
Void Strike hits Mikuru Asahina for 1434 void damage.
You crit Mikuru Asahina for 4809 void damage.

No Spirit Stance, Heartseeker, Penetrated Armor:
Void Strike hits Mikuru Asahina for 3610 void damage.
You hit Mikuru Asahina for 6595 void damage.
You crit Mikuru Asahina for 13275 void damage.
You counter Mikuru Asahina for 4606 points of void damage.

Spirit Stance, Heartseeker, no penetrated armor:
Void Strike hits Yuki Nagato for 2993 void damage.
You crit Yuki Nagato for 9616 void damage.
Vital Strike hits Yuki Nagato for 142374 void damage
Bleeding Wound hits Yuki Nagato for 13996 damage.
Orbital Friendship Cannon hits Yuki Nagato for 126059 void damage

Spirit Stance, Heartseeker, no penetrated armor, Imperil:
Void Strike hits Yuki Nagato for 6176 void damage.
You hit Yuki Nagato for 11453 void damage.
You counter Yuki Nagato for 9320 points of void damage.

Spirit Stance, Heartseeker and penetrated armor:
Void Strike hits Yuki Nagato for 5683 void damage.
You hit Yuki Nagato for 11921 void damage.
You crit Yuki Nagato for 23589 void damage.
You counter Yuki Nagato for 11505 points of void damage.
Vital Strike crits Yuki Nagato for 279914 void damage
Bleeding Wound hits Yuki Nagato for 13996 damage.
Orbital Friendship Cannon hits Yuki Nagato for 305419 void damage

Spirit Stance, Heartseeker, Imperil, and penetrated armor
Void Strike hits Yuki Nagato for 7501 void damage.
You hit Yuki Nagato for 13962 void damage.
You crit Yuki Nagato for 23340 void damage.
You counter Yuki Nagato for 15893 points of void damage.
Vital Strike hits Yuki Nagato for 261331 void damage
Vital Strike crits Yuki Nagato for 414837 void damage
Bleeding Wound hits Yuki Nagato for 13996 damage.
Bleeding Wound hits Yuki Nagato for 16096 damage.(more stacks?)
Orbital Friendship Cannon crits Yuki Nagato for 510916 void damage

Spirit Stance, Heartseeker, penetrated armor, imperil
Scanning Flying Spaghetti Monster...
HP: 2044970 / 2448023
Bleeding Wound hits Flying Spaghetti Monster for 16066 damage.
You counter Flying Spaghetti Monster for 7577 points of void damage.
Void Strike hits Flying Spaghetti Monster for 3899 void damage.
You hit Flying Spaghetti Monster for 6620 void damage.
You crit Flying Spaghetti Monster for 15228 void damage.
Vital Strike hits Flying Spaghetti Monster for 297157 void damage

Keep in mind that, at least when I'm playing, it is fairly trivial to keep Spirit Stance up 24/7, and Heartseeker is kept up nearly always just off of channeling procs from casting regen and the occasional cure. If I decided to keep it up actually 24/7, it would not really cost much more in the way of mana draughts per run, and easily affordable. I included a few categories of damage without Heartseeker and/or Spirit Stance just because opportunities arose to do so. They help give a more thorough picture of the damage output I am experiencing at the moment.

Most interesting though are the normal strikes, critical strikes, void strikes, counters, and bleed damage.

I've also noticed that stacking both penetrated armor and Imperil doesn't seem to significantly increase the damage output.

Spirit Stance, Heartseeker, no penetrated armor:
Void Strike hits Yuki Nagato for 2993 void damage.
You crit Yuki Nagato for 9616 void damage.
Vital Strike hits Yuki Nagato for 142374 void damage
Bleeding Wound hits Yuki Nagato for 13996 damage.
Orbital Friendship Cannon hits Yuki Nagato for 126059 void damage

Critical hit is 9.6k, and void strike is 3k, for 12.6 total. Adding another elemental strike with an infusion would probably put that to about 13k total. I have +75% crit damage (Including Heartseeker). So a normal strike would be around 5500+3k Void Strike for 8500. Approximately 8800 with infusion. Give or take 1k or so here or there due to Overpowering Strikes stacks, higher damage roll, or the lack of those fortunes. 8-9k with these numbers, 7-10k is not unreasonable.

Unfortunately I didn't get a normal strike or a counter without penetrated armor, since I tend to give that out like candy.

Spirit Stance, Heartseeker and penetrated armor:
Void Strike hits Yuki Nagato for 5683 void damage.
You hit Yuki Nagato for 11921 void damage.
You crit Yuki Nagato for 23589 void damage.
You counter Yuki Nagato for 11505 points of void damage.
Vital Strike crits Yuki Nagato for 279914 void damage
Bleeding Wound hits Yuki Nagato for 13996 damage.
Orbital Friendship Cannon hits Yuki Nagato for 305419 void damage

Almost 12k normal hit, with a 5,600 void strike for 17.6k. Crit for 23.6k, 29.2k with void strike. So we're looking at almost, but not quite double the damage done without penetrated armor. So this is a pretty substantial difference in damage output.

If we put that into perspective that I'm interested in discussing though, take Yuki's 550k hp. She takes 61 normal strikes to kill without penetrated armor. 31 with. The usual advice to focus fire and maintain penetrated armor in this case results in nearly halving the number of swings it takes to kill Yuki. At my usual rate of 2.5t/s, that translates to killing yuki in 24.4 seconds vs 12.4 seconds, shaving off 10 seconds every time I kill Yuki/Mikuru/Ryouko. With up to three of them in a round from round 113 on in DWD, that can save 30 seconds per round; shaving off 18.5 minutes for that part of the arena alone. Obviously some of the time Konata will be one of the three SGs spawned, and the savings in time for this choice would be slightly smaller in those instances. I forget the exact rounds 2 SG's show up, but from that round to round 112 you'd save about 20 seconds, and so on. Overall, it probably does make the difference between about a half an hour saved just using penetrated armor.

I included Orbital Friendship Cannon and Vital Strike information, with the latter being a handy way to get bleed damage to include also.

OFC ends up not actually being too useful to me. The reason is that I focus fire on an SG, capitalize on penetrated armor, and by the time she is dead after those 31 turns, all the other monsters are nearly dead from counters. If there are two or more SG's, killing one then the other (62 turns) tends to result in all the non-SG monsters dying off solely from counters (and spike damage). Therefore OFC winds up being just a heavy-hit against One or more SGs, not the "clear the entire round" it's supposed to be. It winds up being the equivelent of about 17 strikes in one button push, but then it creates substantial issues with Overcharge and often forces me out of Spirit Stance for a short while, and wastes me a turn toggling it back on. If I stay in Spirit Stance, it usually takes a full two more rounds to replenish OC high enough to use it again, rendering it useful only about 1/3rd of the rounds in total. It might cut the total turncount down, but at the cost of more maintenance which takes seconds. All the damage it does to squishier player monsters doesn't contribute to kill speed.

So some routine like, "see three SGs in round. Get Imperil or Pentrated armor on all three. Cast OFC," could be done roughly ever three rounds. It could therefore be done 12 times per DWD. The damage output would be roughly 17*3=47t worth of strikes, or 18.8 seconds per cast, less the setup time spent getting ready for it and the OC/Spirit Stance issues afterwards. I'll estimate 3-5 seconds, maybe less with practice. So, -3 seconds or 15.8, we'll round to 16 seconds saved 12 times for 3 minutes, 12 seconds saved. Again we can try this out with the 2SG portion of the arena also, and overall maybe shave off 5-ish minutes total. Much less impactful than the choice to emphasize penetrated armor.

Taking this logic a bit further, I'd like to hypothesize a bit. This build already favors low action speed, granting opponents as many turns as possible relative to the number taken by me. Bleeding Wound does pretty consistently 13996 (14k) damage, aside from one round in which it seems Vital Strike got more stacks into place or something and it did more for that round. Yuki/etc. have 550k hp, weapon descriptions tend to list Bleeding Wound as doing 20% (which would be 110k damage, and kill in 5 rounds), but the observed 14k is closer to 3%. It isn't affected by Spirit Stance, Heartseeker, Penetrated Armor, or Imperil.

I'll have to gather some data to support it, but I suspect normal, void, crit, and counter damage will do similar damage with Imperil and no penetrated armor as it does with penetrated armor and no Imperil. Further, using Imperil instead of penetrated armor would let an elemental strike do more significant damage and become maybe on par with void strikes. The value of this configuration would be using a slashing weapon, like a shortsword. You could keep the damage output of penetrated armor, with slight slowdown due to casting imperil regularly. You could keep the attractive parry stat due to shortswords having reasonably good parry stats. And you could get bleeding wound procs for what amounts to a free normal strike per bleeding enemy turn.

The other variant configuration would be a build that sacrifices parry stat to use an axe for greater normal/crit/void strikes, bleeding wound, and reduced enemy mitigation through Imperil to make up for the loss of penetrated armor. Axes also tend to crit more often on top of having greater ADB. However, the loss of counters due to parry would probably outweigh that increased strike damage, not to mention time spent recovering from the increase in received damage.

While this is done in Hell difficulty, I suspect that raising difficulty won't actually change things. Defenses might rise, but are impacted by penetrated armor/imperil roughly equally anyway. And bleeding wound is based off the monsters' hp, so damage from it would rise too. If anything, bleeding wound favors higher difficulty, as the player's strike damage rises from player progress, while bleeding wound damage rises from stronger/faster monsters.

This post has been edited by Cryosite: Nov 14 2017, 10:49

-

Nice work. Could you perhaps add some context at the start, and then some headers to help readability?

This post has been edited by DJNoni: Nov 14 2017, 15:40

OFC ends up not actually being too useful to me. The reason is that I focus fire on an SG, capitalize on penetrated armor, and by the time she is dead after those 31 turns, all the other monsters are nearly dead from counters.

What I am doing is
1. Using Imperil on every SG.
2. Casting OFC (almost at the beginning of the round) - this makes all regular enemies die at single blow and SG stay half damaged.
3. Then I am attacking SG with my regular non-spiritStance attacks (if spiritStance hasn't depleted - that even better).
4. Few rounds later my OC is regenerated enough to make me activate spiritStance back (if it had been depleted).
5. Then I just killing impireled SG with my spiritStance attacks as usual.
6. At new round, if my OFC has been recharged - I am starting from step #1, if not just cast Imperial and attack enemies as usual.
It is true that using OFC at the middle of the round is a waste - enemies could die easily from counters, but casting it at the beginning of the round is not that bad.
On other hand, I am so used to OFC that cannot imagine playing without it. Maybe if I will give it up and instead would go with spiritStance 24/7 - I will have better clear speed? (IMG:[invalid] style_emoticons/default/unsure.gif)
Also, your research is true for Hellfest. Maybe you will find OFC more usefull on PFUDOR?

weapon descriptions tend to list Bleeding Wound as doing 20% (which would be 110k damage, and kill in 5 rounds), but the observed 14k is closer to 3%. It isn't affected by Spirit Stance, Heartseeker, Penetrated Armor, or Imperil.

According to the wiki, formula for damage looks like this:

CODE

Damage done = Bleeding % * base damage * Stack * 0.4

, so if take Axe with the following properties:

Bleeding Wound: 19.5% chance - 7 turns / 20% DOT

And your base damage is

9,000

The single "Bleeding Wound damage" should be as I thought:

CODE

0.20 * 9,000 * {1} * 0.4 = 720 HP
0.20 * 9,000 * {2} * 0.4 = 1,440 HP
0.20 * 9,000 * {3} * 0.4 = 2,160 HP
0.20 * 9,000 * {4} * 0.4 = 2,880 HP
0.20 * 9,000 * {5} * 0.4 = 3,600 HP

, how you managed to get the 14,000 - I don't know (IMG:[invalid] style_emoticons/default/unsure.gif) Probably I understood the formula wrong...

I'll have to gather some data to support it, but I suspect normal, void, crit, and counter damage will do similar damage with Imperil and no penetrated armor as it does with penetrated armor and no Imperil.

Probably, but Penetrated Armor is stacking on enemies automatically while you attacking them, while with Imperil - you have to cast it manually, recharge, drink Mana Potions to refresh your MP, etc.
Even if "no-Imperil yes-Penetrated-Armor" and "yes-Imperil no-Penetrated-Armor" combinations would deal same the damage to enemies - the former one is way easier to "maintain", if you will.
And Penetrated Armor will easily and freely land on regular monsters too, while casting Imperil on them is a waste of MP because they are easy to die anyway (IMG:[invalid] style_emoticons/default/sad.gif)

Further, using Imperil instead of penetrated armor would let an elemental strike do more significant damage and become maybe on par with void strikes.

Imperil, speaking of which, has a good bonus - you can unlock two abilities, named "Holy Imperil" & "Dark Imperil" which make Imperil to reduce enemy's mitigation against Holy and Dark. So with holy/dark weapon + respective dark/holy infusion you can greatly increase the amount of damage.

The other variant configuration would be a build that sacrifices parry stat to use an axe for greater normal/crit/void strikes, bleeding wound, and reduced enemy mitigation through Imperil to make up for the loss of penetrated armor. Axes also tend to crit more often on top of having greater ADB. However, the loss of counters due to parry would probably outweigh that increased strike damage, not to mention time spent recovering from the increase in received damage.

We had a discussion that over-forged axe in hands of top-level (450+ level) player could let him finish rounds faster, but they were just a discussion, no proof of concepts or data.
The arguments were the same you provided: ADB + Crit from axe > ADB + Crit from rapier.
But once again, you need very good defense to refrain from received damage (you have no Parry from rapier) and you need very high block from shield (like in 95%-100% range, heavily-forged).


Good observation. And thank you for sharing your datas with us (IMG:[invalid] style_emoticons/default/smile.gif)

Axes also tend to crit more often on top of having greater ADB.

Since when do axes crit more often?

Also how can anyone lose 3-5 seconds when casting a single OFC?
Isn't that too much?
Even considering the cases where you get out of spirit stance I cannot see anyone "wasting" more than 2 seconds.

Cryosite, even if you don't want to do some data analyzis (which I understand very well), you cannot affort getting only one value for each situation. There is quite a lot of randomness in the final damage output, so much that in the exact same combat conditions, there can be a difference as wide as around 15-25% of the average, ultimately leading to a maximum damage registered 30-35% stronger than the minimum registered! Well actually I'm getting these percentages out of my hat, but you get the idea. Sometimes it can lead to cases where the max damage registered in one situation is quite close to the minimum of another situation, leading to erroneous judgements.

Even if you just want a "wild guess", I'd recommend you to have at least 5-10 hits in each situation to have a fairly reliable average. You can't possibly have even a "rough picture" without at least that very minimum.

For the PMit stacking (Penetrated Armor and Imperil), it doesn't work all that well because by essence Penetrated Armor, Imperil and the monster PMit are all additive, and the sum of all of them cannot go below 0%. In other words, if a monster is Imperil'ed, there is already a very fair chance that penetrated armor will have absolutely no effect. Just take the monster with the highest PMit in the entire game: a Lvl500 PL2250 Arthropod monster. It gets this as PMit:

Arthropod Lvl500 PL2250
Base End: 100+25*10 = 350
Base Agi: 70+25*7 = 245
scaled_stats = int(0.01 * base_stat * monster_level + (monster_level ^ 1.076675) * 0.3325)
Scaled End: 1750+267 = 2017
Scaled Agi: 1225+267 = 1492
Physical Mitigation = 1 - (900 / (900 + END + AGI / 2)) * (1 - chaos_defense_rank * 1%)
PMit with no chaos upgrade: 1 - (900/3708)*1 = 75%
PMit with full chaos upgrade: 1 - (900/3708)*0.8 = 80%

In the case of no chaos upgrade, just the 3 stacks of Penetrated Armor completely negate its physical mitigation! OK great, it effectively multiplies your damage output against it by no less than 4! And even slightly more in the case of the fully chaosed one.
=> This is where the bullshit of "3 stacks of penetrated armor multiplies your rapier damage by 4" comes from. It is mostly a biased and erroneous extrapolation.

Buuuuut, as I said, this is complete bullshit because it is only an ideal case that has only one representative!
[hvlist.niblseed.com] Hello Kodou Yuki

In other words, your average monster is (far) below that holy 80% PMit. As for Schoolgirls... either their base stats written there are complete bullshit, or they really are pathetic and rely solely on their level for their scaled stats:
https://ehwiki.org/wiki/HentaiVerse_Bestiary#Stats

That is the reason why, with my God damn shortsword, I am able to go below 30 mins to clear PFUDOR DwD: in the end my Imperil'ed schoolgirls have just the same (inexistant) PMit than people with their rapiers. Except that they do have non-existant PMit as soon as I cast Imperil and not after 1-2 stacks of PA that have less than 25% chance of occuring each. Except that I have a slightly higher attack base damage and a proc that deals the entirety of its effect, which results in SGs getting more damage from my direct attacks, getting bleeding wound damage, and my counterattacks against the other monsters being slightly more powerful.

Quite a lot of advantages for a shortsword compared to a rapier in SG arenas, don't you think?

But don't get me wrong, I am not saying that rapiers are shit, we're even far from it. I'm mostly saying that rapiers aren't a magic thing that are just the best overall. They do are extremely effective for their first stack of Penetrated Armor on all monsters and they may allow you not to use Imperil. That may allow them not to use Imperil in the SG arenas, though in the end it will still be slower than with a shortsword, an axe or a club.

As for regular gameplay, yes rapiers may be the best mostly because you (normally) don't fool around with Imperil, which is to their advantage against tanky monsters like arthropods and giants. Though in the end the difference in clear time is not that great, especially when you're not Lvl 500.
Keep in mind that a rapier will hit (in average) a tank 4-5 times before dealing the first PA proc: you only got 20-25% chance to proc PA, never forget that. In the end it still makes 4-5 turns during which one would have dealt slightly more damage with a club, a shortsword or an axe. Including against the other monsters as counterattack damage, mind you.
But yes, once you've procced PA once, you already compensated after the second hit under PA, the first one basically "catching up" with what you would have dealt with other weapons. It other words, you finally started being further in the monster's HP bar after the 6-7th hit than what you'd have been with other weapons. Except that usually after 6-7 hits a tank is usually almost dead.

So when are rapiers really better than the rest? Basically when you have the luck of proccing PA against a tank in normal gameplay at the 1st, 2nd or 3rd hit against it, which happens less than half of the time.

Long story short: rapiers aren't as great and magic as people believe they are. They do are good, even really handy against tanks when you do proc PA on the first hit.

The only thing I'm not sure about, is the efficiency against the SGs: we need to proceed to extensive tests to see if their base stats really are shit like what's written in the wiki, or if they actually are (way) higher, which in the end would definitely make rapiers way better than the rest against them.



Now I'd like to talk about blasting OFC in PFUDOR DwD. You did bring up a good point: it doesn't make that much of a positive improvement. It merely allows one to spare barely 5-10 seconds in a round with 3 SGs, given that he doesn't lose time due to repeatedly failing to imperil one of them (I so God hate when I need to cast imperil 5 times in a row against the same SG). And yes you do often have the issue of having to enable spirit stance again.
But even these issues aside, you're really stripping off more than half of the SG's HP bar at once, 3/4 of it if it's Konata (or even insta kill her if you critted). That alone represents several seconds per SG, which is the source of the small gain in clear time.

What is the global effect in the total clear time of that arena? I don't know, and frankly I don't want to try and notice I'm needing 3-5-7 more minutes when I don't blast OFC. All I need to know is that I do am a bit faster, given that I am properly "trained" to do that smoothly. Because yes, being used to use that tactic is necessary to be efficient. You lose more time than you win if you don't master that technique.

tl;dr
Rapiers are not as magic and almigty as people think they are.

EDIT:
No, axes don't crit more than the rest. I'd even go so far as to say that their higher Burden has a slight negative impact regarding that aspect.

This post has been edited by decondelite: Nov 14 2017, 14:40

What I am doing is
1. Using Imperil on every SG.

...

It is true that using OFC at the middle of the round is a waste - enemies could die easily from counters, but casting it at the beginning of the round is not that bad.

I tried that style for a few clears. It took about 200 turns off the end clear turn count of around 5,000 turns for DWD Hell. Very tiny end difference on low difficulty.

It doesn't matter if you cast OFC at the start of a round on Hell difficulty. If the enemies all die to counter attacks, the only difference in speed it makes is in how fast the SGs die. The trash all dies before the SGs do anyway, and both methods of OFC at start and "Stab the SG and ignore the trash" result in the trash dying at zero extra effort/turns.

On other hand, I am so used to OFC that cannot imagine playing without it. Maybe if I will give it up and instead would go with spiritStance 24/7 - I will have better clear speed? (IMG:[invalid] style_emoticons/default/unsure.gif)
Also, your research is true for Hellfest. Maybe you will find OFC more usefull on PFUDOR?

That is part of my unsure/theory part. Counter attacks are based on player stats, and it will take more counters to kill Pfudor trash than it will take to kill Hell trash.

According to the wiki, formula for damage looks like this:

...

, how you managed to get the 14,000 - I don't know (IMG:[invalid] style_emoticons/default/unsure.gif) Probably I understood the formula wrong...

The source of bleeding wound in my numbers above is from Vital Strike skill upon an SG that has stun. It might be a much stronger bleed than weapon proc. That said, bleeding damage very clearly does damage to things based off their HP though. Back when I was using an axe still, I remember bleeding wounds doing much more against SGs than against trash.

Looks like perhaps using Vital Strike's bleed to get damage information might be misleading in that case. I'll have to test it out with a shortsword. I have one that was IW10'd, but it has the absolute worst potencies possible for 1h. I guess I'll spend 5 amnesia shards, get it reset by someone with Dark Descent, and get it back to IW10, see if it can get some Butcher/Fatal instead of the garbage Overpower and Swift Strikes. It'll take awhile to get the sword set up, but much cheaper in the long run for me to do it than for and end-game player, due to much cheaper gear I'm using to compare against.

Alas, once again though, we get back to the stage where the stumbling block to testing and experimenting is "not good enough gear to compare against our rapiers."

Probably, but Penetrated Armor is stacking on enemies automatically while you attacking them, while with Imperil - you have to cast it manually, recharge, drink Mana Potions to refresh your MP, etc.
Even if "no-Imperil yes-Penetrated-Armor" and "yes-Imperil no-Penetrated-Armor" combinations would deal same the damage to enemies - the former one is way easier to "maintain", if you will.
And Penetrated Armor will easily and freely land on regular monsters too, while casting Imperil on them is a waste of MP because they are easy to die anyway (IMG:[invalid] style_emoticons/default/sad.gif)

That's part of the theory. The same can be said of "Imperil start of round, OFC, sometimes lose Spirit Stance for a few turns, toggle it back on." You have to do stuff other than smash the enemies' faces to maintain that strategy. So it may reduce the turn count at the end, but it makes your run take longer to pull it off. As long as it saves you more time than it costs to do it, it's a good strategy.

The possible tradeoff in this case is: use imperil (slows you down) but get bleeding wound from slashing weapon (speeds you up). If the slashing weapon gives more ATB and Crit% than a rapier, then that speeds you up too.

Test needed: how much slow down, how much speed up.

Imperil, speaking of which, has a good bonus - you can unlock two abilities, named "Holy Imperil" & "Dark Imperil" which make Imperil to reduce enemy's mitigation against Holy and Dark. So with holy/dark weapon + respective dark/holy infusion you can greatly increase the amount of damage.

Also, SG's tend to have -1% resistance to Holy and/or Dark. Would like to see some hard numbers on how much it increases damage output though. There are some additional factors to consider related to "killspeed."

On DWD early rounds with no SG or 1 SG, lots of trash is still alive when the SG dies. They have small amounts of damage from counters, and some trash may die already. But I have to kill them myself, spending time. Yuki/etc. take me 31-ish turns to kill, with penetrated armor. So in 31-ish turns, not all trash dies from Counters. Killing off that trash takes some turns.

Two SG's (Yuki/Miku, Yuki/Ryouko, or Miku/Ryouko for highest HP) take about 62 turns. In 62 turns most or all trash dies. Sometimes I have to hit one or two monsters one or two times. So the round takes 62-66 turns.

Three SGs (Yuki/Miku/Ryouko for highest HP) takes about 93 turns. Always all trash is dead to counters.

Speeding up how fast the SG dies by a lot means trash is leftover. They don't get enough attacks for me to parry/block and counter them with. If I reduce the time it takes to kill Yuki from 31t to 20t, that means I'll only have 40t with two, and 60t with three SGs. Some trash-stabbing will reduce some of those gains.

Still some speed up, with small slow down.

We had a discussion that over-forged axe in hands of top-level (450+ level) player could let him finish rounds faster, but they were just a discussion, no proof of concepts or data.
The arguments were the same you provided: ADB + Crit from axe > ADB + Crit from rapier.
But once again, you need very good defense to refrain from received damage (you have no Parry from rapier) and you need very high block from shield (like in 95%-100% range, heavily-forged).
Good observation. And thank you for sharing your datas with us (IMG:[invalid] style_emoticons/default/smile.gif)

I think for a lower level player on lower difficulty setting, like I'm testing in Hell, Axe might be worth using. A player might not have power armor, and might have plate. If they can get their hands on a good block shield, their plate armor and regen spell might make them safe enough without the parry. I got by with an axe for awhile before I got a good rapier.

Yes, Pfudor is very dangerous. Need very high defenses and probably still spend a lot of potions. But not everyone is end-game. I think this may be another place where a low-level or medium-level player can make very good use of different strategies than high-level players.

So, to get more information, I'm going to raise difficulty a little bit on tomorrows Arena clears to do more test with and without OFC.

I'll also start to work on my shortsword. If I get it to a reasonably useful IW10+forge5, it will be roughly on par with my rapier and be a fair comparison for clear speed. It doesn't need to be an ideal IW10, but needs much less Swift Strikes.

*ehm* i see there's a lot to discuss. and that's fine, really. only a few random suggestions:

• long posts aren't bad di per se. if there's enough to speak about, well... cannot be helped. but organizing them, saying, in lists is never a bad idea. it'll help people to follow your speech better. same going for a visual arrangement for datas (i'm thinking about tables, but whatever can allow just to break the rhythm a bit is good). long logs can be attached to a txt files, while picking only the most significant lines on the main post
• for the same reason, i invite all the subsequent discussions to be introduced by a quote with the exact reference
• again for the same reason, a tl;dr can always attract more audience
• if what decon says it's true, then a bit more datas are needed. especially for things that may or may not be difficulty-dependant and when many variables are involved (difficulty being one of them, for example)
• i partially agree with sapo. i don't see any motivation about why axes crit more often. how comes? they don't have a high crit to begin with (letting apart a Balance suffix), and even their DEX roll is nothing spectacular. if any, i expect for them to crit the least, at least according to range page on wiki
• on the other hand (and not explicitly meant to him), i'll never support the "i-can-cast-this-in-1-second-less-than-you" or the "do-you-even-need-x-seconds-to-cast-this" e-peenis contests. if we're speaking about shaving, saying, 5 minutes out of half a hour it's a thing (which is still nothing in the matter of a day and heavily dependant on your connection, but still a 17% improvement), but if we're speaking about a bunch of seconds out of the same half hours, well...

this being said, i thank all the people who will take a bit of their time to retrieve some datas and (politely) discuss them (IMG:[invalid] style_emoticons/default/smile.gif)

Since when do axes crit more often?

Maybe I'm mistaken? I was under the impression that axes tended to have higher base crit% than other weapons.

Wakizashi of Nimble
Club of Nimble Base
Shortsword of Slaughter
Rapier of Illithid
Axe of Slaughter
All Base crit% 5.73

I stand corrected.

Also how can anyone lose 3-5 seconds when casting a single OFC?
Isn't that too much?
Even considering the cases where you get out of spirit stance I cannot see anyone "wasting" more than 2 seconds.

Even if it took zero seconds, that only means about 30 seconds faster for a DWD run. The estimation, like all the estimations above, are not intended for accuracy, but to lay out the concept.

Also, some people are not terribly dextrous. I suck at "twitchy" games like First Person Shooters, Real Time Strategy, and 2d Fighters (like Street Fighter/etc.). Someone much better with their hands than me would of course be much faster.

The point is not "this is faster, therefore do it." It's an attempt to actually build up the concept of "ok, some things are big improvements, some small. Then we can work on gathering data and improving the accuracy of the numbers.

...

Even if you just want a "wild guess", I'd recommend you to have at least 5-10 hits in each situation to have a fairly reliable average. You can't possibly have even a "rough picture" without at least that very minimum.

There is a very low precision to the data I presented above. It's understood it is very low precious. I would not suggest basing any major decisions off of the data I provided above.

It is useful for a "wild guess" though. Most of those numbers I didn't write down, I just watched my battle log for several days, several arenas, several levels. I copied a few out of my battle log today to give a representative number to show what killing things in the arena looks like for me, which is obviously going to be much different than an end-game player.

No, this is not a huge pile of data that can show the random variance of damage values. Again, others have done that and the wiki has the formulas. But you can clearly see that my void strikes are behaving predictably under different buff and debuff situations. You can see my normal strikes behave as expected. You can see how Penetrated armor and Imperil both still seem to knock the SG's mitigations down to 0 as you describe below. You can see my crits are indeed about what you expect.

The bleed damage is also 100% stable. Except for that one odd one I noticed that did more, it consistently did exactly the same amount many times. Not just with different buffs and debuffs as I recorded, but many bleeds I didn't write down.

Again, I'm aware of how imprecise that data is, and I think it's useful for the topic. More precise data is needed for later on down the road. This stage helps identify which data is most important to that.

and the sum of all of them cannot go below 0%

Agreed. As I noted, Penetrated Armor and Imperil get close to, if not exactly perfectly the same effect: opponent gets little to no defense against strikes, letting you do your full damage. Piling both onto the same opponent rarely, if ever, actually matters.

This is something true down at my level at Hell difficulty, and seems to match what you guys up in Pfudor also experience.

I've noted in my data that having defenseless opponents seems to pretty consistently translate to double damage compared to fully defended opponents. This may be worse, as you noted, with very high defense monsters which may quarter the incoming damage, thus resulting in 4x damage. Silly arthropods.

Except that they do have non-existant PMit as soon as I cast Imperil and not after 1-2 stacks of PA that have less than 25% chance of occuring each.

That's another thing I noticed, but didn't bother to include in my above data: penetrated armor often takes a few swings to apply. I've even noticed on rare occasions where the 5 turns of duration wear off and the SG gets to enjoy her defenses again for a few swings, meaning I hit a bad run of 6-8 swings that failed to proc somehow.

On the other hand, if you crit, according to the wiki, you get a free proc. So 25% chance to proc, combined multiplicatively with your crit% chance. I've noticed that higher-end players seem to have around 60% crit chance, which works out to 70% chance to land penetrated armor per main strike.

Quite a lot of advantages for a shortsword compared to a rapier in SG arenas, don't you think?

It does sound like it to me. Would you mind linking your shortsword?

So when are rapiers really better than the rest?

I'd say they're good against SG's. Three of the four have ten times as much HP as the tankier player-monsters in my experience, and even Konata still has 5-6 times as much.

I'm interested to hear how you fight in DWD with your shortsword, please. I described above my strategy. I hover/spam click over the SGs, letting my counter attacks kill the trash. If any trash is alive after SGs are all dead, I sweep my cursor over them to finish them off. Only reason I ever stop that is to replenish regen and heartseeker, and pop draughts as needed. Very rarely I need to cure.

What does your typical round look like against three SGs, two SGs if that's different, and so on?

The only thing I'm not sure about, is the efficiency against the SGs: we need to proceed to extensive tests to see if their base stats really are shit like what's written in the wiki, or if they actually are (way) higher, which in the end would definitely make rapiers way better than the rest against them.

What extensive testing do you mean? I demonstrated above that Imperil+Penetrated armor seems to not reduce their defenses any further than just penetrated armor or just Imperil. So their pmit is not high enough that either one fails to reduce it to zero.

You can see their HP at any time at the start of a round or using Scan, and see their defenses using scan. You can see how hard they hit you if you fail to parry/block/evade.

Now I'd like to talk about blasting OFC in PFUDOR DwD. You did bring up a good point: it doesn't make that much of a positive improvement.

...

All I need to know is that I do am a bit faster, given that I am properly "trained" to do that smoothly. Because yes, being used to use that tactic is necessary to be efficient. You lose more time than you win if you don't master that technique.

Yeah, that's the basic jist of what I was going for. Every choice has some amount of speed up and some amount of slow down to it.

Even if you're able to do a technique smoothly, it is still slightly slower than doing no technique at all. You really can't beat the speed of just hammering away at your opponents' faces with something sharp and stabby/slashy. But those techniques give "bursts of speed" that let you skip by some stuff. I'm trying to focus on just how much savings that really is.

In the OFC example, you and I, and F4tal all noticed and commented: the speed gain is not from killing the trash really, because that stuff would all die for free anyway. So the HP of those trash don't really count, since they don't actually slow you down any. OFC's damage that does count for speeding you up is only the damage it does to the SG's. Yes, it hits them really hard. We can calculate how fast that is.

Some other choices like going with shortsword instead of rapier need testing. You seem to have experience with it, so I'm really interested in your data.

How much HP does Yuki/Mikuru/Ryouko have at the start of a typical DWD round for you? before you take your first swing, it shows their HP. No need to even scan to get that info, just don't hoverplay at the very start of a round or else you lose the information.

How much does your bleeding wounds do to Yuki/Mikuru/Ryouko? That also should probably be a fixed number that never changes unless you level up (which would raise the SG's hp).

If you feel like offering the damage you do with your normal strikes, void strikes, crits, counters, OFC, that's up to you. I realize that all would be way more work though, so it's not expected.

This post has been edited by Cryosite: Nov 14 2017, 16:01

Imperil, speaking of which, has a good bonus - you can unlock two abilities, named "Holy Imperil" & "Dark Imperil" which make Imperil to reduce enemy's mitigation against Holy and Dark. So with holy/dark weapon + respective dark/holy infusion you can greatly increase the amount of damage.

Holy and Dark Imperil is capped at 25% while Imperil mitigation reduction for fire/wind/elec/cold is capped at 40%.
Unless you are up against SGs Imperil works better with the 4 elements.

on the other hand (and not explicitly meant to him), i'll never support the "i-can-cast-this-in-1-second-less-than-you" or the "do-you-even-need-x-seconds-to-cast-this" e-peenis contests. if we're speaking about shaving, saying, 5 minutes out of half a hour it's a thing (which is still nothing in the matter of a day and heavily dependant on your connection, but still a 17% improvement), but if we're speaking about a bunch of seconds out of the same half hours, well...

I honestly never meant it as a e-penis contest, but since you can bind OFC to a single keypress I don't see how it can take up to 5 seconds (which are a whopping 12.5 turns at his speed).
To me it's just strange to see values that deviate so much from what I would consider a normal cast time of OFC so I just wanted to confirm that with Cryosite.
(edit: Ok, it was a sort of guess-estimation not meant for precision, fair enough)

This post has been edited by Sapo84: Nov 14 2017, 16:04