[TEST] Counter Data

[Armando]

Re: [TEST] Counter Data

In addition to that, I'm pretty sure I'd still Guard sometimes even without Guard skill, just like jobs without shield skill can equip one of them All Job bucklers and still see random blocks here and there.

Callisto: No, I haven't tried parsing my Hit Rate against EP-TW mobs. I probably should at some point, it's definitely worth looking into. Actually, I haven't been on FFXI in like 2 months. As if it weren't bad enough that I can only play on weekends, and I get home very late on Friday, Brawl has been keeping me busy as of late. Even if it's a fighting game, I do just as much testing there as I do on FFXI!

TheBridge: Like the others've said, please point out what variables need to be removed if you want a proper answer from me. You're the only one that thinks the test isn't valid.

[Icemage]

Re: [TEST] Counter Data

There's some gotchas here too. Monks have E rank Parry, so you ALWAYS have a chance to parry if you have a chance to Counter unless you're using a job that has no intrinsic Counter, and no intrinsic Parry skill while wearing equipment that gives +Counter.

i.e. you'd have to do some sort of ridiculously retarded test condition like a WHM/SCH with a Deae Gratia and Avenger's Earring (2% counter). But that's not enough of a % to test with, since the Deae Gratia (aside from being a rather difficult item to acquire) is also a fairly high damage weapon.

It would be a little bit easier to test with a job without such a high Parry skill, however. Maybe I can do a test later on WHM/MNK. My Monk is only level 18, so Parry triggers should be minimal, and I'm sure I can find something suitably miss-worthy like a Lotus Katana or something to test with.

EDIT: There's definitely some sort of level-induced accuracy going on. I just tested with a Lotus Katana (DMG1) with WHM75/MNK18 and I'm getting a ~95% hit accuracy on a Sand Beetle in Eastern Altepa Desert, even though it checks as high evasion and defense (since I have a big fat zero in Great Katana skill).


Icemage

[Malacite]

Re: [TEST] Counter Data

EDIT: There's definitely some sort of level-induced accuracy going on. I just tested with a Lotus Katana (DMG1) with WHM75/MNK18 and I'm getting a ~95% hit accuracy on a Sand Beetle in Eastern Altepa Desert, even though it checks as high evasion and defense (since I have a big fat zero in Great Katana skill).

So I'm not the only one who's noticed this in-game

Told ya the signet thing was hogwash >.>b

[Armando]

Re: [TEST] Counter Data

There's some gotchas here too. Monks have E rank Parry, so you ALWAYS have a chance to parry if you have a chance to Counter unless you're using a job that has no intrinsic Counter, and no intrinsic Parry skill while wearing equipment that gives +Counter.

I don't think H2H can parry at all, actually. But then you're introducing Guard...also, is Parry skill really a requirement to parry? I figured you'd just have minimal proc rate, but I rarely melee with a job that has no parry so I wouldn't know...

And yeah, now that you remind me Icemage, Lmnop had done the exact same thing a couple of months back but after the thread was dead. It's a funny coincidence, because I think he used Lotus Katana as well, and he tested it in Altepa (though he smacked spiders, not beetles.)

If the Crab in my Blinding Potion test was Level 51, it would've had 164 Evasion. With Blinding Potions, I'd have 0 or 1 Accuracy. If our accuracy penalty works both ways and I get +2% hit rate per level, then my Hit Rate should've been 75% + (-165*0.5) + (19*2) = 31%, which would explain the 3.37% counter rate.

[Icemage]

Re: [TEST] Counter Data

I don't think H2H can parry at all, actually. But then you're introducing Guard...also, is Parry skill really a requirement to parry? I figured you'd just have minimal proc rate, but I rarely melee with a job that has no parry so I wouldn't know...

Yes, it's required. I never parry on WHM unless I'm subbing something with Parry skill (RDM, for instance). If I'm WHM/BLM or WHM/SCH, I'll never ever get a Parry.

And yeah, now that you remind me Icemage, Lmnop had done the exact same thing a couple of months back but after the thread was dead. It's a funny coincidence, because I think he used Lotus Katana as well, and he tested it in Altepa (though he smacked spiders, not beetles.)

It's not a surprise. Teleport-Altep is quick and easy, and the Lotus Katana is one of those rare bizarre weapons that any job can use (along with that fake Maul) that has 1 damage, but hardly any job has skills to use.

If the Crab in my Blinding Potion test was Level 51, it would've had 164 Evasion. With Blinding Potions, I'd have 0 or 1 Accuracy. If our accuracy penalty works both ways and I get +2% hit rate per level, then my Hit Rate should've been 75% + (-165*0.5) + (19*2) = 31%, which would explain the 3.37% counter rate.

We're wandering pretty deep into the realm of statistical error here. Let me chug some of my test data and see what I find.


Icemage

[Icemage]

Re: [TEST] Counter Data

Target: Sand Beetle in Eastern Altepa Desert.

Tarutaru 75WHM/18 MNK. 0 GKT skill using Lotus Katana.

Number of times hit: 72 (including 4 critical hits)
Number of parries (54 skill): 6
Number of evades: 322
Number of counters: 8

ENEMY ACCURACY: 21% (assuming all counters and parries would have been hits)
COUNTER RATE: 10% (assuming parries checked before counter)

My attacks missed: 8
My successful attacks: 402 (including 47 critical hits)

NET ACCURACY: 98.05%

Interestingly, it looks like the Sand Beetle had almost the same exact delay as I did (280 on Lotus Katana, but I was wearing some Haste gear - Walahra Turban, Blessed Mitts, Blessed Pants), and it also lost some time from using WS like Power Attack (not counted in my results since it's not Counterable), Spoil, Rhino Guard.

Still not a comprehensive test, but it's all I had the immediate patience for, and gives you at least another raw data point.


Icemage

[Lmnop]

Re: [TEST] Counter Data

I have a theory that Lotus Katana doesn't check accuracy in the normal manor, as far as I can tell. No matter the job, no matter the level, you'll find that you always connect for wee chunks of damage. Hence the Enthunder effect triggered on melee swing, I guess... (for roleplaying reasons. I'm not suggesting Enthunder influences accuracy).

To prove this theory, I think I'll go pick a fight with something terrifying with one equipped...


--------------------------------------------

I am completely and utterly wrong. I just fought (and died against) an EP attercop in AU using all 3 types of wooden katana (wooden, hardwood, and lotus). They were all missing frequently enough to assume they check acc as normal.

What's strange though, is I didn't see any en-animations...

---------------------------------------------

Ok, just fought (and died against) a DC gob in Cape Terrigan. Only the Hardwood makes the En-animation so... disregard this entire post! Back to wacky acc theories with us!

[Callisto]

Re: [TEST] Counter Data

In addition to that, I'm pretty sure I'd still Guard sometimes even without Guard skill, just like jobs without shield skill can equip one of them All Job bucklers and still see random blocks here and there.

I actually can back that up now that I remember, I've guarded with no H2H/Guard skill when whapping at mobs as RDM bare fists.

[Armando]

Re: [TEST] Counter Data

Thanks for the info, the three of you. I'll make some time next weekend to test the level-based hit rate bonus.

Would make some sense considering IT mobs have unnaturally high hit rate even against evasive jobs.

[Icemage]

Re: [TEST] Counter Data

Thanks for the info, the three of you. I'll make some time next weekend to test the level-based hit rate bonus.

Would make some sense considering IT mobs have unnaturally high hit rate even against evasive jobs.

If it makes any difference to you, most endgame LS (including the one I'm in) have given up on Evasion build Ninjas, for precisely this reason. Most of the things we fight are level 80+, sometimes as high as 90+, so even a small difference in levels hampers your chance to evade significantly (and vice versa for monsters).

Against relatively weak NMs like Charybdis, Evasion works pretty well. Against anything above level 75, even tremendous Evasion builds don't do squat.


Icemage

[Malacite]

Re: [TEST] Counter Data

Thus the reason why when I took my NIN to 75 I geared up for def over evasion >.>

That way when Shadows were down at least I wasn't getting steamrolled.