Ok, I started having another go at this, but it turned into a massive wall of text. Let me try this approach instead (which is still a wall of text, but I'm trying my best...):Originally Posted by Dejnov
Let the average dmg per SSR hime per turn = [A*(n-1) + 7.5+B]/n where A is the average dmg from attack and abilities, the 7.5 is from FB (5x + 50% burst streak), n is the average number of turns it takes to burst. A varies a lot depending on DATA rate and red abilities, but I will be generous and set it at 3 (most himes are closer to 2-2.5 in my experience.) B is burst multiplier adders like PF, exceed and Mike AW's ability 2. I will set B as zero for the base case just to make things less favorable to a burst build.
When you replace a generic SSR with Mike AW, you might lower A but also lower n and raise B. I would argue that Mike lowers A by at most 1-ish (her red ability doesn't hit as hard as some other himes but have a bunch of team-wide buffs) and she lowers n by at least 2 (2 if your team has close to guaranteed DA on average, closer to 4 if your DATA rate sucks.) Now, remember, we only replace 1 hime to accomodate Mike, so reduction in A is only 1/5. Mike also raise B by 1x (less on average if you burst much more frequently than the 7T CD.)
- If n before Mike is 9 (9 or 10T roughly on average if you try to get to FB without any DATA buffs or burst fill) then average dmg would be (3*8+7.5+0)/9 = 3.5x.
- With Mike AW, if I reduce A by 0.2, increase B by 0.5 and decrease n by 2-4, we get between 3.54x-3.84x, which is between 4-9% more dmg.
- If you repeat this with what I think are more realistic assumptions of A = 2.5, Mike A reduction of 0 (it can increase A if you time her buff with burst), it's 5-18% more dmg. If you further reduce base n by 1, it becomes 6-24% (i.e. the faster your team, the more valuable Mike becomes.)
- For reference, if you own a soul weapon and have a fluffy friend and full light sub eidolons, going from using Barong to fluffy as your main eidolon gives you a 22% boost (elemental going from 170% to 230%, less if you don't have all light sub eido), so these numbers from Mike are nothing to sneeze at.
For Tish, it's a similar story. Her ability 1's dmg and CD is about average, but the debuffs on it make it on par or better than a lot of other hime's like SSR Arty or LT. Ability 2 boosts atk, crit and most importantly 50% TA, and she can keep it up on 2 himes 5/6 of the time. I will only focus on the TA part. The TA alone boosts A by 1.5 on two himes (so 1.5*2/5*5/6 = 0.5 on average.) If you cast it on the bottleneck himes, it can also easily reduce n by between 1-2.
- With a Mike AW starting point of A=3, B=0.5, n=7, Tish adding 0.5 to A and reduce n by 1-2 will boost dmg by 14-18%.
- Repeat with A=2.5 instead, Tish will boost dmg by 17-21%.
- Remember, we still have not accounted for Tish's atk and critical boost and the fact that fast teams make Mike AW even more valuable.
Lugh is a slightly different animal, but using this framework, she basically boosts A by a lot and reduces n by about 1 or so.
Also, you can extend this framework to evaluate someone like Sol, except you need an extra variable showing how long the team can survive with or without Sol. So, it becomes like [A*(n-1) + 7.5+B]/n*E where E is endurance in turns.
- Sol tends to reduce A and increase n (if you already have Tish and Mike though, it can be neutral on n, since you can just stick her in the last slot as the previous poster mentioned.) Her value is from increasing E.
- If normally you die after the 1st OD at like T3 or T4, and Sol makes you survive till T6 or T8, increase in E is like 100%, and Sol will blow everyone else out of the water (and it's not just heal btw, her atk down is very nice, too.)
- My personal experience with Sol usually falls in two categories once my hp got high enough though:
- For easier content, up to and including advent rag, Sol increases E by so much that the battle has long ended before I use that many turns, so her value is sort of wasted.
- For public raids, I get a guaranteed potion which together with Tish's heal, usually gets me to T8-10 or so, depending on which raid we're talking about. Sol usually just adds about 2-4T more, so E increases by about 20-40% or so, which is a lot, but A takes a hit, and if I have to kick someone like Mike or Tish, A and n takes such a big hit that it's at best a wash. If the raid lags (which it usually does) and you want to try to MVP it, your best bet is to go for high A low E strategy, doubly so during BT.
Btw, lest I be accused of being a Sol-hater (I am not), I am only saying Sol falls behind Mike and Tish in value as your hp gets high enough. If you don't have other heavy hitters like Take or Metatron or Lugh or something, Sol is still very very good and you should keep her on your team. She only got relegated to sub 'cos I do have Take and Metatron (but no Lugh, yet...) Oh, and perhaps paradoxically, I would argue you should mtix Sol before you mtix Metatron or Take, well, at least after you have Lugh. Sol will help keep Lugh alive, so if you're in for a long fight, Sol is most likely better than another heavy hitter 'cos Lugh hits so hard that it offsets whatever reduction in A Sol brings. Utility himes like Sol and Eros are also more useful if you intend to use light for almost all content once you have your basic DPS machine.
Very true, forgot about that one. You can easily fill the whole grid with hammers, too, not just 5 hammers. Metatron is also a hammer user, so bonus if you have her.
Lugh's lance FLB is large assault and medium elaborate, not terrible, large assault is always nice... But yeah Tish's (assault, exceed, ascension), and Mike's (assault, defender) are much better. So I'm even further from the ideal light grid than I thought...
That's an excellent idea. I think I will stick that on Lugh when I mtix her, since it should become 700 regen on her if I understand her assist correctly.
You are right... I can't believe I didn't notice this before given that I have quite a few Thunderbird friends and use them regularly... I always thought I'm getting through trash mobs slower with Thunderbird 'cos of the lower elemental, but it's probably that and lower DATA... My bad.
Sooo, I see Lugh using ability 2 TWICE in a single turn here around the 3:00 mark and get 60 burst from just that (she needs to have bursted already for it to reach that ridiculous level, otherwise it's just 20/use at lvl 75+.) Unless taco one is nerfed... But if my eyes are not deceiving me, tell me how that is not OP and blows every other light himes out of the water for raw DPS. Combine it with Dejnov's regen accessories bonus idea, she's not as frail as people think either.
It depends on a lot of factors, like how many red abilities you have and what your DATA rate is like, but the last time I ran some back-of-the-envelope numbers, I think at around 150%-ish assault and consistently bursting every 4 turns or so, exceed is almost as good as assault. Elaborate still suck though, so if you have a choice, listen to Slashley and don't choose axe (or my gun and sword grid ideas - I consider those sort of free since you'll have excess ori when you're consistently getting all 5 ori every cycle.) If you missed the old light UEs like me, axe grid isn't really an option anyway.
Results 1 to 10 of 149
Reply With Quote