What do you look for when optimizing units?

[Archeleon]

Working on a tool to help with unit optimization and though I have an idea of how to go about it I would like to see what others prioritize when making class choices and what kind of data display would be most useful to you for these decisions. I suppose the main point of uncertainty is whether to base unit/class combination on maximum potential (grind until caps) or more practical play (maybe second seal once to get another round of levels).

https://docs.google.com/spreadsheets/d/1kaQv2vjvVSGkHnhkhsf6KOrIDYQVeyI91WHh3GPRTdw/edit?usp=sharing

This is what I am currently tracking:

Character Wise:

Growth Rates and how they rank in comparison to others.

Stat Cap Modifiers and how they rank in comparison to others.

Personal Skill

Weapon Proficiencies

 

Class Wise:

Growth Rates modified by character growth rate as well as expected stat gain over 40 levels to include class base stats, as well as how they rank.

Stat Caps modified by character modifiers as well as how many levels it would take to reach the cap as well as how they rank.

Class Skills with a loose subjective ranking.

Weapon Proficiencies and how character innate proficiencies increase their ranks.

Combat stats (Ph Atk/Mag Atk, Hit, Avo, Crit, Dodge) based on gaining 40 levels with that class and what they would be with all stats capped.

 

Weapon Wise:

Identify highest ranking weapon of each type (Sword vs axe) and "classification" (Brave vs Ranged) available to each class.

How weapon stats affect combat stats of the class based on the 40 level gain.

 

 

Edited February 6, 2023 by Archeleon

[dragonlordsd]

Unless I've missed something, practical play seems to be the best criteria at the moment. Although you theoretically can grind infinitely, the money limitations in this game seem very severe.

As far as I can tell, the only way to get more gold is by doing gold corrupted, which only spawn occasionally, and only give 1300 gold when they do spawn, which is only half of a second seal.

So I'm not sure in practice how many times you'll actually be reclassing an individual unit, unlike, say, awakening.

[Scaramuccia]

1 hour ago, dragonlordsd said:

the money limitations in this game seem very severe.

You could always ignore donations which solves all money issues you could potentially have.

[Archeleon]

Added the current version of the sheet for feedback.

[dragonlordsd]

On 2/4/2023 at 1:38 AM, Scaramuccia said:

You could always ignore donations which solves all money issues you could potentially have.

Definitely helps a lot. But even without that, you're only directly give enough money for 64 second/master seals (I think. Need to double check how much gold you get from each chapter). There are 36 characters in the game, so that's not even enough gold for one master seal and one second seal for each character. 

[Vicyus]

3 hours ago, dragonlordsd said:

Definitely helps a lot. But even without that, you're only directly give enough money for 64 second/master seals (I think. Need to double check how much gold you get from each chapter). There are 36 characters in the game, so that's not even enough gold for one master seal and one second seal for each character. 

A quarter of the roster come pre promoted making it possible. Assuming you will not advance zelkov/yunaka you only need 20 master seals. Now if you do multiple class changes on base that's different, but most ppl have matched out that going early promote is generally better

Edited February 5, 2023 by Vicyus

[dragonlordsd]

6 hours ago, Vicyus said:

A quarter of the roster come pre promoted making it possible.

Ah, yeah, forgot about that.

Out of curiosity, I wonder how many class changes are optimal? I suppose it depends on the unit. I've seen people proposing 4 class changes for some optimal builds. That said, in theory, you might have to promote down to access specific classes, though I don't know if any optimal builds require that.

[Archeleon]

2 hours ago, dragonlordsd said:

Ah, yeah, forgot about that.

Out of curiosity, I wonder how many class changes are optimal? I suppose it depends on the unit. I've seen people proposing 4 class changes for some optimal builds. That said, in theory, you might have to promote down to access specific classes, though I don't know if any optimal builds require that.

Dont think you really need a lot of class changes, just one to go down a different, optimal class path. You can't learn class skills like you could in 3H. Most of the skill optimization will come from Emblems. Figuring out the best Emblem/Class/Inherited Skills are optimal is the biggest thing.

[Scaramuccia]

3 hours ago, dragonlordsd said:

. I've seen people proposing 4 class changes for some optimal builds.

4 definitely sounds as overkill. Maybe for some very specific strategies/units it make sense, but I would be surprised if it is optimal. base -> master -> good mid game class -> good end game class is 3 changes and few units really need such pattern.

[Vicyus]

On 2/5/2023 at 9:20 PM, Scaramuccia said:

4 definitely sounds as overkill. Maybe for some very specific strategies/units it make sense, but I would be surprised if it is optimal. base -> master -> good mid game class -> good end game class is 3 changes and few units really need such pattern.

That also would only be optimal if your consistently feeding the unit xp, or else constantly switching to another class while leveling in previous classes would just stunt the exp growth of said unit. An example I use is kagetsu. I found his most high end build to be wyvern rider with maxed out killing edge with lyn emblem. He hits around over 50% crit on most mobs. Add in alacrity, he basically crit every opponent, making him an amazing 1v1 independent unit. To do that build, I just need to second seal him to wyvern, while grabbing axe profficency. Optimal time to do this is later into the game, because he's so plug and play that I can wait on getting him exp and use him when the deployment slot gets to 14 around ch 23. That means I don't need to level him way until then, where because he'd be lower level than most mobs he'd gain good exp.

Edited February 8, 2023 by Vicyus

[VirginiaKlug]

On 2/3/2023 at 9:06 PM, Archeleon said:

Working on a tool to help with unit optimization and though I have an idea of how to go about it I would like to see what others prioritize when making class choices and what kind of data display would be most useful to you for these decisions. I suppose the main point of uncertainty is whether to base unit/class combination on maximum potential (grind until caps) or more practical play (maybe second seal once to get another round of levels).

https://docs.google.com/spreadsheets/d/1kaQv2vjvVSGkHnhkhsf6KOrIDYQVeyI91WHh3GPRTdw/edit?usp=sharing

This is what I am currently tracking:

Character Wise:

Growth Rates and how they rank in comparison to others.

Stat Cap Modifiers and how they rank in comparison to others.

Personal Skill

Weapon Proficiencies

 

Class Wise:

Growth Rates modified by character growth rate as well as expected stat gain over 40 levels to include class base stats, as well as how they rank.

Stat Caps modified by character modifiers as well as how many levels it would take to reach the cap as well as how they rank.

Class Skills with a loose subjective ranking.

Weapon Proficiencies and how character innate proficiencies increase their ranks.

Combat stats (Ph Atk/Mag Atk, Hit, Avo, Crit, Dodge) based on gaining 40 levels with that class and what they would be with all stats capped.

 

Weapon Wise:

Identify highest ranking weapon of each type (Sword vs axe) and "classification" (Brave vs Ranged) available to each class.

How weapon stats affect combat stats of the class based on the 40 level gain.

 

 

Thanks for the help. I really appreciate you :)

Edited February 27, 2023 by VirginiaKlug