XeKr

If Marcus > Rutger/Percival, then I’m fine with Allance > Zealot. iirc, we had this discussion in another topic but it has to do with how contributions are valued depending on team size and how forced deployment plays into it (extreme case, Edward). I think I prefer not to assume more optimal/solid teams as there’s a slope effect where characters are good because they work well with “high tier” units but those units are “high tier” in large part because they work well with (or often, are less obsoleted/outclassed by) characters that are good (“high tier”). Is FE9!Jill, a 2nd flier, as superfluous as these Hero Crests/Guiding Rings? Are Rody/Cecille/Ryan/Draug superfluous because of Luke? I don’t feel that strongly about it as I think we usually try to reasonably balance this line of thought but imo it’s messier than assuming non-specific teams.

For Dual Attacks, the AI will specifically move (breaking enemy order) to act as the supporting unit in dual attacks if the supporting unit can’t actually attack anyone that phase themselves. They don’t recognize Guard Stance, so they’ll attempt this even if it doesn’t actually increase damage (it should still increase their hit rate by 10 I think...). If that enemy can initiate an attack themselves that phase, then I don’t think they move specifically for dual attacks, and any that occur (usually combinations of 1 and 2 range units) are coincidences of enemy order. For example in the C3 case, it might be possible for the remaining Archer to move first and hit Jakob with Myrm attack stance, then Myrm hit Jakob with Archer stance for the ko, but the AI isn’t quite that smart. It goes, both the Myrm and Archer will be attacking this turn, and the Myrm is supposed to go first. He chooses Corrin because his sword does less damage to Jakob and his Dagger. If Corrin’s Def is low enough to die to the archer (<8, no Hp Boon), the Archer then sees a kill opportunity and goes for Corrin. Otherwise, he hits Jakob as he does more damage. In your C5 Kaze case, isn’t it just because Merc+Dark Mage kills Kaze anyway? The AI is good about finding the obvious fatal combinations (again the whole reason Corrin needs 8 Def is so the Merc+Archer combination isn’t fatal), and maybe that one it found first. (edit: If your Kaze can survive the Merc+Dark Mage and Corrin is also in range, I'm pretty sure the ai will go for Corrin. Is that situation easily possible with Hp tonic/Sakura aura?) The AI might have some of a self-preservation inclination but they generally still attack those it hits harder. In Conquest, the AI does tend to not attack if it can’t damage your unit.

Eh, in casual play I try to minimize resets to the very best of my ability (like avoid 1% crits if possible), so I would prefer dodge tanking to have more leniency. Ryoma is like 2-3hko’d and fights a lot of enemies, so (given how probability works) it’s kinda risky. Again in Conquest you have a number of very strong units with concrete durability, and often can bait enemies in small groups or through small corridors. Vantage Astra/Crit are nice but aren’t great odds either, it’s probably around a 130% multiplicative boost to “avoid”/chance of survival. Dual guard does complicate things. In more efficient play it’s just something to live with, though he might be kinda underleveled given all the Rescue the game throws at you. And ofc others generally have it worse.

Oh I was playing around with trying to triple refresh Camilla, Jakob, Silas, then sheltering Azura with Corrin and then manipulating lunge for more move, and transfer chaining Azura forward next turn. I don’t think everything quite worked out nicely but eh. ofc girls-only and all that here...still a reliable 3 turns is intriguing. 5000 is like half a statbooster so it could be worth it. Probably not that feasible without Zero though. tbh should allow Anna

Hmm, I actually thought Birthright Lunatic was harder than Conquest Hard because Ryoma to some extent relies on dodging and vantage crits/astra and the lategame can’t be turtle’d very well (and I had a lot of near 20/20 kids even). Only Corrin in certain builds is truly (very reliably) steamrolly throughout the game. In Conquest Hard, Xander, Camilla, Leo (with horse/reverse God) are all completely ridiculous, with a lot of raw durability as well. There’s also less massive waves of enemies, so things are easier to turtle (except endgame). I thought Conquest Lunatic was harder than FE12, because the latter has much simpler math, and it was mostly math that had been part of FE for a long time. Ambushes can surprise you, but otherwise if you knew the AI quirks the positioning felt pretty logical/intuitive to me. Also statbooster’d Kris steamrolls a lot of the game, if available. Both lategames can get crazy, and encourage fast play through them. In Conquest I remember I had to calculate auras, debuffs, percentage-based stuff, attack stances, dual guard meter, check various weapon bonuses, pp only skills, and so on. There were also obscure mechanics like individual character pair up bonuses which seemed random at the time. It was kinda a mess especially if one can’t read Japanese. Though now it’s understood that it’s possible to bypass a lot of the mechanics with smart reclasses/Pair Ups, so it feels a lot easier. And there’s always DLC/My Castle, if one chooses to use it, which can also make things much simpler (much moreso than Rainbow Potion, as strong as it is). The siblings are still the siblings, though Leo now has easier access to Horse Spirit, while Hinoka's Hp got nerfed. Hayato is notable because his bases got buffed greatly from Birthright, his growths were always good, and he can conveniently marry Nyx for Dark Knight. Silas got a durability buff but a Spd nerf, but there's ways to deal with it. Tonics/Pair Up/Practice Katana/etc. Kagero's durability is kinda terrible but Shurikens are strong, especially with access to effective damage on both Horses/Armors. And her high Str isn't as overkill. Compared to say Ryoma, she also has a niche of having very high Res. Don't feel like any maps are particularly challenging, just some can be slow to play through.

Do lunge ninjas end up helping? ;D I think it's Enfeeble staff? Could help for c12 Ryoma but probably not worth

8 turns is a benchmark number because that’s how long it takes for Ryoma to kill the Faceless going for Sakura. Like turns 5-8 are just waiting for him to clean up. All that’s necessary is surviving the middle dv turn 4 ep. Corrin+Kaze and Rinkah standing next to them should work, as the boss typically goes for Rinkah. iirc Corrin+Rinkah with Kaze adjacent can also work as Corrin gets more defense and so +Spd doesn’t double and kill some weaker Faceless (which can expose him to more attacks). This is supposed to facilitate even faster clears as Corrin avoids the Strength Seal, and you can push harder for the top left dv, but I find neither Corrin or Kaze have the durability to keep moving that direction (unless there’s some very clever positioning somewhere). 6-7 is possible with some dodges(/crits), however. For C5 a ridiculous Kaze can shave more turns, otherwise I think around 7-8 is a reliable standard. Kaze with Sakura's aura exactly avoids the 2hko and both Sakura/Azura can take a single hit, so there's a few approaches to it. The limiting factor is your Dragon has no 2 range and Kaze can't one-round the mages on enemy phase (without being ridiculous). I wouldn’t worry too much for Corrin’s exp. Yeah it helps early stat targets significantly and squeezing out any extra exp is clearly useful, but exp gain slows down so fast if overleveled so that he’ll quickly be around the same level, even if skipping many early kills (the bosses are nice though).

Yeah sorry if it seems like we’re/I’m backseating/micromanaging. There are a number of insights wrt to efficiency that people better than me have figured out (though much of the earlygame I did figure out myself). I don’t think they’re very important to blind runs of Lunatic Conquest, but more just curiosities to point out, after the fact. You can note if you prefer no gameplay spoilers or such.

I think it’s interesting to play efficiently with Male, but yeah it’s just suboptimal since Jakob has better durability pre-split, then can reclass to Paladin post-split. Up to you of course. No one really judging your turncounts and figuring out the "puzzles" of Conquest in a first(-ish) run can be rewarding itself. C4 isn't bad to 8 turn if triggering the center DV asap (bait boss with Rinkah), but (to the best of my ability/knowledge) lower requires some shenanigans.

2 in chests (4 uses). I think C9 and C20? But you still need Pass/Entrap/Lunge for enemies in the way.

Is there a way to get error bars on those or is the graph just too messy? tbh the 60-100 region looks pretty good.

For Ryoma, his room is like +20 Def/Res, but it doesn't heal much and Braves/Skills can work. For final boss, he's not hard, but getting to him might be without cheese (Infinite Staff Enfeeble Staff+Negative Chain)

Cool, this should be interesting. Random comments: In Lunatic, you can’t reset for stat growths unless the character joins in that chapter. It’s fixed then. +Hp/Def can 4 turn C1. But it’s harder later. FeCorrin/Jakob can very reliably 3 turn C2, but I’m not sure Felicia can. I think Corrin with 8 Def (I think yours does, as it seems you split the bridge enemies) can 5 turn C3, but the boss hit rates still suck. edit: Oh yeah or 4 turn with crits. Also Jakob with +1 Str can 5 turn without crits.

+Mag Cav is great when the Levin Sword comes around. +Str Cav is pretty good early. Samurai builds don't have to worry about Spd, though they don't have the move or damage of Elbow Room + Defender. I do like +Spd Cav, then buddy Nyx/Odin for Dark Knight. It hits Spd benchmarks easier and is less reliant on Mag growth (and more on massive stacking of various boosts). Also you double more early while Jakob is doing Paladin things and not giving you Spd as a Butler. Though Ninja is arguably the least thinking required, once past the earlygame. edit: Oh and I prefer +Mag Dark Mage on Birthright because you can Horse God some stuff but I don't know if other builds are better. Much easier to make sure you have the resources to skip than to face head on (though it's not impossible).

How much of that can be done reliably? I think there’s no place for rigging when evaluating/ranking characters, for the same reason averages have been used, ever since people started discussing FE seriously. Any lucky/unlucky personal experience has no place in tiering discussion. Personally, I may favor a faster standard than many others, but I’m probably also one of the strongest advocates for highly reliable clears. I think this playstyle is highly applicable to discussing what characters are “good”, because it involves units that can kill a lot of enemies (and complete the game) very quickly and reliably. I don’t think it’s controversial to suggest that’s good in FE. I think Hinoka might be a tiny bit overrated, but I don't think Subaki can replicate everything she does (reliably). The optimal deployment argument is pretty dead (no one likes it), so in situations where Hinoka is needed as more of a combat unit she wins easily.

Can always lower difficulty down. Go for it I say. And if you've beaten L' and L+, you should be fine. Hopefully. >_> (I'm not sure how much you've read about certain lategame stuff)

I’m kinda liking Ninja more and more, though what I tried was still reclassing Jakob first and building Corrin’s weapon rank during C10. Camilla/Jakob still carry until Ninja really starts to matter. Less Shelter makes me kinda sad though. It sucks we can’t go +Str Cav and Buddy/Marry Seal Ninja in time for Chapter 17 (I think. edit: hmm maybe with paralogues..). Apparently Odin saves turns, didn’t you know?

Gwimpage is an advocate for +Mag Cavalier and afaik has the LTC(/speedrun?) records. C8 (I think 5 turns, 3 villages?) is trivial with Paladin Jakob and I don't see how Odin gets 4 levels during it. Jakob both contributes significantly to completing Mozu's Paralogue efficiently (8 move, after all) and wants the exp to snowball exp/weapon rank. Nosferatu obviously doesn't have 8 move and 1-2 range one rounding everything midgame, as Levin Sword Pally does. I'm not sure what you mean by rigging is reliable?

Jakob is best reclassing early to start building the weapon rank. I'm interested to see how -Spd handles the lategame. C10 is rather hard blind, but it's fairly trivial if you have Paladin!Jakob. Even without him, it's also not too bad after studying the enemy stats for a while and using the appropiate tonics for Selena/Camilla/etc.

There is no Lunatic+ but Conquest will probably feel like it if you're playing blind and thought Birthright was at all challenging (and I did feel it a bit lategame, admittedly).

Well I did say depending on how one values Pair Ups. ;P. I also don’t think Effie is that good as a combat unit, given how strong Paladin!Jakob (for female) and +Str Cav Corrin (for both) are early. It’s kinda complicated but the short story is the others don’t fly, so imo it’s a big difference. Also, since Gunter’s classes give Def and he gives more as his character bonus, he doesn’t really lose too much in durability either. It’s kinda a tradeoff of Spd/Res vs. att/flight/hit (and other small differences between Felicia/Jakob/Beruka/etc) +Str Ninja probably doesn’t care about Spd/Res so it’s an easy choice to go for flight. Cav builds are a bit more complicated and gender might factor into it but flight is obviously still strong. Marriage Sealing the other Servants into Wyvern is a possibility, but I’m not sure it’s wise to give up early Ninja (which really wants to reclass asap) or Cav. Male can friendship Silas, but you already lack an extra early mount without Jakob. It might go C after C7, B after C9, A after C11 if the 3 cap from FE13 still applies, and that’s a lot of chapters as a foot unit with no Shelter. Actually reclassing to Wyvern is an interesting decision, though there’s a few issues (Spd early if not +Spd, Mag later if not +Mag and no Elbow Room/Malefic Aura, no swords/Yato. And being a flier against certain later bosses…) [spoiler=shelter hype] A somewhat relevant tangent, I think Shelter also makes a difference, especially in the Ninja build. Corrin won’t have it and Felicia can’t marry Corrin for it. She can marry Silas for it though it takes a number of chapters keeping them together and also gaining 2 levels, hard to do efficiently. Jakob loses +Str/Mag/hit badly, in exchange for +Res/avo, and still doesn’t fly. I’m sure this sounds like hyping Shove (which some people probably will eventually), but multiple refreshes make a much bigger difference than 1 space. Obviously there’s the old rescue-take-drop (no canto) for a few extra squares, but here’s an ideal example scenario. Move Camilla(+her Pair Up) forward. Dance them with Azura. They move ahead. Shelter Azura with Jakob. Have filler1 transfer Azura, switch and Dance for Jakob. (Pair Up someone with Jakob and) Move him ahead. Have filler2 transfer filler 1 away from Azura. Shelter Azura with Xander. Have filler3 transfer Azura, switch and Dance for Xander. (Pair Up someone with Xander and) Move him ahead. Have filler4 transfer filler 3 away from Azura. Shelter Azura with Gunter. Have filler5 transfer Azura, switch and Dance for Gunter. (Pair Up Corrin with Gunter and) Move him ahead. Have filler6 transfer filler 5 away from Azura. Shelter Azura with Silas. Have filler7 transfer Azura, switch and Dance for Silas. Move ahead with Silas. Next turn, filler7 can give Azura to Silas to keep up with the others already ahead. (I think that’s arranged right) Okay that’s ridiculously ideal and probably more slots/spaces than we have (Peri exists too..) but hopefully the point is made. >_> In the obvious practical cases like C20 or C23 you can fly smaller teams over the gaps/wall/etc but it requires the users of Shelter to fly (or be less efficient). @joshcja: You should make a detailed log with turncounts/positionings. It’s very unintuitive that Odin is that good. How is he getting 4 levels in C8? And how reliable are these strats? Especially when you say stuff like “Odin will have the same/more magic as Leo” when 20/20 Sorc Odin averages 29 which Sorc Leo reaches at --/9-10, 20/9 Sorc Odin is around base --/2 Sorc Leo. And he’ll outlevel him somehow even given the new exp curve? And +Mag Cav might want Spirit Dusts given the questionable growth at times and 1 base in Pally. +Spd Cav will want them as well, even in the magic build of going Dark Knight after Elbow Room/Defender for Malefic Aura and the 6 base. It seems you have an entirely different approach though. Re: Mozu. I don't think she's very good, but I will admit she carried reliable clears (dat accuracy) the first time I beat Conquest Lunatic.

I just brute force calculated all the values (floor((3A+B)/4) < disphit) with a meshgrid. Math check later. edit: numbers seem okay, 2nd column matches with true hit chart on SF, and I got the same as Dark Holy Elf for at 75% disp. Obviously need to note averaging where applicable.

I’m okay with her in A, but there’s a noticeable gap between the S tiers (imo Azura, Camilla, Corrin, Jakob, Xander, alphabetically) and the others in Conquest. She’s worse than Leo and Silas. Maybe some others depending on how one values Pair Up/staff botting. >_> (and I suppose how good the latter actually is in Hard) Tbh to me she’s kinda comparable to Gunter. As Pair Ups for better units, Gunter gives more damage/hit and flies (+Shelter is just amazing for efficient play) and is more important as the pocket unit lategame, Effie is more important earlier. 10/1 Effie is pretty comparable to rejoining Gunter, with worse weapon ranks, though she lasts longer. Overall, maybe I’d rather have Effie, but that’s in large part because we have very few (tanky) units available earlygame. And if we’re assigning credit for that, Gunter probably gets some for chapters 2, 3, 15, where he’s invaluable. Keaton’s promoted class gives +10 hit and he’s probably using Beaststone for the Spd which is 10 more hit. But yeah, 1 range is sadness. Do people find the low level staff users mattering much? Rescue is E, which the Falcons can use. Latest!Izana is the best candidate to hex the later bosses. So it's just some token healing early, and the good characters take very little damage then (or are far off ahead of 5 move healers, even with 2 staff range). And in playstyles where healing matters more, if you have Felicia, she heals and debuffs and can Explosive Shuriken midgame and half-replicates Sakura's Personal with her class skill and can get a fast Inspiration maybe.

All the numbers I’ve provided are player-side primary attacks as well. I was about to post something similar to Dark Holy Elf, some kind of weighted averaging. Re: the gba games, I think there are lua scripts that display the RNG, so it should be known. Not sure about 9-12. I’ll poke around in the fixed stat dlc for any useful hit rates. Swords against Axes like in Anna's level with the many berserkers might work for fast collection.

Would you have Keaton in A though? I think she’s better as well fwiw. Also a tangent, I guess I don’t favor grade inflation. Having all the units clustered around S/A/A- doesn’t make much sense. I understand it’s supposed to imply that a lot of characters are “good” and viable in this game, but it doesn’t really draw strong distinctions between students characters. The existence of S already implies we don’t like thinking of C as the “average” grade. If a bunch of characters are comparable in combat or role to each other, I think they should be in the average rank, B. And to be fair, the tier descriptions, however closely people chose to adhere to them, did allude to this (“ahead of the pack” for A). It’s also possible a lot of people voted them S and a lot B/C, but eh like I said the midtiers are a mess. I’m not sure how others are counting online features (which can get out of control quickly), but Forging without other castles is unreliable early. In the likely case you don’t have the right material, you need 5 minerals to convert into the 1 necessary. Usually I play without forging or I’ll build up materials to forge Kodachi/Hand Axes/Javelins for later Ninjas/Mages. Arena refreshes every chapter so you can reliably get a few forges of whatever by then. But again, not early unless you wait many 6(?) hour cycles for the gathering spot. I haven’t really tried Arthur as a combat unit, because his Pair Up stats are just so amazing for Effie, Niles, Selena, Camilla, etc. But if he is a relevant combat unit, I suppose it’s easy to see him above Lazslow or whoever joins later.