THF In Todays FFXI

Treasure Hunter to me is nothing but a placebo effect, i've never seen real results from it and i've been running around with TH6 since Abyssea came out, and TH7 ever since we got Relic+2. Simply from the fact that i went 1/150 on some semi-rare stuff, almost going mad spending cruor on brews, procing each and every time only to result in no drops what so ever.

So disregarding that, you're dealing with an evasive knife wielder. I'm not saying it cant do damage, but on anything worthwhile, evasion can be ignored entirely. Most monsters in the game of worth can cap their accuracy on you regardless. I'd much prefer someone with a bit more oomph behind their attacks, smashing things rather than death by a thousand needle pins.

Just from a combat standpoint, feeding as little TP as possible, while dealing as much damage per hit as you can is the desired outcome, knives simply fall behind in that aspect.

Compared with the damage of others, sure, you wont fall behind that much. Most DD's come relatively close to each other actually if people were to do their best and try. Even PUP's.

Parses can be deceptive and simply cannot be used other than finding out your own damage, since if you knew something was being parsed, you'd do your best and ride your DD abilities and attacks disregarding safety and hate of the tank, while another DD that doesnt know he's being parsed might forsake using Berserk to not make the healers go through too much trouble.

In general, e-peens aside, Thief just isnt as much of a DD in general as any of the two-handers. You cant argue that fact. People want you for Treasure Hunter, maybe even Dagger procs, and that's it.

We must be playing an entirely different game or something.

Followup on the TH effect post I made earlier in this thread.

Did another THMode=None run last night to add to the previous run's data. Also went back and tried to do a more careful analysis of the previous runs' data, controlling mob count and coin drops more carefully (eg: making sure the mob count excluded kills during TE fights, or kills made by other parties, and making sure the coin counts excluded coins dropped from mobs outside the designated cycle, etc). I end up with the following results (along with number of mobs killed each run, to *** margin of error):

TH3 samples:
202 mobs, 2.86 coins per mob
190 mobs, 2.84 coins per mob

TH6 samples:
206 mobs, 2.88 coins per mob
193 mobs, 2.96 coins per mob
206 mobs, 2.97 coins per mob
202 mobs, 2.93 coins per mob

These runs all maintained a similar focus on procs (had other runs that focused on kill speed instead, but party structure was also a lot different, so not really comparable), as well as the same party setups (all thf/sam/mnk, though in a couple cases I did mnk while the other mnk did thf).

So I was mistaken in the estimations I provided in my original post. I was most likely remembering the quick estimates I do at the end of each run, which are subject to the flaws I corrected for here.

As a very rough estimate, it appears going from TH3/4 to TH6 raises coin drop rates from approximately 2.85 per mob to 2.95 per mob, or about a 3.5% relative increase.

Using a simplified drop rate model on the 4 coin slots, 2.85 coins per mob is a 71.25% chance per coin while 2.95 coins per mob is a 73.75% chance per coin, giving a 2.5% absolute increase. I would not be at all surprised if the actual rates were 70% for TH2 (or maybe TH3), and +1% per additional point of TH.


Regardless, TH6 does seem to be having at least a small impact on drop rates; technically still well within the margin of error, but game results tend to be fairly stable once you get more than a few hundred samples. Since there's no penalty in using TH gear in tagging actions, might as well keep it in place there.

One of the coins is 100% if you red proc, so that slot is unaffected by TH.

If you subtract 1 coin from your lists:
If you simplify the drop system down to 3 slots the same way you did above, you get a 64.5% drop rate with TH6 and a 61.7% drop rate with TH3, which is still consistent with Drop Rate +1%.

This is the system proposed by pchan here.

I know this wasnt the point of your data, but id say this also shows how little using thief's knife, in place of say Aluh or STR Thokcha, is really helping your final coin count. Most already are aware of this, but some believe thief's knife is really more efficient simply because you're after drops. Unless you have serious firepower in party already, to the point where youre mostly just tagging mobs, thief's knife has no place in dynamis.

Ya mean he's not just a crazy Frenchie who spouts out some weird ramblings all the time? I am shocked hehe.

I'd be more inclined to believe it's droprate*1.01^TH than droprate+TH% -- or is that what you meant?

The former doesn't mesh as nicely with what you guys said but I definitely wouldn't believe the latter, considering how that would impact stuff with low drop rates.

The model of TH I/II used/widely known was developed specifically using single, non-competitive drops (like Colibri Feathers/Beaks) and seems to break down every time multiple copies of the same item can drop.

It's very obvious that TH1/2 affect drop rates of multi-droppers (even a small sample of Lizards in Crawler's Nest will show you this), but it's unclear how they are affected.

I guess I should have added +3 to the TH levels in my comment explicitly, but that's what I meant.

It would be interesting to see what happened to the Forgotten Touches here too since those are single-droppers.

Yeah, if the information is readily available then I'd like to see the Touches as well. I re-worked bgwiki's TH page to incorporate the new information and generally make it more readable:
http://wiki.bluegartr.com/bg/Treasure_Hunter

Thanks to Merit Weaponskills, many if not most jobs have non-RME options that are really, really close to the RME options themselves, if not straight up superior. "DDs need RMEs because plenty of other DDs have RMEs" is not a necessarily faulty line of logic, but misses the obvious point: the hardest thing to do for 90% of the time in FFXI is to fill the goddamn alliance with the jobs you need, and if an available and willing player is going with respectable sub-NNI gear and an OAT GS on DRK but has no other job anywhere close to that, you bet your sweet *** you're going to war with the army you have.

I call BS! anytime. any place!

Forgottens for the same parses:

TH3
202 mobs, 47 touch, 23.27%
190 mobs, 47 touch, 24.74%

TH6
206 mobs, 51 touch, 24.76%
193 mobs, 45 touch, 23.32%
206 mobs, 49 touch, 23.79%
202 mobs, 48 touch, 23.76%


No measurable impact.

Also, the BG examples on the TH proc system have the difficulty notes reversed.

Edit:
Also, results for a TH0 run, if there's any interest in a baseline. Was purely for relic+2 xp, no thf in the group. Same mobs.

206 mobs, 29 touch (14.08%)

For the TH0 run, that's 1.282/mob with the 1 auto-drop removed already, correct?

edit:
If so then the drop rates pretty nicely fit a function where TH1 = +25%, TH2 = +12.5%, TH3 = +6.25%, etc. such that each level of TH is half as effective as the previous, and the maximum benefit of TH is +50%.

Actually, I shouldn't have even included the coin count since there were plenty of non-proc'd mobs. So, ignore that; can keep it for the touch results.

It's difficult to interpret the effects of TH on Touch drop rate without knowing the groupings, really.

DC Nightmare monsters in Valk can drop:
* Tome Pop item x1
* AF-1 (with a second slot for ToAU/WotG jobs, iirc)
* AF Accessories x1
* Currency x0~4

But I don't know which they can drop at the same time. Rare items are not added to the treasure pool if no one in your party can hold them, but there is evidence that they are still in the loot distributions (you can't force D-ring drops using Pixie Earrings). Drop order is a reliable indicator of slot, in my experience.


On a related note, my original impression from analyzing my city runs was that there were 4 Currency slots:
Slot 1 - Added with Red proc and 100%
Slot 2 - Added with Grellow/Blue
Slot 3 - Added with Blue
Slot 4 - Other stuff
Slot 5 - Always exists and unaffected by procs, around 10%.
So it might not be as simple as 3 independent slots.

He lost to my BLU. ( '-') YOU SHOULD MAKE A BET!

Groupings:

Forgotten is not shared with coins or accessories:
forgotten touch, one byne bill, one byne bill, one byne bill, one byne bill, valor cape

Forgotten not shared with TAU -1
argute mortarboard -1, forgotten touch, Ordelle bronzepiece, Ordelle bronzepiece, Ordelle bronzepiece, Ordelle bronzepiece

TAU-1/Accessory together:
forgotten touch, mirage keffiyeh -1, mirage mantle, one byne bill, one byne bill, one byne bill, one byne bill

Not shared with relic-1:
assassin's bonnet -1, forgotten touch, Ordelle bronzepiece, Ordelle bronzepiece, Ordelle bronzepiece

Relic-1/Accessory together:
bard's cape, forgotten touch, Ordelle bronzepiece, Ordelle bronzepiece, saotome kabuto -1


Have not found any instances of a forgotten dropped with a pop item, but pop items tend to get locked fairly quickly.


Drop order:

Coin [not the guaranteed coin]
Relic -1 (order vs TAU -1 uncertain)
TAU -1 (order vs Relic -1 uncertain)
Accessory
Forgotten
Coin
Coin
Coin

and

Coin
Accessory
pop item
Coin
Coin
Coin


So most likely shares a slot with pop items.

Well you also had brd/cor and I had squat <_<


Oh wait. I did get something. I got high-five'd into the ground multiple times.

They almost certainly re-worked the loot pools when they changed Dynamis, so perhaps they also combined ToAU/RotZ/WotG Relic armor -1 into the same slot. The drop system for that was very strange in the past, with ToAU/WotG -1s only dropping if an RotZ -1 dropped.

If it shares a slot with pop items, that could explain why the drop rate is increased less than expected.

Breakdown by number of coins dropped.

Data from one TH6 parse (202 mobs killed; seems one wasn't proc'd):

Matches the estimate pchan made.


The TH3 parses are more problematic. Seems there were a number of instances where I didn't receive the 'defeated' message, so the parser lists groups of 7-9 items per group (including ~7 coins). So, had to do a manual count to be sure I got all the proper groupings.


TH3/4 parse (202 mobs killed, so I missed a few somewhere, but not re-doing the count)

Doesn't match up quite as well, but it's not -too- bad.


Alternate:

Honestly, from the sound of it I'm once again doubting Thief's Knife has any use, anywhere. If you consider the two situations you'd have:

1. Thief is soloing/doing lower level content and can DPS
2. Thief is there for TH and nothing else

In the first case, the time you waste meleeing TH gear in an attempt to up your TH rate (Say, killing salvage bosses or something) is probably a bigger loss than you just tagging with TH6 and the meleeing it down as fast as possible.

In case 2, if you're going to be meleeing with the sole purpose of raising TH level you're probably feeding more TP and putting yourself in more AoE danger than it's worth, and if you're just RAing to tag TH does anybody seriously give a ***about a 1% increase on the base drop rate?

You could argue that a 1% increase is better than no increase at all, but at the end of the day, we're not mathematical equations, by which i'm asking if we are seriously going to run the same battlefield thousands of times before so we can see that statistical average increase of one extra drop? It just seems like such an absurd way to design the trait.

There are 4 drop slots for coins:

(Coin 1)
Other Crap
(Coin 2-4)

Hypothesis A - Coin 1 can drop with or without proccing and its drop rate is unaffected by proccing. Slots 2-4 are added by proccing and one of them is 100%. The other two have different drop rates.

Hypothesis B - Coin 1 is part of the proc system and the drop structure of procced monsters is more different from unprocced monsters than just adding a drop slot.



Do we have any data that can exclude Hypothesis A at the moment? My reason for proposing it in the first place is because of city Beastman NMs, which (iirc) never drop Singles when they're killed unprocced. It's like their "single slot" is the 100-piece and then they're otherwise normally affected by the proc system.

i could swear i saw forgotten + popitem drop together, but its long ago since i havent done dynamis outside of LS events since... months. Its so boring ._.

As far as I know, no. Since we know you can get a coin drop from an un-proc'd mob, it's most likely the first coin slot. However whether the drop rate is different from the other coins is a complete unknown. Would have to do a manual count of kills where you can be certain whether the first coin dropped or not (ie: only kills with non-coin drops as well). Lot of manual work, but I should probably be able to dig out a few hundred samples from my TH6 parses.

Probably also want a decent sized sample of mobs killed with appropriate TH, but without any procs, to see whether the drop rates were similar, or if procs seemed to affect the rate of that first coin. That'd be a fair bit of work that it would be hard to get someone to do.

I've never bothered checking until tonight but:

I don't understand your last point. If the THF is only going to be RAing the mob, even if thief's knife only increases the drop rate by 0.00001%, it still makes no sense not to use it. Why would you purposefully gimp yourself?

It's true that it will make an "increase", yes, but in a practical sense it's the same as if you didn't get the increase at all, is my point. This isn't like DPS where you swing millions of times over the course of the game and you actually end up seeing that average increase, this is affecting an already low drop rate of battlefields that would literally take years (decades?) of playing this game for you to run thousands of times, if you even bother doing that. Are you going run Legion the 1000 times it's going to take to see another duplus grip? Are you going to kill Dreyruk another 1000 times to see another Gusterion drop when it wouldn't have?

I mean, I'll probably still throw on my Thief's Knife because it's a "why not" scenario in that case, but it's just...such a mind boggling stupid thing to put in a video game it gives me a headache just thinking about it. Am I the only one that feels this way?

Also: There's the possibility that drop rate rounds down just like everything else in the game, so there's a good chance it's actually doing literally nothing when it comes to those low droprate items! Arrrrgh.

Um.. a question about the BGwiki TH page..:


Maybe i just missunderstand it.. but shouldn't the bolded parts be other way around? When you are have like a 0-level gap between current TH you wear and TH on the mob the upgrate is easyer, if the gap is higher upgrate or more "difficult".

Sorry, Motenten pointed that out and I thought he was just asking me to reverse the examples. Now it's fixed.