[Ture]

The idea of dumping the primary mod on most war WS sounds like a stupid fucking idea to me. I like my RR build as it has tons of dex and str (gogo kitty pants), but for SS why would you give away 5.

[Shamaya]

For Thf, the general consensus is to dump the primary mod anyway, and go str merits. Lots of thieves do dex merits too. Nothing wrong w/ it. Lots do str. I did dex and then switched to str a long time ago. But the numbers are really similar. Even if they're similar for war I'd agree too to stick w/ str.

The crit build thing for war sounded so retarded to me. But then I have to remember you guys get a lot more dex for melee than us (weren't thieves the job SE intended to be critting a lot!?). Aurum feet, byakko's, thunder ring (for melee), and hauberkeon, are all dex pieces we can't use =/, and only one of those isn't a standard TP piece for war. Looking at searain and co's stuff (could use more samples!), I agree it looks like it's not terribly hard to cap crit on many merit mobs (especially when mithra, yeahhh) with synthetic's standard gear. My current rule-of-thumb for crit is the general 0-20-50 rule, where 0-20 is the placebo region, and 20-50 is a 30 point attribute spread which accounts for a 15% max-min crit spread. It intuitively looks like 1dex=0.5% crit when in the range of efficacy, but on top of that a cursory glance at searain's samples looks to back this up. Still a bit skeptical myself though.


I'm having a real tough time coming to terms with the "swordchucks" build. It just seems so outlandish (at least coming from my server). I've seen wars swordchuking before but it's very rare. Normally the good ones stay with maneater, including that monster Rayes I mentioned. It's hard to doubt him. But you sound very experienced and a bit convincing, Bryth. In my discomfort what I'll probably end up doing is a full-blown math comparison with either G.colibri or another staple merit mob. I just don't feel comfortable, with close parses especially, unless I know the full DoT/WS dps chock with avg hit rate, pDif, swing vol, piercing accountability, ws rate and avg dmg, and the like. It's shocking to think that Rid/Joy > single Ridill ~> Man/Rid & Org/Rid.

I like GA too. I do feel kinda sad that I'm neglecting my GA build, but it doesn't mean I can't play around w/ war/sam either. That's just an alternative toy though. I wouldn't put in GA merits because I'm only using war for /nin desired and ridill-ok situations. For straight high dmg (or utility) I would play Thf. Or I could level another 2hander (ie bandwagon sam) if I wanted to specialize a 2h job. Not to hate on war as a job. It really boils down to the limitations of combat merits, as always. I think I'll keep my 4evasion merits if 4axe turns out to be pointless. I've already got a perdu and a RC/hastebelt. I particularly want to play with RC. I've built buffed RC/hastebelt parties or used them outside meripos for "mobile zerging" before, and when supported (not easy to do) it's just hilarious. Looks fun so I'll want to try that as well..

[Byrthnoth]

Shamaya, on 14 September 2009 - 11:32 PM, said:

My current rule-of-thumb for crit is the general 0-20-50 rule, where 0-20 is the placebo region, and 20-50 is a 30 point attribute spread which accounts for a 15% max-min crit spread. It intuitively looks like 1dex=0.5% crit when in the range of efficacy, but on top of that a cursory glance at searain's samples looks to back this up. Still a bit skeptical myself though.

This is exactly how I saw it when I plotted the data, but I know some people disagree. It's nice to see someone else reached the same conclusion as me, but I think the posters on BG are convinced it is more piecewise than that. Searain's testing was kind of lacking in consistent low-mid crit rate values, as he and Kwontess weren't so interested in determining the crit rate floor. They were more interested in determining the cap and trying to figure out what set it.


I'm not saying that WAR/NIN with Swordchucks will outdamage Man/Ridill all the time. Situations with low sword Acc would definitely favor Maneater. However, many of those situations might warrant GA usage even /NIN, as it's such an incredible Acc bonus. I mean, at 120ish Dex there is a 30 Acc difference between GA and any one-handed weapon, not to mention the base skill difference. Hi2u "I-wear-3-optical-hats-in-my-weapon-slot"

When you finish your merits, try Swordchucks vs. Man/Ridill on Colibri within a party. Then try Man/Ridill vs. Perdu at Jade Sepulcher within a party.

The damage from Ridill drops off a lot when your hit rate dies. When you start fighting Lurkers, Infiltrators, Skoffins, etc. you can expect your hit rate to suffer. Sure if you sit around up top in Jade and just kill mages and sea puks you'll maybe be okay, but if you're killing fast enough to work your way down into the middle and start pulling from the corner around/with the Skoffins you'll probably start parsing relatively low Acc. Not to mention that as long as you're in a merit situation you're really free to sub SAM. I've never done tests with a Ridill personally, but I know that I was a good enough DD with GA that Ridill WARs would invite me back even when I was still working to cap my merits and didn't have N.head or E.body. Also, Ridill vs. Skoffins isn't as bad as k-club RNG vs. Skoffins, but it's still not my favorite co-DD especially when they can't pull hate.
Pizza has been a huge boon to one-handed jobs though, and I was banned just after they added Pizza so I don't have a feeling for exactly how much it matters. I know Richardd said he was doing pretty well Acc-wise on MNK/WAR with Pizza at Jade, but he had a pretty perfect MNK short of relic.


Edit:
Oh yeah, I forgot to address the single-Ridill thing. That was really more of a joke because /DRG is funny for Ridill but offers no defense. Acc+10, Jump, High Jump, 6% Haste from Shield+Earring. Something like this:

Ridill / Wyvern Targe / ___ / Bomblet
Askar / Ancient / Wyvern / Suppa
E.body / Askar / Toreador / Rajas
Cuch / Speed / Byakko / Aurum
25% Haste, quite a bit of Dex and Acc. With enough Haste from other sources the extra Haste from /DRG would probably overtake most other combinations involving Ridill, especially considering how much higher the Acc and DoT are. And you've got a badass shield to block with amirite? Too bad it isn't really viable due to lack of things to save you from Fang Rush.
DPS - Man/Ridill with 22% Haste from gear, 35% Magic Haste, and a 5/5 Haste Samba - 58.7 (= (1.9*40 + 49*1.15)/( (236+276)*.8*.33/60) )
DPS - Ridill with 25% Haste from gear, 35% Magic Haste, and a 5/5 Haste Samba - 64.4 (= 1.9*40/(236*.3/60) )
Overall, Ridill alone with /DRG would probably have higher Acc than Man/Ridill. Keep in mind that almost 2/3 of Man/Ridill's swings are made by Ridill.

Without Haste Samba it falls into the range where they're about equal damage-potential wise. Either way, one is always obviously more viable.
DPS - Man/Ridill with 22% Haste from gear, 35% Magic Haste - 45.1 (= (1.9*40 + 49*1.15)/( (236+276)*.8*.43/60) )
DPS - Ridill with 25% Haste from gear, 35% Magic Haste - 48.3 (= 1.9*40/(236*.4/60) )

Again, joke build, but I just noticed if you went 10 Acc/5 Atk and 2% Haste with 2 Dex from the MKE hat you could switch to Dusk +1 hands and have even more Acc/Atk.