Treasure Hunter Proc Rate Testing -TH 8 Versus 14 |
||
Treasure hunter proc rate testing -TH 8 versus 14
Offline
does more TH+ gear improve proc rate on hits with abilities like Sneak Attack, Trick Attack or Feint active? the wiki says they only have a higher chance of procing a TH up but not guaranteed?
Asura.Sechs said: » demanding an effort of 2000 hours of samples on a single one. It's not like I was doing anything else between dailies. Offline
Posts: 1109
Odin.Naytan said: » Would be better to test at TH9 than TH8. The current belief based on the one response from SE is having TH up to the current TH on mob is what you need to not receive negative effects to proc chance. If that is 100% accurate, a test on a TH8 mob with THF at TH8 or TH14 should have same results in long(long) run. The belief that the TH8 cap is the cap for everything and any gear that puts you over TH8 doesn't effect anything would need to be tested at at least TH9 to prove or disprove that SE statement, because any TH over TH8 should only matter up to the TH level that is on mob. Im not sure if im getting you correctly when you are saying "better to test at TH9 than TH8." I assume you mean on the mob and if i otjerwise get you correctlu what you are trying to establish is if having additional Character TH above the current MOB TH helps on proccing. of if additonsl TH on chracter only helpes when you TH is lower than the mob's AKA MOB TH 9 Does charcter TH 8 (under) vs TH 14 ( above) make a diffent Instead of MOB th 8 Does cracter TH8 (the same) vs th 14 (above) make a difference because in the later both character TH8 and TH14 equals the MOB TH and above and might just give the same. (you get penalized for being below but no bonus from being above) That is a valid point to test but i dont think we need wait for TH 9 proc to occur for testing. that is an unneededelays instead tag mob with TH8 switch character to TH7 ( below) and test agains character TH14 (above) Again we have a below vs above rate. but its faster testing as we dont need to wait for a TH9 upgrade proc I hope that made sense DaneBlood said: » Odin.Naytan said: » Would be better to test at TH9 than TH8. The current belief based on the one response from SE is having TH up to the current TH on mob is what you need to not receive negative effects to proc chance. If that is 100% accurate, a test on a TH8 mob with THF at TH8 or TH14 should have same results in long(long) run. The belief that the TH8 cap is the cap for everything and any gear that puts you over TH8 doesn't effect anything would need to be tested at at least TH9 to prove or disprove that SE statement, because any TH over TH8 should only matter up to the TH level that is on mob. Im not sure if im getting you correctly when you are saying "better to test at TH9 than TH8." I assume you mean on the mob and if i otjerwise get you correctlu what you are trying to establish is if having additional Character TH above the current MOB TH helps on proccing. of if additonsl TH on chracter only helpes when you TH is lower than the mob's AKA MOB TH 9 Does charcter TH 8 (under) vs TH 14 ( above) make a diffent Instead of MOB th 8 Does cracter TH8 (the same) vs th 14 (above) make a difference because in the later both character TH8 and TH14 equals the MOB TH and above and might just give the same. (you get penalized for being below but no bonus from being above) That is a valid point to test but i dont think we need wait for TH 9 proc to occur for testing. that is an unneededelays instead tag mob with TH8 switch character to TH7 ( below) and test agains character TH14 (above) Again we have a below vs above rate. but its faster testing as we dont need to wait for a TH9 upgrade proc I hope that made sense No this is wrong. You shouldn't test using TH7 because a mob with TH8 already will less likely going to be proc to TH9 by TH7 than TH8. SE already established this point. So whether TH8 is the TH cap and any equipment getting to higher level of TH would have no effect or not would be irrelevant in TH7 vs TH14 because TH14 could be counted as TH8 and would proc easier than TH7 Offline
Posts: 1109
Leviathan.Andret said: » No this is wrong. You shouldn't test using TH7 because a mob with TH8 already will less likely going to be proc to TH9 by TH7 than TH8. SE already established this point. So whether TH8 is the TH cap and any equipment getting to higher level of TH would have no effect or not would be irrelevant in TH7 vs TH14 because TH14 could be counted as TH8 and would proc easier than TH7 Then im not sure why you want to test on a TH9 mob to begin with if we are testing character TH8 vs TH14. Offline
Posts: 409
DaneBlood said: » Then im not sure why you want to test on a TH9 mob to begin with if we are testing character TH8 vs TH14. If it takes less swings (or not) to proc to 9 with TH14 than TH9, that would be something worth testing. I am not sure why it would be TH8 vs TH14, as TH8 should be taking more swings to proc 9 than TH14 per SE. Anyone get an add-on that can track this properly yet? Offline
Posts: 1109
drakefs said: » If it takes less swings (or not) to proc to 9 with TH14 than TH9, that would be something worth testing. I am not sure why it would be TH8 vs TH14, as TH8 should be taking more swings to proc 9 than TH14 per SE. Anyone get an add-on that can track this properly yet? Because that is what we are trying to test if its true.... and the test in OP indicate no to minuscule difference. However there was some valid issues with the test, that we want to fix for next test. In one sentence you state it is interesting to test if TH14 does better. Then the next sentence you declares it as we already know it is true. what is the difference here im missing ? DaneBlood said: » Leviathan.Andret said: » No this is wrong. You shouldn't test using TH7 because a mob with TH8 already will less likely going to be proc to TH9 by TH7 than TH8. SE already established this point. So whether TH8 is the TH cap and any equipment getting to higher level of TH would have no effect or not would be irrelevant in TH7 vs TH14 because TH14 could be counted as TH8 and would proc easier than TH7 Then im not sure why you want to test on a TH9 mob to begin with if we are testing character TH8 vs TH14. Why can't you test TH7 vs TH14? Because TH4 will 100% start at TH8 and the probability of reaching TH9 from TH8 with TH8 gear is greater than with TH7 gear based on SE statement. This is because TH with gear and trait is capped at 8 based of SE statement. The mob you attack with TH8 or TH14 gear will always start at TH8. Now, the probably of getting TH9 on a swing when you are TH8 or TH14 is in question. Hence the need to test at TH8->TH9 proc probably. Specifically, we need to know the probably of a swing at TH8 will proc TH9. Let's call this probably P(x) where x is TH gear+trait. From that we need to know if P(8) is smaller than P(14). We already know that P(8) is larger than P(n) if n < 8. There is only one way to test this reliably. You have to attack a mob at TH8 gear until it proc TH9 then kill it right after proc. Then attack a mob with TH14 (or any number higher than 8) gear until it proc TH9 then kill it. Repeat and compare results. The only way I can see this test can get easier is to do The/Bst on the snails in Ceizak then charm and release so they can repop with a lv 1 weapon. Of course, in an ideal world it would be better to test TH9->10 too because it will have a much better chance of showing larger differences between TH8 vs TH14. Offline
Posts: 409
DaneBlood said: » In one sentence you state it is interesting to test if TH14 does better. Then the next sentence you declares it as we already know it is true. what is the difference here im missing ? In the first sentence I am talking TH9 equipped vs TH14 equipped and if the number swings it takes to proc from 8 to 9. In the second sentence I am stating that SE already said it is harder to proc with a lower TH level equipped than what is applied to the mob. I may be wrong but I took this to mean you should have at least the TH level you are aiming for equipped (ie. Have TH9 equipped if your goal is TH9 on the mob). Leviathan.Andret said: » Of course, in an ideal world it would be better to test TH9->10 too because it will have a much better chance of showing larger differences between TH8 vs TH14. According to SE, proccing 9 to 10 with TH8 equipped should take more swings. I am not sure it would be worth the time just to verify if what SE said is true. Ideally we would test the number of swings it takes to proc from 8 to 9 at multiple equipped levels of TH (8, 9, 14, 20) to see if there is any benefit to having more than the applied level of TH equipped. Offline
Posts: 1109
Leviathan.Andret said: » Why can't you test TH7 vs TH14? Because TH14 will 100% start I corrected the quote to from TH4 to TH14 as i think that is what You meant. and yes for some reason this totally alluded me. Leviathan.Andret said: » There is only one way to test this reliably. You have to attack a mob at TH8 gear until it proc TH9 then kill it right after proc. Then attack a mob with TH14 (or any number higher than 8) gear until it proc TH9 then kill it. Repeat and compare results. But that is EXACTLY my original proposal that i made on the previous page and that Odin.Naytan replied to with "Would be better to test at TH9 than TH8. " Which is what i am trying to get a clarification on as his proposal does not make sense to me. https://www.ffxiah.com/forum/topic/55788/treasure-hunter-proc-rate-testing-th-8-versus-14/2/#3571031 But we know theres a hard cap for what the initial TH is, and thats 8, no matter if you're wearing 8 or 14. If you want to determine if the proc rate beyond 8 is increased when the equipped delta is greater than the current TH on mob, then maxing out TH gear would be the ideal scenario. You're not looking for what the proc rate is based on delta, you're looking for if there is any benefit to wearing TH14 if the cap is 8. Since you're trying to confirm if all this extra TH gear SE has been including is useful, the best course of action is to ramp it up as high as possible: average amount of swings to get to TH9 when wearing either TH8 and TH14 (or higher, I think I saw +19 was max). If the proc rate is affected by the TH-delta beyond the TH8 cap, then it should be discernable in the dataset. Once you have confirmed this at its most extreme scenario, then you can scale back and as-sess how much the delta affects it. Made up numbers inc:
If TH+ has a base proc rate of 0.25%, and each point of delta is a +0.25% proc rate on white damage (no SA, no TA, no feint), you need 200 swings to get one proc at TH8. If you have TH14, thats a delta of 6 (1.5% + 0.25%), which is 57 swings for a TH+ proc. Big difference and very noticeable if it works something like that. Unless of course I misunderstood the past couple posts, in which case disregard. Offline
Posts: 409
Carbuncle.Nynja said: » or higher, I think I saw +19 was max The current cap is 25 (22 from gear, 3 from traits).
Necro Bump Detected!
[30 days between previous and next post]
Offline
Posts: 133
For whatever this is worth in this discussion I know a Thf who uses max th gear, not sure what value he achievs but he farms dynamis Jueno for 5-6 hours a day and gets a LOT of 100 bills. Like 30+
Necro Bump Detected!
[956 days between previous and next post]
In light of the discussion in the Peach thread, I might as well drop my data in the relevant thread here instead of that thread which will eventually get lost to the abyss.
I ran 3 separate parses with a thief not doing any WS's to ensure no ninja TH+ procs. They were in a party actively killing things. Everything within is subject to sample size. I'm not running 50,000 kills per TH tier. The main goal is to find out "do I need any additional TH+ gear once I have my Skulker's Poulaines +3". All TH proc are total numbers within the chat log, not what level the mob died at. TH9 = TH9 TH10 = TH9 + TH10 TH11 = TH9 + TH10 + TH11 The first two were extremes to establish if overcapping has any effect: TH8 Ally DPS 6560 THF acc 96.93% (capped more or less) Kills: 626 over 6 hrs, ~34.5 sec per mob TH9: 353 56.39% TH10: 53 8.47& TH11: 2 0.03% TH13 Ally DPS 6513 THF acc 95.69 (slightly uncapped, I didnt add any acc that was lost from the TH gear) Kills: 693 over ~6h50m, ~35 sec per mob TH9 407 58.73% TH10 83 11.98& TH11 5 0.72% Based on this limited data, it does seem to indicate that TH+ procs are not subject to the TH8 cap. I then ran another parse with TH9 to see how the numbers look. Its established that the delta between THF TH and TH level on the mob affects TH+ proc rate. The expected data from a TH9 parse should look like so: TH9: similar TH10: slightly below TH13 due to the smaller delta of TH8 to TH10 and TH9 to TH10. TH11: should be somewhere between TH8 and TH13 but they dont really mean much considering were looking at sub 1% numbers. I would either need a massive sample size of 60,000 kills, or I would need individual parses where the mob is alive for 3+ minutes. Neither one interests me. Ally DPS 6571 THF acc 97.03% (capped) Kills: 852 over 8h30m, ~36 sec per mob (might have been a pulling issue, it shouldnt be slower) TH9: 465 54.58% TH10: 92 10.8% TH11: 4 0.47% The TH9 numbers are slightly lower for some reason, but the TH10 numbers did land where I expected: slightly below the overcapped TH13 but above the TH8 numbers. It seems like going beyond TH8 in equipment will have a positive effect on proc rate even beyond TH8. The philosophical discussion of is it worth it? Well, as the old FFXI adage goes: ***is situational. Lowmanning dynamis? Not worth it considering the DPS hit from fulltiming inferior gear. But then that opens up the question of why not bring a stronger DD and just rely on TH4 instead. Farming Malignance or Crepu or wave 2 bosses for Volte? You'd probably want to get that TH up as high as possible. Fenrir.Melphina said: » the average total attacks to get a treasure hunter proc. Offline
Any way you could run it on things with much higher HP to see if they proc rate changes much?
Nope.
We already know that the proc rate goes down when your equipped TH is below the mob TH level. I dont care to know the exact amount. We also know that TH+ from non-SA/TA/Feint attack rounds is rather low to begin with. If your goal is to raise TH, you'll max out the gear and atleast have it on for SA/TA. If you're happy tagging 8, you'll tag it and swap to DPS gear. Carbuncle.Nynja said: » The main goal is to find out "do I need any additional TH+ gear once I have my Skulker's Poulaines +3" Although since I can dig this up rather quickly, the four fights where TH11 procced in the TH9 parse lasted: 31 sec 29 sec 37 sec 40 sec So its not like they coincided with drawn out fights where the other DD's were whiffing their WS's. 2003 says you also need to have THF land last hit
Was there a reason you picked TH13 or is that based on the gear you have?
What was available on that character
it shouldnt affect the data though, from a proc standpoint. If the premise is true that the proc rate can only be lowered from its cap (TH on char is >= TH lvl on mob), then the test could have been TH10 or TH18, the rate in going from 8 to 10 in either set should be the similar given sufficient sample size. Needs more data. The only proper metric for measuring this is proc rate at each TH level, which means an addon should be used to track which level of TH each attack round occurs at, and whether that attack round included a TH proc. This allows you to build a proper sample that actually represents the data point in question (proc chance per attack round at a given TH delta).
When you change from TH8 to TH13, you're likely going to be dropping some damage or multiattack, and you didn't publish your sets. Since you can only proc TH on the first hit of an attack round, reducing multihits will increase your chance of a higher proc because mobs will last for more attack rounds. Since these mobs have very low HP, frequency of proc/mob is a very poor metric for attaining this data. Offline
Posts: 345
Thank you for sharing your numbers. It's amazing how people try to mystify TH even now.
Carbuncle.Nynja said: » Lowmanning dynamis? Not worth it considering the DPS hit from fulltiming inferior gear. But then that opens up the question of why not bring a stronger DD and just rely on TH4 instead. Farming Malignance or Crepu or wave 2 bosses for Volte? You'd probably want to get that TH up as high as possible. I thought the current prevailing theory is that volte drops were a D.Ring type of situation? The slot for Volte is guaranteed to drop something, and TH will not improve your odds to get Volte vs. currency. For Crep and Malig the slot is not guaranteed, so TH will have some effect in helping you get *a* drop, but no help at all in getting the piece you want instead of yet another earring. Shiva.Thorny said: » Since you can only proc TH on the first hit of an attack round, reducing multihits will increase your chance of a higher proc. I mean, he said there's a whole party full of people killing these mobs in < 40 seconds. I don't think the THF's auto-attacks are going to swing the kill speed very much. Carbuncle.Maletaru said: » I mean, he said there's a whole party full of people killing these mobs in < 40 seconds. I don't think the THF's auto-attacks are going to swing the kill speed very much. It's still not a good metric, there's too much variation present. If this is something you want to prove, you set up a single THF slow killing apex or something and make 2 samples. It's not that complicated. The data point that matters isn't 'chance of getting TH11 in a single fight', it's 'chance of proccing TH in a given attack round', and we have the knowledge to correctly measure that in 2023. Carbuncle.Nynja said: » TH8 Ally DPS 6560 THF acc 96.93% (capped more or less) Kills: 626 over 6 hrs, ~34.5 sec per mob TH9: 353 56.39% TH10: 53 8.47& TH11: 2 0.03% TH13 Ally DPS 6513 THF acc 95.69 (slightly uncapped, I didnt add any acc that was lost from the TH gear) Kills: 693 over ~6h50m, ~35 sec per mob TH9 407 58.73% TH10 83 11.98& TH11 5 0.72% DPS is within less than 1% of each other, there are 700 kills and almost 7 hours of testing. IDK, I don't see almost any variation here at all, and the results are very stark. A marked improvement in the rate at which TH10+ proc, and almost the exact same rate at which TH9 procs. IDK, if someone else wants to do more testing go for it; I'm convinced enough as-is. I'm not saying his conclusion is wrong, I'm saying his methodology is flawed. Are we assuming that extra TH only helps when the delta is negative? Because if so, delta is 0 going from TH8 to TH9 and he still has an over 2% change in TH9 rate, the same delta for his TH9 to TH10 rate. If it's just random chance producing that 2% difference, then what's to say that random chance on the TH9 procs didn't lead to greater numbers of TH10/TH11 procs due to higher opportunity to proc TH10/TH11?
It has always seemed likely that more TH helps, I'm not arguing against that. I'm saying that the data point being measured isn't the data point that should be being measured, and it's really not that hard to make an addon and properly test it (which comes with the added benefit of being able to report on proc rate itself and impact of TH levels as well). Shiva.Thorny said: » Needs more data. The only proper metric for measuring this is proc rate at each TH level, which means an addon should be used to track which level of TH each attack round occurs at, and whether that attack round included a TH proc. This allows you to build a proper sample that actually represents the data point in question (proc chance per attack round at a given TH delta). When you change from TH8 to TH13, you're likely going to be dropping some damage or multiattack, and you didn't publish your sets. Since you can only proc TH on the first hit of an attack round, reducing multihits will increase your chance of a higher proc because mobs will last for more attack rounds. Since these mobs have very low HP, frequency of proc/mob is a very poor metric for attaining this data. Then you do it??? Im not risking my ***running a bot for 72 hours straight to get an adequate sample size with perfect methodology. I was just trying to *** if wearing more than TH8 is doing anything or is useless. I dont need perfection. Never stays in my mind long enough to bother, and I don't have the time right now. I pointed out the same thing a page ago in this thread; nobody is measuring the variable that actually matters. Why would you go to the effort of testing something, but be offended at the idea of doing that test in a way that actually isolates the variable you care about?
You don't need to bot for 72 hours straight, a 6 hour sample with proper methodology would be worth more than 12 hours sitting in your ML party. |
||
All FFXI content and images © 2002-2024 SQUARE ENIX CO., LTD. FINAL
FANTASY is a registered trademark of Square Enix Co., Ltd.
|