What's Better?

[CDF]

Accounting for double attack with multi-hit weapon skills: +0.02 average number of hits for every +1% DA increase

Example:

3-hit WS, 20% DA: 3.4 hits per WS (average), decomposed to 1 (first hit with fTP not necessarily 1) and 2.4 (fTP 1)
3-hit WS, 22% DA: 3.4 + 2*(0.02) = 3.44 hits per WS (average), decomposed to 1 and 2.44.

Obviously, given 0% DA, 3 hits per WS exactly.

Account for non-100% hit rate accordingly.

[Raen Ryong]

A critical hit at capped pdif is 35% damage increase on average for that hit. If you're really at capped attack, KJ shits all over RR. If you're not, it doesn't. Simple.

I'm curious - where does 35% come from? I could be doing something hideously wrong here, but:

Spoiler: show

Using new pdif damage equations:

New Capping Method of Attack/Defense Ratio

Prior to August 2007, capping of damage multipliers was applied after level correction and pDIF calculation. This allowed Level Correction to be overridden by high enough Attack values, and obviously was not the developer's intention.

After the August 2007 update, the ratio of an attacker's Attack versus the defender's Defense, Ratio, is now capped for the purposes of damage calculation:

* Ratio is capped at 2.0 for one-handed melee weapons.
* Ratio is capped at 2.2 for two-handed melee weapons.

Level correction is then applied to find cRatio, upon which the multipliers to find pDIF work. This entails that Level Correction will always have an effect on damage output.

Ratio is capped at 2.2 for 2handers BEFORE level correction.
=> level correction applied, -0.5 per level
=> this is your cRatio
cRatio is then plugged into the standard pdif formula,

Max pDIF has the following formulas:

Ratio Range Function Value
0 ≤ cRatio < 0.5 fMax(cRatio) = 0.4 + 1.2 x cRatio
0.5 ≤ cRatio ≤ 5/6 fMax(cRatio) = 1
5/6 < cRatio ≤ 2 fMax(cRatio) = 1.2 x cRatio

and Min pDIF has the following formulas:

Ratio Range Function Value
0 ≤ cRatio < 1.25 fMin(cRatio) = -0.5 + 1.2 x cRatio
if fMin < 0 → fMin = 0
1.25 ≤ cRatio ≤ 1.5 fMin(cRatio) = 1
1.5 < cRatio ≤ 2 fMin(cRatio) = -0.8 + 1.2 x cRatio

(there's a big problem with this which I've bolded: cap is apparently 2.2 for 2handers yet those formulae have bounds which do not include this. I'm going to make the broad assumption that they mean to extend that beyond 2 ... not that this matters for mobs above level 79).

=> and then final damage randomiser of 0~5%?

Criticals add 1 pDif with the following condition:

For melee attacks, a Critical Hit raises pDIF by 1.0, and is capped before the final 1.00-1.05 randomizer at 3.0, resulting in an apparent cap of 3.15. This is the only functioning cap on pDIF, and only encountered on Critical Hits.

Now doing some quick and dirty excel work and making one or two broad assumptions,

http://i37.photobucket.com/albums/e5...icalDamage.png

Using 1/(old pdif) as the expression for the increase in damage. Blue box indicates areas where this effect is "dampened" since the Pdif goes on to cap at 3 (for instance, 2.64 => 3.64 doesn't work.. it goes to 3, so it only gets 0.36 of the 1 Pdif boost. The blue box expression is instead (3 - Pdif_Max)/Pdif_Max).

Average is making a big assumption in that it's just averaging the two % increases in damage... this is assuming the pdif function selects numbers evenly throughout its range which is again an assumption.

Now I get the feeling I've done something completely wrong and going up the wrong track, but it seems that the per-hit increase in damage from a critical is much higher than your figure of 35%, favouring RR even more. Please could you point out if I've done something horrendously wrong?

Note: these figures are for when your Attack is capped; don't take them out of context!

[Thorny]

was a quick calculation for capped cratio(2.2)

seems like you're under the impression ratio caps before correction, wondering where the evidence for that is? Gekko numbers on kirin/JoL make it seem pretty clear to me that isn't the case..

[Raen Ryong]

I got it straight from wiki which obviously has inherent flaws; am trying to find the evidence behind it now. I remember reading some threads about it (think there were some on BG too iirc)?

EDIT: Could you show the Gekko numbers etc? If unexplainable, it'll very simply show that the model I was using above either doesn't work or I've misused it.

[Thorny]

running quick numbers with 150 str and hagun i'm getting an upper cap of 549 for gekko without a DA on kirin, so i guess the question becomes: can anyone produce a kirin zerg parse with dia3/angon/minuets to examine ws damage? Obviously substituting other mobs is fine, but Kirin is a pretty common mob to have zerg buffs on and has a drastic enough level correction for that to be obvious. If ratio was capped before level correction then damage would have a very very low maximum on those mobs and either I'm blind or I'm missing something.

[Raen Ryong]

If we are to presume said model above is true, I think that it would have to do with how you apply the "attack modifier" from weaponskills like Tachi: Gekko and such. I agree that it seems really low otherwise...

[Thorny]

If we are to presume said model above is true, I think that it would have to do with how you apply the "attack modifier" from weaponskills like Tachi: Gekko and such. I agree that it seems really low otherwise...

Can apply the same to penta in bahav2, I know for a fact I've seen over 1k on there. Penta removes the issue of attack bonus negating level correction, but raises the question of bahav2's level. I still can't see how, based on practical results, ratio could cap before level correction.

[Raen Ryong]

It'd be useful if we could get some recent WS figures so we can properly see whether the new "commonly accepted model" is in fact flawed. Since Greater Colibri are pretty well established to be Level 81/82, if there exists a singlehit WS without a pdif bonus (maybe something like Slice? People probably haven't properly tested this though), with zerg level buffs there should be a considerable difference between the old model and the new one (assuming the new one to be correct)!

[BroodIfrit]

WAR new to 75, just a few questions about a Rune Chopper build.
Assuming I can get the 26% Haste cap, what headpiece and body should I be using?
As far as headpieces go, I have Askar Zucchetto, Ares' Mask, Nocturnus Helm, and Maat's Cap. Bodies I have Adaman Hauberk, Askar Korazin, and Zahak's Mail. Also, with Grip, should I be using an accuracy grip, or a Claymore Grip?

[Nightfyre]

What are your haste options in other slots?

[Cream Soda]

For mnk on ixion, would it be better to go full str build for DK or would it be wise to toss some acc in there to make sure the second hit lands?

Was looking to swap genbu for 25 ws acc anwig and osode for usu body

Edit:
Also, Dune boots vs Kyoshu Kyahan

[Motenten]

~~~ stuff requoted below; this is kept as link to original post ~~~

Well, you're changing a whole lot of stuff in this -- race, gear, food -- but there's still some mistakes you've made.

He uses an average round till 100TP of...9.09, which is completly retarded.

It's including over-TPing using a naive formula, whereas yours assumes no over-TP at all.

Even with his assumptions: 88% accuracy, 5% DA, 15% Kick rate, 13 STP during melee phase, 12 during WS, I've got 8.44

With those assumptions, typical Asuran TP return will be 15.8-16.8. That requires 16 melee hits to reach 100 TP. Over-TP can be considered as: 50% of all hits after the first in the round plus chance of a second round landing before the weaponskill (high risk at high haste) which I'll model as 60/round delay (143 in this case).

That means 1.564 to 1.592 over-tp hits per weaponskill (2 to 5 KA merits), giving a full total of 17.564 to 17.592 hits. At 2.245-2.275 hits per round, that means an average of 7.82 to 7.73 rounds per weaponskill. Scaling for 88% acc means 8.89 to 8.78 rounds per weaponskill. 9.09 is reached with a fixed addition of 2 over-tp hits per cycle.

Also, for whatever reason, he doesn't boost with af gloves.

Boost is calculated from base attack, not final attack (as you seem to assume). Simple verification:

Get 400 attack. Boost with AF gloves gives you +75.
Add Berserk to reach 500 attack. Boost puts you at 575, not 593.
Use a kabob for +60 attack (460 total; remove str as needed). Boost increases your attack to 535, not 546.
Use Berserk with food effect active. Total attack will be 400+60+100+75 = 635.

In the case of the quoted comparison, kabobs were used on both (and thus must be removed as a factor), and Footwork likewise needs to be removed when calculating the Boosted attack.

you do not use focus in a crab sushi build

Not necessarily correct, but not too far off. Depending on gear, it may still be possible to add a few percent with Focus. You have an extra 3 acc as a Mithra, though, which cuts a bit out of the available adjustments.

Gear sets optimized for birds

You also changed body to Usukane, and have a 5/5 Usu build. That changes the balance a bit. You gain an extra 1% haste, a notable amount of extra accuracy, an extra fStr (partly because of the change of race) and enough Store TP to reduce your x-hit. The haste and reduced x-hit are most notable because the Footwork setup can't take advantage of them. Adding White and Black Tathlum to the setups along with Usu body means you've added an extra ~25 mil worth of gear just to shift the comparison.

The other comparison with Vegetto's Usu build already showed that it pulled ahead of the Footwork setup compared to the more generally available gear.

Damage per fist using cRATIO instead of pDIF:
with base crit rate of 12% with 29 dDEX, and =6% from destro.

Fist =(35+18+10)*(401/296-0.35+0.18 )=74.63
kicks=(35+0 +10)*(401/296-0.35+0.12)=50.61

Asuran Fists

Asuran=(35+18+9+13)*(526/296-0.35)*8.1*0.95=823.57

Slightly off, since the mob def is 295, not 296.

Fist:
35+18+10 = 63
401/295-0.35 = 1.009
(63 * 1.009 * .82) + (63 * 2.009 * .18 ) = 74.907

Kick:
35+10 = 45
401/295-0.35 = 1.009
(45 * 1.009 * .88 ) + (45 * 2.009 * .12) = 50.805

And for Asuran, after adjusting the def, average is 827.0.

number of rounds untill 100 TP:
with 95% acc, 15% kick, 5% DA, 21 STP during melee, 12 during asuran: 7.018 ( compared to his 9.09 that's +30% WS damage)

That assumes 0 over-TP hits. 15 hits needed to reach 100 TP plus expected .625+.964 additional hits (given 140 delay) means 16.589 hits per weaponskill. Average number of rounds is then 16.589/2.25/.95 = 7.761.

DPS=60*(74.63*15.005*0.9286+50.61*15.005*0.0714+82 3.57)/(7.018*348*(1-0.60))=117.78 damage per seconds, or 35,333 damage in 5 minutes.

You also seem to have ignored JA delays for Boost and the weaponskill itself.

Damage is:
7.61 * ((74.907 * 2.1) + (50.61 * .15)) * 95% + 827 = 2,019 per cycle
7.61 * 140 + 180 = 1,245 delay per WS cycle
18,000 delay per 5 minutes = 14.458 cycles
14.458 * 2019 = 29,191 damage in 5 minutes.

Compared to the original value of 24,876, it's a gain of 17.3%. Coincidentally, the simple comparison of Veg's Usu build was also a 17% gain relative to the quoted comparison.

Just noticed hsi assumption of focus merits instead of kicks ( bad choice of merits..).

There were no Focus merits assumed. Merits are +5 Counter, +3 Chakra, +2 Kicks.

[Meat]

For mnk on ixion, would it be better to go full str build for DK or would it be wise to toss some acc in there to make sure the second hit lands?

Was looking to swap genbu for 25 ws acc anwig and osode for usu body

Edit:
Also, Dune boots vs Kyoshu Kyahan

why not go sam?

[miokomioko]

Pretty sure Dune Boots > Kyoshu Kyahan always since Dunes essentially put DMG+30 on your feet... ?

[Cream Soda]

why not go sam?

You have to scan to get points for camping. I have no movement speed on sam so i scan on mnk/rng. Too much trouble to do a full swap on gear sets.

Pretty sure Dune Boots > Kyoshu Kyahan always since Dunes essentially put DMG+30 on your feet... ?

and kyoshu put 25, it's not that much of a diff, considering you add 20 acc and 20 attack into the equation

[Motenten]

Looking only at base damage, going from KKs to Dunes is ~3.25% (varying slightly with fStr and WSC).

Accuracy only matter for the second hit, so +acc is valued at about 1/3 its increase from base. As such, you'd need to get about +10% damage relative to base accuracy for the +acc alone on KKs to balance it out, which occurs when your accuracy without the shoes is about 86%.

The +20 attack works for both hits, so no extra modification there. To gain 3.25% on cRatio from 20 attack would mean starting out at (does a bunch of math...)

att = 615 + 0.6984d

an attack of 615 + 70% of the mob's defense. Basicly, not happening. Therefore if attack isn't capped, it will -always- provide more benefit than the extra damage of Dunes. Also, you'll get more than 20 attack while Footwork is up (~22 attack with the Footwork bonus alone).

Overall, I'd only use Dunes if you're capping attack and at 90%+ accuracy. You can get away with slightly less than cap, but it's unlikely that you'll know -exactly- where that point is on any given target. You can generally tell when you're solidly past that point, though.

[mdkuser]

Motenten: the rounds per ws I have take into account all possible retrurns of asuran fists, and also take into account TP overflow, it is an exact formula similar to what you can find for 5 or 6 hit builds for 2handers. I did not simply divide the hits needed till 100 by the average number of hits per rounds... ( although this may look like it for 21 STP builds, but that's not the case, because you need 14 hits, not 15, starting from max asuran return). I just list all possiblilities of DA and KA occuring and find an average number of rounds, then do the same comparison for each possible TP returns of asuran fists. For example , the average rounds needed after a max asuran fists is 6.84, and 14 hits are needed. IF you do the division 14/2.13 you get 6.57 which is a different figure.


A simplified way do the the rounds maths is to assume 100% accuracy. 8 hit asuran fists returns 17.4TP; In the 21 STP build ( 5.9 TP/hit) you need 14 hits to reach 100, and in his 13 store TP build ( 5.5 TP/hit), you need 16 hit. He explicitly says that the 16 hits will be 8 rounds, which is compeltly false; its actually somewhere between 16/2.25=7.11 and 16/2=8, randomly, and its average value can be computed:



Make a list of all sequences of DA and KA leading to 14 hits ( and 16 hits), there is quite a lot, then list the probability of each possibility ( its tedious, ive got it on a spread sheet); with 5% DA and 15% KA rate, I've got 7.41 rounds needed to get 16 hits, and 6.52 rounds needed to get 14 hits; That's a bit less than one round differences, as expected, since you have more than 2 hits per rounds on average.

Of course in practice it is more complicated as acc max is 95% so there can be infinite sequences, but the maths work out; and you do not just divide 6.52 by 0.95 to get the rounds, because you just never have the same number of hits per round, and that's what the alla guy is doing ( another approximation). For instance here, with 95% accuracy, and starting from a 8 hits asuran fists, with 95% acc, 15% KA, 5% DA, you need 7.78 rounds to get 16 hits, and 6.84 to get 14 hits. 7.78 is NOT 7.41/0.95, and 6.84 is NOT 6.52/0.95, the first digit generally coincide, but not the second, so it is a first order approximation (1% ish error).

[Chen]

You have to scan to get points for camping. I have no movement speed on sam so i scan on mnk/rng. Too much trouble to do a full swap on gear sets.

scan with SAM gear in inv, but on naked MNK/rng bar shoes?

[ZhiGaruda]

why not go sam?

MNK is a very formidable DD vs. Ixion, but only if you are also TPing on it (not sure why you wouldn't be)

It can definitely rock any other DD job in terms of DoT in a 10+ minute kill.

[Cream Soda]

scan with SAM gear in inv, but on naked MNK/rng bar shoes?

Like to keep my chakra gear on me at all times on mnk, would def be useful here if I actually claimed it, too.