First, let me just bring this all together for the sake of convenience. To skip to the new stuff, scroll down to the first bolded text.
Ok, I haven't read the new responses before posting this. I promise to get to them in time. I've been running some numbers and actually going out and testing to verify the formulas available on FF Wiki. So, this won't be a direct response to any post; but it's a mathematical argument, so it really does stand alone. I found the results enlightening, and I'll be completely forthright and transparent about the determinations.
First off, here are the sources. As I said, I've actually put them to the test, since you shouldn't assume that something on the internet is true then copy/paste it.
Regarding TP:
http://wiki.ffxiclopedia.org/wiki/TP
Regarding Delay:
http://wiki.ffxiclopedia.org/wiki/Delay
Regarding Dual Wield:
http://wiki.ffxiclopedia.org/wiki/Dual_Wield
Regarding Haste:
http://wiki.ffxiclopedia.org/wiki/Ha...atus_Effect)
Regarding Attack rating (pdif):
http://wiki.ffxiclopedia.org/wiki/Ma...l_Damage_Guide
If I have to go over every bit of the above links, this long post will be overly long indeed. So if I don't spell it out entirely, check one of the above to clarify.
I'm going to be comparing a THF/WAR with a Blau to an THF/NIN using the top-notch gear. I'll compare the Holy Grail of Thiefdom to what I'm sure many of you have disparaged or at least argued against, because I want to give YOU the utmost edge in your arguments. Assume that the only difference in gear is Blau +shield (I didn't even give a shield with bonuses) for /war and Blau/SK + Suppanomimi for /nin. (Again, I could have replaced the Suppa with a useful earring for /war, but I didn't.) Accuracy is taken to be 100% because they would have the same Accuracy, and it's a needless calculation to multiply both sets of numbers by an arbitrary percentage.
THF/NIN
Dual Wield adds together the delay of both weapons, then subtracts a percentage based upon the level of the Job Trait/enhancing items. In this case it's 20%, the best a THF/NIN can get.
178+150= 328; 328 x .80 = 262.4 effective delay
The TP formula isn't that hard, but there's a table associated with it; so I'll just post the result here as 4.5 TP per dagger hit. (It's 4.59 by the formula, but this is another time where the game *always* rounds down and so you lose the 0.09 TP.)
Weapon Delay/60 = number of seconds between attack rounds. Divide that number into 60 seconds to get the number of attack rounds per minute.
262.4/60 = 4.37 seconds/round; 60/4.37 = 13.73 rounds/min
Because you get 2 swings per round with DW, we have to multiply it by 2, getting 27.46. But remember the number of rounds/min, it's important later on.
27.46 attacks/min x 4.5 TP/attack = 123.57 TP/min
THF/WAR
Now that you have the basic idea, I'm going to speed up.
178 delay = 4.9 TP per hit
178 delay/60 = 2.97; 60/2.97 = 20.20
20.20 attacks/min x 4.9 TP/hit = 98.90 TP/min
Now that looks like an impressive difference, but looks are deceiving. Unfortunately many people stop at this point, but the story is less than half over. As they say, the devil is in the details. You forgot Double Attack (DA) and Triple Attack (TA). (I will not use TA to denote Trick Attack here.)
THF/NIN
DA and TA have a chance to trigger with the main-hand each ROUND of attack, not with each dagger swing. Ridil and KC work differently, but that's another story. So, when you slow down your attack rounds, you are gimping DA and TA even if only a little bit.
DA 0% = 0 additional attacks
TA 5% x 13.73rounds/min = 0.69 additional TA events/min; You get 2 extra swings with TA so, 0.69 x 2 = 1.38 additional attacks
27.46 + 1.38 = 28.84 Total attacks; 28.84 x 4.5 = 129.78 TP/min
THF/WAR
DA 10% x 20.20 = 2.02
TA 5% x 20.20 = 1.01
20.20 + 2.02 + 1.01 = 23.23; 23.23 x 4.9 = 113.83 TP/min
So THF/NIN wins the TP war with an extra 12.3% TP. So I was certainly wrong about it not giving any bonus, but the issue is of course damage. Can the extra Attack Rating make up for this? The answer happens to be, "Yes and then some," but I'll cover that at the end.
Adding in Brutal Earring (5% DA), Homan Body (2.5% TA), and 5 TA merits.
THF/NIN
DA 5% x 13.73 rounds = 0.69 attacks
TA 12.5% x 13.73 rounds = 1.72 events x 2 = 3.44 attacks
27.46 + 0.69 + 3.44 = 29.98; 29.98 x 4.5 = 142.16 TP/min
THF/WAR
DA 15% x 20.20 = 3.03
TA 12.5% x 20.20 = 2.5 events; 2.53 x 2 = 5.06
20.20 + 3.03 + 5.06 = 28.29; 28.29 x 4.9 = 138.62 TP/min
So as we add in some of that elite gear, DW still has an advantage at +2.49% TP/min.
OK, I'm finally back and I'm going to add in the haste gear. I'm going to do two different calculations, one with 15% Haste from the spell and no merits/exceptional gear (because that will reflect the situation prior to commonly available haste gear), and another situation with capped haste gear of 25% and the Haste spell.
Haste spell: 15%
THF/NIN
Delay: 328 x (1.00 - .20 - .15) = 213
TP/hit = 4.5 (unchanged by haste)
213/60= 3.55; 60/3.55= 16.9 rounds/minute and 33.8 attacks/min (base)
TA= 16.9 x 5% = 0.85 events; 0.85 x 2 = 1.7 extra attacks
33.8 + 1.7 = 35.5 attacks/min; 35.5 x 4.5 = 159.75 TP/min
THF/WAR
Delay 178 x (1.00 - .15) = 151
TP/hit = 4.9
151/60 = 2.52; 60/2.52 = 23.8 attacks/min (base)
DA= 23.8 x 10% = 2.38 extra attacks
TA= 23.8 x 5% = 1.19 events; 1.19 x 2 = 2.38 extra attacks
23.8 + 2.38 + 2.38 = 28.56 attacks/min; 28.56 x 4.9 = 139.94 TP/min
So, /war is 12.4% TP less than /nin. Surprisingly, this isn't much different than with no haste ( and no +DA/TA gear). Again, the attack bonus from /war can more than make up this difference.
Haste 40%: 25% gear + 15% spell
This will be done with merits and bonuses from gear to reflect an end-game elite THF.
THF/NIN
Delay: 328 x (1.00 - .20 - .25 - .15)= 131
131/60=2.18; 60/2.18= 27.52 rounds; 27.52 x 2 = 55.04 attacks/min
DA: 27.52 x 5%= 1.38 extra attacks
TA: 27.52 x 12.5% = 3.44 events; 3.44 x 2 = 6.88 extra attacks
55.04 + 1.38 + 6.88 = 63.3 attacks/min; 63.3 x 4.5 = 284.85 TP/min (Even if you don't like THF, you have to admit that's pretty damn cool.)
THF/WAR
Delay: 178 x (1.00 - .25 - .15) = 107
107/60= 1.78; 60/1.78= 33.71 attacks/min
DA: 33.71 x 15% = 5.06 extra attacks
TA: 33.71 x 12.5% = 4.21 events; 4.21 x 2 = 8.42 extra attacks
33.71 + 5.06 + 8.42 = 47.19 attacks/min; 47.19 x 4.9 = 231.23 TP/min
Here, /war gains 18.8% less TP than /nin.
Picking up on new material here.
Let's just take the last scenario above and use it for comparison of pdif. This is a bit hazier, since you really need to know what you are fighting to make damage calculations. Plus it is a pain in the ass. In one of the links had someone kindly figured out the increase in DPS for each point of attack increase against Greater Colibri and against Kirin. I'll use that.
With gear, a courel sub and 8 dagger merits, a Lv 75 Hume THF/NIN would have an attack rating of about 415 for his TPing set up. The same character as THF/WAR would have an attack rating of about 426 without any attack buffs up. Berserk will give +105 attack 60% of the time; 105 x .6= +63 on average over time. Warcry will give +22 10% of the time; 22 x .10 = +2.
So for the average additional attack above what /nin has: 11 + 2 + 63 = 76 attack.
Looking for 415 on the chart vs. G. Colibri we see that it's just above 409 and therefore each point of attack will give +0.257% DPS.
76 x 0.00257 = 19.5% over and above /nin
With /nin, you'll get a whopping 34% more dagger hits. Half of those hits will be with a dagger that has 20% less base damage than the Blau that /war is using. I honestly can't think of a perfect way of reconciling this, but the best I can think of would be:
34/2=17; 17 x .8= 13.6; 17% + 13.6% = 20.6%
At a 20.6% actual damage increase just from extra dagger hits, /nin can overcome the Attack rating advantage of /war. That doesn't even take into account the TP increase or the added damage from the SK. (But you should remember that an additional 1% of TP gain does not mean an additional 1% of DPS. For Warriors it's about a .5% increase, and a bit more than .5% for THF and SAM.)
However, that is just on a mob with relatively weak defense. Referring to the same table, if the mob checks just normal DEF (not high or low, pdif somewhere in the range of .6 to .7), /war's advantage rises to 29.9% from higher attack. I don't think this catches up either, but it's pretty close. If the mob checks as high def, then the attack give an advantage of 46.7%, which is certainly going to overcome the advantage of DW's faster attacks and even the added effect of the SK.
To summarize, I'll admit I was impressed at just how much of an advantage /nin could get at the high end. However, if you look back at the lower end of the gear and haste spectrum, you'll see a definite advantage for a warrior sub (and of course on high defense targets). As I said before, you can't let people think that they can just slap on a ninja sub and eat a meat dish and they'll be uber. It takes quite a bit of gear and merits to make that work. Haste gear and the SK dagger don't fall from the sky. I'd even say that thf/nin before you get access to haste gear is not the best option when talking strictly about damage. For those without that gear, or too low of a level to access that gear, /war is going to be superior in many xp situations. But if you do have the right gear, I admit it looks like /nin is going to be better more than half the time and at least viable the rest of the time.
Finally, DA and TA cannot process on both the main-hand and off-hand at the same time. I know Wiki says it can and that Alla has an old debate in the posts beneath the description of TA; but it is not true. I've played around with thf/nin even more than thf/war just due to soloing and farming. I've gotten NIN/WAR to lv 74 now. I'm not the kind of person who watches TV or goes afk while partying, I'm staring at the screen. I've never seen my character or anyone else get 4 attacks/round as nin/war or war/nin (other than with a Ridil) and I've never seen 6 attacks from a thf/nin. What you are seeing is back-to-back DA or TA, or you were just fooled by lag. Watch closely and you'll see that only the main hand gets more than one swing. Am I the repository of all knowledge? Hell no, but if you want to prove me wrong then fraps it and post it. But I think you'll spend the rest of your life making home movies of your character beating up EP mobs without getting the video you want.
