Spell Interruption Rate questions

[pchan]

As form my eye-balling it seems the earlier they hit you, the greater the chance of interruption.

Yes this is pretty obvious, also if you cast stoneskin and move right away it will interrupt, but if you wait till the spell is almost cast and move there is a decent chance not to get interrupted and actualy cast when moving.

[el_tio]

Yes this is pretty obvious, also if you cast stoneskin and move right away it will interrupt, but if you wait till the spell is almost cast and move there is a decent chance not to get interrupted and actualy cast when moving.

I'd just assume that'd be a function of the 20% or so at the beginning and end of any spell that a spell will always go through. Also, if you move and then move back to the spellcast starting point you won't be interrupted.

[RyaWHM]

Yeah, it's that kinda-big difference between what you see, and what's actually happening. I think the easiest way to really get a good understanding is if you twobox. You just get used to that delay over time, but even just comparing yourself /following someone, to having someone /follow you...

= you, = that guy



When you /follow:
You're right behind that guy!



When someone /follows you:
________________ He's several steps behind!



This is also roughly the distance you can run, if you time it perfectly, at the end of your spell, before it actually goes off! Because as I understand it, you've done a "valid" movement, it just hasn't registered yet. By the time it does, the spell is already done!

A lot of BLM or RDM solo videos, you can see them standing to cast, then running a step, then finishing. Usually when I see it done, and even when I do it, it's not quite the maximum distance you can run (earliest you can run) -- but from trying to run farther, in my experience it's risky because sometimes I can run really far, and sometimes a shorter distance than I have succeeded with before.....fails.

Could this be some rhythm of communication between you and the servers? Has anyone else noticed this, maybe in relation to how laggy it is?

----

Anyway, if I do test this next part, I'll prolly just use Utsu:Ichi on mid-level mobbies, because I'm comfortable enough with it to know those really BAD times to start casting (sure to get hit!) ... so I can just use that.

[Auspice]

Interesting; I looked into a spell interrupt build a while ago and everyone I knew said the cap was 50% and not to bother. Also, the extra inventory I'd have to lug around wasn't feasible so I ditched it. Here's what I came up with for my RDM:

Druid's Rope - 10%
Hermit's Wand - 25%
Karasutengu Kogake - 15%
Muse Tariqah - 10%
Nashira Turban - 10%
Solitaire Cape - 8%
Warlock's Tabard - 15%
Willpower Torque - 5%

That's 98% interrupt down right there.

[bungiefan]

There's a brief delay between the server processing the command and your system showing the effect of it. Sending a spell cast command to the server, you don't get confirmation right away, though it is casting on the server. What you see of mobs and other players is a couple of seconds behind what the server sees. That's why you have trouble attacking a mob that is moving unless you aim ahead of it and unlock the camera, and why mobs can hit you even though they seem to be farther away than attack range (they are closer than they appear because you are seeing where they were about 1-2 seconds ago).

[Suiram]

Never having 2boxed, it's still fairly easy to get an idea for when you can move and still let a spell fire. I usually go by the % of casting bar, although there's no hard-and-fast rule that I could impart here because it depends on a lot of factors like which spell you're casting (your cast time), what job/subjob (Fastcast), and what % your spells fire at (when I played on PS2 they seemed to fire around 75% without FC, now on PC it's more like 71-72). But you just do it enough and watch the casting bar while you do it and eventually it's easy to get the hang of.

[Raineer]

There's a brief delay between the server processing the command and your system showing the effect of it. Sending a spell cast command to the server, you don't get confirmation right away, though it is casting on the server. What you see of mobs and other players is a couple of seconds behind what the server sees. That's why you have trouble attacking a mob that is moving unless you aim ahead of it and unlock the camera, and why mobs can hit you even though they seem to be farther away than attack range (they are closer than they appear because you are seeing where they were about 1-2 seconds ago).

I wish more people would learn this.

Also, the delay can be pretty dramatic. Running 3 or 4 accounts in the same house shows this sometimes to a ridiculous extent. You think you're running next to someone but I've seen up to 25' distance between them on another screen.

[Guppy]

We can't do the Aquaveil tests until we figure out your base rate, for the reason see the Example in my previous post. In order to figure out your base rate, we need you to figure out the MINIMUM amount of spell interrupt gear you need to wear to get every single spell past the mobs. This is the critical first step.

Well, suppose we assume there's no funky caps split between gear and merits.

We already have gear setups that can yield 99% equipment -interrupt. Just add up AF+1 boots (25%) + 2x wands (50%) + 3 blocks of 8% down (Choice of Solitaire Cape, Max. Merits, Several types of Obi, Magnetic Earring).

If we get no interrupts testing this, it doesn't prove anything -- a 1% chance could just mean we got constantly lucky with our rolls (increasingly less likely with larger sample size), or that the base resist rate isn't 0%. But if we get hit even once, we then know the base rate is 0%.

[Kaylia]

Was looking on Warrior to see if I could get a nice setup for emergency utsu.


Merit: 8%
Magnetic Earring: 8%
Mountain Gaiter: 5%
Resolute belt: 8%
Woodvile axe/muse tariqua: 15%

44%...lame.

[RyaWHM]

I wish more people would learn this.

Me toooooo! It's frustrating seeing someone try to claim a mob with a melee swing, by basically just /following...and he's like "WHY CAN'T I HIT HIM?!?!"

Sometimes I try to explain it, but usually it's not worth trying

[zagex]

Interesting; I looked into a spell interrupt build a while ago and everyone I knew said the cap was 50% and not to bother. Also, the extra inventory I'd have to lug around wasn't feasible so I ditched it. Here's what I came up with for my RDM:

Druid's Rope - 10%
Hermit's Wand - 25%
Karasutengu Kogake - 15%
Muse Tariqah - 10%
Nashira Turban - 10%
Solitaire Cape - 8%
Warlock's Tabard - 15%
Willpower Torque - 5%

That's 98% interrupt down right there.

warlock's tabard is 10%, +1 is 12%
also, magnetic earring is 8%, merits can be another 8%

[Guppy]

Was looking on Warrior to see if I could get a nice setup for emergency utsu.

I've heard some folks have tried SCH for Aquaveil-ga, and that even without any enhancement gear, it works pretty well for NIN and /nin. One more reason to figure out just how much -Interrupt it gives, I guess.

[Arkanos]

Hey Rya, if you need any help on testing this stuff, give me a /poke. ;x (Cyferis here)

This could be really, really nice for RDM/NIN. You can easily get 74% interrupt rate down without sacrificing any haste gear (2x Hermit Wand, Solitare Cape, Willpower Torque, Magnetic Earring), and can get the remaining if you sacrifice haste/fastcast slots; if Aquaveil is anything greater than 25%, and given that it lasts 15 minutes with composure, it could make getting shadows up on rdm/nin with lowest possible recasts even more trivial than it is now. And this isn't even counting merits. Need to test just how much Aquaveil gives now...

[KittyFantastico]

I'm a spell interruption noob here, but if you're going to devise tests -- does the level difference between you and the mob hitting you matter? I seem to recall reading somewhere in the past that it did-- but I also can't remember where that was or find it again. But it seems generally true that a mob much higher level than you even with a weak hit is more likely to interrupt than one far below you that hits you three times. (I used to think when I first started it was how much damage you took while casting, until I read at some point the level difference thing....)

[Blaksen]

The max amount of spell interruption rate down for a Nin is 78% or 86% with merits. So depending on the base rate a nin may possibly be able to become immune to interrupts. However doing so results in the loss of pretty much all haste equipment unless Nin gets a really high base rate on HNMs. I think the only things you could keep on is Blitz ring and Loquacious Earring. This would make recast down be at 37% with haste and double march. I'm not a Nin though so would it be worth it for an ichi macro?

Pld can only hit 55% or 65% (If they use Muse Tariqah and I am unsure if the switch would be worth it because if I recall its a size 2 shield) or 63% / 73% with merits. However a Pld can also get 22% more if they use the latent earring bringing the max up to 77% / 87% or 85 / 95% with the shield but considering the loss of haste and shield skill it seems mostly impractical for a Pld to try.

[chiyio]

The max amount of spell interruption rate down for a Nin is 78% or 86% with merits. So depending on the base rate a nin may possibly be able to become immune to interrupts. However doing so results in the loss of pretty much all haste equipment unless Nin gets a really high base rate on HNMs. I think the only things you could keep on is Blitz ring and Loquacious Earring. This would make recast down be at 37% with haste and double march. I'm not a Nin though so would it be worth it for an ichi macro?

Pld can only hit 55% or 65% (If they use Muse Tariqah and I am unsure if the switch would be worth it because if I recall its a size 2 shield) or 63% / 73% with merits. However a Pld can also get 22% more if they use the latent earring bringing the max up to 77% / 87% or 85 / 95% with the shield but considering the loss of haste and shield skill it seems mostly impractical for a Pld to try.

In Before Chroal Roll

[Thunder]

Has anybody tested this on higher level mobs? Couldn't it be just a level correlating thing?
Zero interruptions seems kind of broken to me if it's applicable to.. everything.

[RyaWHM]

does the level difference between you and the mob hitting you matter? ... But it seems generally true that a mob much higher level than you even with a weak hit is more likely to interrupt than one far below you that hits you three times.

Has anybody tested this on higher level mobs? Couldn't it be just a level correlating thing?
Zero interruptions seems kind of broken to me if it's applicable to.. everything.

So far, the highest I have tested my full set on is a Lv.79+ VT Yagudo, and the results were encouraging; I didn't get interrupted at all, but I didn't test that particular one for very long. But the fact that Ichi wasn't interrupted is very encouraging if you ask me! :D

As far as plain old level versus level, that will be easier to prove when I do some more testing, but it seems to me like level isn't very important, just skill.. so your level indirectly affects it just 'cause your skills cap differently at different levels...

My theory is that at ~C skill you have ~50% chance of interruption on EM. More skill/lower mob = lower interruption rate.. Lower skill/higher mob = higher interruption rate

...but that's just the feeling I get from the various tests I've done -- needs more proof!

[pchan]

I tried the chigoe too, with 81% spell interruption and aquaveil, 3 chigoes beating on me doing 300 + damage during each stoneskin (no phalanx obviously), 30 attempts (only,lol) and zero interrupt ; looks like aquaveil > 19%.

[RyaWHM]

Wait, what job/sub and what skill level? You were testing interruption of the spell Stoneskin? (gonna take a wild guess and say RDM, since your total was so high)