Random Question Thread XI: You're Wrong

[rog]

So does tp/hit go down on my weapons when I put on suppanomimi?

Yes, just like sword strap. Unlike haste, DW affects the base delay. Tp is calculated by the average delay of your weapons: ((x+y)*(1-dw%))/2, then just plug that into the tp formula, like you would for a single weapon.

[Byrthnoth]

So does tp/hit go down on my weapons when I put on suppanomimi? Im not entirely sure if thats what you meant by dividing the delay in half?

Like some people have told me if Im getting like 5.6 tp/hit on my mainhand and put on suppa, SE made it so I'll start getting like 5.4~5.3 tp/hit to make up for my icnreased attack speed, and that there used to be a floor on tp/hit after dual wield is calced and people would use that to their advantage by going over that floor and turning dual wield gear into haste, but Im not sure what to think of that.

I suppose I'll just have to test it myself by dual wield identical daggers and seeing if my tp/hit is going down after suppa...

Yes, your TP per hit will drop because your effective delay has dropped. However, that doesn't mean that your TP gain will decrease.

Here: http://wiki.ffxiclopedia.org/wiki/Tactical_Points
Look at the formulas below the "Base TP" heading. Those equations are what you would stick the "New Round Delay/2" number into. Right below that table there are three graphs. The middle one shows the relationship between single-hand delay and TP/second.

I didn't actually realize that Dancer gets Dual Wield IV at 80. That, combined with AF3 Neck and the possibility of Auric Dagger (please both be potent), means that OA2-3 dagger may not be nearly as amazing as I thought. We're going to need to keep a few different daggers around so we can balance our TP gain with our DoT to meet the party's needs.

[urat]

So now has anyone figured out by how much your tp/hit goes down based on dual wield, or does it vary % by % due to floors and what not so much its just more effective to find it empirically?

Or is it straight up 5% dual wield lowers tp/hit by 5%

Edit: Posted at same tiem as byrn, reading the page as I type

Edit again: Ah shit I totally get it now, I didn't know there was a direct corrolation programmed into the game of how much tp/swing a weapon has based on its delay.

I thought the tp/hit of a weapon was separate from delay and a variable on a weapon by weapon basis over at the SE database.

Everything makes sense now thanks a bunch guys :D

[Cesaria]

Quick one - anyone worked out the repop time of the pre-popped NMs in abyssea yet? Can't see it anywhere immediately from scanning through the specific threads.

[urat]

Quick one - anyone worked out the repop time of the pre-popped NMs in abyssea yet? Can't see it anywhere immediately from scanning through the specific threads.

It looks like about an hour

[tsola]

Quick one - anyone worked out the repop time of the pre-popped NMs in abyssea yet? Can't see it anywhere immediately from scanning through the specific threads.

Seems to have a window of 1-3hours. A friend mine told me it took Megamaw Mikey 1.5hours~ to repop for him the other today. Today I camped it and it took 2hrs 50 mins to repop.

[Onorgul]

Any Monks noticed a higher Counter rate? I just hit 79 and have obviously been using Perfect Counter (killing Puks for Wargfangs trials), but have noticed a pretty big spike in my Counter rate. It's all eyeballing, though.

[Phoenyx]

Any Monks noticed a higher Counter rate? I just hit 79 and have obviously been using Perfect Counter (killing Puks for Wargfangs trials), but have noticed a pretty big spike in my Counter rate. It's all eyeballing, though.

I'm only lv77, but I've racked up any perceived increase in counter rate to the fact that I've been soloing a significant amount more since the update than I did prior to it. YMMV, and all that.

[Francisco]

It seems about the same to me, eyeballing - but from time to time it seems higher.

Conclusion: Counter Rate is higher, except when it isn't.

[Onorgul]

I'm only lv77, but I've racked up any perceived increase in counter rate to the fact that I've been soloing a significant amount more since the update than I did prior to it. YMMV, and all that.

I specifically mean that in the time since I've hit 79, which was about an hour ago, my Counter rate seems to have jumped. It could just be statistical randomness, but when I was 78, I didn't seem to Counter much at all. I'm personally dubious that SE would have given us more Counter, considering how broken the mechanic is already, but I won't turn down an extra 5 points or whatever if someone can find it.

[Ocara]

Id the OT2-3 h2h don't triple on kicks, and using them with footwork is their purpose, is there any point in upgrading them from OAT until it is fixed/confirmed to not be broken? Do the triple kick procs work as double kicks thereby increasing double kick rate a little?

[Phoenyx]

I specifically mean that in the time since I've hit 79, which was about an hour ago, my Counter rate seems to have jumped. It could just be statistical randomness, but when I was 78, I didn't seem to Counter much at all. I'm personally dubious that SE would have given us more Counter, considering how broken the mechanic is already, but I won't turn down an extra 5 points or whatever if someone can find it.

Certainly worth looking into. I'll see if I can get a parser working while I'm working on my trials. I figure the number of mobs I'll fight from lv77 to lv79 wouldn't be insignificant so any increase should be easily detectable.

[Nightfyre]

Id the OT2-3 h2h don't triple on kicks, and using them with footwork is their purpose, is there any point in upgrading them from OAT until it is fixed/confirmed to not be broken? Do the triple kick procs work as double kicks thereby increasing double kick rate a little?

DA rate is indeed noticeably higher when using OAThrice.

[Darkmagi]

Aprox 75% iirc. Really massive DA rate.

[Motenten]

OA2-3 with Footwork seems to transform 1% of 3x procs into 2% additional 2x procs. With the 50:30:20 split, if the 3x were simply truncated to 2x you'd only have about a 50% multi-attack rate (before KA effects), whereas it seems to be more like 70%.

Onorgul: I was actually parsing myself for just this issue a few days ago (level 80). I refrained from using Perfect Counter at all for a while, just to see. With a few hundred hits taken it was mainly hovering between 11% and 14% counter rates (I have 5 in merits), so there didn't seem to be any real increase.


On the issue of accuracy vs haste, accuracy is easy enough, and already covered. Haste is penalized by JA/WS delay (just WS for most, Boost+WS for mnk) in proportion to the total unhasted delay between weaponskills (the longer such delay, the less impact WS delay has).

Imaginary setup:

You hit once per second.

It takes 20 hits (always), and thus 20 seconds, to get enough TP to weaponskill.

The weaponskill itself has a two second delay.


Now add 50% haste. You're attacking twice as fast (once per half-second).

It still takes you 20 hits to get the TP needed to weaponskill, which now is 10 seconds.

It still takes 2 seconds to perform the weaponskill.


Total damage of melee+weaponskill remains identical regardless of haste.


If you take any arbitrary window of time, are you doing twice as much damage?

Without haste, in 1000 seconds, without any haste you perform (1000/22) = 45.45 melee+ws cycles.

With haste, in 1000 seconds, without any haste you perform (1000/12) = 83.33 melee+ws cycles.

Therefore adding the 50% haste increased your damage by 83.33/45.45 = 1.8333 = 83.33%.


This is less than the basic haste calculation would imply, which is a 100% increase in damage for hitting twice as fast.

In this case, the base unhasted delay was 1200 (20 seconds * 60 delay/sec), which is actually fairly low. Most non-multiattack weapons will have a base delay of 2500-3000 per weaponskill, roughly, which reduces the amount lost. At 3000 total base delay, 50% haste would be an 89.3% gain for mnk, 92.6% gain for anyone that doesn't use a preparatory JA.


The formula I use for the value of additional haste is:

[BaseCycleDelay * (1 - OriginalHaste) + JAWSDelay] / [BaseCycleDelay * (1 - NewHaste) + JAWSDelay]

[rog]

Accuracy would be affected in the same way though, since hitting more often results in more ws/time. The penalty to haste would also be affected by acc for the opposite reason.

[Mercilessturtle]

Working towards OAThrice h2h for mnk, but what are the best magian scythes now? Im guessing OAThrice and full damage, but how does the sTP one stack up with a potential 5-hit? And how does Plaga scythe compare?

Oa3 isn't very good, assuming the 50:30:20 rate that has been reported. Comparing it to a Penitence:

111 dmg 528 delay vs 51 damage 528 delay Oa3
111 vs 51 is +105% TP and +62% WS
Oa3 gives you +63% TP, so you would need like +104% WS damage to make it equal assuming 50:50 tp:ws. I'm not sure how to work out the exact WS increase due to TP overflow, but I don't think it will be a 100% increase.
Plaga is meh, you can consider it the same damage as a str magian scythe during TP phase, but with a higher delay. That darkness damage would have to be pretty high to make up 20 attack and the delay difference during TP phase, and it does nothing for WS damage.

Assuming fully upgraded level 80 scythes the order is something like this:

OAT -> Redemption -> STP -> STR -> Penitence or double attack Maleficence

And while OAT has a substantial lead, the difference between redemption, STP and STR magian scythes is pretty small. The exact numbers depend on your TP:WS split, your pdif, and what gear you gave up for the 5 hit build on the STP scythe.

[tenkiki]

Anyone know the fast cast bonus for the teal set?

[urat]

On the topic of OA# weapons, do these weapons have a %chance per hit that tests for next hit when the previous returned a true?

In other words, is there a static %^hits that will tell us the chance that an occasionally attacks # weapon will attack n amount of times, IE for K club it could be something 50% so the chance of it attacking three times is 0.5^(3-1) (since Im pretty sur ethe first of the hits is guaranteed to proc XP)

Or is it a changing chance per proc, something like 50%-30%-20%-10% or something, or is it a static 50-25-12.5-6.25

Thing is I am working now on a program for calculating tp/m in C++, but it currently has no support for OAT etc weapons, it assumes the weapons attack once each.

If there is a static % for a weapon it would be a helluva lot easier to input a simple

cout << "Is your weapon a multi hit weapon? 0=no 1=yes";
cin >> choice;
if (choice == 1) {
cout << "Input maximum amount of hits per round the weapon may perform:";
cin >> maxhits;
cout << "Input %chance mod on hit proc tests for this weapon:";
cin >> hitmods;

then I can just do a calc like

chance of 5 hits in a round is = hitmods^(5-1)

But if its a changing %chance then, well, fuck. I hate using lists but that'd be the only way to deal with that.

Eventually I want the ability to parse the weapons as well based on input accuracy and weapon stats allowing you to get a more realistic comparison of weapons using a slight random mod.

[rog]

IE for K club it could be something 50% so the chance of it attacking three times is 0.5^(3-1) (since Im pretty sur ethe first of the hits is guaranteed to proc XP)

Kc is extremely far from an even distribution. Wiki has it listed as 5:15:25:25:15:10:3:2, which looks fairly accurate from my experience with it. And kc can most certainly single hit, it's just very rare.