What are the best/ worst classes (and class types) in general?

[dragonlordsd]

3 hours ago, lenticular said:

Meanwhile, Engage tops out at a 30% class growth rate (for berserker), and has a lot of classes give 20%, some of which are at basic tier. Even in a hypothetical case of a character leveling up from 1-20 in a basic tier and then 1-20 in an advanced tier, they're still going to end up with a substantially higher average growth. Lord (20) -> Successeur (20) averages to 20. Axe fighter (20) -> Berserker (30) averages to 25. Lance Armour (15) -> General (20) avergaes to 17.5.

Overall, in these circumstances Pannette spends the entire game with an effective strength growth rate of 75 (45 peronsal + 30 class), whereas Dimitri has an average growth rate across the whole game of 69.8 (60 personal + 9.8 average from class) and only manages to top out at a 75% growth for his final ten levels.

I've seen this arguement a lot, and it's a false equivalence. Unless you're hard reclassing everyone into a berserker, which is a lot harder to do than 3 houses, almost none of your characters will have access to that 30 berserker growth. For most of the game, you're limited to less second seals than you have playable characters.  Most other classes are at more typical growth rates . Wyvern Lord is a 5 percent difference between Engage and three houses, as are warrior and paladin. Berseker is the only exception to this, and as it's the only class with such a massive difference, it's just an exception not a rule. Yes, the class growths are higher, but they're not that much higher.

[Ragnell17]

On 2/12/2023 at 3:12 PM, Azz said:

As for mages, Lindwurm, Sleipnir and Mage Knight are goated. No need for Sage since Ivy will probably have the S rank tome anyways and Mage Knight has one of the best class skills in the game and again, there are enough staff users that Sage is eh. Sleipnir could use a better tome rank but Horty will probably spend more time staffing than attacking. High Priest is unneeded cause the S staff is stupid and you'll probably never get it in an efficient run and Martial Master is bad cause arts are bad. Royal Knight is an option and works fine as a staff bot, namely Mauvier works well as a staff bot.

As for the dragons, they are good support classes and despite Divine Dragon being fairly mid, I think the boosted emblem effects, especially with Byleth makes it worth it.

I think one thing that you can do with Sages that give them their own niche is having them with the Byleth Emblem. As mysticals get Thyrsus which adds +2 tome range. Sure Instruct Magic isn't as good as Instruct Spectrum, but the universally useful tome range more than makes up for it. Basically turns Thoron into Bolting and Elsurge into a guaranteed hit thunder tome.

Edited February 14, 2023 by Ragnell17

[Scaramuccia]

1 hour ago, dragonlordsd said:

I've seen this arguement a lot, and it's a false equivalence. Unless you're hard reclassing everyone into a berserker, which is a lot harder to do than 3 houses, almost none of your characters will have access to that 30 berserker growth. For most of the game, you're limited to less second seals than you have playable characters.  Most other classes are at more typical growth rates . Wyvern Lord is a 5 percent difference between Engage and three houses, as are warrior and paladin. Berseker is the only exception to this, and as it's the only class with such a massive difference, it's just an exception not a rule. Yes, the class growths are higher, but they're not that much higher.

I fought similar. But after checking math it isn't a big difference even for an "average character". In 3h I would call 45 str character as good, but not exceptional and in Engage same applies 35 str. This is 10% per level. In 3h best classes are 15, good are 10. In engage even if you ignore Berserker - there are several 20 and 15 is everywhere. So we have at least 5 difference. But don't forget leveling - in 3h you use many classes before getting your endgame class. And they all have lower growths. In Engage you promote much faster. As a result you get more level ups with better growths - there is an example of math for it in the thread and it shows that this effect is approximately 5% per level.

Suddenly we get 10 vs 5 +5. Which looks like a balance. So growths are not the reason why weapon mt is more valuable in Engage than in 3h. 

[lenticular]

1 hour ago, dragonlordsd said:

I've seen this arguement a lot, and it's a false equivalence. Unless you're hard reclassing everyone into a berserker, which is a lot harder to do than 3 houses, almost none of your characters will have access to that 30 berserker growth. For most of the game, you're limited to less second seals than you have playable characters.  Most other classes are at more typical growth rates . Wyvern Lord is a 5 percent difference between Engage and three houses, as are warrior and paladin. Berseker is the only exception to this, and as it's the only class with such a massive difference, it's just an exception not a rule. Yes, the class growths are higher, but they're not that much higher.

I'm not just talking about Berserker though, and I'm not sure why you think I am. My main point is this: just comparing the growth rates of end-game classes is misleading, because you spend so little time in those classes, in Three Houses.

Yes, you are correct that there is only a 5% difference in the strength growth between Wyvern Lord (Three Houses) and Wyvern Knight (Engage), but in Three Houses you are only actually getting those growths for a quarter of the game, maybe a third. Meanwhile, in Engage, a unit is spending upwards of half the game in its final class, usually over three quarters, and in many cases its entire existence. Three Houses class growths are brought down significantly by the low or non-existent growths on their beginner and intermediate classes.

Let's do a couple of like-for-like comparisons. First, Louis and Raphael.

For Louis, he's recruited at level 6 as a Lance Armour. Let's then promote him to General at level 15. This is later than I'd like to promote, but maybe you didn't have enough Master Seals to promote him earlier. We'll then take him all the way to level 20, second seal him, and raise him again to level 5. He levels up 9 times as a Lance Armour and 23 times as a General. In total, he expects to gain 9*(0.40+0.15)+23*(0.40+0.20)=18.75 points of strength over those 32 level ups, for an average of 0.59 strength per level.

For Raphael, let's assume that we recruit him at level 1, then we promote him to Fighter at level 5, Armor Knight at level 10, and finally Fortress Knight at level 20. We then carry on until he hits level 40. (You could turn him into a Great Knight at level 30 if you wanted to, though it would make no difference to this calculation since Great Knight and Fortress Knight have the same strength growth.) Under these circumstances, he expects to get 4*(0.50+0)+5*(0.50+0.05)+10*(0.50+0)+20*(0.50+0.10)=21.8 points of strength spread over 39 level ups, for an average of 0.56 strength per level.

Next, let's compare Chloé and Petra.

For Chloé, we'll recruit and promote at the same levels of Louis, starting her as a Lance Flier and promoting to Wyvern Knight. She expects to gain 9*(0.25+0.10)+23*(0.25+0.20)=13.5 points of strength over 32 levels, for 0.42 per level.

For Petra, her class progression will be Commoner to Fighter to Pegasus Knight to Wyvern Rider to Wyvern Lord. This gives her an expected 4*(0.40+0)+5*(0.40+0.05)+10*(0.40+0)+10*(0.40+0.10)+10*(0.40+0.15)=18.4 points of strength over 39 levels, for 0.47 per level.

So, overall, Louis can expect slightly better strength growths than Raphael, but Petra can expect slightly better strength growths than Chloé. Which is hardly surprising. Louis is a better unit than Raphael is, and while Chloé and Petra are both good units, Chloé feels like she's the somewhat more likely of the two to struggle with her strength and need help. But in both cases, the difference is small, coming in at less than 5% effective growth rate over the course of the game.

While there are differences between the two games in how important weapon might is, it isn't due to growth rates.

[67chrome]

On 2/11/2023 at 11:20 PM, Tyrannate said:

In a game where low % rates don't feel nearly as impossible to achieve (and new players have been noticing this as well) having a weapons with 

20

105

10

12

Is just an incredible, incredible stat line to have.  

Lol.

That sword is such a potato.

It's +3 Mt, +5% Hit, -1 Spd vs. the Silver Sword.

 

Regal Blade and Hagakure Blade are miles better, and it's not like Blazing Sword or Fates are well-known for having good S rank weapons.

[FashionEmblem]

18 hours ago, DaveCozy said:

I wanted to come back to this thread and say one thing:

Lunar Brace+ from Eirika + a refined brave weapon + Pincer attack is actually really insane. Even in Maddening it's enough to ORKO bosses if you set it up right.

So now I do think Halberdier has some legit usefulness in Engage. Sleeper class imo.

Sleeper class by giving them the skill that costs 5000 SP? Even the base version costs 3000 SP—that's easily as much SP as a character will see an entire playthrough, if at all (depends on how much favoritism you give them with access to emblem rings). Lunar Brace(+) is good on every physical attacker, especially on Maddening thanks to chonky enemy stats. There's a reason it's so expensive—other notable skills in Speedtaker, Pair-Up, Vantage++, Wrath, and Dual Assist+ all cost 2000 SP vs. Lunar Brace's 3000 SP—the devs know it's good and price it accordingly.

While there is a niche to getting a brave attack out of a stronger weapon with the Pincer skill than a low might brave weapon, 5000 SP is a lot of favoritism and can make anything good. For example, Dual Assist + Hero with Lunar Brace costs 5000 SP and will give you chain attack chip throughout an entire map and have extra might to kill things in their own right. I'd say you can get more bang for your buck with 3–4000 SP and combo skills, such as pick 2 of Pair-Up, Vantage+(+), and Wrath (get the 3rd from your equipped emblem) or Canter and Speedtaker/Dual Assist +/Lunar Brace. And, once again, 3000 SP is a charitable estimate for spending most of the game with an emblem equipped, instead of being forced to rotate in a bond ring; 4000 SP is essentially never letting go of an emblem (difficult to do without DLC bracelets, unless you are Alear whose class is all about wielding emblems and the plot gives them the most access).

Sorry to come off a bit harsh. You are correct this is a good combination, just an expensive one that isn't relevant to the main story campaign without feeding all the DLC SP books to your single Halberdier or postgame grinding of SP books rewards from the Tower. Maybe if NG+ is ever implemented to carry over skills or SP after a playthrough.

Also, comparing against Hero so they can both be lance back-up units, Brave Assist takes off 20% HP, in addition to potential chip damage on a target from the Hero's own attack. More realistic base Lunar Brace is 20% of defense. HP is a larger stat pool than defense, so the numbers are just bigger. And the Hero gets to do this just by hitting level 5—no SP involved. The reason it may not feel as strong is because Hero is supporting someone else's kill and doesn't get to bask in the glory themselves.

Edited February 14, 2023 by FashionEmblem

[DaveCozy]

3 hours ago, FashionEmblem said:

Sleeper class by giving them the skill that costs 5000 SP?

Oh no, dear goodness no. You don't inherit it. You use Lunar Brace+ while engaged with Eirika lol

Definitely do not buy Lunar Brace, it's way too expensive. It also means you can only realistically use 1 Halberdier to great effect per playthrough.

Just as a reference. These are the true damage calcs that get added to your Halberdier's damage output with Eirika and other abilities in the mix. Assuming bond level 18 with Eirika, and quadding with Pincer Attack + Brave Lance:

((Enemy def × 0.3 rounded down) x4) + 5x4(blue skies) + 12(next to Alear) + 4(next to Magic Dagger spoiler unit) + your poison stacks (+4/8/12) + chain attacks

It's still expensive because it needs a Brave weapon, which are expensive in general to refine and engrave, but realistically you are only doing this with one unit per team. And you are also having that unit carry Eirika too. It's still a bit of a sink, but a lot more manageable and also relevant for the story like this.

Edited February 14, 2023 by DaveCozy
true damage calc added

[FashionEmblem]

3 minutes ago, DaveCozy said:

Oh no, dear goodness no. You don't inherit it. You engage and use it while engaged with Eirika lol

Definitely do not buy Lunar Brace, it's way too expensive. It also means you can only realistically use 1 Halberdier to great effect per playthrough.

Just as a reference. These are the true damage calcs that get added to your Halberdier's damage output with Eirika and other abilities in the mix. Assuming bond level 18 with Eirika, and quadding with Pincer Attack + Brave Lance:

((Enemy def × 0.3 rounded down) x4) + 5x4(blue skies) + 12(next to Alear) + 4(next to Magic Dagger spoiler unit) + your poison stacks (+4/8/12) + chain attacks

My apologies, I should not have tunnel visioned on inheritance and remembered just equipping the ring provides this skill. Eirika is good on pretty much all physical attackers, but actually gives magic boost statwise so I think she works well for a hybrid characters. Running her on Fogado and it's nice to have a bit more magic bolstering his Radiant Bow while Lunar Brace bolsters his everything else. I guess in my head I autocompleted to run Eirika on speedy character that can already double rather than relying on Pincer Attack. Chloe also does good magic damage and appreciates the lunar brace. I can imagine a reclassed Céline into a more physical class can do similar work, though she's not as speedy. I imagine it synergizes even better with Martial Master since it has a lightweight brave option, but you need characters with the stats to pull off arts.

[DaveCozy]

2 hours ago, FashionEmblem said:

My apologies, I should not have tunnel visioned on inheritance and remembered just equipping the ring provides this skill. Eirika is good on pretty much all physical attackers, but actually gives magic boost statwise so I think she works well for a hybrid characters. Running her on Fogado and it's nice to have a bit more magic bolstering his Radiant Bow while Lunar Brace bolsters his everything else. I guess in my head I autocompleted to run Eirika on speedy character that can already double rather than relying on Pincer Attack. Chloe also does good magic damage and appreciates the lunar brace. I can imagine a reclassed Céline into a more physical class can do similar work, though she's not as speedy. I imagine it synergizes even better with Martial Master since it has a lightweight brave option, but you need characters with the stats to pull off arts.

no worries, it's fair to point out the SP cost. I am glad you brought it up, because it gave me a chance to clarify.

Eirika definitely is a competitive emblem too. The advantage of Halberdier is that it's guaranteed to double regardless of speed, provided Pincer Attack is learnt and set up.

Setting up pincer attack is also fairly easy as well, utility like Canter exists to make that manageable. I feel there's some merit to putting Eirika in one due to that, specially when some Maddening enemies and bosses late-game can hit upwards of 30+ AS (I think the ch24 boss has 40 AS) and doubling gets really difficult without Speedtaker or setting up a Draconic Hex.

[FashionEmblem]

1 minute ago, DaveCozy said:

Setting up pincer attack is also fairly easy as well, utility like Canter exists to make that manageable. I feel there's some merit to putting Eirika in one due to that, specially when some Maddening enemies and bosses late-game can hit upwards of 30+ AS (I think the ch24 boss has 40 AS) and doubling gets really difficult without Speedtaker or setting up a Draconic Hex.

Chapter 17 taught me that Draconic Hex is going to be pretty vital on Maddening. Happy to learn sooner than later. I'm thinking maybe two characters inherit it alongside the Corrin user (Yunaka in my save file).