TH testing: final results

[mcb]

TH+ items in for the duration vs TH+1 items on for 1 attack then swapped out?

[Mojo]

I started an online effort, but got lazy because I wasn't receiving enough help.

Treasure Hunter Data

Anybody can edit this now, so if you feel like helping go ahead and put information into the spreadsheet. Take a look over it before you start adding shit though so you know how I had it organized. There were some interesting correlations made between "no drop" and "drop", which is why it's recorded the way it is. i.e., a mob has a base drop rate for dropping anything (say 1/3.) TH1 seemed to increase this by 1.5 and TH2 to 2.0 (so from 1/3 > 1/2 > 2/3) while TH2 also increased the number of drops within a drop (if that was possible.) There was speculation that TH+1 items increased multiple drop rates, but I don't really have any information on that. Either way, it may better explain oddities within the OP's data (inconsistencies between Treasure Hunter across mobs.) Maybe I'll put some more work into it at 90 cap since I'll actually have TH2 on a high level job.

[Mojo]

I'd be interested in your data, Thorny. Please do post it. Also, I thought I might add that the "no drop" "drop" correlation between TH1 and TH2 can be seen across all of my data. It's hard to tell with the OP because that type of information didn't seem to be recorded.

[Thorny]

I use logger and during all downtime kill lesser colibri for minor exp/beastmen's seals. This includes any character I can reasonably log on given #/pcs, so almost always have a character there. It seems people are interested, so when I get back home I'll pop my external in and have a miniscript count number of feathers/beaks/kills on THF char's logs vs mnk char's logs. I'd be interested in seeing further testing on things with varying droprates, but it should be a sufficient sample size to establish effect on droprates in that range.

[LinktheDeme]

Theory that my friend was working on that items had a max amount that could drop in a certain time interval. He started working on it when ASA came out and everyone heavily camped the zones. Nothing would really drop then but a bit after when the zones were less crowded the items would begin to drop. He found similar results when he went to crawlers nest with a thf and killed all the Exoray's in the room for mold

[Shenrien]

50%, mob either has it, or it doesn't, that and TH doesn't effect R.ex items.

Bring on the hate!

[FailureMidgard]

Just wanted to clear up, re: stacking. No matter what, this is irrelevant and people need to drop it, it's like any other job trait, you can't have TH2 without TH1. You can't have attack bonus 2 without attack bonus 1.

Attack Bonus 1: +10
Attack Bonus 2: +12
Attack Bonus 1+2: +22

vs

Attack Bonus 1: +10
Attack Bonus 2: +22

Doesn't matter how you look at it, because you cannot get 2 without 1, period. And a Drk/war doesn't get +10 twice last I heard.

[geno3302]

Just wanted to clear up, re: stacking. No matter what, this is irrelevant and people need to drop it, it's like any other job trait, you can't have TH2 without TH1. You can't have attack bonus 2 without attack bonus 1.

Attack Bonus 1: +10
Attack Bonus 2: +12
Attack Bonus 1+2: +22

vs

Attack Bonus 1: +10
Attack Bonus 2: +22

Doesn't matter how you look at it, because you cannot get 2 without 1, period. And a Drk/war doesn't get +10 twice last I heard.

Except you actually get two separate traits in your ability list "Treasure Hunter" and "Treasure Hunter II" not like all other traits where you get hidden increases. The general interest is in how the two work together, I'm not a THF so any THF feel free to correct me, one of the theories is that Treasure Hunter gives you a second roll while Treasure Hunter II actually increases the drop rate. The interest is not only how the two abilities function but also how the Treasure Hunter +1 items increase the abilities: Do they boost some mechanic with the first trait or do they increase the drop rate with the second trait ect.

[Byrthnoth]

I use logger and during all downtime kill lesser colibri for minor exp/beastmen's seals. This includes any character I can reasonably log on given #/pcs, so almost always have a character there. It seems people are interested, so when I get back home I'll pop my external in and have a miniscript count number of feathers/beaks/kills on THF char's logs vs mnk char's logs. I'd be interested in seeing further testing on things with varying droprates, but it should be a sufficient sample size to establish effect on droprates in that range.

It'll also be useful to see what the slot distribution is. Feathers (I think) have two slots they can drop in per Colibri, and Beaks have one. Seals have one (but a 5 minute timer). It would be a little hard to do (would need to have timestamped kills and write a more complicated script), but very interesting to see if TH affects seal drop rate.

[Nevex]

Did you check if Treasure Hound Kupopower was active during these tests?

[Klutz]

Inb4 Treasure Hunter is a lie.

[Tagus]

Omg, I spent an entire day of my life killing Bats for virtual science.

Hahaha... thanks for the effort man. I lol'd at that statement, mostly because of the familiarity most of us probably have with that kind of realization. Your sample sizes might not get you an A+ if this was a college dissertation, but for a video game the proportions of your percentages are enough for most of us!

[Enedin]

Did you check if Treasure Hound Kupopower was active during these tests?

Oh yeah, it was never up. There were days when I had a few hours planned the coming week to test TH, and then I saw Kupower show up. >.> ruined that lol.

@Tagus: thanks

@Thorny: looking forward to those Colibri drop rates. Now all we need is a THF to kill those off for TH2.

I'd be interested in your data, Thorny. Please do post it. Also, I thought I might add that the "no drop" "drop" correlation between TH1 and TH2 can be seen across all of my data. It's hard to tell with the OP because that type of information didn't seem to be recorded.

I started my testing on mobs that can't drop the same item twice, for simplicity. Some of those mobs had other items they could drop, which I didn't record, like Bugard Tusks.

On other mobs, like the Stirges (which were very kind to me and my results), I DID record all the drops, but I didn't note down any double drops.

And then there's the mobs you're looking for I guess: the mobs that only drop one item, like the... uh... I think. Not sure XD been away from FFXI for too long to know. Well, easy to look that up if needed.

[Mojo]

Perhaps you've misunderstood me. When I reference the "drop" "no drop" occurrence, it's not restricted to mobs with single drops. For instance, the Dhalmels I parsed could drop a wide variety of items. The base drop rate (that is, if the Dhalmel dropped anything at all) started low around 20-25% with a drops per kill ratio of .25. Adding TH2 (my sample size for TH1 is too small to remark on) put this rate at around 50% (the expected value) despite the drops per kill ratio ending up at .81. You seem to have organized your data on an item by item basis, thus that type of information has been lost.

[FailureMidgard]

Except you actually get two separate traits in your ability list "Treasure Hunter" and "Treasure Hunter II" not like all other traits where you get hidden increases. The general interest is in how the two work together, I'm not a THF so any THF feel free to correct me, one of the theories is that Treasure Hunter gives you a second roll while Treasure Hunter II actually increases the drop rate. The interest is not only how the two abilities function but also how the Treasure Hunter +1 items increase the abilities: Do they boost some mechanic with the first trait or do they increase the drop rate with the second trait ect.

And yet, as the OP said, you can't get TH2 without TH1, which makes this entirely a moot point. Which, as in my example, IF YOU COULD GET ATTACK BONUS 2 WITHOUT ATTACK BONUS 1, it would need to be tested also if in fact AB2 overwrites AB1, or if AB2 is just 12 attack. As it stands though, this is a stupid tangent that for some reason got multiple posts, and I'm saying it should be dropped, as the best thing it's doing is confusing people and making them think there's some merit to a true THF plus a /THF.

[Mojo]

Eating a Lucky Egg will give you TH2, but not TH1.

[Enedin]

Perhaps you've misunderstood me. When I reference the "drop" "no drop" occurrence, it's not restricted to mobs with single drops. For instance, the Dhalmels I parsed could drop a wide variety of items. The base drop rate (that is, if the Dhalmel dropped anything at all) started low around 20-25% with a drops per kill ratio of .25. Adding TH2 (my sample size for TH1 is too small to remark on) put this rate at around 50% (the expected value) despite the drops per kill ratio ending up at .81. You seem to have organized your data on an item by item basis, thus that type of information has been lost.

Oh yeah, it's just two ways of looking at it. Mine assumes that each item has a set %, while you say that SE coded it differently, on a "any drop" basis.

How does your system deal with one-item mobs? If it's the same as your multi-item mobs, then you can in fact use the one-item mobs' data in my chart, since there is no other data to be extracted from that mob, anyway.

So far, I'd like to think that I'm in the right direction, because of Miraun's calculations. But any idea has merit, and an idea might lead to another, like I triggered Miraun's idea.

Where's my Math-signal when you need Mathman...

[Miraun]

The problem Mojo is that with the potential method that I had posted on the first page is that mobs with multiple drops may drop dependencies that we're not aware of... It's a lot more controlled with a single drop item.

Taking something like a Dhamel for your example.
Up to 3 hides, 2 meat, Giant Femur, and/or Paprika Grass. (according to otherwiki... canyon)
Going off of the proposed structure that I had theorized on the first page:

For all we know, it could be a roll for a single hide drop, then a roll for a double drop, then a roll for triple, but only if the preceding roll succeeds does it continue with the subsequent roll. Or it could be three independent rolls on one hide.

The difference? The distribution of how many hides to get.

Under each hide being independent instances: (just assuming a straight 20% Drop rate)
51.2% chance of no drop
38.4% chance of 1 drop
9.6% chance of 2 drops
.8% chance of 3 drops

If there's some strange hierarchical structure of drops that are subsequently dependent, it could vastly change the results:
80% chance of no drop
16% chance of 1 drop
3.2% chance of 2 drops
.64% chance of 3 drops


Yeah, that doesn't even begin to accommodate for the data on the first page, just a concept... May also get a fourth roll only if you failed the first three rolls. There are just a lot of potential complications that make dealing with multiple drops more difficult to come up with a pattern for easily (assuming sample size less than a few billion)

[Jex]

Im with thorny on this one, i mean good work and this is cute and all but TH testing w/o TH4? really?

That and im sure most of us are not worried how TH effects random mobs around vana'diel

[Enedin]

Im with thorny on this one, i mean good work and this is cute and all but TH testing w/o TH4? really?

That and im sure most of us are not worried how TH effects random mobs around vana'diel

If you're not interested, no need to reply.