Suggestions needed: How to perform a detailed test?

[Sechs]

Since I am a noob, I guess this thread belongs in here.
Inspired by some curiosity and the fact I didn't see any test about it, I wanted to perform a test to discover (supposing the test will be successfull) how TP bonuses for WSs are managed by game mechanics.
For example the modifier. We know the bonuses we get at 100, 200 and 300% TP. What happens, say, at 150%? You get something in between 100 and 200, or you get the 100% bonus because you haven't reached 200% yet?
This goes for WS which have a damage increase with TP, for those that increase crit hits or accuracy or others, I wouldn't know how to test.
When the WS reports "Damage increase with TP" or something like that, I think this line refers to the different damage multipliers that you get.
So, how to perform such a test? I have ideas and questions, please help me create a perfect scenario for this test:


1) Which WS?
Since my main job is Monk and all H2H WSs are at least 2 hits, I wanted to use Staff WSs, probably Full Swing? My Staff skill isn't capped yet (217ish?), but since I'll be going to fight low level enemies, I won't get skillups and hence this shouldn't become a variable in my test

2) Where to test?
I was thinking to test with very low level enemies, like in Ronfaure or Gustaberg. This way I won't have problems soloing them, it will be quick to kill, and I will minimize miss rate. Any cons in fighting such low level mobs? Their range outside should be level 1-5, and since there is such a big level difference between my level and theirs, this range shouldn't constitute a big variable over my total damage (which will probably most likely be always capped, or close to that)

3) Which setup?
Avoiding any item granting me Double Attack, Triple Attack etc. Kick Attack won't be an issue because I will be using Staff. About the Staff... I was wondering if I could use something with very low damage but very low delay as well (like Treat Staff II, only 2 damage?). Would that create problems, not allowing me to reach caps etc?
I was considering to sub Samurai for Meditate, always keeping Hasso and any other ability down, so that that won't change my basic stats. Lots of +haste gear to be able to engage another enemy asap.

4) Minimum number?
What is the minimum number of killed enemies you suggest to have a consistent result?
I was considering killing some with around 100% TP, some with around 150% TP, some with around 200% tp.

5) How to analyze?
I would be creating a macro saying in /p or in /echo how much TP I have the second I use the WS. Then I would be logging the text, filtering only those lines along with the consequent damage line. This way I should be able to see how much TP I had, and how much damage that TP produced.
Supposing I manage to get a consistant number of kills, I should be able to see if at 150% TP I have a sensible difference compared to 100% TP. If there is a difference, it means the TP modifiers are applied to the damage formula in a progressive way (so in theory each single TP you get, can virtually produce more damage). If there is no such difference, it means the modifiers are applied in a scaled way, so using WS at 199% is the same as using it at 100%. (at least for the category of WSs I described above).



Any comment? Any suggestion? Tell me what's wrong in what I wrote and how can I create the best possible conditions to perform this test.

[Delekii]

You picked a terrible time to do it - pdif is, at the moment, almost completely uncontrollable, meaning that it is going to be quite difficult to work out whether damage variation is coming from pdif, tp, or both. I'm pretty sure this imformation is already known (though possibly not specifics), but good luck to you in any event.

[Sechs]

So I guess I should wait another patch to fix... (but wasn't pdif fixed in the recent 2nd patch, the Pankratium one?)

And yeah, common knowledge says the TP increase (the multiplier) is "stepped", so 199% TP WS is the same as doing a 100% TP WS.
But nobody ever did more detailed tests or demonstrated it, so I Wanted to dig through it.
Sometimes people may have the attitude of "let me wait 2 more hits so I get 200%" before using their WS. It could be useless, and they wouldn't waste their 193% TP, supposing this test will show that TP multiplier bonuses aren't "stepped".

I mean... it's stupid and useless I know, but I wanted to understand. Guess I'll just wait after WotG

[Suiram]

I am pretty sure that someone did test on a number of weaponskills (Nagamaki maybe?) and reported that it's a continuous, piecewise-linear function, that is to say, it's a straight line from 100 TP to 200 TP, followed by a straight line (with perhaps a different slope) from 200 to 300.

Edit: viewtopic.php?p=586836#p586836

[VZX]

Yep, Nagamaki did it. It was done

Still, it doesn't close the possibility to research further
What would be interesting is to test 'How TP affects the accuracy of WS with "accuracy varies by TP"'

[Sechs]

Yep, Nagamaki did it. It was done

Still, it doesn't close the possibility to research further
What would be interesting is to test 'How TP affects the accuracy of WS with "accuracy varies by TP"'

<3 <3 <3 <3 Nagamaki

That's good, less wasted time for me. I think I asked for such info like... 8 months ago? But back then there was no test about it, I only found general comments. It's good to see Naga-sama tested it.

As for the TP<>accuracy relation that you mentioned (as well as the TP<>CriticalHits on some other WS) it's interesting, but I can't imagine a way to measure it. For the accuracy one... probably you could just have a big range of sample (like over 500 performed WS?) on the same kind of monsters, and see how much the % of miss is higher at 100% compared to 300%, with the same gear of course. That way you might be able to estabilish an approximate (very approximate) amount of +acc bonus from the WS doing a simple comparation.
AverageWsAcc300%TP will be a value higher by X compared to AverageWsAcc100%TP, you look at this difference, and then you take into consideration (there are already several tests about this) AverageAcc (not talking about WS, just average Acc), then you check the already existing tables to see how much +accuracy gear you need to reach AverageAcc+X % (with X being as described before). When you've done that, that amount of +acc gear should more or less be close to the Acc boost you got from the 200% additional TP in your WS.

Something like this could work, the problem is in creating a sample large enough... for normal blows you just fight the same kind of monster for hours. For WS it takes time... you need to get TP, to use a WS etc, and we're talking about High Level monsters (you can't do a test like this on low level ones...), and that would make it even slower.
Just the idea of gathering a sample large enough for this kind of test gives me a headache.



As for the Crit hits thing... I don't think you can test it since there is no way to accurately tell when a WS (single hit of course) hits for critical. It doesn't show on screen.
In theory you could "guess" from a "higher than average range" damage, but that would be too inaccurate imho.
Btw, once again thanks for linking me that Naga-sama post
Now I know for sure that there is no need to "wait" for 200% if you're, say, at 190% TP, you can use your WS and you will receive a (minimal) boost over 100%, it won't be the same as using it at 100%. Not like we're talking about something really useful lol, I know! But still, it's always good to know how things work

[careswww]

I am not 100% sure but I remember someone doing some tests with rampage(during the time the whole DEX/CRIT relation was discovered), resulting in 200TP = +20% crit on rampage or something like that.