Yoshi P talks about Warriors- from Producer Letter Live @ TGS

[HiirNoivl]

Here's something else that's interesting.

If you Raise your Crit Rate, everything has a Chance to Crit, this includes Fracture and Vengeance and your Auto Attacks. IF you use Vengeance on Top of Blood Bath then you will be Hitting Crits not only when you do damage but when you receive damage. Vengeance goes from being a 50 potency attack with Damage received to a 75 potency with damage received and that translate with more HP received.

It makes Bloodbath actually worth it.

Geez what about Storm's Path going from nothing to something, too. I never used Storm's Path because it was such a weak Cure... Dang, I may break it out for some Berserk action now.

I don't think you haters realize what you're missing. 24.78 was the crit rate I had over all, there were some spikes in there where my Crit Rate was up to 57%. Some other spikes where it was 27%... I bet that was with Internal Release.

This will result in the best garuda kill yet!

It already has! That was my first test! Where I went UP in DPS when I had gone down in ilvl on my weapon. I mean... Just blew my mind lol

This is a basic lack of understanding of game mechanics.

Base crit rate is 10%. Wrath gives 2% crit per stack. so Wrath V gives 10% crit. which is *gasp* twice the default.

This is wrong.

http://valk.dancing-mad.com/?page_id=202

According to this calculator made by people way smarter than me, if you don't add Crit Rate your chance of Critical Hit is 5.15 %

With all my melds and everything my chance of Critical hit is 12.63 %.

With Wrath V: 15.74 %

With Internal Release: 22.59 %

Yet I result in 24.78% Crit Rate on Average....

Why? I don't know. Hidden modifier?

[Onisake]

This is wrong.

http://valk.dancing-mad.com/?page_id=202

According to this calculator made by people way smarter than me, if you don't add Crit Rate your chance of Critical Hit is 5.15 %

With all my melds and everything my chance of Critical hit is 12.63 %.

With Wrath V: 15.74 %

With Internal Release: 22.59 %

Yet I result in 24.78% Crit Rate on Average....

Why? I don't know. Hidden modifier?

why? because the calculator is wrong...that's why. it's off by 4% >.>;; is it really that hard to figure out?

I trust data I collect only. unless you show me all the parsing data you have to show that that calculator is correct, i'm going to go with the data i've collected on my own time and the data i've seen from others. DATA not just a random calculator value that I have no idea if its credible or not.

The data he provides for his number generator is limited, at best. he admits he extrapolates his data. doesn't explain why, and just points to another website. Furthermore this data has been around since beta. I have not seen any updates to his data since that time. his damage formula is off of 4 data points... in otherwords: I don't trust it.

[CurlyBruce]

...

His data points were extrapolated because it's asinine to test crit rate at literally every value possible (341>342>343) He picked his plateaus and noticed it had a linear correlation which was corroborated by another persons separate test. He admits that his formula is most assuredly not the real formula but for what we need it works. The reason Hiir is parsing a different crit rate is simply due to margin of error in his sample size. He probably was parsing different Garudas or w/e encounters and didn't actually sit at a Training Dummy for 1000-2000 hits which is what you need to have an acceptable though not ideal margin of error in your results. Parsing my own results at a training dummy now coming up to 600+ hits and my crit rate within the predicted % from what Valk's calculator says it would be. Incidentally, it is also correctly calculating what my average auto-attack is with my current STR and DTR, so food for thought.

[Onisake]

...

Someone doesn't know how statistics work.

How many attacks do you think you get off in an average boss fight? in a 10min fight with 2.5s GCD you can get a maximum of 240 attacks. margin of error:20% means 200 attacks. Considering he said average crit rate i'm assuming he's collected a couple thousand data points. meaning i wouldn't expect him to be more than .5% off. meaning his crit rate is roughly 25%. Looking at the data he provides later, he has an 'average low' of about 17% and an 'average high' of about 32%. averaging those two gives 24.5%. nice to see someone who doesn't pull data out of their ass for once. Margin of error, his crit rate is 24-25%. Given real data is more likely to be on the lower side than the higher side due to human error and other facts (such as missing attacks), i'm going with 25%.

...

I wouldn't agree that crit rate > STR/det for overall damage. there is absolutely no question whatsoever, that your main stat for damage will increase your average damage way more than crit rate will.

However War's mitigation doesn't come from sustained damage. it comes from burst. given we can control these bursts somewhat, (Berserk, internal release, bloodbath, infuriate) I can buy that crit may be better for mitigation than raw str. However it's just plain false that crit > str for damage. The other war supplying data out damages you on average. but this is could be his DDs are not as good as yours. the mobs only have so much HP.

For short bursts, crit is theoretically better. War mitigates by healing in bursts. So I'll buy that crit may be better for mitigation than STR, provided you can figure out a way to consistently crit when you need to (Bloodbath is up, when you IB)

Also, Steel cyclone has ALWAYS been good. it's just not OP as F--- like it was in 1.0. 2xSC=>2x overpower is my standard opener for trash/herds of mobs. I don't worry about mitigation here as much as holding threat. The only ability in WAR's kit that is 100% utter trash is holmgang. Even unchained can be situationally used to good effect. BB, zerk, veng, unchained, infuriate, overpower spam is great for pulls of 6+ mobs. holmgang, I just...I have no idea when that ability is useful.

Right now, I'm stacking in all Vit (with the assumption that EhP>self healing), is it more viable to have maybe 50/50 in str/vit due to our skills relying on our dmg as well? Or favor Str even more than vit like you have?

So far, in my runs, I haven't had any issues (nor my healers) with keeping me up or with aggro (I'm only up to Garuda HM though, first kill last night and I screwed myself out of the axe, but that's a different story).

Hiir found something that works for him. you need to find what works for you. It's not just about individual performance, its what the people you are with are comfortable with.

I help run an end-game FC. that being said, we have a lot of new players who don't know their stuff yet. so we need to train them. if I carry too hard as tank, they dont' learn. so I go for VIT over STR on paper STR looks so much better than VIT. but in reality, that extra ~450HP is the 1 or 2 seconds you need to get healed. Can't count the number of times i barely survive a buffed Triumverate on hydra. That alone is enough for me to say VIT>STR. but later on when everyone is on the ball, STR may be better than VIT.

[Lhun]

*snip*STR may be better than VIT, holmgang is useless*snip*

Quick note on holmgang: Use it when bosses use moves that make them run to other players and ignore hate stack. Also, use it immediately when they turn to do a move on someone (the kind where they turn right back after), because it gives you a big block of time where he cant hit you, to heal yourself and take a breather. I sometimes stun, run behind him, and Holmgang. This increases the stun duration, essentially.

Partially. STR increases parry mitigation percent. High dex is important too, it increases parry rate, along with parry stat direct.

At 50, aim for the following stats. See my above posts for more information.

405 STR(363 if you can't gear to 405: Don't go over 363 in that case, put it in Vit instead. There's no difference between 363 and 404)
211 Dex. This is where the parry rate hits a minimum of 12% on dex alone, before the Parry stat is taken into account. This also helps crits, which ALSO can make Thrill of battle and Second wind, and Inner Beast all crit, which is a GREAT thing to have happen - have mantra and convalescence up when you do these. It also helps accuracy, which can make sure that stuns land.

Blood Bath is 25% of total damage as health, and doing more damage means you get more back. It does not stack with Convalescence, but it does stack with Inner Beast.

To be clear: I don't use feather foot.

Butcher's Block is by far the highest hate generation in the game.(same as RoH, 5x damage Except war does more damage, so it's the best one in the game.) Because of that, you don't need to do it as often, and you combo it with sunder anyway (which is 3x damage on top) - so make sure you're doing DPS abilities and other things while you fight - don't get stuck doing the same combo over and over.

I combo Convalescence and Mantra with Second wind and Thrill of Battle, alternating. I combo Bloodbath with Vengeance exclusively unless I'm in a pinch. I do a heavy>maim>storm>fracture combo once every 30 seconds. I'll often use unchained for this if it's an easier fight/midboss, and Infurate right after. It comes up often, but I'll generally save Infuriate for self heals or finishing him off with SC. I use Foresite during the off times, or if I'm waiting to stun. I'll usually stun, then run behind and holmgamg.

Creating macros for all this makes it a lot easier.

[Kitru]

Butcher's Block is by far the highest hate generation in the game.(same as RoH, 5x damage Except war does more damage, so it's the best one in the game.)

Actually, it's not.

First off, 8% is not "by far" (BB is 280 enmity; Halone is 260; 280/260 = 1.077). Secondly, Medica II is the highest hate generation in the game. It's not even a contest. Just *try* keeping aggro off of a WHM that's tossing out Medica II like crazy: it can't be done. The initial ping has an enmity modifier of .8 and the HoT is at a 1.0. Plus, it hits 8 targets and generates enmity on *everything*. A tank's worst nightmare is a WHM that doesn't know how to manage its enmity because, no matter what that tank does, that WHM is going to be tanking.

[Tronic]

Actually, it's not.

First off, 8% is not "by far" (BB is 280 enmity; Halone is 260; 280/260 = 1.077). Secondly, Medica II is the highest hate generation in the game. It's not even a contest. Just *try* keeping aggro off of a WHM that's tossing out Medica II like crazy: it can't be done. The initial ping has an enmity modifier of .8 and the HoT is at a 1.0. Plus, it hits 8 targets and generates enmity on *everything*. A tank's worst nightmare is a WHM that doesn't know how to manage its enmity because, no matter what that tank does, that WHM is going to be tanking.

When in tanking stance the enmity they cause is almost equal: 210x5 for BB and 208x5 for RoH, when taking into account dmg reductions.

[Lhun]

Actually, it's not.

First off, 8% is not "by far" (BB is 280 enmity; Halone is 260; 280/260 = 1.077). Secondly, Medica II is the highest hate generation in the game. It's not even a contest. Just *try* keeping aggro off of a WHM that's tossing out Medica II like crazy: it can't be done. The initial ping has an enmity modifier of .8 and the HoT is at a 1.0. Plus, it hits 8 targets and generates enmity on *everything*. A tank's worst nightmare is a WHM that doesn't know how to manage its enmity because, no matter what that tank does, that WHM is going to be tanking.

You need more warriors in your life. We're talking about single target emninity here, not group total eminity. This doesn't count at all.

Math: Damage Enmity = Healing enmity x2. Or, 1 point of damage = 2 points of hp healed - IN HATE GENERATED.

SE provides us a tool for this. Provoke brings hate to +1 of the highest hate generated person on the team. He should have INSTANTLY GRABBED HATE BACK, and kept it.

Butchers block is not "8%". How on earth are you getting that number? Butcher's block is EXACTLY 5x the damage dealt in emninity generated. Pld has a wet noodle and no dps potential. Warrior hits for MOUNTAINS of damage with BB, and it's stacked with sunder, which is 3x damage as emnity, RIGHT BEFORE the other skill is popped. Your argument is invalid.

[CurlyBruce]

...

Funny thing about data points is you can't just use 10 groups of 200 samples and call it 2000 samples. He keeps posting his Crit Rate and DPS against bosses which aren't alive for more than 10 min, but why doesn't he just go smack on a test dummy for a while. If he really wants an accurate sample of his average crit it needs to be in a controlled environment where no other factors come into play. Go to Coerthas, Whitebrim. Smack on a level 50 Test Dummy for 42 minutes (2.5GCD * 1000 = 2500s / 60s = 41.6min). You'll get at least one auto-attack between each ability so that will give you a sample size of around 2000. Initiate the fight with Infuriate and spam your butcher's block combo (this is to keep wrath at full and has the added benefit of testing whether the damage calculator is accurate by not diluting the test with Maim bonus or Storm's Eye). Spam Internal Release whenever it's up. Post parse results.

For the record I parsed my crit rate without Wrath V or Internal Release with just 30 extra crit rate and got 7.22% which matches the Valk-Dancing Mad calculator. As far as crit rate from the actual stat goes it's accurate. The only thing that remains to be seen is whether % Crit Boost from abilities is a flat bonus or a multiplier. Judging by the fact that other abilities that increase a stat by a % (Berserk/DivineSeal/Foresight/etc.) actually update the character screen to show the new stat I would argue that the % Crit Boosts are a flat boost which is why the calculator is wrong when you add those in. Of course it's all just conjecture until someone parses it properly.