[milich]

think for a second. if focus skews results by having a lot of ACC sometimes, and not much ACC other times, but WS ACC looks to be capped, how is focus relevant? it's not.

the presence of a well constructed hypothetical test does not excuse irrational questioning/critique. respectable analysis does not consist of "lol there's variables, this must not prove anything". find real variables and explain why they're relevant, and boom, you have something to contribute.

edit: and please stop overstating the importance of nagamaki's test. he himself wrote directly after the results that his sample size was too small for him to draw any conclusions (same for his test on AF). i agree that my not-even-a-test-but-analysis-of-a-long-party-parse shouldn't have its importance overstated either, but that WS dmg comes from somewhere. i don't have hidden effect "defies dmg equations and amount of ACC to bust crazy asuran fists on greater colibri and mamool".

[Tyche]

Focus is relevant to your melee ACC, ebcause that is what you are using as a comparitive baseline. Maybe focus was up for half of your WSs, and was only up for 40% of your melee damage. You don't know, hence, variable. You jsut seem to fail to comprehend the variables. The questioning isn't irrational or irrelevant. "Respectable testing" does not consist of multiple variables that can skew results in any way possible. Prevelant "real" variables have been pointed out. You know how to run this test, and by your own admission, you don't wish put in the time. Why are we still having a conversation? And I'm not overstating his testing, im stating his testing is better constructed and more thought out that yours. Forgive me for taking yours with a grain of salt and trusting his.

Edit: Im not dismissing the results, you know that. You even said i wasnt critiquing the results, but the testing procedure. His testing is a hell of a lot more accurate, and isnt variable filled. That doesn't mean his is correct and your's isn't. It just means he had a better constructed test. That's what I've been saying this whole time. Construct a better test and come back with results.

[milich]

well, first of all, you're choosing to trust someone who said "don't trust this, it's just a start". and further, you're ignoring the fact i would need to focus *100%* of my WS in order to get capped ACC from a base of below 87%.

the TP mod would destroy my analysis if it were that powerful, but we all should know that it doesn't, since we haven't been storing tp to 150% and taking off all our ACC gear. if asuran fists could crit (it can't; i've done nagamaki style tests to show that in fact), it would also destroy my analysis.

anyway, it's not important to me that you or path agree with my testing, or come to comprehend the folly of what you're saying about these variables. clearly, my analysis has no use for you, which is fine. i'm sure your AFs hit hard enough. i won't continue this getting-to-be-silly argument unless i do tests along the lines of those you require.

if anyone else would like to discuss the parse i posted and whether or not it means what i think it does (without overestimating the relevance of focus and AF's TP mod), i'm interested in that. i'd also be interested if someone has information about asuran fists' TP mod that destroys my analysis.

edit: i made this post before your edit. i think i overestimated your confidence in nagamaki's results. still, as i tried to get across in this post already--but in case i'm not clear about it--i respect the disagreement, and don't see any reason to continue arguing. you clearly understand that (if not why) i don't acknowledge the variables you're presenting, and i understand that you require a better set up test. i'm only interested in all this to see whether or not i should keep gorget on, so the only reason for us to go back and forth further would be that a tighter test has been performed.

[Tyche]

He never once said "Don't trust these numbers". He said "I'll let people interpret the numbers. The margin of error was also calculated for that test. When the MoE was completely shifted toward the favor of the gorget, it was at that point that it on par with the PCC. The chances of the MoE actually being shifted that far over are slim to none. This goes without saying that someone also adjusted the calculations to support the theory of it only affecting the first hit, and naga's testing was spot on with their calculations. This also goes without saying that the raging fists/sidewinder test wasnt the only test supporting the gorget not working for multi hit theory.

So yeah, i'll interpret the numbers as naga said, and i'll choose not to completely ignore the data, and certainly not dispute it with an uncontrolled test.

[milich]

well, if my WS average is skyrocketing due to magic influences, i'll keep gorget on out of superstition then.

[Yhelothur]

Enjoy your unnecessary macro swap.

Get specs/PCC at least, jesus.

[Wrish]

Unnecessary stalemate here. The critique should have been over the complex interpretation of WS damage formulas and assuming they would be accurate to predict AF damage. Clearly, they predict your average AF damage with 100% accuracy as under your measured AF average, so something is amiss. While comparison to past WS accuracy could be relevant, that is hampered by eyeballing and recalling from memory... and possibly by other changes to gear and party setup.

If I were you I'd simply replicate the parse with an alternate neck piece in the macro. Same party, same macro swaps otherwise. That's the easiest way to exclude any defects or miscalculations in damage formulas, and it's more applicable to the rest of us. We'd all be interested. Maybe it'll disprove the accuracy limitations... perhaps even the fTP one.

TP, focus, and even warcry should balance out between long parses so long as the players maintain reasonable focus. Be sure you record as many factors as possible though. Recording the average AF damage is the easiest part because parsers automate it.

[milich]

Enjoy your unnecessary macro swap.

Get specs/PCC at least, jesus.

i'd be more hurt if i thought you understood the OP.

---

wrish,

mm, i should do some parses without shadow gorget. the dmg equation is accurate (that's a given), and my interpretation of it isn't particularly complex... anyway, i mentioned eyeballed pre-gorget ACC only in relation to the absurd claims about AF's TP ACC mod, and tangential autobiographical stuff about why i believe this or that (ie my eyeballed pre-gorget ACC has nothing to do with the parse or its analysis).

---

i'd still appreciate more parses/data from anyone else (say, for example, some 85%ish ACC or lower parses, your equips, and your dmg/hit and dmg/WS), or further information as to how to project ACC from average WS dmg. ideas for why my data are not reliable are OK too, as long as they aren't "such and such varies in such a way that clearly doesn't speak to your results, but i'm sure it magically invalidates your analysis."

[ZahlSechs]

I think we need Aurik to come and save us all
I always thought gorgets gave a % bonus to acc, noticeable, and a very little bonus to damage too.
Do they really apply only to the first hit?
When were all those tests I read long ago concerning PCC and Faith Torque losing to Ele gorgets wrong? Or is this one wrong?

To get real tests for monk anyway you would need to:
1) be sure you don't get any buff
2) don't eat food
3) use the same equipment
4) use raging fists (acc doesn't vary with tp)
5) never use focus or any other abiloty
6) avoid /war (we don't want double attack)

This way you can almost always tell how many hits land (except maybe when you get hit at the very same exact time you use WS).
I wouldn't be surprised if Ele gorgets are broken and work only on first hit, seeing how many broken things SE created, I woudl rather be surprised by how nobody ever noticed, or rather demonstrated the opposite, all this time.

That would probably mean that the difference is so little, that in the end it doesn't matter that much what you use. But sure, if we find out they do not work, it will just be Inventory space +2 for me, since I would just keep Faith Torque up all the time...

[Yhelothur]

milich :

Enjoy your unnecessary macro swap.

Get specs/PCC at least, jesus.

i'd be more hurt if i thought you understood the OP.

Talking about yourself in the third person still doesn't make you right.

Read the thread on BG. If you don't believe it, then get the hell outta here.

I noticed a HUGE difference with PCC, and my soil gorget sits in mog house unless I'm farming mobs that are lower level. Then it shows improvement. That's about it.

[Tyche]

I don't see how replicating the parse with a different neck peice is going to help. You aren't going to get better results since ACC for WS was seemingly capped. Still going to have variables, no matter how small they may or may not be, they'll skew results. Zahl pretty much laid out the test. If you are looking for info on the asurans ACC mod and/or the speculated WS ACC being calculated differently, then that would be easy to test too. Until you know exactly how TP effects AF, AF is going to continue to be completely useless for this test. What if at 100% TP you get a 10ACC bonus over your melee damage? I haven't the slightest idea how it effects it, admittedly, neither do you. Too many unknowns for they way you want to conduct the test. You've spent an ample amouunt of time refuting what people have pointed out to be problematic with your parse. Apply your efforts in a more productive way and do the test the correct way.

[ZahlSechs]

Yhelothur :
I noticed a HUGE difference with PCC, and my soil gorget sits in mog house unless I'm farming mobs that are lower level. Then it shows improvement. That's about it.

As much as I apreciate your and any other useful post adding more comments and input to this big question, let me remind you that over the last months/years we've seen countless people posting about how they noticed "huge" differences using elemental gorgets over PCC or Faith Torque.

Everyone in KI and other forums was posting with the usual attitude "Gorget > PCC > Faith, if you don't understand this u're teh n00bz gtfo".

Which is actually the same thing which, up to a certain degree, is happening now but in the opposite direction.
What I'm wondering is: how is this even possible?
I don't want to say that this new discovering isn't true, it might be very likely that Faith/PCC > Gorgets, I'm not completely against believing this.
I'm just saying that what's happening is indeed very bizarre, and either people have been fools all this time, OR the difference (whether Gorgers or PCC/Faith are the winners) is very little and not "huge".

[Tyche]

In some cases, hedaches can be cured with water pills if the user beilieves it's headache medicine. It's called a placebo effect. Point to one test from when the gorgets were first released that supported them in a multi-hit environment. I don't recall ever seeing any for multi-hits, just a lot of people eyeballing. Read the entire thread, there's probably four or five total, extensive tests for the gorgets, all of which supporting single hit WS, and none of which supporting multi-hits. All the numbers are there for interpretation. Mob level, mob vit, pdif, etc. Read the whole thread, i recall you said you skimmed it in a different thread. Why are people so afraid of change?

[ZahlSechs]

Not afraid, just suspicious.
I read the whole thread btw, and yes, I'm definitely changing my mind. I'll be keeping my Faith Torque on from now on when using Raging Fists and Asuran Fists, will probably still use gorgets for like... Howling Fist and Dragon Kick (not like I use them often lol). Maybe I'll just let my Soil and Thunder gorget in MH for Inventory+1.
I don't have anything else to say, other than suggest everybody to go there and read it, it's pretty much explained in a good way, hard to argue.


No neckpiece
229/300, 76.33%

Thunder Gorget
233/300, 77.66%

Peacock Charm
259/300, 86.33%


Basically Gorgets REALLY work and REALLY are better (damage and acc wise) than anything else, but that bonus applies ONLY to the first hit. So PCC and Faith Torque are better for our Raging Fists and Asuran Fists.
Gorget are, probably, better when you use all other 2hit WSs like Dragon Kick, Howling Fist etc.
Sorry if I behaved suspicious, it's not that I didn't want to change my mind... I just like to understand things on my own, rather than believing others on their words.

Kudos to Nagamaki, and thanks Tyche for linking the thread.

[Wrish]

Yhelothur:
I noticed a HUGE difference with PCC, and my soil gorget sits in mog house unless I'm farming mobs that are lower level.

Tyche:
I don't see how replicating the parse with a different neck peice is going to help. You aren't going to get better results since ACC for WS was seemingly capped.

We need more scientific statements here or there will be silly contradictions. If Tyche's claim that accuracy is capped is true, then it would be impossible for PCC to make noticeable difference, and vice versa with the counterpositive. Additionally, Tyche's claim of capped accuracy can't be good for PCC, can it? If your accuracy is capped, then clearly a spike necklace, chivalrous chain, or elemental gorget would all be better in that case. But there's in fact no evidence that accuracy is capped; it's only a suspicion.

Perhaps Tyche is implying there shouldn't be madrigal because accuracy is already at 85%+ and the TP mod could bump that up to cap.

Still going to have variables, no matter how small they may or may not be, they'll skew results.

However, only a variable that can be shown to discriminately skew results one way and not the other can be a valid complaint against a parse of any length. The complaint for a short parse is merely small sample size. 200 battles is 200+ AFs, a statistically relevant size.

I understand the worry over TP and concerns over the experimenter. If a mob is near dead, then one could just turn their back while it's finished and then unload AF shortly onto the next mob.

milich:
the dmg equation is accurate (that's a given), and my interpretation of it isn't particularly complex... anyway, i mentioned eyeballed pre-gorget ACC only in relation to the absurd claims about AF's TP ACC mod, and tangential autobiographical stuff about why i believe this or that (ie my eyeballed pre-gorget ACC has nothing to do with the parse or its analysis).

The problem is that without the eyeballing, we're left with a damage formula that clearly contradicts reality (predicts lower maximum damage than observed average, perhaps because pDIF is misapplied) and a weapon skill with unknown accuracy. The damage equation can't be taken for granted if it already mispredicts what happens.

further information as to how to project ACC from average WS dmg

Well, we could always macro out TP and HP before/after WS, simultaneous with a running parser, and simply discard the TP results where HP took a hit (and merit mobs hit hard enough).

But there is really no substitute for comparing parses between the same character, same player.

[ZahlSechs]

Wrish :
If Tyche's claim that accuracy is capped is true, then it would be impossible for PCC to make noticeable difference

Concerning this, there is another post (in the same thread on BG) talking about accuracy and possible "caps" and "diminishing returns" from a certain value on. I suggest you to give a look to it.

A little extract

Nagamaki:
Finally, one thing about "diminishing returns". It doesn't seem to apply in the effective accuracy sense in tests so far, although like I mentionned some more high level tests are needed to properly confirm the theory there. However, effective accuracy in itself does have diminishing returns the higher it gets, and obviously any time you have more accuracy than the amount that gives you 95% is, that accuracy is wasted and replacing it with a different stat would be more useful. Even below cap though, while +10 acc will seemingly always add 5% to your effective accuracy, the increase in DoT caused by that 5% will depend on what your effective accuracy was at beforehand. If it was at 50% then going to 55% is a relative boost of (5/50) 10%, while going from 90% to 95% is a relative boost of 5.55% (5/90). Its important to make this calculation if you're looking for how much a certain amount of accuracy will boost your DoT, as opposed to just directly going with the effective accuracy amount.

Copyright Nagamaki™

[Wrish]

The cap for accuracy is a game-universal constant slightly past 95%.

The usual context of diminishing returns is a steadily decreasing marginal return with a limit approaching zero. The damage versus accuracy graph does not resemble this. Marginal returns decrease to a limit of marginal accuracy * 1.05 before abruptly dropping to zero. Before the drop-off point, it can hardly be said that accuracy isn't worth it, as an additional 1% accuracy still translates to greater than 1% damage bonus.

However, in a comparison with other factors as haste and attack, this concept would certainly be applicable. The way I like to put it is that the lower your base accuracy, the more important it may be to wear accuracy instead of haste/attack gear in a given slot. That was the initial support for the people using sushi when it came out, and you now see the empirical backing for it.

[Tyche]

Re: When i said his ACC was capped...through the OPs deductions playing with the Pdif averages and what his average asurans should be, he deduced he had capped ACC on his WSs with the gorget. I was just saying that if that were really true, and he wanted to test PCC, the results wouldnt differ. To get an accurate test, you need to be below cap because you may not be exercising full potential because of limitations. That's why naga used NIN/MNK to run the tests, NIN H2H skill isnt that great.

[aurikan]

I'm deliberately not reading this thread.

There's some evidence that shows gorget acc could only affect the first hit of a multi-hit weaponskill. From what I've seen it's not conclusive, because of the sheer amount of data require to shrink the error margin.

I still use my gorget, for now at least.

[Azerath]

That's a lot of reading in those links, took me a while. It's nice and all that people are going through such tests and all but, don't they have a life or at least something better to do? I mean they are going through such trouble for what? To find out how something so miniscule and insignificant. How much of our total damage is provided by our WS'? Is there such a significant difference between PCC/Faith/Gorget to make people spend countless hours performing boring tests? I would assume we all would say "No." Using either of the three will NOT make or break you in any situation.

Yhelothur
Enjoy your unnecessary macro swap.

Get specs/PCC at least, jesus.

lol I find it so funny that you have to jump at every oppurtunity you get at letting people know you have a PCC. From making a thread, making several posts and making references that you have one. And didn't you only get the PCC cause your friend left you gil and X number of items? lol... *slow clap*

Yhelothur
Read the thread on BG. If you don't believe it, then get the hell outta here.

I noticed a HUGE difference with PCC

Haha I would wait to see more evidence to get a clear result and then make such a bold statement. But then again, there will always be people spouting "someone said this on BG, there for they are right!!!!!"

"HUGE" difference?? How much is "HUGE" care to post your findings?