Dexterity's impact on critical hits

[CDF]

In Searain's data, dDex51 was 23.79 avg (actually I thought this was 50 b4 I went and glanced at it). Meanwhile his dDex50 came out to 23.03. What do you think is more likely--that (dex-agi) caps at 50 and our inaccuracy is an artifact of sample size--or that 50(dex-agi) brings you right about at the cap but not quite, and that another point is needed to top off your crit?

The number of crits and the number of total hits were never given... or were they?

The slope for that region is estimated to be 0.982% w/ interval estimate (0.931%, 1.033%)

[Shamaya]

イエーーーヽ(・∀・）人（・∀・)ノ ーーーイ おつかれさまでした
also, don't forget Deadgye Masa :3. And I believe the JP's blogging name is "TTaru"


Look for, "checking on Level Correction", the exciting sequel!
And "enhances critical hit rate", the spinoff series.

[Starr]

It's nice and all but what conclusions can you make from the data? Could you tell me at which point it would be worse off going for a DEX build rather than a STR/Attack build on Drakesbane or Raging Rush?

[mdkuser]

the conclusion is that DEX builds are not good unless you know your target's AGI precisely AND can stack huge amounts of dex.

[pchan]

If your question is which is better ? 50 str or 50 dex, the answer is probably 50 str if this doesn't cap your fstr. Anyway since all dex build trade either attack or accuracy or haste, they are rarely worth it. Keep in mind that going from 9% crit to 24% crit is only 13.76% increase in melee dot with 1.0 pdif (or ws damage if you use it for WS) and much less with larger pdif.

[Starr]

the conclusion is that DEX builds are not good unless you know your target's AGI precisely AND can stack huge amounts of dex.

We knew this way before this testing was done though.

@pchan, that would never be a relevant situation due to STR gear having DEX on though, give me a more plausable one.

[Deadgye]

Job well done guys! Have some tea! ( ^-^)_旦～

[Masamune]

My answer to Starr's question will be simple:

Can't answer your question without knowing exact crit% of those crit based WSs.

On that matter, i again PMd a few forums users, but always getting more or less same answers: "i'm sorry i have no numbers/test links, maybe can eyeball/read somewhere X/Y/Z% for 100/200/300TP".
Only locations i didnot search yet are the Japanese forums/blogs, and don't know the translation for "Raging Rush" or "Drakesbane" in JP.

As for opening a new testing thread here, im honestly not that excited, given the lolcomments i got @ Allak and litterally dead @ KI... So far, all I have is a 387 samples test for RR 100TP, which can't answer in any way Starr's question. EDIT2: and also a 200 sample test Drakesbane 100TP from Charlottie @ wiki DRG sub-forum, and the well known Evisceration thread (100tp only too) from Banalaty @ Allak THF sub-forum.

EDIT: one full fanny pack for all contributors !

[Starr]

tl;dr testing was pointless but a bit of fun?

[Masamune]

for you only who can't see further than the end of yur taru nose.

[Starr]

The testing was 100% pointless but for those of you who enjoyed doing it, it is ever so slightly more accurate than what we knew before but it does nothing to change the way people gear or play since it has been found out. I hate the way people like you always avoid trying to defend yourself because you know you cannot.

[pchan]

What's better for capped accuracy, ebody or ares ? Now you got your answer.

[Masamune]

Why do you think i necrobumped this crit testing, do you seriously think it was only for debunking a few 1% difference there and there compared to old values ? Think harder.

EDIT: and take the time to read the thread (instead of going directly to last page) i'm pretty sure will help you think correctly.

[CDF]

Critical hit rate bonuses on weapon skills: separate issue.

Some things you can glean from the data:

Large DEX bonuses (basically Byakko's Haidate, Hachiryu Kote, and Frost) will get you at least +1% crit rate if your (DEX - AGI) is > 6. What that translates into improving damage on average depends on knowledge of pDIF averages for non-crits and crits for your given "cRatio."

Crit rate (and hit rate) are basically integers, so you shouldn't use non-integer values if you want to be as accurate as possible in doing calculations.

"Stacking DEX" hypothetically isn't an all or nothing proposition. If somehow you have enough DEX to be at the high end (unlikely except for weapon skills) without making absurd equipment choices, you might then be concerned with marginal changes in DEX compared to other alternatives (maybe Hachiryu Kote vs. Hecatomb Mittens), but you don't have to go all the way to 24% if it means unpalatable trade-offs.

Then again, I am not sure if it's been established that the crit rate bonuses on weapon skills are independent of auto-attacking (TP) crit rate. If it has, then it hasn't been established whether the bonus applies to secondary hits, as far as I know.

[Starr]

What's better for capped accuracy, ebody or ares ? Now you got your answer.

I knew that before this thread was ever made, by about a year. I read the entire thread Masa and it does nothing to discredit what I am saying, if this thread does nothing to change the way people gear or play then what point was it other than a bit of fun for those who parsed? Answer me, don't beat around the bush with bullshit answers for once.

[pchan]

I knew that before this thread was ever made, by about a year. I read the entire thread Masa and it does nothing to discredit what I am saying, if this thread does nothing to change the way people gear or play then what point was it other than a bit of fun for those who parsed? Answer me, don't beat around the bush with bullshit answers for once.

You are a big idiot really, people now know exactly at which point commonly used gear start give dex crit, so that they can accurately compare 2 set of gears : few exemples

- is raja really better than toreador in X situation
- is usu body better than haubergeon +1 for Y situation
- If I use my 6-hit ws that can crit, usu leg vs random acc gear
- askar vs turban (lol)
- augmented dex5 ring to +6 acc VS mars's ring

You might not care to know which gear makes you better, but a lot of people actually do. This thread is giving the exact benefit of dex towards crit%. HUGE improvement.

[Masamune]

Well... since apparently you can't guess the applications of this testing, even after reading the whole thread and CDF answer, then i'll try one last time :

- Using inaccurate crit% leads most calculators around to wrong results.
- The method used here can be used for other testings (check Zanshin one by Motenten)
- Those crit% values are also the basis for various crit related stuff, like PahluBody Motenten mentioned (did you really read the thread?) or later an crit based WS testing.

If you can't understand to what those 3 conclusions leads to, then you are hopeless.

[Starr]

You are a big idiot really

Oh the irony, are you too stupid to realise you are commonly known as the most idiotic poster in BGs history?

Masa.
You still haven't answered my question but to answer your points:
-ok, so you need to know a mobs exact DEX to make your gear land on a point that does increase your crit% by 1 but noone is going to do that cus they're going to aim for dDEX 50 or nothing anyway so it does nothing to change the way people gear or play.
-the method? the gold fish method? that wasn't even originated here but on alla a long time ago from another test?
-wat? You could find the crit % from pbody using capped dDEX on lv1 mobs then minus the 24% we already knew years before this thread was made. Also this testing does nothing to change how people will gear for WS.

I understand them yet they are still pointless.

[Byrthnoth]

For me it was just nice to see this testing finally get completed so definitively. Searain's data was very useful, but it wasn't nearly this precise.

It always sucks when someone is like, "Hey guys, I'm new to WAR and just got my Haidates. Should I WS with Haub/Haidates/Rajas/Heca hands/head/feet/Warwolf's belt and Thunder Ring or Flame Ring?" And you have to either bullshit a yes/no answer because no one really knows or give an explanation non-answer to them. The truth is, we still won't be able to answer a question like that because it's so target and buff dependent, but now we can at least say "Yeah, probably on Lolibri."

[Starr]

The truth is, we still won't be able to answer a question like that because it's so target and buff dependent, but now we can at least say "Yeah, probably on Lolibri."

My arguement is that this has not changed since this thread ever existed based on the basic testing we had before and even more so if we had found that JPs testing.