I've been meaning do parry tests for a while, so here's a topic to post my tests. I'll be editing in my tests as I do them. All misses are discounted, as misses are calculated first.
The target for these tests is Mimas, a level 36 Gigas NM, so parry rate based on parrying skill is should be capped, but not confirmed.
709 Skill, 2100 JP, Inquartata +10, Parry Rate +0%, NM target
1110 attacks
613 parries
55.23% parry rate
709 Skill, 2100 JP, Inquartata +10, Parry Rate +5%, NM target
1910 attacks
1148 parries
60.10% parry rate
709 Skill, 2100 JP, Inquartata +0, Parry Rate +0%, NM target
1483 attacks
682 parries
45.99% parry rate
Parry Testing
|
Parry Testing
I would like to suggest one additional test, in regards to the functionality of parry rate+5%. That being a test vs a mob sufficiently high level to assume that base parry is floored.
The point of this being to show if parry+ mimics Inquartarta's mechanics in regards to the parry floor or not. That being how Inquartarta is added after the floor is applied. If parry+ does not share this trait, then it would potentially be useless(or at least have reduced effectiveness) on anything high enough level to floor RUN's base parry rate from skill. Also, why this particular NM? it doesn't seem to have much going for it in regards to this test beyond being a MNK mob. It doesn't hundred fists a lot or anything does it? The sample gathering must take a really long time, especially considering that it certainly has floored hit rate. I realize that having already gathered the first sample, than the others need to be on a mob of at least matching level, so it's hard to change at this point but this NM really doesn't seem to be a great option for testing.
[+]
I chose Lamina because it's a monk, is in an area without a time limit, is reliably up, and doesn't have any status effects that prevent parrying. I didn't really see any immediately obvious better options for testing while afk, as any NMs that I examined that have enough accuracy to not whiff constantly or attack notably faster are either in timed battles, can kill me afk, or have annoying gimmicks like en-statuses that prevent parrying.
However, it seems on Lamina, it seems that I might not have capped parry rate from skill, meaning that the small level range it can have (58-60) might be screwing up my results. I might need to choose a lower level mob (or a mob that is known to have a zero level range) and start over. Edit: Switching to Mimas, and starting tests over. 1 Inquartata = 1% parry despite what the devs said.
http://noranekofantasy.blog.fc2.com/blog-entry-568.html Base parry (5~25) + traits/gifts (19) + HQ Erilaz Leg Guards (2) + HQ Turms Leggings (5) + augmented Ogma's Cape (8) should give 39~59% parry rate.
[+]
So, at least for NMs (don't know about normal mobs), +1 Inquartata in gear = +1% parry rate, and +1% parry rate on gear = +1% actual parry rate.
[+]
We should be able to determine the rate on normal mobs pretty fast. Mass pull testing etc.
Also, should submit a bug report to SE. Either they'll say 'oops', or maybe they'll fix it. Or you know, ignore it like they do most things. I think I missed something, why do you say there's a bug somewhere?
Didn't Geriond test show that Inquartata and Parry+X% work exactely as expected, that is it converts to +X% parry rate regardless of the target and its level? Because SE specifically stated that +1 Inquartata = 2% parry rate. Whereas we're observing a 1:1 exchange ratio. When in game results directly contradict dev statements that's immediately bug report worthy.
While we haven't confirmed the rate on normal mobs, SE didn't say a damn thing about a lower rate on NMs. So I say call them on it and ask them to fix this ***. Also, this hasn't shown us anything in regards to disregarding the target level. The tests were on an extremely low level mob, thus we're not going to see how Inquartata interacts with the parry floor in these tests. Oh, I missed the "SE said it's 2%". Where did they say that? Official forums? JP official guide?
Your comment makes sense now, sorry for forcing you to write this last post of yours! Concerning the level thing, other tests were being made for that specific purpose during SoA's early life. It was a small sample but it showed pretty similar parry rate overall % despite of trying against different level targets. I don't remember if targets were NMs or normal monsters though... Wonder if we can dig that test out. Atm I can't really remember if it was here on FFXIAH or somewhere on one of the old BG collective threads. Dev post. At the time Erilaz legs were the only Inqaurtata gear so it makes sense that they specifically referenced them rather than making a general statement about inqartata+.
Also, dev post stating that Inquartat isn't affected by mob strength(level.) That said, SE keeps saying ***that's wrong recently. So I'd like to see testing confirming this in any case. And as I mentioned earlier, a test on the parry+ augment to see if it also ignores mob level(is applied after the parry floor.) This is the first Iquartata(god I hate typing this word) test that I'm aware of. Link. And I'm not sure that anyone has bothered with a detailed test since. If you know of one, link it so I can reference it on the BGwiki page. It had a test on lvl 0 mobs, than another on lvl 100 mobs. This was done pre-ilvl, so no ilvl parry+ on weps. While the test seems to have proven accurate, there are various issues with it, such as the the tester's parrying skill being uncapped during the test and skilling up during it. As well as the fact that it was simply assumed, and never verified that the RUN's base parry rate was floored at 5%. And finally for the first sample you have to assume a parry cap or 20% for the samples to match. whereas we've seen 25% base parry rates(Although, I'd never seen 25% parry until ilvl parry skill was added to weps, so that may have been a valid assumption at the time.) Basically I just like having things verified whenever there's reason to doubt. Like SE telling us ***wrong. The two dev posts and the test post were all linked from the BGwiki inquartata page. Always good to check the references if you aren't sure about the validity of data. Ragnarok.Martel said: » This is the first Iquartata(god I hate typing this word) test that I'm aware of. Concerning the test I'm not sure if it's the one you linked. I know for sure it was a very old test done way before they added Erilaz set, it was performed with just the job traits. Taim Meich's test might be the one I was thinking about. Granted I seemed to remember a more vivid discussion and further data from other people, but I could be wrong on that in the end. It's probably something silly like each level of the Inquartata job trait gives +2%, so the devs looked at that and said "oh yeah, Inquartata+1 means another +2%" on the forum when in fact "Inquartata+1" from gear literally just means +1% parry and not +1 level to your job trait.
[+]
So. Recently it seems to be a theme that something gets tested, then turns out to not be as potent as we had thought. Well, prepare for additional disappointment.
I went out to do a test to simply confirm that Inquartata(hence forth INQ, cause *** that word.) ignores monster level, by applying after the floor. Well, I do feel I confirmed that. But I also verified, via ridiculous sample sizes, that Master RUN only has 19% INQ, rather than the 21% we'd thought. Basically, we seem to have stuffed an extra INQ tier in there somehow. According to JPwiki 99 RUN has 11% INQ, and master RUN has 19%. The last non gift tier was at 90, according to JP, while we have one added at 99. Though BGwiki does have a verification tag on that. I'd actually noticed the JPwiki values awhile back. I just thought they were wrong. Now. Data. And story time, cause I like feel like it. For the TLDR people, well, I put the conclusion above in bold text, so you can go home now.The thing I'm wondering about is how so many previous tests lined up so perfectly for 21% INQ on master RUN... I mean, Geriond's tests were just spot on for that value on all three tests. +/- a percent, but you expect that. The only thing I can think of aside from freakish chance is that maybe the parry cap isn't the assumed 25% that fits with those tests. I mean, I've seen parses with 26% and so on. But you tend to dismiss small variances. I think I may need to do a test to try to verify the current parry cap.
[+]
Is it at all possible that the minimum base parry rate is 3%, and not 5%?
I'm planning on doing some parry testing without ilvl parry skill on Ascended Chigoes later this week (regarding parry floors and the Ambu Parry augment). Asura.Geriond said: » Is it at all possible that the minimum base parry rate is 3%, and not 5%? This is very possible, we just assumed 5% cause lots of other stuff have a floor around there and we had to use some value for floor. Asura.Geriond said: » Is it at all possible that the minimum base parry rate is 3%, and not 5%? I'm planning on doing some parry testing without ilvl parry skill on Ascended Chigoes later this week (regarding parry floors and the Ambu Parry augment).
[+]
Out of curiosity, has it already been eliminated that parry doesn't use hard percents and instead uses x/256 or x/1024 values?
I've not heard of anyone testing specifically for that, but... Is there any reason to believe that it does?
Ragnarok.Martel said: » I've not heard of anyone testing specifically for that, but... Is there any reason to believe that it does? Was just thinking that something like a x/1024 basis might possibly explain the surprising discrepancy, and feels very SE-ish. I was thinking it should be a straightforward task to eliminate alternate bases as well. If it's not something that someone better at statistics than myself can crunch out w/ your existing data, then I don't think there'd be a pressing need to do new tests for it. For the non-RUN testing, would THF be a good choice? It can equip Turms feet too, so you could double-check the value from Inquartata+4/+5 on a job that has none baseline. I have no idea how to get Kparser setup, or I'd donate some time to the cause. I don't know exactly what level mobs you'd need to floor parrying. But if you tried THF on mass pulls of the level(131+ apex mobs) I tested, I think you'd get flattened.
You could do 1~2 mobs, perhaps. but then it'd be veeery slow sample gathering. In this case, There's little need for more inquartata gear testing. Between mine and geriond's tests the potency of inquartata+ from gear is well verified both on NMs and normal mobs. It's the base trait value, and the exact parry floor that's being questioned atm. So, next test. Just one for the moment. Since the parry floor was called into question after my last round of tests, I went to confirm.
PLD/BLU. 658 Parry skill. 0 Inquartata+. 131~133 mobs.  Nearly 10k sample. 5% parry rate. So unless you think RUN gets its own parry floor, and it's lower than PLDs, the floor is 5%. Thus based on my earlier tests, Master RUN Inquartata+ is 19%. Which sucks. But it is. This test was soooo much easier on PLD. I mean, RUN is pretty good as holding mass mobs, and Turms hands are hax. But PLD is just better at dealing with tons of mobs(provided you can get them all in front of you.) At least when it can attain a decent block rate. Which in this case was 99.34% with reprisal up.(Think I had Block+8 and shield skill+15)
[+]
So what did we have left to test in here?
We've confirmed/reconfirmed -Inquartata trait potency -Inquartata+ gear potency -The parrying floor Could test the parry cap, but I'm not sure how relevant that is. Would mostly be a matter of curiosity, as you'll never have capped parrying on mobs that matter. Oh, did we ever get a floored parry test with the parry+ augment to see if it works the same as INQ?
[+]
Necro Bump Detected!
[394 days between previous and next post]So about that parry augment test.. about a year later..
Anyway. PLD/BLU. 658 Parry skill. 0 Inquartata+. 131~133 mobs. I can't get kparser to work anymore, so parse adddon it is.   I have my parse addon configured to break up defensive parses by subslot and reprisal up, which is why there's 2 versions of me there. Doesn't really affect anything here. Just add all hits/blocks together to get the non-parry sample. Small sample, but still pretty conclusive. In the previous test on PLD I was well acquainted with the parry floor. With the parry+5% augment, I'm 5% above the floor at 10% parry rate. Either I just so happened to have just barely floored parrying on PLD.. hah. Or the Parry+ augment is basically Inquartata. Parse addon export data
[+]
All that is left to test now is that its bonus are halved on ITG targets.
Thanks for your test Martel! Hope someone will add this on BG-Wiki as in some time from now it could be hard to find all this wonderful information lost within the depths of FFXIAH How is that rumor even a thing? Look at the first post in this thread.
55% parry vs an NM. Next test. Added +5% Parrying rate. Gets 60% parry rate. Last test, removed +10 Inquartata and 5% parry rate, gets 45% parry rate. These clearly indicate that both Inquartata and Parrying rate+ augment are 1:1 parry rate on NMs.
[+]
Asura.Sechs said: » All that is left to test now is that its bonus are halved on ITG targets. Ragnarok.Martel said: » How is that rumor even a thing? Asura.Sechs said: » Hope someone will add this on BG-Wiki ... Ragnarok.Martel said: » Look at the first post in this thread. ... clearly indicate that both Inquartata and Parrying rate+ augment are 1:1 parry rate on NMs. Beat me to it. But seriously though, thank you for taking the time to test Ambuscade Cape Parry+% augment Martel. 'Tis appreciated! I'm not actually sure where the augment info would go on the wiki... I guess you could add an entry for it on the Inquartata page? It's not called Inquartata, but it appears to be functionally identical.
[+]
Ragnarok.Martel said: » Because SE specifically stated that +1 Inquartata = 2% parry rate. Whereas we're observing a 1:1 exchange ratio. When in game results directly contradict dev statements that's immediately bug report worthy. |
All FFXIV and FFXI content and images © 2002-2024 SQUARE ENIX CO., LTD.
FINAL FANTASY is a registered trademark of Square Enix Co., Ltd.