Jump to content

Lunatic+ playlog/guide/walkthrough - COMPLETED


Recommended Posts

some of the tactics that will be necessary going forward. In fact, let's list out a few of them right now; some are map-specific, but others represent things that are important for the rest of the game:

  • The enemies on the bottom floor don't attack unless you end your turn in their range. This means that you can walk up to Kellam, Talk, Pair Up, and walk out of range on Turn 1 without having to deal with combat. This lets you spend as much time as you need to get your crap together. Top-half enemies work similarly.
  • Learn to love the "Remove" command. It will unequip a weapon. This is how we deal with Counter. Note that if you trade afterwards (even if it's not a weapon that's being traded), a unit will automatically re-equip the topmost usable weapon.
  • Enemies with Pass can ruin your day when it comes to setting up choke points, but while they can pass through your units, they still need an empty space to land on at the end. This means that you can make choke points work by clogging up every available spot on the other side of your tanks. This comes in handy on this chapter, which has natural chokes available.
  • While Pair-up is important, you also need to know how to effectively use the Switch and Transfer commands. When you are operating in tight quarters, or need to provide a temporary stat boost, or need to pass someone down a line, there is no substitute. You always move the Support unit, so Switch will get the right person in that slot.

I really like how you did this, since I tend to use these strategies without explaining them, but doing so makes it much more clear and easier to pick up on for players inexperienced with Lunatic+.

As for your characters, I am really impressed with how you got Miriel to level 8 already and I am surprised that Avatar surpassed Fred as early as chapter 2 (when in my run I am at Chapter 11 and Fred still has a significant lead).

Anyway, I am sort of worried about your Chrom being underleveled, but I suppose if you've done it once with essentially an Avatar bow solo, you could do it again. Just out of curiosity, what is the exact path you plan on taking with your Avatar?

Link to comment
Share on other sites

  • Replies 431
  • Created
  • Last Reply

Top Posters In This Topic

I really like how you did this, since I tend to use these strategies without explaining them, but doing so makes it much more clear and easier to pick up on for players inexperienced with Lunatic+.

That's the idea. I feel as though at least some of the frustration with Lunatic+ is due to people trying to make vanilla Lunatic strategies work, when this mode is specifically designed to unravel such things. You need employ some truly insane tactics to make the most of your army, but some of those moves are counter-intuitive, so I figure it will help to spell them out.

As for your characters, I am really impressed with how you got Miriel to level 8 already and I am surprised that Avatar surpassed Fred as early as chapter 2 (when in my run I am at Chapter 11 and Fred still has a significant lead).

It's very easy to train a 2-range unit in Ch3 because the only 2-range threats are the three Archers that you didn't already kill on the LHS. I wanted to train Chrom, but it would have been seriously difficult to do so, and I'll have opportunities for him later on. I very nearly trained Virion instead of Miriel, but he has no future on this team because of the complete nonsense that is his class set. Archer, Wyvern Rider and Mage? What?

I think that Louise is as far along as she is, just because I intentionally stuffed her full of everything that I could. Prologue and Ch1 were used to level her up; she got almost every resource. Frederick would have only gotten a few levels if I had done the same with him, and his Silver Lance would be gone by now.

Anyway, I am sort of worried about your Chrom being underleveled, but I suppose if you've done it once with essentially an Avatar bow solo, you could do it again. Just out of curiosity, what is the exact path you plan on taking with your Avatar?

Grima is as troublesome as a rainy day to this crew. He has Pavise+ (which the archers will ignore), and Lucina is Good Enoughâ„¢ with Parallel Falchion. Consider that Warrior!Morgan with Bowfaire, a STR tonic, 49 capped STR, A-rank GK Lucina Pair-up and a max forged Brave Bow has 81 atk. This is 15 damage a shot (17 with Rally STR) to Grima, leaving me to come up with only 31-39 extra damage to finish him off. Lucina as a GK with say, 40+2 STR (not even close to the cap) does 14 damage per Dual Strike, meaning if I proc DS on three out of four hits, Grima is dead. This is with one attack. I can still follow-up with a couple blasts from Avatar's Longbow (5x2), or give her a Brave Bow too (6x4). Training Lucina is very easy because I pass Veteran down to her; she made it to Level 13 Great Lord without ever being the Lead unit (just sat in Morgan's pocket for the whole game).

But I will train Chrom. I need Charm to make this work smoothly, so at the very least he needs to hit Lord level 10. Even though things are slow-going now, I expect it will pick up once he is able to be Louise's support partner more often (she uses Frederick a lot at the moment).

Louise's class path is probably going to be Tactician -> Mercenary -> Bow Knight -> Sniper. Bow Knight is just a promotion of convenience, because of Mercenary, although Bowbreaker is legitimately useful once enemy hit rates get to 150+ (the bow strategy results in a lot of focus fire from archers/snipers on Enemy Phase, for obvious reasons).

Edited by Interceptor
Link to comment
Share on other sites

Grima is as troublesome as a rainy day to this crew. He has Pavise+ (which the archers will ignore), and Lucina is Good Enoughâ„¢ with Parallel Falchion. Consider that Warrior!Morgan with Bowfaire, a STR tonic, 49 capped STR, A-rank GK Lucina Pair-up and a max forged Brave Bow has 81 atk. This is 15 damage a shot (17 with Rally STR) to Grima, leaving me to come up with only 31-39 extra damage to finish him off. Lucina as a GK with say, 40+2 STR (not even close to the cap) does 14 damage per Dual Strike, meaning if I proc DS on three out of four hits, Grima is dead. This is with one attack. I can still follow-up with a couple blasts from Avatar's Longbow (5x2), or give her a Brave Bow too (6x4). Training Lucina is very easy because I pass Veteran down to her; she made it to Level 13 Great Lord without ever being the Lead unit (just sat in Morgan's pocket for the whole game).

But I will train Chrom. I need Charm to make this work smoothly, so at the very least he needs to hit Lord level 10. Even though things are slow-going now, I expect it will pick up once he is able to be Louise's support partner more often (she uses Frederick a lot at the moment).

Louise's class path is probably going to be Tactician -> Mercenary -> Bow Knight -> Sniper. Bow Knight is just a promotion of convenience, because of Mercenary, although Bowbreaker is legitimately useful once enemy hit rates get to 150+ (the bow strategy results in a lot of focus fire from archers/snipers on Enemy Phase, for obvious reasons).

That sounds like a good plan. I haven't really thought ahead to Grima and I have no experience with endgame Luna+ to base anything upon, so I have no idea what endgame Luna+ stats look like. Do you think it will be reasonable to get Avatar to cap stats (everything I've heard seems to imply that hitting caps is pretty reasonable before endgame, but those people might not have been using a full team for all I know)? Also, I wasn't sure how the money situation looks in end-late game since max forge braves are expensive, but considering you've done this before I'll take your word for it since you've done this before.

I'm sure being able to pick up armsthrift earlier would have helped. I really regret that I wasn't (and probably won't be) able to get it on my Avatar, who was barely salvageable even with a relatively early promotion.

Link to comment
Share on other sites

Lunatic+ Grima's stats are the same as in Lunatic and the only difference in skills is Pavise becomes Pavise+ (a meager 20% proc rate difference). The nearby goons will have skills that will make things problematic if one can't kill Grima in a single PP/EP set, though.

I've like to point out that when I did it, I have Avatar with an unforged Brave Sword going at him with Chrom (I don't think he was quite at max Str, but he did have Basilio's Rally Strength and a Gaius' Confect) wielding Exalted Falchion as support. I didn't have any trouble taking him down that way (although I had to get the kill as an EP counterattack). Theoretically, an unforged Brave Bow with Parallel Falchion support should be roughly equivalent (if not better) due to a lack of Aegis+. Forging it would only help seal the deal.

Link to comment
Share on other sites

That sounds like a good plan. I haven't really thought ahead to Grima and I have no experience with endgame Luna+ to base anything upon, so I have no idea what endgame Luna+ stats look like. Do you think it will be reasonable to get Avatar to cap stats (everything I've heard seems to imply that hitting caps is pretty reasonable before endgame, but those people might not have been using a full team for all I know)? Also, I wasn't sure how the money situation looks in end-late game since max forge braves are expensive, but considering you've done this before I'll take your word for it since you've done this before.

As KTT said, Lunatic+ Grima is the same as vanilla, except for the Pavise+ swap (which makes little to no effective difference, because Grima would already have max SKL and Rightful God anyway).

Honestly, a bow-using team is better positioned to take out Grima than basically anyone else. Grima's DEF is the same as his RES, and a Brave Bow is better than Waste or a Celica's Gale (because of HIT/mt, respectively). An underrated benefit of bows is the HIT rate: people laugh at skills like Prescience or +20 HIT, but Grima has 90 avoid! He is no joke to land four consecutive attacks against.

I had no trouble maxing everything but RES and MAG on my Sniper Avatar, but the real MVP of Lunatic+ lategame isn't Avatar, it's Morgan. Morgan has better caps (because of being a child), and levels more quickly (Avatar can pass him Armsthrift, so he doesn't waste time as a Mercenary).

As for money, it's no issue on this team. First, I have three Armsthrift users (Avatar and Morgan with an easy 100% proc rate, Gregor with a pretty good one), and two units with infinite PRF weapons (Chrom and Lucina). Secondly, Gregor has to spend some time as a Barbarian in order to promote directly to Warrior, so he picks up Despoil. Since the strategy of this team revolves around Player Phase kills, that means small Bullions basically rain from the sky even with his modest LCK stat. This limits my expenses to tonics, Physics, Levins, and random consumable heals as needed. That's a lot of money left over for arts and crafts. In my Endgame Lunatic+ file, I have a forged Brave Bow, forged Longbow, and still have 10k gold in the bank, plus a pile of sellables that I could use to raise some quick cash.

Link to comment
Share on other sites

Honestly, a bow-using team is better positioned to take out Grima than basically anyone else. Grima's DEF is the same as his RES, and a Brave Bow is better than Waste or a Celica's Gale (because of HIT/mt, respectively). An underrated benefit of bows is the HIT rate: people laugh at skills like Prescience or +20 HIT, but Grima has 90 avoid! He is no joke to land four consecutive attacks against.

I think that's more the class that the skills come with. A Sniper is already pretty accurate. Like even against Grima, if we assume the following:

+Def/-Luk Avatar

48 capped Skill

44 capped Luck

Skill Tonic

Luck Tonic

Skill+2

S Support (30 Skl, 30 Luck, no Skl class bonus)

Brave Bow

We get (144 + 44 + 6 + 2 + 6 + 9 + 3)/2 + 15 +70 = 192 hit. This negates all of Grima's avoid without any of the Sniper's other skills and Skill+2 could easily be replaced by a Rally or taking a support partner who increases skill (like a Pally). Note that a unit could lose 21 hit and still have 93% true hit against Grima. Most other enemies will have much worse avoid. Hit+20 would be great on any class that doesn't already have an insane amount of hit (as shown when the Lunatic and up enemies start toting the skill and axe users see a huge benefit). Your Lucina will punctuate this quite nicely because those skills will really benefit her rather ugly 36 Skill cap (plus as a support, won't have the A or S rank bonuses).

The issue I see is that two of the skills the Sniper gets aren't really helpful to the unit until they go to a different class (granted, Prescience could help a Bow Knight with that 5 Skill gap, but the class is already using two of the most accurate weapon types in the game). They're effectively long term benefit skills that require going through a class line that has a really awful enemy phase. I probably wouldn't even consider going through it on any mode other than Lunatic+.

Link to comment
Share on other sites

All true; the Sniper class really isn't that great outside of Lunatic+. The skills are more useful outside of the class than they are in it.

The leavings of Sniper are pretty important for Morgan, though. His SKL cap as a Warrior is 41 in this file, he's not going to have an S support with his partner (max of A-rank with Chrom or Lucina), and regardless of anything else he wants Bowfaire for the extra 8-10 damage on Grima. His HIT is going to be 170 with a Brave Bow. That will get modified by Pair-up and possibly forged +HIT, but he's still short. Not short enough where I actually care (very easy to repeat a Grima attempt), but even if he had 93% true hit, that's still a 25% chance to miss at least one of his four attacks.

Link to comment
Share on other sites

I always park Henry or Tharja right next to Grima to decrease his Avo by 25 (Hex + Anathema). With your projected numbers for Morgan and Lucina, something like this should boost your success rate even higher.

Link to comment
Share on other sites

All true; the Sniper class really isn't that great outside of Lunatic+. The skills are more useful outside of the class than they are in it.

The leavings of Sniper are pretty important for Morgan, though. His SKL cap as a Warrior is 41 in this file, he's not going to have an S support with his partner (max of A-rank with Chrom or Lucina), and regardless of anything else he wants Bowfaire for the extra 8-10 damage on Grima. His HIT is going to be 170 with a Brave Bow. That will get modified by Pair-up and possibly forged +HIT, but he's still short. Not short enough where I actually care (very easy to repeat a Grima attempt), but even if he had 93% true hit, that's still a 25% chance to miss at least one of his four attacks.

True. We're talking about a Morgan with 50+ enhanced Speed, though, right (42 + 2 for Tonic + 3 for 30 Spd on Lucy + 4 for Rally)? Since bosses like to act first on EP, shouldn't this give you 8 attacks overall?

Also, if you've got someone to spare who has Bow Knight levels (given your strat, I'd assume so), Rally Skill can also help shore that up.

Link to comment
Share on other sites

Lunatic+ Grima's stats are the same as in Lunatic and the only difference in skills is Pavise becomes Pavise+ (a meager 20% proc rate difference). The nearby goons will have skills that will make things problematic if one can't kill Grima in a single PP/EP set, though.

I know that Luna and Luna+ enemy stats are identical, I was referring to Avatar's stats... oh well, it isn't important anyway since I guess most people's Avatars hit caps...

But good to hear that Grima isn't really that hard anyway.

Link to comment
Share on other sites

I'll be doing Paralogue 1 before Ch5; I did test runs of both, and keeping MaribellexRicken alive without a Rescue staff just requires too many things to go my way in terms of enemy skills. The upshot is that Mercenary Louise should be fine there, because of Levin Swords (basically can handle the left hand side by herself).

My current dilemma is what to do with Miriel; she's got nothing but options. Ideally I feel as though I want her to eventually be a Valkyrie for Dual Support+, but there's also an argument for the skills from Dark Mage (since they work even as the Support unit). I could go Mage -> Dark Mage -> Troubadour -> Valkyrie, which gives a ton of Support unit utility, but entails a lot of reclassing. She'd be a decent wife for Gregor, and give me a Laurent with Armsthrift and workable base stats. I could also tap her as the primary Dark Mage/Sorc of the group, or swing through Sage to pick up Tomefaire (to pass to Laurent, most likely).

Currently leaning towards the DM -> Troub path, since it maintains her offense for the next few critical chapters.

Edited by Interceptor
Link to comment
Share on other sites

I'm liking Mage->Sage->Sorceror since it doesn't require too much reclassing but eventually gives her much-needed durability. Going Sorceror first would be a waste IMO since it gives arguably inferior skills first

Edited by Walhart
Link to comment
Share on other sites

Sage to Sorc is a good thought for firepower, but I'm a little concerned about possible advancement speed if I reclass from a promoted class to another one. Going from unpromoted level ten to another unpromoted class only adds 4 cumulative levels, but going from Sage 15 to Sorcerer 1 adds 17 of them, so she'd be getting EXP like a 20/18 unit at 20/15/1. Also I'd miss out on Hex and Anathema unless I put Tharja or Henry out on the field.

I guess in this scenario I would make Lissa my Valkyrie for Dual Support+, though I wouldn't know who to pair her with. Either she'd be an old maid (not a problem, since I wouldn't use Owain anyway), or just hang out with Chrom, I suppose.

Link to comment
Share on other sites

Sage to Sorc is a good thought for firepower, but I'm a little concerned about possible advancement speed if I reclass from a promoted class to another one. Going from unpromoted level ten to another unpromoted class only adds 4 cumulative levels, but going from Sage 15 to Sorcerer 1 adds 17 of them, so she'd be getting EXP like a 20/18 unit at 20/15/1. Also I'd miss out on Hex and Anathema unless I put Tharja or Henry out on the field.

Hmmm... that's a good point. I wouldn't worry too much about Hex or Anathema though, although since your run is (kinda?) dodgetank oriented, maybe it would be more important here. Also, normally I'd say that Miriel's offense is important because there aren't really that many good offensive ranged attackers, but since you are doing a bow-heavy run there probably won't be any shortage of those...

I guess in this scenario I would make Lissa my Valkyrie for Dual Support+, though I wouldn't know who to pair her with. Either she'd be an old maid (not a problem, since I wouldn't use Owain anyway), or just hang out with Chrom, I suppose.

I might be missing something, but Dual Support+ seems sort of useless on a devoted healer (assuming that's what your Lissa's role is) and IMO healers generally don't benefit much from pairs. Although, making her a support for Chrom sounds like a decent idea since you don't seem to be using him much.

Link to comment
Share on other sites

You're right about Dual Support+; I really need to have it on a non-healer if I want to see any use out of it at all. So I'm going for the gusto, and taking Miriel down the Mage -> DM -> Troub -> Valkyrie path. If it gets rocky, I'll skip the Troub part and use Lissa there instead. For right now, Miriel is a DM, and Lissa will stay as a Cleric for the time being. I may eventually promote Lissa to a Sage, and then reclass her to Falcoknight as a finisher (purely for the mobility; I don't have a flier for lategame, and that hurt me last time).

Hex and Anathema will be useful for Gregor mostly (he had some accuracy problems during training last time), but it will help the crit-based offense of the other units as well. Between Solidarity, Anathema, and the extremely high Support ranks possible when fighting shoulder-to-shoulder, my team has a lot of listed crit chance with that setup. It can be helpful when dealing with Aegis+ and high DEF units, since my bow team doesn't really have a good answer to that.

Anyway, OP updated with Paralogue 1:

[spoiler=Paralogue 1: Sickle to Sword]

This is a good chapter to catch up on training. Donnel is unsalvageable, naturally, but there are opportunities for Chrom here.

Keys to the chapter: train Chrom, get Avatar to a point where she can reclass, pick up the Rescue staff, remember to keep a Levin Sword handy, and don't get greedy or careless.

I chose to deploy Louise, Miriel, Kellam, and Lissa. Chrom and Donnel crash the party.

The immediate goal is to clear out the starting zone. Frederick can OHKO the nearby thief, possibly with help from a Pair or tonic, so make sure that the thief doesn't have Pavise+ (which is the only reason you might want to reset here). In my case, LouisexChrom can severely injure the nearby Barbarian (with magic or Levin), and if Chrom doesn't finish him off with a Dual Strike, MirelxKellam surely can with Fire/Thunder. Donnel and Lissa can be Paired and moved safely out of the way. The archer will most likely go for Miriel here, which is fine; she can take a hit (Kellam ensures that she will survive even if the archer has Luna+), chunk him a good bit on retaliation, and Lissa heal her back up next Turn.

This leave you with two archers (one wounded), and four barbarians either close by or incoming. There is plenty of running room to the north to kite them backwards, especially since the further enemies will be slowed down by the forests. This allows you to deal with most combinations of possible skills. Beyond that, a few strategic notes for this chapter:

  • If you happen to get a map where nobody out of those early enemies has Pass, you can set up a choke pretty easily. The ones that blocks the northern wall entrance on a forest is especially useful.
  • Use Frederick mostly to weaken things for other people to kill, but don't be afraid to have him KO something if necessary; there are plenty of enemies on this map.
  • Miriel and Avatar are especially good at baiting archers and other 2-range enemies to attack them through the walls. It's easy to make this safe because of Lissa. Abuse it whenever possible (the archers in the boss room are a good place to start; draw them from the right or the bottom).
  • Some enemies have an aggro radius that is smaller than their attack range. For example, the even though one of the Barbarians in the "pillar room" has a 2-range weapon, he will not move until you are in attack range of his 1-range partner (at which point they will both move to attack).
  • Take care to only attract what enemies that you actually want to fight. They battle in groups, and you don't want to get overwhelmed and force Frederick to have to kill things. Be patient, and peel off loners wherever you can.
Don't worry too much about the Killer Lance thief; it's not worth taking risks to kill him off for a weapon that's not going to be used (and is available to buy from half a dozen Spotpass teams anyway). If you can get to him safely, that's fine, but the value there is mostly for the EXP, not the weapon. He won't steal Rescue, which is the real treasure here. Carefully wind your way towards the final room, picking off stragglers wherever possible, and finish with the boss.

I wanted to give the boss to Chrom, but he hits far too hard for that, so it went to Louise instead. I reclassed Miriel to Dark Mage once she hit level 10. Lissa will stay in her base class for now.

Turns: 52

Heroes: Miriel & Kellam

           Level HP ST MG SK SP LK DF RS Wlvl
Louise     17.39 29 17 13 9  14 17 20 7  C Swd, B Tome
Chrom       6.95 24 9  1  12 11 9  9  1  D Swd
Frederick   3.05 30 15 2  13 11 6  15 4  D Swd/Axe, A Lnc
Lissa      12.56 22 5  12 9  8  16 6  7  B Stf
Miriel      3.53 28 2  9  11 12 12 8  12 B Tome

Supports   B - Louise|Frederick
           C - Chrom|Sully/Lissa, Louise|Lissa/Chrom/Miriel, Miriel|Kellam, Frederick|Lissa
I wish I could have gotten Chrom a little bit further along, but I gave him the safest kills that I could. There will be another opportunity to get him some serious training in the next Paralogue. Louise may be able to squeak out one more chapter as a Tactician, we'll see.
Link to comment
Share on other sites

I can go waaaay faster than this, but there's no value in cutting Turns when you're going for reliability. They are extremely short Turns for the most part, one movement or one heal in many cases. I'm just posting the numbers for the lulz, the chapters don't really take much more time than normal.

Link to comment
Share on other sites

Too late to say this, but the Hammer and the short axe on the pillar room barbarians are completely pointless: The short axe barb will attack on 1-range anyway for more damage, and will attack first. The hammer barb will be one space behind him and with a steel axe equipped.

Link to comment
Share on other sites

I hadn't noticed that, but one of them had Pass in my final run, so I ended up having them bracket in Avatar anyway. The Hammer guy got choked out, the Short Axe guy walked through to attack Miriel, iirc.

AI can be pretty weird sometimes. The aggro radii of the enemis here are definitely some kind of custom BS.

Link to comment
Share on other sites

I hadn't noticed that, but one of them had Pass in my final run, so I ended up having them bracket in Avatar anyway. The Hammer guy got choked out, the Short Axe guy walked through to attack Miriel, iirc.

AI can be pretty weird sometimes. The aggro radii of the enemis here are definitely some kind of custom BS.

Lunatic+ is just a bunch of custom BS

Link to comment
Share on other sites

I hadn't noticed that, but one of them had Pass in my final run, so I ended up having them bracket in Avatar anyway. The Hammer guy got choked out, the Short Axe guy walked through to attack Miriel, iirc.

AI can be pretty weird sometimes. The aggro radii of the enemis here are definitely some kind of custom BS.

I've noticed a few different patterns that make me think there's just a variety of universal scripts now. Here's what I've got:

-Leeroy Jenkins mode, where just getting into range of one unit will pull the unit and every other member of a group it may be tied into. Sometimes these guys won't even be grouped up.

-Suicide bomber mode, where they just aggro and try their best to get themselves killed on the nearest player unit, no matter where they are on the map. Typically the enemies on this script will be aggroed at the start of the map, but there seem to be a few groups on a turncount timer before exhibiting this same behaviour. Oh, and reinforcements will do this too.

-Lazy mode, where the boss or Mire user will hang out and not really care if you're beating the crap out of his buddies three tiles away.

-Almost sort of coordinated mode, where a player unit must be within range of multiple units (threshold seems to vary... chapter 2's second wave being a good example of this) before it will pull the group.

Also, attacking a unit at all will cause it and any associated group to aggro, even if the player unit Galeforces or gets Rescue Staffed out of attack range.

Groups also aren't necessarily restricted to those closest to the unit. Distant enemies on the flanks can be linked in so that when they aggro and attack from a different direction, it kind of resembles a tactical move (Gangrel's Spotpass Paralogue is a good example of this).

Link to comment
Share on other sites

I've noticed a few different patterns that make me think there's just a variety of universal scripts now. Here's what I've got:

-Leeroy Jenkins mode, where just getting into range of one unit will pull the unit and every other member of a group it may be tied into. Sometimes these guys won't even be grouped up.

-Suicide bomber mode, where they just aggro and try their best to get themselves killed on the nearest player unit, no matter where they are on the map. Typically the enemies on this script will be aggroed at the start of the map, but there seem to be a few groups on a turncount timer before exhibiting this same behaviour. Oh, and reinforcements will do this too.

-Lazy mode, where the boss or Mire user will hang out and not really care if you're beating the crap out of his buddies three tiles away.

-Almost sort of coordinated mode, where a player unit must be within range of multiple units (threshold seems to vary... chapter 2's second wave being a good example of this) before it will pull the group.

Also, attacking a unit at all will cause it and any associated group to aggro, even if the player unit Galeforces or gets Rescue Staffed out of attack range.

Groups also aren't necessarily restricted to those closest to the unit. Distant enemies on the flanks can be linked in so that when they aggro and attack from a different direction, it kind of resembles a tactical move (Gangrel's Spotpass Paralogue is a good example of this).

Someone should make a topic for posting info about the AI and trying to figure it out a la the Growths topic. Maybe someone on Casual could use battle saves to figure out what certain enemies do under certain circumstances. I mean, it'd be a lot of work, but if enough people contribute...

I don't know, I haven't really thought it out, just putting it out there.

Because the AI is just too damn bizarre.

Link to comment
Share on other sites

Aggro behavior is pretty well defined.There are set patterns and triggers that Kuroi mostly covered..

Specific AI movements appear random if there's a choice. You can do the exact same positioning and the AI will attack different units or the same ones from different sides or move in different directions (I think it's most obvious in Chapter 4). There is some logic to this, but there's also a definite RNG component as well.

Link to comment
Share on other sites

Aggro behavior is pretty well defined.There are set patterns and triggers that Kuroi mostly covered..

Specific AI movements appear random if there's a choice. You can do the exact same positioning and the AI will attack different units or the same ones from different sides or move in different directions (I think it's most obvious in Chapter 4). There is some logic to this, but there's also a definite RNG component as well.

I agree, there is definitely a rng component involved at times. As far as compiling info, I was referring to figuring out which specific enemies/groups in each chapter fall under which of the categories KTT described (although this is obvious in some cases as with the bosses).

Link to comment
Share on other sites

It took a little while to find a strategy that would be consistently repeatable, with a high reliability, but I put one together. OP updated with Ch5:

[spoiler=Chapter 5: The Exalt and the King]

This is one of the many "difficult because you get rushed" chapters. You have to deal with fliers that bypass potential choke points, and reinforcements that pour in from the forts. There are two basic ways to clear it: either you throw caution to the wind and go full-tilt offense, or you sit back and turtle in a corner. We'll be doing the second thing, since it's more reliable, and also makes it very easy to get a lot of solid training (particularly for 2-range attackers like Miriel). It has some complicated movements, but all of them are necessary to get the enemies to move the way that we want.

Keys to the chapter: bring enough healing for a marathon, set up a stable defensive corner, and eliminate the most dangerous threats before doing anything else.

As soon as you load the chapter, take a look at the two barbarians closest to you (one below the fort, one against the west cliff); if either of them have Luna+, you may want to restart. Hawkeye is acceptable, if somewhat annoying. Any other skill is perfectly fine. We'll have to endure attacks from them for a long time before we can kill them, so it's important that they not be too dangerous offensively. That's why we have the "stable" requirement: you don't want to be spending actions trying to keep the front alive, while you are having Miriel deal with the Riders.

We're deploying Avatar, Fred, Kellam, Sumia, Lon'qu, Lissa, and Miriel. Sumia needs to start directly above Chrom.

First, Pair RickenxMaribelle and use Lissa to Rescue them down the cliff. Secondly, we need to set up a decoy: move Sumia three tiles north, and then three west, onto the cliff. The reason for this, is to divert the mages temporarily so that the melee units get to the corner first (mages hit too hard). Then, it's time to set up our turtle shell (don't dally or try to kill anything). The defensive formation we're going to build will be in the lower right corner, and looks something like this:

CCCC
  12
 345
Where C is a cliff, and the bottom + right are the edges of the map. The numbers represent unit pairs:
  • Avatar + Chrom - two enemy facings; one ground, one cliff (i.e. a Wyvern Rider). They are good anti-Rider units in an emergency, which is why they are here.
  • Miriel + Sumia - faces only Wyverns, with Sumia included to ensure doubling.
  • Frederick + Kellam - two facings, both on the ground.
  • Lissa + Lon'qu - able to heal both of the forward positions, not in danger of being attacked by anything but 2-range units. Lon'qu included for defensive purposes (don't want Lissa doubled).
  • Ricken + Maribelle - as long as this formation holds, they are not targetable by any enemy on the map. Maribelle is able to heal Lissa and Miriel, and Ricken is able to help Miriel deal with difficult Wyverns (like ones with Counter, for example).
Sumia is going to be nearly dead and in a lot of trouble after the first Turn, so move her to the right (there's only one safe tile to land on), and next Turn Pair-Up with Miriel. You can swap Lon'qu and Sumia around with Transfer if Miriel needs the extra +1 SPD for something.

Frederick and Avatar will want DEF tonics here. Even though Lissa can heal back all of the damage, the extra DEF makes it harder for the enemy to find fatal combinations with their back line attackers. Keep emergency Vulneraries on hand. Don't get careless about healing to full each turn: the enemy will swap out positions for a kill if it can (i.e. sending in a Mage). It's entirely possible that Avatar is tanky enough that she doesn't even need to be unequipped (both enemies attack Fred instead), but use caution if you do that.

Take your time letting the Wyverns come to you, taking them apart with Miriel, using Elwind when needed. Counter can be dealt with by unequipping Miriel, and killing them at 2-range with Avatar/Chrom (the reason that they were put in this spot in the first place). Remember to trade away Avatar's weapon afterwards. Take care not to allow a situation where Miriel will kill something on Enemy Phase after taking a hit; another Rider will potentially fill in the spot, and she's 2HKO'ed here.

Once all the Riders are dead (including reinforcements), it's time to slowly unwind the gigantic traffic jam of enemies on the ground. There are a few ways to do this safely, but I prefer to take out the 2-range mages first, since they are the most dangerous. To pull off this maneuver, have Avatar take Sumia from Miriel (needs the +RES), have Miriel kill the enemy to the left of Avatar, move Avatar into the space left open, move Lissa to where Avatar used to be, and then Frederick where Lissa was. Now you look like this:

CCCC
 142
  35
If you're smart about clearing a spot (Levin works well for this), mages will keep walking in to blast Avatar or Fred. Keep killing, trading, and healing, until there are nothing but 1-range enemies left (don't forget that one barbarian has a Short Axe), which are simple to finish off.

Finally, once the enemies that move are all dead, you can move out to clean up the rest. The remaining Wyverns (including the boss) act as a single group, so once you aggro one you will aggro them all. Be prepared for that. The final three enemies -- DM, barbarian, myrm -- don't appear to be linked.

Miriel cleaned house in this map, mostly due to Chrom being unable to really participate in meaningful combat. I had to use a second Rescue charge to save Miriel from being out of position on the final set of Riders (damn the boss), but c'est la vie.

Turns: 60

Heroes: Miriel & Lon'qu

           Level HP ST MG SK SP LK DF RS Wlvl
Louise     20.00 31 18 14 10 14 20 20 8  C Swd, B Tome
Chrom       8.15 26 10 1  13 12 11 10 2  C Swd
Frederick   3.17 30 15 2  13 11 6  15 4  D Swd/Axe, A Lnc
Lissa      18.24 25 7  13 12 10 20 6  11 A Stf
Miriel      9.31 32 2  12 14 17 13 9  15 A Tome

Supports   B - Louise|Frederick/Chrom
           C - Chrom|Sully/Lissa, Louise|Lissa/Miriel, Miriel|Kellam/Frederick, Frederick|Lissa
Louise is now ready to be reclassed to Merc. I still need Miriel's offense for just a little bit longer, so although I could reclass to Troubadour in the middle of the next map, I'll be holding off until after Ch6 is done (the last truly difficult Lunatic+ map). Edited by Interceptor
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...