Low Man Delve Discussion

This was true prior to the ilevel boosts, stun accuracy isn't an issue now.

Rdm is the stunner of choice now in a group that doesn't have a cor due to dia3

In a perfect world RDM will be great, it's not a perfect world. Using RDM to stun lowman fractures lowers your margin of error. A strong group could use a good RDM just fine, but PUG's ("people who don't know better") would be better off using SCH. If you can't find a competent SCH, you can't find a competent RDM anyway.

Chainspell is woefully inadequate because stunlocking lowers the amount of stuns you can land in total. Even if you pop it and sit on it and just stun like normal, you're better off with a 2HR SCH. You lose Embrava/Kaustra and 4 second stuns for 3 minutes.

Spontaneity would be a decent argument as well, but with such a long cooldown it's rarely worth mentioning.


Insult me all you want, it doesn't change the facts.


This is undoubtedly false, it excels in some content, but not all.

So what sets do you all use on rdm stunning?

Accumulating resistance is plausible given your observations (at least in the case of severe magic accuracy deficits) but does not account for the "wall" occurring within ~40 stun attempts (give or take a couple). This could be a biphasic situation with constant increase in magic evasion followed by a sudden jump in magic evasion that can't be overcome merely by stacking magic accuracy. Clearly there is an observable deficit between Apamajas II and a good item level weapon but not between a good item level weapon and the additional magic accuracy bonuses available to SCH over RDM. Focalization + Venabulum + Klimaform (with Venabulum offering the the vast majority of the macc bonus) is not buying an additional minute of Stuns (subject to data).

I already said SCH's advantages in recast reduction relative to RDM lay primarily in the safety margin offered by Alacrity. NOT JA recast reduction from Dark Arts. NOT JA recast reduction from equipment. Still, you don't need Alacrity-stun for Meteor at all unless recast is gimped (like if magic haste is down) and Alacrity disguises it. Meteor is slow-casting like Stonega V et al. And actually pre-item level AF2+2 boots are irrelevant for Delve bosses (anyway they are Alacrity only; new AF feet are better without Alacrity) b/c they simply don't spam moves that fast. Muyingwa is not spamming TP moves/magic every 4 seconds. Sufficient is sufficient. Stun is either ready or it isn't. It can't be "more" ready.

RDM equipment - magic accuracy from item level weapon is obviously the dominant factor "all else being equal." Any marginal bonuses from accessories (earrings and rings amount to ~10 magic accuracy at most) are insignificant in comparison with respect to your purported claims about how lacking magic accuracy in those slots will cause resists to occur sooner, e.g., you get to ~40 stuns and you're done. +10, +20, even +100 is not likely to buy even 5 more stuns. You're not getting to 50+ stuns. (Actually recast reduction from earrings and rings is also marginal. This is merely an example that RDM can achieve adequate recast reduction for TP moves only on Tojil.)

For TP moves on Tojil only, you do not need 10 seconds (or less) for Tojil, period. This is utterly obvious from observing the time interval between consecutive TP moves (not between Dispelga and the next TP move).

If I'm not on SCH or WHM, I'm on GEO backline for NM debuffs (excluding Luopans). Although I'm currently working on the new GEO relic armor, I have a 100% success rate on enfeebles unless the NM has a natural resistance to the spell. Ensuring that Eft (as an example) stays silenced (in particular), para'd, slow'd reliably makes life for the SCH's much easier. This is particularly true on Tojil when I consider it my sole purpose to ensure he stays silenced, blinded and slowed (he is resistant to Paralysis). Stunners have better things to do than waste their stuns on magic. My point though is my magic accuracy set is pre-reforge, so it is possible to be reliable on enfeebles without the new gear.

Last night I was in a PUG that failed >> 3x << in a row for the same reason(s) - we hit the resist wall at or around the 50% HP mark. The SCH's stuns began to resist and about a minute or so later my Silence's (with me on GEO) started resisting, but not in the sense that you would think. They would stick, but unstick about 1-2 seconds later rather than Tojil resisting the effects of the spell outright.

With the technical information you, Conagh and Spira posted, would this then mean that my magic accuracy was calculated to be higher than the resistance wall, but barely since it didn't stick for longer than a few seconds?

With regards to the stun/resist wall, what determines when a spell is resisted outright or it does land, but for an extremely short time?

I don't anaylse the log after a fracture run, this is obviously eyeballing about the 10s recast. Quick look at some random youtube video it seems like (looking at the timestamp) the shortest interval between 2 TP moves is 11s. But of course timestamp only displays up to the seconds, so whether it is 11.0 or 11.9 I wouldn't know. is your 11.13s recast good enough? maybe, but it is certainly tight, esp if you happen to have a lag spike and your previous stun failed to stun the move and was a second later, you might well have lost your chance to stun the next TP move.

Is there a 11s interval between stunning a meteor and a TP move? you're saying yes, i'm saying i don't know since i never timed it or looked at the log. So I assume you're right here.

But I'm sure that if turtle do a back to back tremor, or the peiste do a back to back mist, a sch can stun both but a rdm most likely fail, as T1-5 TPs a lot more often than the mega boss.

Another thing to note is T1 hennetiel, it would be quite critical if you failed to stun a dispel move and that dispel happen to remove your mnk's Formless strikes. A sch with alacrity will ensure that not happening (unless your mnk(s) take so long on the crab that you run out of strats...). Of course you could argue that taking a blue mage in to terror it will stop it from happening.

Allow me to quote myself:



I think there isn't much more value in this debate and there's more than efficient info here for the OP to know whether to take a alternative stunner (if he can find one with good enough gear , that is ^^)

First, you cannot have 100% success rate on enfeebles since outright resists are possible. Silence, paralyze, blind appear to follow a three-level model of resists (no resist, half-resist, full resist... well this is more extrapolation from things with known potency or fixed duration, like Drain, Aspir, Slow, etc.), the caveat being duration of unresisted silence/paralyze/blind follows a probability distribution (uniform) so it is harder to distinguish between "full" duration and a half-resist.

Silence lasting 1-2 seconds may be one falling in the half-resist bin and/or cumulative resistance. It is difficult to know because the mechanics are opaque.

The conceit behind the "wall" concept is that there seems to be a critical point for the stun accuracy/evasion check after which stun is obviously less accurate. For example, I might stun ~40-45 consecutive times with Atinian Staff (I lost this data though and I never looked into the mechanics of increasing resistance) then suddenly miss 5 straight. I never get to observe 10 consecutive misses though since the boss usually dies before then.

On the other hand, with further consideration I can see that the above could easily be reconciled with the concept of "gradual" resistance as well. Suppose evasion to stun increases by 10 points for every stun attempt (this is completely made up; no one knows). For example, Apamajas II has 70 magic accuracy (assumed) compared to Atinian Staff, which has 203, so all else being equal stunning with Apamajas II may lead to outright resistance "a lot sooner," though I have not observed this personally (until possibly recently doing 6-man delve) as Focus and/or Languor have been confounders (doing delve pre-update) and really I have not gotten to the point of "absolute" resistance (and noted stun counts).

If one were not even using Apamajas II to begin with, under this model failed stuns would obviously attenuate the effect of "salvage" attempts to increase magic accuracy (Focalization, increasing macc from equipment) and there is not a lot of "headroom" for SCH to stun for more than a handful of times after resists start to occur (further stuns increasing evasion under this hypothetical model).

So going back to RDM vs SCH magic accuracy, under this "cumulative resistance" model SCH obviously would have more stun accuracy headroom due to RDM's 104-dark skill deficit but I doubt this would manifest as an important difference when resistance to stun appears to increase rapidly. It would be interesting to find out though. It's true 1-2 more stuns might be the difference between winning and wiping. That may be an important difference to others.

Again the principal advantage of SCH for recast is Alacrity and for Delve in general superior magic damage to kill bosses marginally faster (well bee not really), not magic accuracy.


I guess YMMV as to how much tolerance in recast time is acceptable. In any case March +4 or better provides more recast slack if Apamajas II is undesirable for bosses (~11 seconds assumes March +3). SCH can get ~9 second recast without Alacrity and without Apamajas II with certain reforge armor as well.

1-5 NMs, again Alacrity is an advantage for back-to-back TP moves occurring in ~6 seconds (peiste moves), which does NOT happen on bosses. These days I'm pretty sure SCH has enough recast reduction to stun kraklaw moves without Alacrity, and RDM can use Apamajas II on that without macc issues.

Sorry i do not know how to make those gear boxes, but here is a RDM stun set to look at. (4 second recast (if 93% is true cap) with good Macc)

Weapon: Venabulum (or excalibur 119 to be fancy)

Grip/Shield: Any/Mephitis Grip

Ammo: Any/Aureole/Kalboron Stone

Head: Atrophy chapeau +1

Neck: Orunmila's Torque

Earrings: Any/Lifestorm Earring+Psystorm Earring

Body: Hedera Cotehardie

Hands: Hagondes Cuffs +1

Rings: Any/Omega Ring+Sangoma Ring

Back: Any/Merciful Cape

Waist: Witful Belt

Legs: Mes'yohi Slacks

Feet: Vitivation Boots +1

Gives 25% Gear haste, 50% Fast cast(25% recast), and 43% with haste and marches(Ghorn)

As for resists, you shouldn't have any problems in 6/7 man, mobs do have lower Meva in low man and much more room for error

Going over a few things alluded to earlier in this thread.

Recast reduction cap

TL;DR the cap is still 80%. 9 seconds appears to be the minimum recast for Stun without Alacrity. 8-second Stun without Alacrity is not possible (Timers plugin should not be considered reliable for testing this.) With Alacrity, 4.5 seconds may be the minimum recast for Stun. I have never observed a 3-second recast with Stun with Alacrity even though ~3s Stun recast would be predicted based on Alacrity bypassing the 80% cap. Note that bee can use TP moves ~7 seconds apart (based on timestamp differences), so you might have to wait a few seconds on recast to stun back-to-back TP moves (no Alacrity) occurring < 9 seconds apart.

I was under the impression that the normal (w/o Alacrity) recast reduction cap is higher than 80% (BG wiki article, BG post with test conditions), but it seems the results aren't valid as they were based on the old Recast plugin as opposed to recast shown in menu. Menu recast is reliable for checking whether the 80% cap still applies or not, provided that you print it with lua to ensure you get the actual recast value instantaneously after a spell is cast. You could create a lua addon file containing this script to print Stun recast in chat:



Until now, I couldn't check this myself since I didn't have sufficient gear. As a refresher on estimating recast, Magian staff recast goes into the Fast Cast term.

With 30 FC, capped gear and magic haste, and 18% JA recast reduction (Dark Arts and reforge AF feet), recast reduction is approximately (1-(1-30/2/100)(1-704/1024)(1-.18))*100% = 78.2%, corresponding to Stun recast of ~9.8 s. This would show up as 9 in recast menu, and does.

With additional Magian as recast, recast reduction increases to (1-(1-(14+30/2)/100)*(1-704/1024)*(1-.18))*100% = 81.8%, corresponding to Stun recast of ~8.2 s (would show up as 8 in recast menu if the cap is higher than 80%). This should show up as 8 in recast menu, but I still get 9, indicating that the 80% recast reduction cap still applies.

Using atma in Abyssea, Caster's Roll and/or reforge AF2 head in addition to the above should further confirm the 80% cap as well.

Regarding the effect of Alacrity, when combined with an active weather effect (edited this in; see next post for clarification) it allows the 80% cap (9 s with stun) to be bypassed.

If you are at 80% recast reduction before Alacrity, Stun recast would theoretically be 3.15 seconds with AF2+2 feet and weather (3.06 seconds with reforge AF2+1). Under the above conditions where I am definitely (with Magian staff) at 80% recast reduction w/o Alacrity, I am NOT getting 3 second recasts with Alacrity based on menu recast. My impression is that the maximum recast reduction is 90%, but this would have to be confirmed using Klimaform (180 base recast).

Stun resistance observations (6-man Delve, so no Focus/Languor)

201 INT, 420 dark skill, and 79 magic accuracy before Eminent Staff (INT +12, magic accuracy skill +213) (I could get ~35 more magic accuracy from equipment though, not including Lehbrailg +2 +/- macc augment)

For the last bee I did, Stun observations were as follows (using Eminent Staff):

Stun (36x)
Resist
Stun
Resist
Resist
Stun
Resist
Stun
Resist
Resist

This is the first time stunning by myself experiencing resists indicative of stun evasion clearly exceeding stun accuracy, with no 2nd stunner acting as a potential confounder.

Another time I did Tojil alone I had 39 stuns with no resists (using Eminent Staff). Using Apamajas II, I stunned bee 27 times without resist, and on Tojil I stunned 26 times followed by 2 resists.

Since there is not really a place for GEO in 6-man delve (although you certainly could bring a 7th), a SCH with a decent ilevel weapon could expect to stun bosses ~35 times with no resists (perhaps 40+ with magic accuracy to replace Focus and/or Languor), and maybe ~25 times with Apamajas II (needs confirmatory testing), noting the 143 magic accuracy difference between Magian recast staff and Eminent Staff. RDM would be expected to sit somewhere between those extremes (I don't have RDM so I can't check.) SCH might be expected to eke out a few more stuns with Focalization (~25-30 macc total with 5/5 Focalization)

Subject to further testing/confirmation.

Just for clarification

From my experience, alacrity doesn't allow you to bypass the cap, AF2+2 (or reforged) with thunderstorm does. (i.e. what bypasses the cap is Weather: Enhances "Celerity" and "Alacrity", not just alacrity itself )

i.e:

alacrity without thunderstorm and with relic boots-> 9s

alacrity with thunderstorm and with relic boots -> ~4s

No they dont, the only thing adjusted is HP, thats it, thats all, JUST HP

Is it possible to win without stun? Yesterday I was fighting shark as soon as shark came in it did some kinda damage I put on PDT set the other two dd died. It was me versus shark for 10 minutes getting cured and using PDT set then later another dd unweakened joined me for like 2 minutes. Got it down to like 53 percent.

I think if we fight in PDT set it's possible to win without stuns? Asking cause we waited for stunner for like 3 hours

shark and bee can be done without stuns if you're good, tojil pretty much requires it.

What's the setup for shark without stuns?

Monk X 3? Cor bard white mags? The first 5 NM is no problem.

i use apamajas 2, vivid strap, incantor stone, nahtirah hat, vanir cota, sch af +1 hands, swith cape, witful belt, orvail pants +1 , relic 109 loafers, psystorm/lifestorm earrings, prolix ring and angha ring ad dark torque for stun set. Can prob get better gear for neck and replace angha but i don't get resisted often. Honestly i don't remember the last time tojil resisted either to be honest. Stunning now is far far easier than what it used to be.

I haven't done it without a stunner, but with 4 monks you can keep Mantra up 100% of the time. With 3 monks, you'd want to wait 20 seconds between mantra going down and using the next mantra.

Without a stunner, I would probably go MNKx4 (or x3 with 1 COR), BRD, WHM, GEO. Put GEO in the second group and rotate bard in there as necessary.

brd whm variety of DDx4 (bring a samurai with piercing set would be nice to abuse both slashing and piercing weak time) would probably work. You'll need to make sure your whm is on point as curing will be the biggest strain in the battle without stuns, since guillotine will auto-kill you if your DDs end up with <50% hp after the attack (this tends to be the usual culprit of wiping), also be careful with the hp down move since that lowers your buffer for the auto-kill. The majority of shark's TP moves are actually magical dmg (according to bgwiki) so maybe fighting in hybrid-MDT/MDT sets is what you should be looking for.

a bard with ghorn ballads will help immensely (and maybe use a song for scherzo may help to reduce the dmg for gullotine with max HP down effect on). If no go on ghorn maybe replace a DD with Cor.

Has anyone beaten any of the mega bosses for Delve with a thf in a low man or 6 man group? Just curious (a very well geared thf).

thf is an asset in ceizak. piecing weak mobs, feint to offset zone effect, flee/hide to pull might work as well.

Not even sure why you ask about thf. I'm sure people have. Of course, you can do low-man delve with any DD. PUP, DNC, DRG, BST, etc. Usually the people that use those jobs can come on them because they have almost all the best gear/weapons for that job so it can perform just fine. Delve is just more ideal with certain jobs, like MNK on Tojil, SAM on Bee, etc. Personally, if I was shouting for a group, I'd probably rather go with 5 people than take a THF I didn't know. Then again, from my experience, a THF on Sylph is about as useful as a RDM (no offense to career RDM).

As for Delve:
Low man isn't any more difficult than 18-man. From my experience. In fact it's easier depending on your definition of low man. 3 people? 6? 8? 10? I'd be really impressed to see 3 people beat it. 6 may be the same or slightly harder than 18 man. 7 is definitely easier/faster than 18 man.

However, if you're looking to do low-man delve for weapons because you don't have 119 weapons yet, then you are going to struggle. I prefer low-man delve because gather time is significantly faster and clear time is about the same or faster. However, the people I run with are just there for relic craftables and plasm to buy relic craftables. We've already got our gear from Delve so we're either using REM weapons or at least ilvl 119 weapons. Our bards have daurdabla/ghorn.

So I think low-man Delve is better for those that have already got the gear from it and are trying to 119 their relic gear for various jobs. If you and your group aren't at that point, you may want to consider 18-man, or at least supplement your 6-man group with a Geo and/or Corsair.

wut? You referring to the trash mobs? I don't think the bosses are weak to piercing.

Other than Arcane Orbs, anything that flies, is weak to piercing.
For Ceizak thats Butterflies, Gnat, Moth, and Mega Bee.

Some friends and I are planning a 6 man bee run. Someone mentioned that to make decent time on the bee itself the DDs need to be at at least 950 acc. I'm hovering around 930 WITH sushi, and I had thought my gear was pretty decent (119 skirmish 3 weapon and several 119 skirmish pieces). Basically, I'm just wanting to know if the bee's evasion is as big of a problem as my friends made it out to be. I guess I can invest in some more acc gear but I wanted to see what other people have experienced. Thanks!

my group has won all 3 areas as 5. we brought buyers as 6th of varying jobs including a thf who afk'd for his. vere 119 monk, oat monk or tsumaru Sam , 4 song bard, whm, and sch. sch does a good amount of damage. we've even recently been taking 7 people (2 buyers) because it seems to only gain 5% hp with 7. I see no reason PUG's with 3 decent dd's, bard whm sch x2 or a geo couldn't do these:

I am planning to make a static myself for BEE on Bahamut. Why bad and DDx4 best? Do the sams i get have to have relic bow? Would like to get the kinks iron out before I shout for members.

Also is 7 man much of a difference in HP from 6?

9 is about as high as you want to go if you're trying to keep NM hp low.

I think someone said at one point in this thread that they parsed and found that adding a 7th meant 5% extra HP, so not too much if that's true. I'm curious about our job layout too. I know we have an aegis PLD, WHM, SCH (or BRD), DRG (myself), and two other DDs, one I think is a RNG, not sure about the other, possibly a THF or PUP with rng frame. We are planning on beading 345 for our first win, though some of us already have the win.

if you're beading 345, you could probably win with melee rdm's.

parse isn't accurate, it's level and hp go up every time the zone message displays

wouldn't worry about how many people you add as long as they fill useful roles.. 3 dd + brd and cor buffs & dia is enough damage to win even with 18/18

lol agreed.