What's Better?

[mdkuser]

I was wondering about two pieces.

Usu Pants vs Pup relic legs/(+1).
Hades +1 vs Destroyers when acc not capped of course.

Hades sainti +1 are better than destroyer in any situations, even capped accuracy. destroyer is generally better than the NQ though, depending on how much accuracy is needed.

[Argettio]

can someone point me to a mathy askar v. turban thread? this came up recently with some folks and i'd like to point to more reference material.

It comes up in this thread once every few weeks, although I couldn't give a direct link.

[Shinryuu]

can someone point me to a mathy askar v. turban thread? this came up recently with some folks and i'd like to point to more reference material.

It's easy, you calculate the returns of each different factor going from a blank hypothetical item (in this case, a hat with 4% haste).

Turban gains 1% haste. Haste affects TP damage and WS frequency. Haste return is 1/(1-haste). To see Turban's increase, you do (new haste return-old haste return)/(old haste return)*100.
Example: you have 50% haste, so the return is 2 (200%, twice as much damage, etc.). Adding one haste:
(new return (1/(1-.51) = 2.04) - old return (=2))/(old return) *100 = 2.04% damage done for TP and WS.

Now Askar hat. 4 STR means your weapon gains 1 base DMG through fSTR. This only goes for the TP phase (you're either switching to another hat on WS, or you're using askar for WS even when using turban for TP). Say 50% of your total damage comes from TPing (can parse this), and +1 DMG means a 2% increase (for an old base DMG of 50, 51 is a 2% increase), the total damage increase from the STR is TP damage * DMG increase = 1%.

Now the DEX. 2 ACC is 1% hitrate. Hitrate affects TP damage and WS frequency, so no need to multiply w/ anything. Return is calculated as follows: (new hitrate-old hitrate)/old hitrate*100. Say you have 80% hitrate, which now goes to 81%:
(81-80)/80*100=1.25% more damage for TP and WS.

DEX can also give critical hitrate. Say this is the last 4 DEX you need, means 4% more critrate. Calculate your average pDIF. A crit is pDIF +1. Say you have 20% critrate pre-askar and you have a normal pDIF of 1 your average pDIF is .8*1+.2*2=1.2. A crit increases your pDIF to 2, so 4% extra crit results in a new average pDIF of .76*1+.24*2=1.24. Going from 1.2 to 1.24 is an increase of 3.3%. Multiply this by the factor of TP damage out of total damage (crits don't go for WS, same as STR, see reasoning above). Say this is still 50/50, the crit means an additional +1.67% damage.

In this hypothetical, Askar clearly wins (1.67%+1.25%+1% > 2.04%).

Yes guys, this indeed also means +crit is a fairly weak stat to increase damage. Destroyers suck, by the way.

[mallister]

was looking for more of a 2h perspective (specifically drk), where the fSTR is 1% or less, the accuracy from the dex isn't fully used (if at all) and that 4 dex isn't breaking you any sort of tier. in that case it's 1% haste vs 1fSTR and the accompanying 2-3 attack and pretty clear in turban's favor. i know the maths was looking just for some more examples or discussion.

for my playstyle turban's a clear winner unless we're lowmanning something and i'm not getting buffed at all. folks are surprisingly aggro about this subject...whatever. i guess they're fond of their hats?

[mdkuser]

For DRK it makes no doubts turban wins, because DRK isnt a DEX job (mainly because it cannot use byakko's haidate), because high dmg weapon means less benefit from 1 fstr, etc.

For SAM I did maths here with my gear to figure out that askar "wins" when accuracy is below 82% ish ( against a target having 80 vit which was already high), at which point its needed to use pizza or sushi with turban.. so askar doesn't really win anything. I was thinking about farming askar head at some point for a dex build on SAM, until I realized DEX builds are stupid because they only work on specific mobs/camps.. You generally will not get a lot of crits boost from 4 dex, unless you are fighting really weak mobs, in which case acc is capped and turban wins. Crits is only a small part of your melee dmg, which is a small part of your total dmg..

[Shinryuu]

Mallister, what exactly are you looking for then? If you know the math you can apply it to any situation.

Well, I guess if you want a playstyle example, here comes: if I ever got an askar zuchetto, I'd toss it (WAR, DRG).

DEX builds aren't stupid! It's just that haste and ACC are superior and most melee struggle (fail) to cap either. I like my WAR's DEX very much thx but yeah when it's some ridill WAR dinging 118 DEX with like 15% haste and bream sushi +1 doesn't cap its hitrate in merits...

[Nightfyre]

was looking for more of a 2h perspective (specifically drk), where the fSTR is 1% or less, the accuracy from the dex isn't fully used (if at all) and that 4 dex isn't breaking you any sort of tier. in that case it's 1% haste vs 1fSTR and the accompanying 2-3 attack and pretty clear in turban's favor. i know the maths was looking just for some more examples or discussion.

for my playstyle turban's a clear winner unless we're lowmanning something and i'm not getting buffed at all. folks are surprisingly aggro about this subject...whatever. i guess they're fond of their hats?

There was an extended discussion in the first What's Better thread here. There's an earlier post quoted but I didn't find it on the page previous and didn't care to look farther back. Second post down starts it off, goes for several pages.

[mallister]

thanks guys ~

i'm good to go on the issue, for me it's pretty clear and really always has been. some of the guys i talk to in game and on other boards have different views on the subject, was just looking for some meat to add to the discussion.

[Shinryuu]

Views, huh? It's funny. This is a game where damage is based on formulae. These are known. Purely from a damage perspective, you can't have views, you just have to do the math. More is better. So many people make me sad.

[TybudX]

I like my WAR's DEX very much thx

There's a big difference between a Mithra just happening to be a few DEX off capped crits and any other race having to make poor trade offs. ; ;

[Tortalius]

Aight another question for you.

I have 2 Pup builds, and I'm not really sure how to figure out which one is more useful. One's an acc build, other is haste build. Full 8 h2h merits no dex/str merits.

Here're the 2 sets.

Same in both sets: Hades/Turbo animator (2dex)/chiv chain/brutal/hollow earring/rajas/uthalum/pantin cape/pantin legs (5 acc 2 str 2 vit).

Haste: Turban/Goli/pantin hands +1/swift/Enki feet
Acc: O hat/relic body or ACP body(10 acc)/enki hands/potent (or life)/pup af + 1 (5 str 5 acc).

Against merit mobs/limbus/dyna. Which would be more likely to help?

pup/war and mithra if that helps.

[mallister]

lol well i was trying to say it nicely...

[ringthree]

Views, huh? It's funny. This is a game where damage is based on formulae. These are known. Purely from a damage perspective, you can't have views, you just have to do the math. More is better. So many people make me sad.

We know the damage formula for our side of the equation, but there is very little known about the mob side of the equation (e.g. stats on anything higher than Greater Colibri).

For example, I don't think that any crit calculations take into account level correction. Most of the crit % calculations are based on extrapolated information from Greater Colibri testing.

Does level correction exist for calculations such as accuracy and crit %? Does level correction effect crit % from stats vs. crit % from Dex/Agi? No one knows for sure.

Using formula to figure out damage is a great tool to improve performance, but it's not a simple as you are making it out to be. I am not saying you are wrong at all, just that there are still unknowns and people should take that into account when working out the math.

[Tsuko_Asura]

For DRK it makes no doubts turban wins, because DRK isnt a DEX job (mainly because it cannot use byakko's haidate), because high dmg weapon means less benefit from 1 fstr, etc.

Not that I'm disagreeing with your conclusion, but it was my understanding that the journey to capped critical hit rate was a linear progression. Is that incorrect?

[ringthree]

Not that I'm disagreeing with your conclusion, but it was my understanding that the journey to capped critical hit rate was a linear progression. Is that incorrect?

No one knows for sure. Many of the assumptions behind crit hit rate are guesses at best.

When I graphed out the information that someone else provided it appeared to be a curved graph.

I have asked several times for more information about Dex for crit hit rate. I usually just get linked to the old crit hit thread that the person linking the thread obviously hasn't read all of.

If there has been more testing on crit hit rate, I would love to see it!

[Neosutra]

Hades sainti +1 are better than destroyer in any situations, even capped accuracy. destroyer is generally better than the NQ though, depending on how much accuracy is needed.

No.

In fact, what the fuck were you thinking?

[Limlight]

errr. what? hades better than destroyers?

I think my head is going to explode.

[ringthree]

No.

In fact, what the fuck were you thinking?

Didn't we just go over this?

[mdkuser]

that other thread in standard was about NQ vs destroyer. HQ beat destroyer, its got more base dmg and less delay, 9 acc..

[pchan]

+6% crit makes you cool on a parse. 10 acc or 20 att is better damage.