[X] My Assistant

Has anyone done before a test of 2h + heavy? Probably a little bit unsafe, but maybe shielding or protection sets can pull it off (or a mix of both).

iirc someone was doing something like that but with mithril pieces, but I can't remember who. Can someone refresh my mind? the only thing I remember is that he was a well known veteran from the catgirl club.

I don't play much anymore but after reading all these posts I feel like returning to play more actively

It was sickentide who played with the 2H Heavy style. However, I don't remember him posting any more detailed results (just one or another print at the end of an Arena and/or Grindfest).

It was sickentide who played with the 2H Heavy style. However, I don't remember him posting any more detailed results (just one or another print at the end of an Arena and/or Grindfest).

I forget things easily, thanks basara. I hope someday he share his results or resume his efforts in case he stopped his research for some reason. It would be interesting to know more about that particular style.

Another test like: "I did it because I was curious". (IMG:[invalid] style_emoticons/default/heh.gif) (IMG:[invalid] style_emoticons/default/tongue.gif)

Based on some previous messages posted by BlueWaterSplash, about the relationship between the condition of the equipment and its respective bazaar price, I decided to do a simple check of this.

I spent the last few days using my Peerless Shortsword Slaughter, and I wrote down the condition values (numeric and percentage), and their respective price at the bazaar. As the amount of data is large I will not post the complete table; just a few excerpts later, to exemplify some facts. Graphically, this was the result obtained:

(IMG:[imgur.com] https://imgur.com/ktUtLIO.jpg)

From what can be seen, this confirms what is already on the Wiki, about the degradation of the equipment being a fixed value for each monster killed. As a result, the price at the bazaar also decreases linearly. In fact, I realized that the price variation is even more visible than the value of the equipment degradation, because on some occasions the value of the condition has not changed, but the price has dropped slightly (after I played an arena with few rounds). The image below shows the condition in percentage, the condition in absolute values, and the price at the bazaar.

(IMG:[imgur.com] https://imgur.com/ETBzM4E.jpg)

The initial condition is 800 and the starting price is 5,420 credits.

One thing to be corrected on the Wiki is regarding the minimum condition of the equipment. There is written "0%", but the minimum is "1%", as shown in the following image.

(IMG:[imgur.com] https://imgur.com/HmTeDRW.jpg)

I continued to play, and even suffered a deliberate defeat, but the condition does not become 0.

The minimum price was 632 credits (image below). But I realized that this value did not change after a certain point, even with the continuous degradation of the equipment. For this reason I did not add these points to the chart. The lowest point was that of the first data above that minimum value. But using the equation given by the linear sequence, we can see that the minimum value of 632 credits is reached when the condition of the equipment is 80, that is, 10%. This minimum value, of course, should vary depending on the type and quality level of the equipment.

(IMG:[imgur.com] https://imgur.com/lY9FZDd.jpg)

Another thing I realized was that when I was below 50%, the drop in attribute values was linear (as it says on the Wiki), but it was not continuous like the drop in price (as shown above). Basically, the attribute values fall 1% for every 0.5% reduction in the condition of the equipment.

(IMG:[imgur.com] https://imgur.com/Fjta2kj.jpg)

Finally, one thing that I think is wrong in the Wiki is the formula for equipment repair cost. By the formula we have:

Scraps Needed = 1 + rounddown((100 - condition%) / quality_mod
Scraps Needed = 1 + rounddown((100 - 1) / 1.4
Scraps Needed = 1 + rounddown(70.71)
Scraps Needed = 71

But I only needed 70 to repair the sword:

(IMG:[imgur.com] https://imgur.com/x96YYj9.jpg)

Obviously I (or someone else) would need to check this with other equipment to check the formula's validity.

I forget things easily, thanks basara. I hope someday he share his results or resume his efforts in case he stopped his research for some reason. It would be interesting to know more about that particular style.

oh, i have plans to finalise my research, don't you worry. unfortunately, i am currently roadblocked by two compounding factors: IWs being a bitch on the pieces i need for my research and a lack of free time. the latter factor will be alleviated in may, when i will be done with final exams

here's a brief summary of my findings so far:

mithril helps you survive at lower leves, but as your player and upgrade levels increase, you can swap mithril for savage. this is the goal of my current efforts, which will be followed by a thorough examination of mithril vs. savage

for arenas, longsword and estoc give similar performance. for IW and fest i recommend use of a mace

for mace style, use of Imperil and OFC is a must. for the other two, very little, if anything, is gained over simply powering through in spirit stance

for a balance vs. slaughter comparison, i have all the needed pieces, but IWs...

...

Glad to see you're still working on it and thanks for the brief summary. I'll wait patiently until you finish all your tests and disclose all your findings in detail, I'll make sure to read it with great interest. Oh! And good luck IW'ing all those pieces.

I got a new cookie and silver award, and run a grindfest, but 2020 cookie doesn't seem to give an exp bonus.


Yesterday I had 7 cookies = 140% exp bonus and acquired 5.76b in a GF.

So

base_exp (including goldstar bonus) = 5.76b / 2.4 = 2.4b


As I got an additional cookie and award, I should have gotten 8b exp.

expected_exp_bonus = 8*20% + 75% = 235%
base_exp 2.4b * 3.35 = 8.04b


But I actually got only 7.56b exp.

7.56b / 2.4b = 3.15
current_exp_bonus = 215%


How did the number 215 come out?

7*20% + 75% = 215%


The same case happened in 2018.
2017's cookie bonus was not applied at all through 2018.

https://forums.e-hentai.org/index.php?s=&am...t&p=5691912

I got a new cookie and silver award, and run a grindfest, but 2020 cookie doesn't seem to give an exp bonus.
Yesterday I had 7 cookies = 140% exp bonus and acquired 5.76b in a GF.

So

base_exp (including goldstar bonus) = 5.76b / 2.4 = 2.4b
As I got an additional cookie and award, I should have gotten 8b exp.

expected_exp_bonus = 8*20% + 75% = 235%
base_exp 2.4b * 3.35 = 8.04b
But I actually got only 7.56b exp.

7.56b / 2.4b = 3.15
current_exp_bonus = 215%
How did the number 215 come out?

7*20% + 75% = 215%
The same case happened in 2018.
2017's cookie bonus was not applied at all through 2018.

https://forums.e-hentai.org/index.php?s=&am...t&p=5691912

Congratulations on your silver award

That's why Parry istn't that useful for 1H-Mage.
Specail Thanks to lololo16.

Nezu edit: big image -> attachment

This post has been edited by Nezu: Feb 14 2021, 01:05

After a year I was able to get the last radiant piece I needed to run one of the last tests I'm interested in: how can an optimally equipped non-imperil elemental mage perform compared to holy?

For some background on the non-imperil setup, I chose to use my PSWD with an Elementalist robe, because it's the closest you can get to 1.0 with Willow - but it only comes to 0.966, with 600 base prof. I plan to also test it with PSRD, to see if the 1.0 prof factor is more important than Willow's CR, but unlike my earlier propositions regarding the effect of CR on average damage, I think CR probably plays a bigger effect on turn times than I previously thought, so I'm expecting that test to perform worse. However, if it does turn out to be better, the optimal setup would be elementalist pants rather than robe.

Some stats on the different equip setups:

Non-imp elec: 44,764 arcane score, 32,939 magic score, 0.966 prof factor
Non-imp holy: 43,462 arcane score, 31,984 magic score, 1.0 prof factor
Imperil elec: Unknown - I salvaged some of the gear pieces used in it since then, as I planned to replace them anyway. Prof factor was around 0.85 and the arcane score was similar to the non-imp elec set, I believe.

As you'll see from the results I choose not to use gems or draughts (except spirit draughts as non-imperil). I have my Monsterbation HP threshold set to 0.4 for the non-imperil setups - it's a hundred or so turns faster with it set to 0.3, but there are more deaths to unexpected spirit overconsumption from spirit shield eating multiple crits when at low health, then triggering spark in the same turn. I am currently safe(ish) to play holy on any day, with maybe occasional deaths on Fridays.

Holy 1 is non-imperil on Sunday (holy day). Holy 2 is non-imperil on days that aren't Sunday or Friday. Elec 1 is non-imperil on Saturday (elec day). Elec 2 is my standard imperil set on days that aren't Sunday or Friday, for performance comparison against a very average imperil run.

All of these numbers are an average of 10 runs of PFUDOR grindfest:

CODE

=================================================
|       | Holy 1  | Holy 2  | Elec 1  | Elec 2  |
-------------------------------------------------
| Turns | 5615.2  | 5798.3  | 6,345.7 | 5,987.4 |
-------------------------------------------------
|       |         |         |         |         |
-------------------------------------------------
| T3    | 1,001.2 | 1,031.5 | 1,151.8 | 989.8   |
-------------------------------------------------
| T2    | 1,452.9 | 1,501.2 | 1,696.3 | 716.4   |
-------------------------------------------------
| T1    | 2,074.3 | 2,136.3 | 2,461   | 93.8    |
-------------------------------------------------
| Imp.  | 0       | 0       | 0       | 3,408.6 |
-------------------------------------------------
|       |         |         |         |         |
-------------------------------------------------
| HP    | 113.4   | 115.7   | 122.7   | 101.4   |
-------------------------------------------------
| HE    | 24.1    | 23.5    | 21.5    | 4       |
-------------------------------------------------
| MP    | 107     | 110.6   | 118.7   | 125.3   |
-------------------------------------------------
| ME    | 11.1    | 11.7    | 0       | 10.8    |
-------------------------------------------------
| SD    | 16.6    | 18.2    | 20.9    | 0       |
-------------------------------------------------
| SP    | 64.5    | 65.7    | 59.7    | 52.1    |
-------------------------------------------------
| SE    | 9.1     | 10.9    | 3.7     | 3.2     |
-------------------------------------------------
| LE    | 10.9    | 10.1    | 5.8     | 0       |
-------------------------------------------------
|       |         |         |         |         |
-------------------------------------------------
| Cure  | 417.2   | 430.8   | 392.4   | 218.5   |
-------------------------------------------------
| F.C.  | 192.1   | 205     | 176.4   | 179.3   |
-------------------------------------------------
|       |         |         |         |         |
-------------------------------------------------
| A.F.  | 4.2     | 4.9     | 5       | 3.8     |
-------------------------------------------------
| Regen | 12.6    | 13      | 13.7    | 9.6     |
-------------------------------------------------
| A./G. | 13.1    | 14      | 12      | 8.2     |
-------------------------------------------------
| Life  | 26      | 27.7    | 22.3    | 21.8    |
-------------------------------------------------
| Inf.  | 51.2    | 53      | 45.6    | 31.7    |
=================================================

(A.F. = arcane focus, A./G. = avatar/gods sccrolls, inf. = infusions)

My non-imperil elec set is as close as I can get to perfect; it has very very high rolls, I have all relevant profs maxed, it's max forged (obviously), I am DD9, etc. There's not really much room for it to improve (pending results on the PSRD test). In comparison, my holy set is visible in my sig - there's a PHOH, yes, but you'll notice my cap at time of posting is still just a high-roll mystic, my pants are only 66% EDB, and I'm a pretty long way away from being able to run 4+1 holy (568 prof at the moment, need at least 590 with a peerless robe for the average damage score to break even with 3+2). There is a lot of room to improve it, and I anticipate a gain of about 200-300 turns depending on how much it improves survival.

However you'll also note the consumable usage. I don't really like spending all those elixirs - especially the last elixirs, which I try to use if my spirit is low after sparking (very efficient to restore both HP and SP at the same time, right?), and otherwise as a backup when SP elixir is on cooldown. To that end non-imp elemental is more viable than I expected, though still miles behind holy.

I'll try to update this when I run some other tests, to provide some idea of what turn counts mage can offer if you have enough equipment etc.

This post has been edited by Nezu: Feb 14 2021, 09:22

Thanks for showing me the crude axe. The most important thing to know is that it is condition 1/203 (1%) and sells for 102 credits.

As a wild guess, perhaps the base cost of every equipment is 100 (nice number) and that axe was worth 2 credits in that state?

The initial condition [of peerless shocking shortsword of slaughter] is 800 and the starting price is 5,420 credits.

The minimum price was 632 credits...the minimum value of 632 credits is reached when the condition of the equipment is 80, that is, 10%.

Sorry for the very late response, I am very busy in real life this year but fortunately HV 0.87 has delayed everyone's research. Unfortunately HV 0.87 changed the equipment cost formulas, but in a somewhat documented way, so we will need to be careful of it in the future.

Thanks for the terrific data! I wonder if we could generalize the equipment sell cost to be the starting price x condition % + 90 credits?

---------------------------------------

Back to the related topic of trying to learn how to calculate PXP0 from all an equipment's base stats. I have various formulas for this that seem close but generally have errors. (This topic is related because we're also interested in learning how to calculate bazaar cost from PXP0, or vice versa).

Magnificent Shade Breastplate of the Fleet (352 PXP0, 4 PABs)
Magnificent Shade Leggings of the Shadowdancer (351 PXP0, 4 PABs, suffix adds crit chance)
Magnificent Savage Shade Boots of the Shadowdancer (335 PXP0, 3 PABs, suffix adds crit chance, prefix adds crit damage)

I recently noticed the above 3 isekai equipments in auction. I didn't bother to write down all the stats, but note the relatively large differences in PXP0. It's always been known that having an extra PAB greatly increases PXP0. It's a whole extra stat (as opposed to just a higher or lower rolled stat) thus a very big addition into the PXP0 total.

The interesting observation is that adding the crit chance stat or crit damage stat may not affect PXP0. So maybe those two stats don't go into the mystery PXP0 sum calculation (although they do still presumably benefit from the PXP0 quality bonus contribution to their scaled stat, in my view that can be considered an entirely separate calculation and phenomena).

Alternatively, it might be possible that crit chance and crit damage do go into the mystery PXP0 sum calculation in a lesser weighted amount compared to a PAB, but there would have to be some reason for it, and I cannot find that reason. If we look at the equipment ranges base rolls column on the left, crit chance and crit damage have large base rolls. Note that this "base roll" number is different from what we traditionally consider the "base stats" that we see in equipment popups. If we consider the "base stats" then crit chance and crit damage have comparable values to "base stats" for PABs.

The only way in which PABs are greater than crit chance and crit damage, are when they scale. Because many probably realize that PABs (and ADB/MDB) scale a lot with level, while most other stats like crit chance and crit damage only scale slightly with level. But PXP0 is a fixed value per equipment, so I do not think the game would use scaled stats to calculate PXP0.

Thanks for the terrific data! I wonder if we could generalize the equipment sell cost to be the starting price x condition % + 90 credits?

The current formula is:

CODE

round( 400 + 1.2 * $base_value * $quality_factor * clamp($wear_factor, 0.25, 0.75) )

Clamp is a php function that establishes a minimum & maximum value, so it's capped between 25% and 75% essentially. Base value is just 1 for common equipment and 4 for rare equipment.

With that, you should be able to calculate quality factors and relate that back to PXP.

This post has been edited by Nezu: Mar 15 2021, 02:00

Can we get the analogous previous version formula? I presume it would have been similar to:

round(90?? + $base_value * $quality_factor * clamp($wear_factor, 0.1, 1.0)

Can we get the analogous previous version formula? I presume it would have been similar to:

round(90?? + $base_value * $quality_factor * clamp($wear_factor, 0.1, 1.0)

It was this:

CODE

round( 500 + $base_value * $quality_factor * max(0.1, $wear_factor) )

In the previous patch the base value used to range between 1.0-1.3 for normal equipment, and 2.7-3.5 for rare equipment. Quality factor was unchanged.

Your clamp would be technically correct as the 1.0 involved would be 100% durability (I think).

This post has been edited by Nezu: Mar 15 2021, 04:54

I'm pretty sure Nezu is posting the formula for the buy price and BlueWaterSplash is looking at the sell price.

IIRC the difference is a factor of 5.

Thanks sigo8, you perfectly answered an important question that had been bothering me!

Thanks for showing me the crude axe. The most important thing to know is that it is condition 1/203 (1%) and sells for 102 credits.

As a wild guess, perhaps the base cost of every equipment is 100 (nice number) and that axe was worth 2 credits in that state?

The initial condition is 800 and the starting price is 5,420 credits.

The minimum price was 632 credits...the minimum value of 632 credits is reached when the condition of the equipment is 80, that is, 10%.

I wonder if we could generalize the equipment sell cost to be the starting price x condition % + 90 100 credits?

I just realized I made a small math error, so indeed the base global equipment sell price was a nice number of 100 credits, not 90 credits.

5420 credits - 100 credits = 5320 credits at 100% condition. At 10% condition this would become 532 credits + 100 credits = 632 credits. Bingo!

So now what's left to wonder is what the $base_value are. Well, according to Nezu this part of the formula is actually simpler now (just 1 or 4) whereas I'm not sure what determined the exact value within the ranges of 1~1.3 or 2.7~3.5 in the prior game version.

And then we can figure out various $quality_factor and try to relate them to observed PXP0.

I'm pretty sure Nezu is posting the formula for the buy price and BlueWaterSplash is looking at the sell price.

IIRC the difference is a factor of 5.

It's just a generic 'equipment value' formula. This was given while we were rebalancing sale value of equipments, for what it's worth.

Well, according to Nezu this part of the formula is actually simpler now (just 1 or 4)...

I should hope it's right, because it's not actually according to me, it's Word of God. (IMG:[invalid] style_emoticons/default/tongue.gif)

@BlueWaterSplash

Just to let you know that the current patch has changed equipment prices. The value of my Peerless Shortsword dropped from 5,420 to 4,432 credits.

And, I don't think I intend to repeat those tests. (IMG:[invalid] style_emoticons/default/heh.gif)

I should have posted this a month ago, but laziness overwhelmed me. (IMG:[invalid] style_emoticons/default/heh.gif)

Comparison between different sword combinations for DW-shade set.

(IMG:[imgur.com] https://imgur.com/6mx1Qtk.jpg)

Arena chosen: 150 rounds
Equipment forge: Club Slaughter, Rapier Balance, Rapier Slaughter and Shortsword Slaughter (full forge); Shade Fleet Set (forge level 50)
Hath Perks: Dæmon Duality V and Innate Arcana V (Spirit Shield, Protection, Spark of Life and Haste)(slot 5 not assigned) (no other "game perk" more)
Supportive Spells in continuous use: Regen and Heartseeker
Deprecating Spell in continuous use: Imperil
Shock Spike Shield
Use of Featherweight Shard and Infusion on swords

To avoid variations due to the type of element and also the day of the week, I tried to use infusions in order to always have the same elements in the swords, that is, 2 Holy + 1 Fire + 1 Elec.

Although the statistics of the sets are not exactly from the same time I performed the tests, the numbers have not changed significantly. The test dates, for the same set, are not continuous because in some cases I had to retake the test due to connection problems or inconsistent results (very different from the others).

Links of my equipment can be found in my signature.

[ 1 ] Club Slaughter (Main Hand) + Rapier Balance (Off Hand)

(IMG:[imgur.com] https://imgur.com/hbdrjmw.jpg)

[ 2 ] Club Slaughter (Main Hand) + Rapier Slaughter (Off Hand)

(IMG:[imgur.com] https://imgur.com/8r9rnMj.jpg)

[ 3 ] Club Slaughter (Main Hand) + Shortsword Slaughter (Off Hand)

(IMG:[imgur.com] https://imgur.com/neSPSUw.jpg)

[ 4 ] Shortsword Slaughter (Main Hand) + Rapier Balance (Off Hand)

(IMG:[imgur.com] https://imgur.com/qBrEciU.jpg)

[ 5 ] Shortsword Slaughter (Main Hand) + Rapier Slaughter (Off Hand)

(IMG:[imgur.com] https://imgur.com/a6SHcYu.jpg)

[ 6 ] Rapier Slaughter (Main Hand) + Shortsword Slaughter (Off Hand)

(IMG:[imgur.com] https://imgur.com/uX6apiS.jpg)

[ 7 ] Rapier Slaughter (Main Hand) + Rapier Balance (Off Hand)

(IMG:[imgur.com] https://imgur.com/Fnc4EL0.jpg)

Simple conclusions:

Comparing the results of [1], [2] and [3]
Even with a higher ADB of the set [3], the absence of Penetrated Armor made a negative difference of more than 300 turns. Between sets [1] and [2], the first has the advantage of having more Offhand Strike on hit, crit chance and crit damage, compensating for the lack of ADB (this also applies in comparison to the set [3]). This highlights the importance of using a Rapier Balance as an Off Hand weapon.

Comparing the results of [4], [5] and [6]
Comparing sets [4] and [5], the situation is similar to what was observed between sets [1] and [2]. Even the turn difference was very close (121 between [4] and [5]; 125 between [1] and [2]).
The comparison between the sets [5] and [6] was made to see what would happen if the swords were to be inverted. The observed difference was not very large. The set [6] was worse mainly because using a Rapier Slaughter as the Main Hand produces a weaker attack.

Comparison between [1] and [4]
In the past, when I purchased my Shortsword Slaughter, I thought it would be superior to a Club when used as a Main Hand for the DW style. This thought came from the comparison based on the 1H style, where Bleeding Wound is much more efficient and useful than Stun. However, seeing in practice, for the DW style the exact opposite occurs. And the difference was big (365 turns; the same occurred between [2] and [5] with 361 turns).

About the set [7]
I decided to test the use of two Rapiers, and from the results you can see that they were terrible. This set lacked attack power.

Note 1: The most important of these tests is the comparison between the results, not the absolute result itself. My results do not serve as a parameter to define the limits of each set, because there are players with better playing techniques than I do, and therefore their results would be better (less turns) if they performed the same tests (but I believe the conclusions would be the same).

Note 2: The main differences between these tests and the old ones are: {A} now I have Innate Arcana (before I triggered manually all spells, which obviously made me lose more time and have to pay more attention when the effect of some of them ended); {B} I upgraded from 25 to 50 in my Shade Fleet set; {C} in the results tables, I changed the way of representing time, because only now have I learned how to format the excel cell correctly (IMG:[invalid] style_emoticons/default/heh.gif) (aiming at calculating the average time, which I used to do manually). It took me a while to realize where the error was. (IMG:[invalid] style_emoticons/default/tongue.gif)

Comparison between [1] and [4]
In the past, when I purchased my Shortsword Slaughter, I thought it would be superior to a Club when used as a Main Hand for the DW style. This thought came from the comparison based on the 1H style, where Bleeding Wound is much more efficient and useful than Stun. However, seeing in practice, for the DW style the exact opposite occurs. And the difference was big (365 turns; the same occurred between [2] and [5] with 361 turns).

Yeah, the stun is a real luxury for DW because stunned monsters can't parry. Club is less useful for 1H because 1) overwhelming strikes, 2) monsters can't parry counters, and 3) 1H is already getting stuns from counters as well.

-- skip all -- (IMG:[invalid] style_emoticons/default/heh.gif)

Impressive piece of research !!

Kudoz and thanks alot, Basara. There's a lot of time in this, and a lot to study; thank you again (IMG:[invalid] style_emoticons/default/wub.gif)