Jump to content

FE6 HM 0% growths, with commentary (complete)


dondon151
 Share

Recommended Posts

  • Replies 492
  • Created
  • Last Reply

Top Posters In This Topic

https://www.youtube.com/watch?v=lq7g1iLPQVU

Chapter 24 is completed in 7 turns.

Chapter information

[spoiler=]

Important weapons

maltet.pngMaltet (マルテ)
S rank lance | 1 rng | 11 WT | 16 MT | 80 hit | 0 crit | +5 skl to user when equipped

Maltet, the divine lance, rounds out the Divine Weapons that are used in this run. There’s not much special about it - it’s actually quite underwhelming - but Milady can use it nonetheless to dispatch enemy manaketes with great efficacy.

Mechanics notes

About this chapter: 7 thrones

Chapter 24 is the climax of Roy’s campaign, and what better way to show that than to use the same objective as the previous 29 chapters… 7 consecutive times. Roy is busy in this map, spending most of his free actions sitting on thrones, which somehow reveals the previously hidden contents of rooms.

7wzEYcx.png
The chapter 24 map is so big that its miniature version doesn’t even fit on the screen.

The enemy AI, in depth

The strangest occurrences in this chapter are that on several different occasions, an enemy manakete elects to not KO a unit when given the opportunity. Why does this happen?

Chapter 24 has enemies with AI byte 3 = 0x02 and 0x0A. Luckily, there's a pattern to how the AI is assigned.
0x02 = Jahn + manaketes on throne and left/right of throne (care about counterattack damage)
0x0A = manaketes with elixirs + all reinforcements (do not care)

The enemy AI calculates a value called Targeting Priority (TP) in order to determine whom to attack. AI byte 3 = 0x02 cares about counterattack damage and receives a negative TP modifier from it. AI byte 3 = 0x0A does not care about counterattack damage and doesn’t receive any negative TP modifier. All AI bytes do care about whether a target can counterattack, and targets that can’t will receive a positive TP modifier.

We’re going to ignore AI byte 3 = 0x0A here because it’s not important. Please take note that AI byte 3 = 0x0A calculates TP differently than AI byte 3 = 0x02.

Here’s the TP breakdown for AI byte 3 = 0x02, also courtesy of Gryz. Beware of numbers ahead:

Taking the attacker’s damage into account: is the attack lethal?
If yes, add 50 TP.
If no, add TP = 2 * damage * hit/100 [capped at 40]

Regardless of above:
Add TP = 20 - defender’s remaining HP after attack [floored at 0]
Add TP = map bonus [unimportant]
Add TP = defender class bonus [unimportant]
Add TP = turn number

Can the defender counter at all without his weapon breaking?
If no, +10 TP.
If yes, subtract TP = 2 * damage * hit/100 [capped at 40]

If yes above:
Subtract TP = 20 - attacker’s remaining HP after counterattack
Subtract TP = map bonus [unimportant]


So for the case of Milady and Perceval on turn 2 enemy phase, the TP calculations work out as follows:

Milady’s TP

If attack not lethal: add TP = 2 * damage * hit/100 = 40 [capped at 40]
Manaketes hit hard and are accurate, so the above expression easily caps at 40.
Add TP = 20 - defender’s remaining HP after attack = 16
Add TP = turn number = 2

If defender can counter: subtract TP = 2 * damage * hit/100 = 40 [capped at 40]
Divine Weapons also hit hard and are accurate, so the above expression easily caps at 40.
Subtract TP = 20 - attacker’s remaining HP after counterattack = 20

Milady’s TP = 40 + 16 + 2 - 40 - 20 = -2


Perceval’s TP

If attack not lethal: add TP = 2 * damage * hit/100 = 40 [capped at 40]
Add TP = 20 - defender’s remaining HP after attack = 0 [floored at 0]
Add TP = turn number = 2

If defender can counter: subtract TP = 2 * damage * hit/100 = 40 [capped at 40]
Subtract TP = 20 - attacker’s remaining HP after counterattack = 20

Perceval’s TP = 40 + 2 - 40 - 20 = -18


When TP is calculated to be zero or negative, the value calculated above is ignored! Instead, the game sets TP to what was calculated at the “if attack is lethal” step, which in this case is +40 for both Milady and Perceval.

Now that they’re tied in TP, the AI determines which target to prioritize by looking at… deployment order. Just like status staff AI, the tiebreaker here is deployment order position. Perceval is lower in the deployment order than Milady, so the AI targets him. Simple!

How about the case of Shin/Cecilia and Niime on turn 4 enemy phase? The TP calculations work out as follows:

Shin/Cecilia’s TP:

If attack not lethal: add TP = 2 * damage * hit/100 = 40 [capped at 40]
Add TP = 20 - defender’s remaining HP after attack = 17
Add TP = turn number = 4

If defender can’t counter: add TP = 10

Shin/Cecilia’s TP = 40 + 17 + 4 + 10 = 71


Shin and Cecilia have identical HP + def, so the tiebreaker between them is once more deployment order - Cecilia is at a lower position than Shin, so Cecilia gets attacked first.

Niime’s TP:

If attack is lethal: add TP = 50
Add TP = 20 - defender’s remaining HP after attack = 20
Add TP = turn number = 4

If defender can counter: subtract TP = 2 * damage * hit/100 = 40 [capped at 40]
Subtract TP = 20 - attacker’s remaining HP after counterattack = 20

Niime’s TP = 50 + 20 + 4 - 40 - 20 = 14


The TP calculations here are straightforward; Niime clearly has lowest TP of all potential targets. So if you bothered to read through all of that, that should unravel the mystery behind the AI behavior in chapter 24.

Edited by dondon151
Link to comment
Share on other sites

Holy damn. You captured 2 thrones on the same turn. Like wtf how is that even possible. This chapter was the most entertaining one yet. Okay, it's not as cool as C21 but it's still bomb.

The most surprising part was that you turned on animations.

Link to comment
Share on other sites

Huh, that AI deal is funky. But I do recall long ago the same thing happening like one of my first HM runs

I had Karel using Durandal and at low HP but a Manakete attacked Echidna using Armads instead because she did less than Karel or something wacky looking at the calculations..

Link to comment
Share on other sites

how many more turns do you think this chapter would have taken if their AI was not retarded and OHKO'd Niime or killed Miledy when possible?

i don't know, probably just one more turn. many things would have to be changed in order for things to work.

when i planned the chapter, i actually didn't know the whole deployment order tiebreaker thing. i guess i was really lucky that perceval was below milady and cecilia was below shin. if it were the other way around, i would have gotten stuck.

also we didn't get a chance to point this out but when cecilia used the angelic robe on turn 4, that was the last of our 42 stat boosters. all of them ended up being important, go figure.

Edited by dondon151
Link to comment
Share on other sites

https://www.youtube.com/watch?v=HoBiZ5gijc8

The final chapter is completed in 1 turn.

Final turncount: 157

Final in-game time on epilogue save: 4h 11m 24s

Top 5 units (battles):

1. Marcus - 103
2. Milady - 82
3. Rutger - 80
4. Shin - 77
5. Zealot - 71

Top 5 units (wins):

1. Rutger - 58
2. Shin - 46
3. Milady - 45
4. Perceval - 38
5. Marcus - 37

Thank you for watching...

Edited by dondon151
Link to comment
Share on other sites

Now that this is finished, I can mention the TAS project I hinted at earlier in the chapter. So, while a 0% growths TAS would be basically a much faster version of what dondon has done (not much in the way of improvement via strategy, at least), I figured it may be far more worthy of a TAS if I were to add a SS Rank requirement into the mix as well. Makes things entirely different, for sure, also means that critting everything isn't necessarily the best way to go. Not something I think anyone has attempted yet, but I think for a TAS it makes sense to do as having access to easy RNG manipulation makes things 'too easy' in some ways, with a ranked run the complication would come more from the strategy rather than the manipulation. At least, that's the theory.

I expect to lose a few turns compared to dondon, for reasons of pushing the requirements, also by being more limited on boots purchasing to meet the Funds rank.

I actually wanted to start this a while ago, but I figured I would wait for dondon's playthrough to finish so I can compare, and use some of the ideas here as a basis for the strategies I will use. Going to be a different kettle of fish, for sure, but part of the reason the normal TAS took 3 years to complete was that you can only plan certain things from a completed playthrough as a basis (stuff like weapon ranks, supports etc).

Link to comment
Share on other sites

My top ten moments, terrible puns and english included:

#10

Brave Bow users tearing Frey, Narshen and Sigurney apart. Where is your throne bonus now, IS?

#9

The trivialization of chapter 12. Even got time for the arena and the boss ended up as exp fooder for Tate. Echidna just took so long to get dressed or whatever. Even Geese ~60 Hit attacks were of use. Weird that Tate didn't attack Klein from 2-range, though. Thats what she does in my runs.

#8

Garret and Percival teaming up and infiltrating Ilia Castle. Now I want to know what was really going on there, where is Tarantino when you need him?

#7

Thany taking a ballista hit. Because 17 HP/6 Def count as solid in endgame.

#6

Douglas beating the chapter 22 hero, thus ending the war. He didn´t even wear his boots to walk the -*gets shoot by Igrene's Silver Bow, thanks to her natural weapon rang*

#5

Manaketes refusing to kill anyone in chapter 24. Silly Jahn, thats why you don't create soulless war dragons that can't act on they own. And you couldn't even cap your level after 1000 years of training?

#4

Just chapter 7. Saving Treck as soon as possible right under the nose of the enemies and with such a swiftness, it´s like we are recruiting FE9 Astrid. Marcus and Zealot luring every enemy, just so that those can be chipped to death by the rest of the team. Also the bossrushing to prevent the reinforcements, basically one-turning the chapter without one-turning it, if that makes sense at all.

#3

Chad blocking like ten units in chapter 5. Bet the nomad in front of him was his long lost father and the situation was akward as it was bittersweet.

#2

Chipping Henning Oakenshield with the light brand and finishing him off with a surely super accurate steel blade crit. The true antagonist of the game, truly suited to guard Durandal, the true legendary sword of FE6. Wait, no, that would be the first Killing Edge.

#1

One of the most epic chapters of the whole franchise? Let's spend most of the time with selling everything and buying boots! Who thought a secret shop with an unlimited amount of stat boosters would be a good idea? Probably the same guy who banned the base shop on hard mode. Murdock being the easiest General to beat count as blashphemie, too. Now waiting for the sequel where he returns and tells Percival that he let him win/used too much warp powder

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...