Dexterity's impact on critical hits

[Mojo]

Over 100 damage I think.

[Starr]

(sound so like Starr dont yu think lol?)

I bet I could come back in a year and idiots are still QQing about me, why you crying Masa? You pissed off cus I used to laugh at you on Hades?

[Andradi]

Sounds more like a harmless ribbing to me.

Then again, you do seem to enjoy blowing shit out of proportion.

[Starr]

Sounds more like a harmless ribbing to me.

Then again, you do seem to enjoy blowing shit out of proportion.

na, harmless ribbing is done between friends, I doubt Masa is doing it as a harmless ribbing after having to blist his sorry ass back on Hades.

[TybudX]

Here comes Starr to shit up another formerly decent thread.

[Starr]

Oh right yeah, I was the first one to start shit talking in this thread.

[pchan]

semi-/nod to Siniroth, i was thinking same as you, even further with taking euvhis flowers as example: they change litterally jobs between closed/opened "modes", resp. thf/drk.
So only check tests when Hpemdes have mouth opened can prove if yes or not they change job from drg/drg to some other combination under that "mode", which means obviously stats changed.

You only need to perform an evasion check, right ? Unless both evasion and agi change..

[Masamune]

Correct Pchan, a simple evasion check should be enough. Then from there can work out its "new" job under opened mouth mode => evasion + agi comes from calculations.
Difficulty would be to keep it under mouth opened mode so can perform those evasion checks, and do that for at least 3 levels of Hpemdes.

and lol Starr i honestly didnot thought yu would be THAT sensitive about my comment rofl.
and 2. it's me who necrobumped that thread to actually contribute, with numbers, not useless comments like yurs.
3. it's also me who blisted you, not the opposite way, after yu made a whole assault PT wait for yu for more than 30 mins while mister were doing a rank evaluation CS which were way too long, to end up threat us all of noob and other insults, needless to say everyone included me blisted yu as a consequence.
4. i dont care poeple like you, only those giving actual numbers, ways to test are the ones interesting and worth reading, guess too hard to understand ?

[Siniroth]

It won't close its mouth while slept, sleep it in mouth open mode and mess with checks np

[Starr]

3. it's also me who blisted you, not the opposite way, after yu made a whole assault PT wait for yu for more than 30 mins while mister were doing a rank evaluation CS which were way too long, to end up threat us all of noob and other insults, needless to say everyone included me blisted yu as a consequence.
4. i dont care poeple like you, only those giving actual numbers, ways to test are the ones interesting and worth reading, guess too hard to understand ?

Hey genius, I had to unblist you for that and you clearly made all that shit up but say what you want, I'm sure more people had you blisted than me lol.
You don't care about people like me but you talk about me when I'm not even posting in the thread? ok, great logic there.

[Mojo]

Well, upon further thinking about this problem, I've been trying to quantify exactly how close the parsed critical hit rates will be to the actual critical hit rate values. I have a way to approximate it by simulation but I haven't been able to figure out purely mathematical method to do it. I think if there is a desire to derive an accurate formula that there should be enough data for the parsed critical hit rates to be within a 0.5% interval (or +/-0.25%) of the actual values at least 90% of the time. This means that if the actual value of the critical hit rate was 15% under some conditions, the parse size would need to be large enough so that whatever parsed critical hit rate would be between 14.75% and 15.25% at least 90% of the time (hope that makes sense.)

After looking over data in the thread, it's very clear that it's nowhere near the level that it should be. For example, there were 2041 strikes and the critical hit rate came out to be 12.10% for one of Masa's data points. The odds of the true critical hit rate being between 11.85% and 12.35% is only around 30%. Even with around 10,000 strikes, this chance only increases to 50%.

I'll keep trying to solve this problem but as of now those are the most accurate numbers I can provide. It's still sufficient to say that more testing should be done, especially around end where critical hit rate is changing so much. Otherwise it's really anyone's game as far as a critical hit rate model goes.

[Yugl]

So, if I'm keeping up with this correctly, we just need someone to parse their critical hits while hitting the hpmede (W/e) with about 60ish STR? I play on PS2, but if someone else doesn't mind afking and parsing while I wail on one with SCH or DRG (Would have trouble reaching the STR unless I augmented a few items on DRG) I'm up for doing the test. How should the person parsing set up their parsers (Remember, I'm on PS2, so you'll have to explain this carefully)?

[Masamune]

Actually, i calculated i would need minimum {Lowest Hpemde VIT -22} str to hit the fSTR lower cap = -weapon rank of a dmg1 weapon = -1. That points at race/jobs lvl75 combinations capable of pulling as low as 71-22=49str ! or on a more realistic scenario, on highest lvl Hpemde vit=75, would need as low as STR53(which is what deadgye posted).

Which means as elvaan 5str merited, i can't do such test since i can't reach that low str on my current jobs having the most dex gears options...

Also, another bad news is the RR crit% tests are pretty much dead as it looks last time i checked this morning @ here, KI and Allak... and i don't have the motivation to continue it if noone contribute.

[pchan]

Also ... They don't open their mouth at all until you actually deal damage to them, that is wrong info, which means the test will work if your str is low enough.

[Siniroth]

That...was already mentioned?

I'm pretty sure that they don't open their mouths unless you damage them, so if you're talking about 0 damage hits here then you shouldn't have a problem.

Open at 30 seconds after taking initial damage

[Thunder]

Need to use a lower level job then, Masamune?

Or koenig gear.. but then you'd have to get a lot of dex+ to offset that large negative dex.

[Yugl]

Looking at FFXI Stat Calculator, without a sub, SCH's base STR is 55 (60 with merits). With Errant, I'm not sure if I can reach exactly 53 STR, but I can get damn close. That said, I still need someone to PM me (Can post here too, but chances are I'll notice much better if you PM me) that is willing to afk while I wack at Hpmedes. Have you calculated their EVAs so that I can cap my ACC on SCH (Trick Staff)?

[Thunder]

You need to also be able to reach high dex values. I don't know if 50 over(65/67 base?) is needed but it'd sure help, I'd imagine.

[Byrthnoth]

Naw, any value 10+ over the 65/67 base will help. Values around dDex=35 are possibly the most useful, but anything over 10 is good.

Also, the evasions for Hpmedes are just a few posts up:
http://www.bluegartr.com/forum/68786...ml#post3183441

[Masamune]

Thank you for contributing Thunder & Yugl. Since byrthnoth answered both of you, i'll just add this:

- no need to reach exactly 53 str, read few posts above i explained need ANY str in such way it makes {Yur STR - Hpemdes' VIT}=-22, so you need to be sure which level of the mob yu plan to whack on, so you know (with the vit values provided a few posts above) if you are indeed @ the fSTR lower cap.

- choose a job having a lot of DEX gears options to play with. because if you take for example SCH like proposed above, yu'll prolly end up with abysmally low dex also compared to Hpemde's AGI, ie unexploitable dDEX.