Thread: Prelude to Violet Patch 4.4 Notes (Preliminary)

Originally Posted by Dualgunner

Unless they keep making fights mandate a caster (like Ultimate) nerfing SMN and bringing the other casters in line with nerfed SMN only puts MCH into that fourth slot.

I'm actually kinda interested in this statement, so I'm going to go through some rDPS numbers. This is all VERY rough so I hope that someone with a better understanding of the methodology of speedkill compositions can comment on this :blobheart:. The numbers after the buffs or nerfs are just estimated with RDM buffs being 2.5% and BLM buffs being 4%, rounding more or less to the nearest 50 rDPS.

Here's a few Chad parses to look at,

http://www.xivrdps.com/encounters/QY9dbMBxjCHTtvD3/29 (Aya Liz, #4 MNK, 7209 rDPS) #3 Fastest Run of Chadarnook.
http://www.xivrdps.com/encounters/vZ6mfd1VN4rycnhJ/1 (Laqi Thish, #12 BLM, 7163 rDPS) #78 Fastest Run of Chadarnook. This would be 7450 rDPS after the Fire IV Buff.
http://www.xivrdps.com/encounters/mwcWkHDxJ4bM97Yp/15 (Nixillis Nihiru, #2 SMN, 7108 rDPS) #12 Fastest Run of Chadarnook. this would be 6964 rDPS after the radiant shield nerf.
http://www.xivrdps.com/encounters/YdCAc78aDbz1Hhy6/6 (Giga Catastor, #5 RDM, 6700 rDPS) #79 Fastest Run of Chadarnook.This would be 6850 rDPS after the RDM buffs.
http://www.xivrdps.com/encounters/rqZDbkd1CMhRajW8/8 (Nox Aero, #1 MCH, 6624 rDPS) #37 Fastest Run of Chadarnook.

A couple disclaimers, xivRDPS seems to act weird about some buffs, but I'm reasonably certain these numbers are about right. Secondly this is chad so of course MNK is really high with quad phys and triple melee. I'll post a few on Guardian as well. Also I'm not the best informed raider on the planet so I'd have to run these by the people on the Balance to get a more robust understanding of what these numbers mean in terms of speedkilling, but I digress...

Without diving too deep into these reports (you can click 'Original Log' in the bottom left to see the original fflogs report for yourself) this seems about right. MNK, BLM, and SMN are all rather potent jobs, and in a fight like this with incredibly high uptime it makes sense that they'd be doing this well. This more or less speaks for itself though; if the best MCH in the world is ~500 rDPS behind the SMN here then I think it's rather clear the other three are more of a benefit. MNK has a bit of an edge here over the other three jobs, and outside of rDPS the only thing it really contributes is Mantra. It is very popular for Chad though so I'd put my money on it being the best choice here even if the numbers are close.

Note, I didn't take the best DPS for these jobs, just more or less the fastest with these comps. I might have made a mistake sorting but in this case where all the examples are top 12 or better I think I'm safe.

Also l m a o at how close RDM and MCH are here. This is the only result here I'm legitimately surprised by.

Guardian is an ugly duckling, but for the sake of a better comparison here are some of the best fourth slot parses for those;

http://www.xivrdps.com/encounters/wtFRGfVnW6kjZ9YA/13 (Laqi Thish, #3 BLM , 6585 rDPS) Unranked by speed.* This would be 6800 rDPS after the Fire IV buff.
http://www.xivrdps.com/encounters/K87BJwpxhfaYN2zb/40 (Menhera Girl, #3 SMN, 6652 rDPS) #1 Fastest Run of Guardian. This would be 6543 rDPS after the radiant shield nerf.
http://www.xivrdps.com/encounters/dnP23JQm71f4rq8X/8 (Lisa Armstrong, #3 RDM, 6202 rDPS) #19 Fastest run of Guardian. This would be 6357 after the RDM Buffs.
http://www.xivrdps.com/encounters/38FDRgfLyYzPVk4b/28 (Aya Liz, #1 MNK, 5985 rDPS) #22 Fastest Run of Guardian.**
http://www.xivrdps.com/encounters/Fd4R6VfmQAk3WYMx/3 (Kimchi Samgyupusal, #9 MCH , 5904 rDPS) #19 Fastest Run of Guardian.

Alright this is where things get interesting again. MCH isn't terribly strongly represented in the top Nth% of Guardian, in the top 50 it represents probably less than a fifth of all reports. Most of them have SMN, which probably makes sense given how high that #1 is. It'd be tied with Laqi after the radiant shield nerf, which to me means that after the patch things will be switched (Modest BLM buff, Minor-ish SMN nerf).

To build from that, MCH again is the lowest, but still more or less tied with the MNK parse. It's worth noting (hence the '*') that the Aya parse was a 3/1 triple melee set-up. It's more or less catered to Aya in comp, but it is still a remarkably fast run so I have included it regardless. Post-patch then it would indicate to me that it makes more sense to go with the casters over either of the physical options.

It's really annoying generating all these logs but I'll do Godka for the sake of it. It's worth noting a few things at this point before I do that:

So far, 2 things have become abundantly clear. BLM is under-represented in speedkills for a reason that these numbers themselves do not indicate. BLM is essentially tied with SMN and yet there are no Black Mages in speedkill groups. Hell, just finding a BLM in the speed rankings is a pain so I just grabbed another log from the same BLM. It seems strange to me that despite the fact the two jobs are fairly close they don't have nearly the same representation.

Secondly, MCH is still rather useful in that it has refresh. Groups may still choose to take the rDPS hit from bringing a MCH if it means more mana and thus easier DPS for their AST. A lot of these logs show their Astros dropping well under 5% MP over the course of the fight, so I'd hazard a guess that if their AST had to heal more then MCH might become more prevalent. Speaking of healing more...

Here are some godka logs;

http://www.xivrdps.com/encounters/ThzXjdCkybcY8NKx/19 (Leith Noir, #3 SMN, 6964 rDPS) #2 Fastest Kill of Godka. (by 7s). This would be 6779 rDPS after the radiant shield nerf.
http://www.xivrdps.com/encounters/W1FwqZhKtnY47pyA/11 (Lisa Armstrong, #6 RDM, 6590 rDPS) #53 Fastest Kill of Godka. This would roughly be 6750 rDPS after the RDM buffs.
http://www.xivrdps.com/encounters/VJG1ckvqPX6NbRFD/6 (Kuru Mi, #15 BLM, 6551 rDPS) #84 Fastest Kill of Godka. This would roughly be 6800 rDPS after the Fire IV buff .
http://www.xivrdps.com/encounters/WcGXfrzJ9YQCayLA/6 (Ember Lunalesca, #8 MNK, 6385) #26 Fastest Kill of Godka.
http://www.xivrdps.com/encounters/tg9xZYj8w4DGdfab/27 (Shasha Pony, #5 MCH, 6303 rDPS) #1 Fastest Kill of Godka.

So as expected 11 of the top 20 ranked parses use MCH, 10 use SMN, and one uses RDM. Well that last one surprised me, the rest didn't however. Casters are still over MCH by a sizeable margin with MCH and MNK more or less being equal. If you consider how the 'After Buffs/Nerfs' DPS is going to change, I think it's fairly reasonable to assume that casters are going to be quite comfortably ahead of MCH, with BLM eclipsing SMN quite readily.

Going through these logs it's pretty easy to see where MCH has the edge regardless; AST MP. Astrologians have the roughest time managing their mana levels, more so than any other healer. As such it's no surprise that in a fight like Godka often times the healers are flirting with sub-3k mana levels quite often during the fight. Low-piety builds to emphasize DPS further add to this issue, as most of these Astrologians have less than 18k MP, so their natural MP ticks are a bit weaker as well. Aside from that, however, MCH doesn't really offer much else. It has dismantle, but melee can already manage Ultimate Embrace with feint, and a caster can manage LoJs and Forsakens with Addle. The coverage is lower with Addle, but still gets the job done.

All in all though, I think it's fairly clear that MCH isn't in quite the cosy spot it appears to be in. If the mobility issues of casters can be worked around sufficiently to allow them high enough uptime -- and that in turn doesn't affect the DPS of the rest of the group -- I would hazard to say that SE may in fact be able to offer five reasonably viable options to the fourth raid slot.

Now, to be clear, I'm using pretty cream-of-the-crop comparisons here. The lowest ranked person's parse I used here was #26, which is a rank I can only personally say I've reached on WAR a couple times in Savage fights. These are also incredibly high peforming parties, with the lowest rank (not including Laqi's pug) being #86 on Godka. I think that these numbers will apply even further down the rankings list in parties with relatively even party compositions and performance, though the absolute difference between choices might shift a bit. There's certainly opportunity for other casters to show dominance here though, and I'm interested in how they'll flex their magical muscles.

Oh god I wrote this in reply to a post 12 hours ago what have I done.

tl;dr

MCH is still highly sought after for its dismantle and its decent raid buff, but realistically given these numbers I think it's fairly acceptable to assume that there's going to be a legitimate argument, even a compelling one that the casters could wholly replace it in the next tier. SMN, BLM, and even RDM are looking like at high levels they could effectively make the difference in utility. These numbers are obviously caught in the high-end of raiding, so there is certainly some variance in these numbers so I wouln't draw anything more than a trend from them, however...

Going into next tier just from this it would seem

High Uptime 3 Melee Special:
BLM
MNK
SMN
RDM
MCH

Low Uptime Special
BLM
SMN
RDM
MNK
MCH


Addendum 1:

Aya on the Balance said that MNK on Guardian is crap but they haven't put much effort into trying to improve their time there, so there is a chance that with more work MNK could perform better in that fight. Nobody has really tried though.

You also forget one thing called party synergy. How much those jobs benefit from the buffs, like mch benefit 230 dps from trick attack while blm not that much etc. Another point is mch can raise DRG utility by another 300+ dps. BLM partysynergy is still pretty bad compared mch, but their personal dps is gonna be really absurb and they can probably even surpass mch in speed kills after that patch. Monk still have too good party synergy and personal dps so I think blm can not reach monks level as long mechanics allow triple melee.

Originally Posted by Sunako

snip

That's why I talked about rDPS, and not pDPS. The raid DPS contribution of the different jobs is their pDPS - raid Buffs + the amount they buffed party DPS. Rolled into pDPS are all the interactions you describe. BLM might not benefit as much from party buffs, but that doesn't matter of they do more DPS with buffs removed than a MCH anyways. The xivrpds logs I linked show the difference in the raid DPS contributions if you wanna click on 'em.

http://www.xivrdps.com/encounters/vZ6mfd1VN4rycnhJ/1
http://www.xivrdps.com/encounters/rqZDbkd1CMhRajW8/8

If you look at these two different reports it's easy to see on the right hand side. Nox definitely has higher raid contributions than the BLM, but when you consider the DPS they bring it's still lower than the BLM.

Originally Posted by Sunako

I mean even if mch rdps looks pretty low, mch still bring lot more rdps for other members than what blm would. Like drg would not get 600 rdps from piercing without mch.

If the rDPS gain on piercing is 600, 300 per ranged phys, and then you replace the MCH with a BLM, you gain rDPS (6624 => 7163) and then lose 300 from the piercing loss. Just doing this comparison, you still would gain 240 DPS.\

edit: I take this back entirely because this is just a misunderstanding in full:

Saying another job's rDPS goes down because another job is introduced is immaterial so long as the job that gets slotted out has higher rDPS than the job it replaced. Sure, a DRG might do 300 fewer rDPS than a BLM because it's not getting that piercing, but that 300 DPS it lost is just rolled into the normal rDPS of the BLM. It hasn't disappeared, you can't count it twice, it's just DPS not affected by the piercing. DRG's rDPS drops, but Overall DPS does not change because the effect of piercing was removed in the calculation. Raid buffs just drop, but the end rDPS of the group remains the same if you sum the damage on the right hand side. Or it goes up in the case of BLM vs MCH on O6S.