Author Topic: Theorycrafting! (Because I like competitive metagames too much)  (Read 133619 times)

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4378
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #525 on: October 13, 2015, 09:41:58 AM »
OK, let me try and break down my thought process into individual criteria, because this might be easier in categories...

Ordering of mastered (or to some practical degree mastered) skillsets...

1. Math Skill
2. Summon
3. Time
4. Draw Out
5. Yin Yang
6. White
7. Jump
8. Dance
9. Item
10. Punch Art
11. Sing

And everything else is stuff you're probably not going to use once you have mastered skillsets unless it comes with the class you're in right now.

Ordering of gamebreakingly undercosted skillsets

1. Summon
2. Black Magic
3. Item
4. Math Skill (1850 is undercosted for what it does yes)
5. Time (Demi is super inexpensive.  And while some parties don't like it, Haste is super inexpensive too).
6. Yin Yang (It's very laser focused, but tends to have strong skills for one fight, like Silence Song or Life Drain).
7. Talk Skill (that 100 JP invitation)
8. White Magic (600 JP for a nuke is a lot cheaper than all the other nukes.  It only crowds out a small number of other nukes, so that's fairly niche, but still)

R/S/M

1. Squire (Definitely the largest headache to balance)
2. Chemist
3. Time Mage
4. Wizard
5. Thief
6-11. Geo/Monk/Archer/Knight/Oracle (in...some...order)

I don't really feel the need to split these into earlygame/lategame R/S/M because skillsets are nearly always the focus, and R/S/M are nice buffs.

Miscellaneous grinding options

1. Mediator (Chapter 3 chantages, Faith buffing)
2. Thief (Chapter 3 chantages)
3. Move-Find Item I guess
4. Some level-up-down classes or something (Mime/Bard/Ninja!)

Carriers

1. Wizard
2. Ninja
(fairly large gap)
3. Geomancer
4. Knight
5. Various mages (Priest/Summoner for speed reasons)

Penalty classes (classes that can be painful to get JP in)

20. Calculator
19. Mime
18. Bard
17. Thief
16. Squire
15. Samurai
14. Archer
13. Monk (particularly when no abilities or female)
12/11. Armor classes  when clothing gets good (Knight/Lancer)
10. Chemist (specifically for people with spells pre-gun)

(Technically Mediator and Dancer aren't great numerically, but in practice, Dancer will always have access to dance, and you have a gun by the time you have any particular incentive to go mediator).

Am I safe in reading the cliff notes of this as

Quote from: metroidcomposite year 2000AD
what class should Ramza be?
Quote from: Excal then and forever
Monk
Quote from: elfbot2000
Samurai!

Quote from: the metroidcomposite of now and the future
Hmmmmm
Quote from: DankHolyElftron2020
You should reaaaaaaaally be thinking about Samurai.

I believe that is correct (I think you got the year right too).  You did miss the part where I was like "So how do I make Ramza a Dark Knight like Gafgarion?" though.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4378
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #526 on: October 13, 2015, 04:49:49 PM »
Alright, so last time I made a list like this I had "check points" which were something like 300 JP, 700 JP, 2000 JP, and 2000 JP was considered "endgame".  2000 JP is like...Chapter 2.

If I change these check points to be something like mid chapter 1, mid chapter 2, mid chapter 3, mid chapter 4...then yeah, Math Skill happens sometime in Chapter 2, and largely dominates things from then on.

However there's still room for other setups; you don't want to train five calculators at the same time; Math Skill isn't ideal for rescue missions (and you'd really like a Ninja for Roof of Riovanes to just remove any reliance on AI).

I will say, though, robes kind-of matter for chameleon robe.  Which means if you have a Ninja, and want to train a strong secondary, there's another reason why you might want to go for Lancer/Ninja rather than Monk/Ninja.  (Jump also lets you just time your jumps as Ninja to avoid mathskill).

So anyway, Calc -> #1


Next...what have we learned from laying out the criteria I did in the previous post?

I'm noticing I put Oracle both higher on the low cost skillset list than Priest, and higher on the mastered skillset list than Priest, so Oracle should move up above priest (+5 spaces or so >_>).

I'm redoing my criteria, adding more checkpoints to make Samurai sound justifyable and it...still kind-of sounds iffy.  The class is up there on the "your stats suck" scale.  It requires more JP than Mathskill.  You use it on a magic user, so it's not like Jump which goes on a Ninja who might not want to dip into Calc.

SnowFire

  • DL
  • Denizen
  • *
  • Posts: 4964
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #527 on: October 13, 2015, 05:29:41 PM »
For whatever it's worth, I'm definitely with Elf for not really being down to hype early Chantage.  (How is this even reasonable in C3?  I don't recall enemies pre-Meliadoul that'll have it...  do certain random encounter generics have it?  I thought the early perfumes were from breeding Uribos then poaching the results then hoping for the best which fuk.dat).  Even assuming it IS easy, Chantage breaks the game badly enough that relative power rankings aren't super-important anyway, except for the likes of Wiegraf / Roof / Elmdor where it's either Ramza alone or else Reraise doesn't really stop how you'd lose the fight.  It'd be like trying to rank FF7 characters assuming you grab Knights of the Round as fast as possible; fine, you win, we get it (even if KOTR was somehow tied to one character).  Mediator does have the run & gun, sure, but Chemist does the same but better, so eh.

I'm not sure how much this is worth since all FFT SSCs have been done, but I will agree that Lancer is pretty good for a physical Ramza build vs. Wiegraf, which is a classic sticking point for physical Ramza.  Run around like a coward sniping from 8 range & all, and still useful vs. Velius.  Sure Lancer falls off hard in C4 and its skillset isn't useful outside the class, but "good at one of the hardest fights" is a pretty decent niche.  (Maybe STILL only worth a horrible rating but that'd speak better of the competition rather than worse of Lancer, IMHO.)

Dark Holy Elf

  • DL
  • Denizen
  • *
  • Posts: 8161
  • Well-behaved women seldom make history
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #528 on: October 13, 2015, 05:40:53 PM »
There is a 1/3 chance that a Uribo will spawn in a certain one-shot story battle (Zigolis Swamp) in chapter 2. Invite it, breed it until you get a Porky, and poach those for Chantages, which you can buy from the fur shop starting in chapter 3. This is the only way to get it before Chapter 4. As I said before I don't think much of this.

I pretty much agree re Lancer as well.

Erwin Schrödinger will kill you like a cat in a box.
Maybe.

SnowFire

  • DL
  • Denizen
  • *
  • Posts: 4964
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #529 on: October 13, 2015, 06:58:47 PM »
Yah, I know the Uribo farming trick (although I've never personally done it), I was just assuming that maybe there was a less hair-pulling path out there.

Grefter

  • Villain.
  • DL
  • Denizen
  • *
  • Posts: 10386
  • True and Honest. Smarter. More aggressive.
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #530 on: October 13, 2015, 09:43:52 PM »
That is the least frustrating way of getting them, soooooooo...

Also since they don't come here to comment, I will put words in their mouths

Quote from: metroidcomposite
I'm noticing I put Oracle both higher on the low cost skillset list than Priest, and higher on the mastered skillset list than Priest, so Oracle should move up above priest (+5 spaces or so >_>).

Quote from: RICHARD and Laggy
:)
NO MORE POKEMON - Meeplelard.
The king perfect of the DL is and always will be Excal. - Superaielman
Don't worry, just jam it in anyway. - SirAlex
Gravellers are like, G-Unit - Trancey.

Dark Holy Elf

  • DL
  • Denizen
  • *
  • Posts: 8161
  • Well-behaved women seldom make history
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #531 on: October 13, 2015, 09:59:18 PM »
The least frustrating way of getting them involves poaching Porkies directly in END, but that's too late for them to be worth anything besides being shiny, at least in FFT original.

Erwin Schrödinger will kill you like a cat in a box.
Maybe.

hinode

  • Enough expository banter! Now we fight like men!
  • DL
  • Denizen
  • *
  • Posts: 1352
  • And ladies! And ladies who dress like men!
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #532 on: October 14, 2015, 01:23:03 AM »
Quote
Penalty classes (classes that can be painful to get JP in)

20. Calculator
19. Mime
18. Bard
17. Thief
16. Squire

Mmm. In paper this makes sense, but in practice it's never felt like an issue. Early C1 Squire isn't much of a liability if it is one at all, and then later on you can get lots of spillover from special characters like Agrias, Orlandu, C4 Ramza, etc. I guess if there was a hypothetical uberskill that required 2000 Squire JP it would matter more, but in vanilla FFT you can get the Squire skills that matter easily enough.

Laggy

  • ReDux'd
  • DL
  • Denizen
  • *
  • Posts: 1147
  • Generations of suffering & all I got was a stick
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #533 on: October 14, 2015, 10:09:22 AM »
Oracle has 90 MP mult in LFT because they got innate Move MP Up.
<Eph> When Laggy was there to fuel my desire to open crates, my life was happy.  Now I'm stuck playing a shitty moba and playing Anime RPGs.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4378
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #534 on: October 14, 2015, 03:37:24 PM »
Quote
Penalty classes (classes that can be painful to get JP in)

20. Calculator
19. Mime
18. Bard
17. Thief
16. Squire

Mmm. In paper this makes sense, but in practice it's never felt like an issue. Early C1 Squire isn't much of a liability if it is one at all, and then later on you can get lots of spillover from special characters like Agrias, Orlandu, C4 Ramza, etc. I guess if there was a hypothetical uberskill that required 2000 Squire JP it would matter more, but in vanilla FFT you can get the Squire skills that matter easily enough.

I agree with pretty much all of this.

But well you see, sometimes you want 2100 Squire JP so that you can unlock Mime!!!!!1  (Been there, done that in Chapter 4...)

And it also mostly it locks squire out of any serious consideration for other uses of JP.  Like...

Is it worth it to have Accumulate on a Monk?
A. no.
B. That's 300 Squire JP.  On top of the 400 Squire JP you're reserving for Gained JP Up and Move+1.

Is Counter Tackle going to be your first reaction ability for most of your characters?
A. Probably not.
B. You can change the class of your guests, you know.  If you already have the 400 Squire JP you want, You can make them Knights, usually giving them better stats, and getting you JP spillover for Weapon Guard; since Knight starts with 150 JP on average, you only need ~50 spillover.  Unlike Counter Tackle where you'd need an extra 180 JP after Move+1 and Gained JP Up.  (And Weapon Guard is usually preferable for mages).  Chemist is another popular guest switch job once Guns show up, because you tend to want some extra Chemist JP, but tend to have a lot of characters who don't really feel like being Chemists.

Like...yeah, you basically get everything you want from Squire basically for free, so it's really only a way of making Mime even worse.  But it exists!
« Last Edit: October 14, 2015, 03:40:55 PM by metroid composite »

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4378
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #535 on: October 14, 2015, 04:22:25 PM »
Even assuming it IS easy, Chantage breaks the game badly enough that relative power rankings aren't super-important anyway, except for the likes of Wiegraf / Roof / Elmdor where it's either Ramza alone or else Reraise doesn't really stop how you'd lose the fight.

Well, yeah, I'm assuming Chantages only break the game 1/3 of the time, if you bring a Mediator to the Chapter 2 fight.  (Because I'm not going to actually reset that fight; that would involve having a reset, which would be UNACCEPTABLE!  This is a zero reset zero casualty challenge!  It's even harder than the zero casualty challenge!!1!1!1!1)

Battles like Roof and Yardow are also why Ninja is fairly high.  What are Math Skill wizards actually pretty bad at doing?  Saving Rafa, since you can lose before you get a turn, and have issues with Friendly Fire.  With Ninja you always get a turn.

But mostly, this is a list about picking out optimal setups.  Oracle was misplaced and shouldn't be below those two; already fixed that.  I think we've established that Samurai probably should be lower if anything.  I think we've established that Lancer is king of the physical skillsets, synergizes best with Mathskill out of the list of physical skillsets (and I'll also add has a decent use for the Excalibur).  So...sure, Lancer over Thief and Mediator--I can get behind that.

But like...I'm skeptical about the rest.

Monk?  If you're mastering punch art you need to justify doing that over mastering jump--it's not a quick dip skillset.  And have fun not wearing Chameleon Robes.  In terms of a quick dip into Monk for R/S/M...Move+2 is a bigger deal than Martial Arts/Counter.

Dancer?  I do think they move up some with the addition of more checkpoints; don't think they move up that much.

Geomancer?  I'd rate Move+2 over Attack Up, and their ability to carry is pretty niche in vanilla FFT (LFT buffs their stats by 10%-15% while nerfing Wizard and Ninja stats for a reason.  Geos still have swords, so Agrias cares).

Knight?  Like...only in maybe 1/2 to 1/3 of playthroughs will you get good mileage out of Knight Swords on a generic rather than a special character.  Weapon Guard is good, but below Move+2.

Dark Holy Elf

  • DL
  • Denizen
  • *
  • Posts: 8161
  • Well-behaved women seldom make history
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #536 on: October 14, 2015, 09:34:31 PM »
Ninjas aren't particularly valuable in Yardow. The enemy ninjas have 9 speed, so even your ninjas won't outspeed that (the only +speed equip you can get that they don't have is the Green Beret, and that at best will make up for the fact that the ninjas have ~25 levels with Ninja speed growth and probably 5-10 levels on you to boot) so they don't prevent the (very slim) chance that the ninjas gank Rafa with unusually powerful Throws. Meanwhile, pretty much anyone can prevent Malak from finishing her off (the rare times he even has any chance of doing so), and she is pretty much never in the Summoners' range.

Math Skill also honestly deals with the roof pretty well! Protect or healing on Rafa, some status or other on Lede (paralysis should work offhand, or just killing with damage, though you'll probably have to be Level 18 to manage that (so that you can go Wizard). Priest with Green Beret + Sprint Shoes will be the 9 speed required at any reasonable level, Ubersquire as well most likely.

This is why I'm not terribly comfortable with 100% assuming we just blow everything up with Math Skill, it kinda makes all but a very small number of other jobs pointless. Despite some on-paper advantages, Ninja is honestly one of them. (Their biggest thing is being the fastest possible carrier for Math Skill, but I'm not sure that a ~10% edge over Priest is worth the trouble of unlocking them.)

Erwin Schrödinger will kill you like a cat in a box.
Maybe.

Laggy

  • ReDux'd
  • DL
  • Denizen
  • *
  • Posts: 1147
  • Generations of suffering & all I got was a stick
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #537 on: October 15, 2015, 12:58:54 AM »
I'm bored at work, so I figure I may as well throw in my 2 cents. Here's how I'm going to gauge the classes.

1) I don't assume that this is hardcore min-max world where Math Skill has beaten the game by the start of C3. While I'm going to give Math Skill some credit for its capability to reduce every fight to a single ability, the truth of the matter is that is not how most people play FFT. It's a little difficult to articulate this because this is ranking jobs on their effectiveness at beating the game, but it also assumes that you aren't setting up your entire party to optimize/synergize the way a CT5 Holy strategy demands. IOWs, there is some fun factor being weighed in where the more degenerate/boring a strategy is, the more I'll squint at giving it credit. People simply enjoy blowing up their enemies with giant space rocks and/or angry thunder gods, or punching demonic blobs and goats for absurd damage, in a way that watching Holy cast 10 times in a row as the only fight interaction doesn't provide.

2) Jobs are not ranked SCC-style or in a vacuum. Synergy with other jobs (and how easy it is to achieve said synergies) are definitely taken into consideration.

3) The amount of time investment (JP grind, generally), and overall the efficiency of the job, is a strong factor that is weighed as heavily, if not more heavily, than end/aftergame value which assume an overabundance of JP.

4) The difficulty of execution is factored in somewhat, but generally assumes a player is well-versed in FFT. This mostly is some wiggle room for CT calculations, Teleport, etc.

---

1. Summoner
Minimal investment: Easy to unlock. Early summons are absurdly damaging, efficient, and handily destroy the game. At 4 CT it is kind of really hard to mess Ramuh up even late.
Overall endgame value: Does technically get supplanted by Math Skill. But the rest of their skillset remains relevant (run your big summon of choice for assassinations, Golem is great utility, Moogle is solid healing, etc.) and you'll almost certainly never feel lacking in options nor power using Summon.

2. Wizard
Minimal investment: Coast on Bolt early on, grab Magic AttackUP later. Utterly wrecks the first two chapters due to stats and abundance of magic boosting equipment.
Endgame value: Best carrier in the game for the top skillsets, Summon and Math Skill. There isn't much that a Wizard with Black/Summon doesn't rip to the ground effortlessly for pretty much the whole game. Also the best user of Draw Out in the aftergame if desired.

3. Chemist
Minimal investment: 90 JP for better than half of the skillsets in the game. The immediate assumed default secondary for everyone early on, and probably still always set on at least one PC for the guaranteed, easy revival. Auto Potion is the best reaction in the game and comes at a dirt cheap 400 JP relative to its power.
Endgame value: Phoenix Down only gets better the longer the game goes. Auto Potion remains overpowered from start to finish. They simply don't provide any offense in a game where killing everything quickly is generally more efficient, but they are amazing at making sure you don't reset due to silliness or random mistakes.

4. Squire
Minimal investment: All that guest JP spillover and time you spend in Ubersquire is so hard...
Endgame value: Yeah, this is all up in that Gained JP Up and Move+1. Yes, this is so utterly prevalent that they place this high despite basically being ignored past C1 otherwise. Footnote - I recognize that Gained JP Up technically lets you access the good broken stuff (Summon mainly) faster, but past that delay it doesn't really matter too much because you don't actually -need- that much JP to break the game over your knee. Past that point it's just mostly for fun and playing with variety for the sake of variety. Which is all fine and good, of course, but is the main reason I have it below Chemist.

5. Calculator
Minimal investment: I'm always a bit surprised at how tame Calculator's unlocks are relative to its potential. They're pretty much equivalent to Ninja and Samurai, and makes Mime even more sad.
Endgame value: Newsflash, Math Skill is still insanely good even when you don't optimize your entire party for it. It is kinda painful to grind up Calc JP but it doesn't take -that- long and doesn't take a lot of spells for it to be really, really good, either. The sheer variety and power that the skillset presents is undeniable, on top of being instant and free.

6. Ninja
Minimal investment: In pretty much no way, I think, does Ninja even touch Calculator for effectiveness as the other usable late-unlocked job in the grand scheme of things. But on the other hand, Ninja is great out of the gate and needs less grind overall.
Endgame value: It gets perhaps more credit than it objectively deserves by being one of the few actual viable physical options in FFT. Still though, gamebest speed and OHKOs are things no one ever passes up, and it does edge out the mage jobs in some fights where going first and killing something immediately is helpful.

7. Time Mage
Minimal investment: Haste is nice, although it does present some additional difficulties to those using spellcasters. Demi is often overlooked despite it being a fairly easy to grab boss killer (I dunno, it just isn't that exciting). They suffer a bit from being pretty much all utility, which doesn't actually end fights. But overall they're a well designed job where everything feels usable and impactful - can't really go wrong with hasting your team and slowing the enemies and such.
Endgame value: Teleport rocks, of course. Short Charge is really more of a way to play around with most of FFT's spells to make them bearable rather than coming down to use the few insanely broken ones (see: Ramuh, Bolt) except for the Meteor setup, which does of course crush a large segment of the game if you're up for it.

8. Priest
Minimal investment: I'm honestly not at all impressed by Priest as a healer. Item is way, way, WAY better early on... like it's not even close - Potion and PD cost less than any two abilities on Priest, are instant, cheap, and just generally far more effective (don't need to worry about faith, cast time, etc.) Raise is kinda nice despite these drawbacks because it returns targets with HP but I still don't feel it comes close. Cure is good as inexpensive MT healing, but this is FFT - you're usually better off just killing stuff before it hurts you. By the way this is what Holy does.
Endgame value: 600 JP to put your faith in the light. The steam engine for Math Skill's omnipotence. Still great as an assassination spell on Wizard or whatever. It is pricy, but you don't normally need to cast it more than once. They also have speed which can be kinda nice sometimes if you need to sync up with enemy speed on a mage class.

9. Oracle
Minimal investment: Most people I know make Paralyze or Sleep the go-to spell on Oracle when they first pick it up. It's easy to see why, as they are strong debilitating spells with some aoe that just make a problem go away. Silence Song is amazing when it is relevant. Life Drain is fast and kills zodiacs dead. Sticks are cool, though honestly not that great IMO.
Endgame value: Most of it's up in Life Drain, I think, although again people don't tend to run 'Life Drain x4 to end the fight right away' unless they're doing a challenge or something. Otherwise, Oracle is a bit awkward because status just ends up being less efficient and slower than blowing things up in FFT, and that's the rest of their skillset (I mean, I am ranking the trio of Cure/Raise/Holy above all of YYM, which is kind of harsh but I think is actually true in practice). A lot of it just isn't accurate enough to be worth using or investing in. However, it is still an FFT mage, and thus is better than every other physical job not named Ninja.

10. Mediator
Minimal investment: Comparisons to Oracle are obvious. On the other hand, Invite can be absurd, they don't have charge times and they don't need to worry about faith. They're also gun users and can use Black Robe + elemental guns, which is nice. But their skillset is pretty much a two-trick pony (Invite, Mimic Daravon) aside from...
Endgame value: Br/Fa twinking. It's there, it's a thing. It's not really ever necessary and serves more to deal with OCD compulsions to perfect your party members, I think. They do certainly offer more than Oracles with dedicated use but I'm pretty sure I'd take an Oracle over one in just generally playing through the game.

11. Geomancer
Minimal investment: You only actually need a few Elementals for the skillset to be decent. You don't ever set it, but hey, you're not using Geomancer for its skillset anyway. Obviously a solid carrier for Agrias (much, MUCH better than Knight IMO, at least pre-Excalibur) and Attack UP pretty much cements this.
Endgame value: If Agrias didn't exist and wasn't so popular, Geomancer would be so much farther down the list. I mean, -yeah-, they can be mage carriers with 4 move and shields, thanks to Rune Blade/Aegis/etc. but I don't actually think these benefits, nice as they are, really can beat out the far more obvious carrier, Wizard, which comes with native Bolt and just way more damage. But I appreciate that they exist as an option, nonetheless.

12. Monk
Minimal investment: WHY ARE MONKS 3 MOVE. I remember mc saying over and over how much this aspect of them was making her cry when we were playtesting them in LFT. An interesting and varied skillset - said variety having a lot of appeal and making Monk quite popular with new FFT players, I find - hampered by being overpriced in JP costs, having annoying vertical tolerance issues, and being rather ineffective by the time you do actually buy them, with one notable exception (Earth Slash).
Endgame value: Earth Slash is indeed quite good, though pricy. Hamedo has... uses, though it's still just way too expensive. No hats is of course an utter killer in a game where hats give you the god stat, speed. Martial Arts Ninja is a thing. I had a very hard time trying to decide whether Punch Art or Jump win out when mastered, but Monk's just more fun. And yeah, Martial Arts Ninja edges them out for me.

13. Lancer
Minimal investment: Yeah, they're kinda eh... 3 move again (why), not fast, armored class. But hey, spears are all right (beat out swords for a good while, the extra range makes up for the mobility somewhat; they're pretty underrated).
Endgame value: Mastered Jump is good and definitely competing for best physical skillset. It's just really boring.

14. Knight
Minimal investment: Weapon Guard! Kinda usable in C1 if you aren't Bolting everything to death for... some reason! And generally everyone will have Knight unlocked, unlike some other jobs (hi Thief), because FFT magic/physical sides of the job tree are ~balanced~.
Endgame value: Well there are better reactions, like I don't know, Auto Potion over there. Admittedly RSM tend to be a bit of the backburner so you may end up just using a freebie Weapon Guard for quite same time, and it's all right at that. Excalibur can make one work sorta maybe. But otherwise, uh, why would you, ever, terrible skillset, terrible equipment, ugh.

15. Samurai
Minimal investment: It isn't.
Endgame value: I like Draw Out a LOT design-wise. But I pretty much have to echo NEB on this: it is forcibly staggered progression, expensive gil wise, expensive JP wise, the job itself is terrible on stats and equipment, and no one here has bought Blade Grasp hype for over a decade. In the super lategame department why on earth would you run this over Math Skill, objectively, other than because it's less time consuming to click through the abilities? If you throw out Math Skill and really don't want to worry about charge times then sure, it's a thing and it's pretty cool... strong self-centered MT damage, healing, buffs, and an 8 range nuke that supercedes every damaging DO before it (<_< >_>). How popular it seems to be with players is a testament that they had a good idea here going, but botched pricing it way too much.

16. Thief
Minimal investment: You get Move+2 when you go to Ninja! Steal Heart is cheap and usable, easily the best (the only?) skill they have available.
Endgame value: I have approximately zero respect for stealing shenanigans as a serious exercise because of the absurd timesink. You do it if you want to be a collector, but that's really about it. Everyone knows how bad they are otherwise. So they're a Ninja engine and little else.

17. Archer
Minimal investment: There is literally nothing here worth looking at that isn't Concentrate. Oh my god, Charge is so bad (yes yes I know you can make it work a bit early on but why would you stay in this job beyond the JP you need for Concentrate is beyond me.)
Endgame value: Concentrate's cool on your Ninja. But Move+2 is guaranteed and Concentrate isn't. Bows were cool at Dorter 1 when that enemy Archer was wrecking your shit with his Longbow, and then quickly become irrelevant.

18. Bard
19. Dancer
20. Mime
Too lazy to explain these in-depth, but they're pretty obvious (way too late, don't do enough being that late. Bard has comedy option Move+3 for those who don't like Teleport and it's easier to JP grind with them. Both have some okay early abilities in Life/Angel Song and Nameless Dance, but meh.)

Redeeming Mime is the reason I made LFT.
« Last Edit: October 15, 2015, 01:17:13 AM by Laggy »
<Eph> When Laggy was there to fuel my desire to open crates, my life was happy.  Now I'm stuck playing a shitty moba and playing Anime RPGs.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4378
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #538 on: October 15, 2015, 06:36:16 AM »
Yeah, I definitely remember some moments of "Laggy is buffing Lancers again in LFT, not because they're underpowered, but just because nobody in the playerbase is actually excited about them."


Serious, question, though--about either Laggy's list or mine--if we're valuing Monk primarily for their lategame contribution with mastered Punch Art as a Ninja secondary...should we value Samurai higher?

Draw Out is better than Punch Art; tends to deal more damage at most points in the game (if you're in a class with real MA; obviously not always in the Samurai class itself--although sometimes yes in the Samurai class as well.  Chapter 2 says hi, with Koutetsu dealing 84-96 on Samurai class, and Earth Slash dealing 30-48 on Monk class.  Wave fist isn't much better at 40-60 from Monk class).  Both Draw Out and Punch Art pass the "fun"/"appealing" barometer that Laggy cares about.

Like...don't get me wrong, I'm all for calling Samurai trashy.  I've done a Samurai SCC, and they ARE kinda trashy.  But like...by "trashy" I mean "they're not Summoners."  Not "They're even shittier than Monks."  Even in an "I'm stuck in this mediocre class training a skillset" scenario, I'd have to think about which one is worse.  Yeah, Samurai have shit equipment and poor stats, but Monks don't have hats, so like...everyone's a loser.

Ninjas aren't particularly valuable in Yardow. The enemy ninjas have 9 speed, so even your ninjas won't outspeed that (the only +speed equip you can get that they don't have is the Green Beret, and that at best will make up for the fact that the ninjas have ~25 levels with Ninja speed growth and probably 5-10 levels on you to boot) so they don't prevent the (very slim) chance that the ninjas gank Rafa with unusually powerful Throws. Meanwhile, pretty much anyone can prevent Malak from finishing her off (the rare times he even has any chance of doing so), and she is pretty much never in the Summoners' range.

Math Skill also honestly deals with the roof pretty well! Protect or healing on Rafa, some status or other on Lede (paralysis should work offhand, or just killing with damage, though you'll probably have to be Level 18 to manage that (so that you can go Wizard). Priest with Green Beret + Sprint Shoes will be the 9 speed required at any reasonable level, Ubersquire as well most likely.

This is why I'm not terribly comfortable with 100% assuming we just blow everything up with Math Skill, it kinda makes all but a very small number of other jobs pointless. Despite some on-paper advantages, Ninja is honestly one of them. (Their biggest thing is being the fastest possible carrier for Math Skill, but I'm not sure that a ~10% edge over Priest is worth the trouble of unlocking them.)

Mmm...these are all good points I hadn't thought of.  I will say that without Math Skill, there's not a whole lot useful that a Priest can do with that initial turn.  (Kiyomori if you have Draw Out!  Maybe if you have a couple Priests like this with Equip Sword you can stab an Assassin to death before she moves).  Although...maybe just getting the turn, and charging so that the Assassins attack you to interrup the charge is enough.  (Or charge a fatal spell on Rafa so they think there's no point in attacking her.  Meteor without Short Charge!)

Yeah, ok, point taken, Ninjas aren't as special as I thought they were.  Mostly this makes me wonder if Ninja is too high at 6th.  Like...certainly there's always been an argument for Time Mage above Ninja.  Maybe Priest/Oracle are worth considering too.

Grefter

  • Villain.
  • DL
  • Denizen
  • *
  • Posts: 10386
  • True and Honest. Smarter. More aggressive.
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #539 on: October 15, 2015, 08:00:18 AM »
Are you asking, should you think about Samurai?

I dunno, maybe we should ask Elf.  Elfman, should metroid be thinking about Samurai?

Also Laggy, really can't you just handle

Quote
1) I don't assume that this is hardcore min-max world where Math Skill has beaten the game by the start of C3. While I'm going to give Math Skill some credit for its capability to reduce every fight to a single ability, the truth of the matter is that is not how most people play FFT. It's a little difficult to articulate this because this is ranking jobs on their effectiveness at beating the game, but it also assumes that you aren't setting up your entire party to optimize/synergize the way a CT5 Holy strategy demands. IOWs, there is some fun factor being weighed in where the more degenerate/boring a strategy is, the more I'll squint at giving it credit. People simply enjoy blowing up their enemies with giant space rocks and/or angry thunder gods, or punching demonic blobs and goats for absurd damage, in a way that watching Holy cast 10 times in a row as the only fight interaction doesn't provide.

This by just going "Yes Calculator #1 ok thanks now lets talk about the fun things".
NO MORE POKEMON - Meeplelard.
The king perfect of the DL is and always will be Excal. - Superaielman
Don't worry, just jam it in anyway. - SirAlex
Gravellers are like, G-Unit - Trancey.

Ranmilia

  • Poetry Lover
  • DL
  • Denizen
  • *
  • Posts: 1687
  • Not a squid!!
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #540 on: October 15, 2015, 10:57:18 AM »
Dancer at rock bottom surprises me.  Sure, they're a pain to get to, not good at bosses, and inconsistent.  But once you're there, they make it much easier to gain JP in anything else (especially Calc!) and the skillset *can* YOLO RNG solo any non-boss fight.  No matter how bad the rest of your team is, "have someone Nameless Dance and protect them a turn or two" is always an out that gives you a floor % to win any fight it can work on - I've certainly seen multiple casual runs fall back on it.  On Hatbot-style runs I'd much rather be randomizing to Dancer than to Bard, Knight, Archer or Thief. 

Also on the LFT end, Dance went through quite a lot of revisions!  Turns out skills that autotarget every enemy on the battlefield and can't be defended against are very hard to balance.  There is not a lot of middle ground between "useless" and "oh wait, we made Slow Dance 100%, now two Dancers and a Mime win every fight in two turns." 

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4378
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #541 on: October 15, 2015, 04:18:46 PM »
Dancer at rock bottom surprises me.  Sure, they're a pain to get to, not good at bosses, and inconsistent.  But once you're there, they make it much easier to gain JP in anything else (especially Calc!) and the skillset *can* YOLO RNG solo any non-boss fight.  No matter how bad the rest of your team is, "have someone Nameless Dance and protect them a turn or two" is always an out that gives you a floor % to win any fight it can work on - I've certainly seen multiple casual runs fall back on it.  On Hatbot-style runs I'd much rather be randomizing to Dancer than to Bard, Knight, Archer or Thief. 

Also on the LFT end, Dance went through quite a lot of revisions!  Turns out skills that autotarget every enemy on the battlefield and can't be defended against are very hard to balance.  There is not a lot of middle ground between "useless" and "oh wait, we made Slow Dance 100%, now two Dancers and a Mime win every fight in two turns."

Yeah, I believe I mentioned they should probably move up given that I'm adding additional checkpoints after "you have 1850 JP and therefore know mathskill, gg". 

In particular, here's what I'm seeing as potential uses:

1. Calculcator with Dance (other than taking a long time to get your turn, Calc's dance about as good as anyone).
2. Samurai with Dance (for most of the game, Samurai have range 2; Dance gives them range, matches the side of the job tree they're on, and you probably want female samurai.  This might actually be their best non-Math secondary; Dance is also strong early while Draw Out is strong late, so that works out nicely).
3. Spellcasters with Dance (if you want a no-MP skillset to stick on MP hungry spellcasters, Dance is...pretty good, maybe 3rd best, after Mathskill and Draw Out.  That said, if your spellcaster is MP hungry because on turn 1 they murder half the enemies with Meteor...fewer living enemies makes Dance weaker, so ehhhhhh, the synergy here is questionable)
4. Chemist with Dance (Possibly the best non-Mathskill skillset on a Chemist.  Gives Chemist strong stuff to do on the first couple turns before they become a phoenix down bot; that said, if you already have Dance you may not stick in Chemist much past the 400 JP for Auto Potion).

I've also certainly done Ninja with Dance, but I'm not sure how much I would consider that a serious setup.  (Dance really doesn't make much use of Ninja speed.  Female Ninjas deal less physical damage.  It does give Ninjas more range, so that's cool.  And there is some synergy in that Ninjas usually want to clean-up or pick off stray enemies, not yolocharge into the middle of all the enemies.  But...yeah, I think Jump and Punch Art are still going to be the primary choices here, although if you happen to be going with a female Ninja I think it's a solid option; takes less training than Jump, and female Ninjas are not using Punch Art ever).

Dark Holy Elf

  • DL
  • Denizen
  • *
  • Posts: 8161
  • Well-behaved women seldom make history
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #542 on: October 15, 2015, 04:55:49 PM »
I'm just not sure what classes you'd move Dancer above. Dance is cool but the class is still a pain to unlock and as a skillset it does have some big weaknesses. I have a hard time seeing Dance > Punch Art by a big enough margin to make up for Monk's other advantages, or Dance > Draw Out period, so really I guess I'm saying that I don't really see Dancer rising above anything besides maybe one or more of Knight/Archer/Thief.


Quote
Yeah, ok, point taken, Ninjas aren't as special as I thought they were.  Mostly this makes me wonder if Ninja is too high at 6th.  Like...certainly there's always been an argument for Time Mage above Ninja.  Maybe Priest/Oracle are worth considering too.

My last post wasn't really intended to discredit Ninjas, they do do some things legitimately well. Fast, high, 0-CT damage is useful! Just, their niche gets badly crowded out by Math Skill, which can be almost as fast, and is also instant high damage (or similarly potent effects). However, I think it's very reasonable to talk about a metagame where Math Skill isn't seeing use and that's where Ninja is actually the best at their niche. Them vs TM is debatable, but I do think they should stay above Priest/Oracle.


I think Laggy undersells white magic healing somewhat (in particular, I think Cure is better than Potion; being able to heal multiple people is really useful, and CT on Cure isn't usually -that- big a deal except on Priest itself during that weird window in chapter 2 where a 7 speed mage is a Bad Thing) but it doesn't really matter because his broader points about healing not being super-decisive in FFT are spot-on.

Erwin Schrödinger will kill you like a cat in a box.
Maybe.

Laggy

  • ReDux'd
  • DL
  • Denizen
  • *
  • Posts: 1147
  • Generations of suffering & all I got was a stick
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #543 on: October 15, 2015, 06:46:20 PM »
Cure is better than Potion, but the PD vs Raise comparison just blows any advantages that Cure have out of the water to me, basically.

Dancer's just way too late. You have to go down the wrong side of the job tree for that gender to unlock it for a skillset that's pretty much inferior to the dead-easy mage stuff. Dance isn't bad in itself, but the question is, generally, why bother to begin with?

I could squint and see moving it above the classes that exist basically only to supplement Ninja while unlocking it (Archer/Thief) but even then the accessibility thing rears its head.

EDIT: Re: viability of Punch Art over Draw Out, you get to access Monks and use them for a period before not wearing a hat is utterly crippling, and they aren't item-locked out of their skillset - you can just beeline straight for Earth Slash. For Samurai they are terrible no matter when you use them. And MA Ninja w/ Earth Slash alone is pretty okay, you don't really need to master Punch Art at all (I mean you can take Chakra and Revive for giggles but usually you're just going to be chipping from 8 range when you can't get into melee for the OHKO.)

Remember that Koutetsu is 2 range and Earth Slash is 8 range, too. It kind of evens out, and Koutetsu looks especially worse when the still-common Bolt in C2 has 5 range - at least in some situations the extra aoe and ranged chip on Earth Slash can be relevant.
« Last Edit: October 15, 2015, 09:26:49 PM by Laggy »
<Eph> When Laggy was there to fuel my desire to open crates, my life was happy.  Now I'm stuck playing a shitty moba and playing Anime RPGs.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4378
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #544 on: October 16, 2015, 06:41:54 AM »
Hmmm...ok, spent a bit of time thinking about how I'd want to do checkpoints, if I was measuring stuff.

A chapter 1, 2, 3, 4, 4 checkpoint split sounds reasonable enough.

Something like...level 5 in Chapter 1 (playthroughs are often a bit lower than this, but whatever round numbers), level 10 in Chapter 2 (roughly where my SCCs tend to beat the chapter; 10-11 usually), level 15 in Chapter 3 (Most SCCs aren't level 18 for Velius/Wiegraf unless they intentionally grind a few levels because they're scared), level 20 in early chapter 4, level 30 in later chapter 4.

Store checkpoints would be...

Ch1: Sand Rat (the thing about Lenalia is that you literally need to backtrack to even use it; ain't nobody got time for that).

Ch2: meeting at Lionel Castle (Barius Valley has the same problem as Lenalia where you need to detour to even get stuff--detour pretty far if you want Cross Helmets)

Ch3: Orbonne Monestary (Yardow has some of the same problems outlined above.  Also, people do buy Bracers, but not necessarily in Chapter 3.  Most expensive thing in the store at any point, only available for three fights in the Chapter).

Ch4-1: I think Betha Garrison?  I considered chapter 4 start here, but like...not that much changes?  It's two store checkpoints away from Orbonne, and two of not the most important store checkpoints.

Ch4-2: Endgame, with some consideration for sidequests and stuff.

What do these level 5, level 10, level 15, level 20, level 30 numbers mean for JP at each point?

Well...level 5 in Chapter 1 means that you...gained 400 exp.  You'll have a mixture of 10ish exp actions (interacting but not killing) and 20 exp actions (killing).  Overall, let's say a 2:1 ratio of standard:kill.  This means 30 actions overall.

At Job level 1, with Gained JP Up, an action gains you 15 JP.  This goes up with job level, so 18, 21, 24, etc.

Another thing to consider is the overshoot effect.  Battles on average last, let's say, 3 turns.  So...let's say you only want 200 Chemist JP.  Sometimes you'll take exactly the number of actions you need, sometimes 1 extra action, sometimes 2 extra actions.  On average you waste 1 action per every multiclass.

With all this in mind, the average number of actions to reach a job level and then stop are...

JL2 (200 JP in class): 4.3
JL3 (350 JP in class): 12.6
JL4 (550 JP in class): 22.2
JL5 (800 JP in class): 32.6

So...if you started the game with Gained JP Up, you could almost get one class to 800 JP.  Note that these values depend on player level too, so you gain JP faster at higher level--the above turn counts mostly apply to Chapter 1.

So like...6 actions to get Gained JP Up into JL2 Chemist, JL2 Wizard, JL2 Time Mage, JL2 Summoner...that's 23 actions to get Summon.  Still about 7 actions left; could get one of those classes very close to JL3; maybe learn Moogle or something.


For the Chapter 2 checkpoint, there's actually more levels gained; 5->10 instead of 1->5.  This means probably 37 actions.  Additionally, actions are now worth a little bit more:

JL2: 3.8 actions
JL3: 10.9 actions
JL4: 19.2 actions
JL5: 28.5 actions

So...question: can you get Calculator?  Let's say Chapter 1 you went for JL4 Wizard.  You can proooobably swing that in Chapter 1 using guest spillover to shortcut you Gained JP Up.  Then this chapter you'd need JL4 priest, JL3 Oracle, JL3 Time Mage...that's already about 40 actions.  With a lot of help from spillover you can probably swing this, but if you're not getting much spillover, you won't be online with Calc yet at this checkpoint; you actually just barely won't have it unlocked.

(For what it's worth, if you're going dedicated Monk, you'd be about at roughly job level 7 now, 1550 total JP, so you could get Earth Slash, Wave Fist, and...one of Chakra/Revive.  Lancer would be worse-off due to stricter unlock requirements, but probably Job level 6--1150 JP, which means like...Level Jump 8, vertical Jump 3 is doable by this point...but only if you skip all the intermediate jumps and start grinding Lancer in Chapter 1 when they don't have spears...).





Anyway, might actually run some calculations on the various checkpoints later.  Just brainstorming what a set of checkpoints might look like, and thinking about which checkpoints make sense to check.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4378
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #545 on: October 16, 2015, 07:17:29 AM »
I have a hard time seeing Dance > Punch Art by a big enough margin to make up for Monk's other advantages, or Dance > Draw Out period, so really I guess I'm saying that I don't really see Dancer rising above anything besides maybe one or more of Knight/Archer/Thief.

Sure; I mean, it's noteworthy that I did say "pick up Dance to help you train in Samurai"; I'm not advocating for Dancer above Samurai here.  And Dance vs Punch Art is...probably pretty close in an abstract "if you were gifted one mastered skillset, which is better" comparison, so yeah, even if Dance is a little better, Monks are a whole lot more convenient.

But like...if we're doing later game no-mathskill comparisons, yeah, I probably see Dancer above Bard and Archer.  Maybe one or both of Geomancer/Knight.

Quote
EDIT: Re: viability of Punch Art over Draw Out, you get to access Monks and use them for a period before not wearing a hat is utterly crippling

Umm...but Monks are bad at several points.  Like...chapter 2 they're bad just because it's not LFT and Power Sleeve doesn't show up until half-way through Chapter 3, so their damage is trash.

A few physical classes still manage to be alright in Chapter 2 granted.  Lancers are jumping for like...108 damage (double to triple Earth Slash damage).  Ninjas can physical attack for about 80 with daggers, and throw balls for about 64.  All while being 8 speed (Going on 9 speed).  And can buff this with a support ability.

Quote
And MA Ninja w/ Earth Slash alone is pretty okay, you don't really need to master Punch Art at all

Is it though?  Like...I'd hazard a guess that Throw outdamages Earth Slash...certainly throughout Chapter 2, and probably during a decent chunk of Chapter 3 (I'd have to run the numbers).  And like...it's pretty similar range (6 range; 7 with Battle Boots.  But not height or line restricted).

Like...yeah, sometimes you'll chip two people with Earth Slash.  Have a cookie for that.  But maybe your Ninja should go equip a useful skillset that does something your class doesn't already do...like Item for Phoenix Down, or Time Magic for Haste.

Quote
Remember that Koutetsu is 2 range and Earth Slash is 8 range, too. It kind of evens out, and Koutetsu looks especially worse when the still-common Bolt in C2 has 5 range - at least in some situations the extra aoe and ranged chip on Earth Slash can be relevant.

Uhh...Koutetsu totally has better AoE, more likely to catch multple enemies at once when compared to Earth Slash.  Like...yeah, the range on Earth Slash is arguably a bigger deal because you can stay safer.  But Koutetsu is pretty good at AoE.
« Last Edit: October 16, 2015, 07:21:58 AM by metroid composite »

Dark Holy Elf

  • DL
  • Denizen
  • *
  • Posts: 8161
  • Well-behaved women seldom make history
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #546 on: October 16, 2015, 07:26:38 AM »
I do think your levels are erring slightly on the low side for what it's worth. You're proposing that we gain only 5 levels over the course of chapter 3, so that's... 45 Exp a battle per PC (assuming no randoms, which isn't really reasonable given the crossing of Zeklaus Desert thrice and potential backtracks for equipment), in a chapter where average enemy level is like 26 (i.e. every action which targets an enemy should be netting you almost a quarter of a level, so the average battle ends in 1-2 rounds? It's possible (certainly fights like Wiegraf 2 and the roof generally will, the other 3 assassination maps are possible but depends somewhat on playstyle) but that's certainly a very offence-oriented setup. I also don't really agree that most SCCs don't hit Level 18 for Riovanes, certainly most of the lower-offence ones do. Of course, ones which high-end play most closely resemble (Summoners!) typically don't, which may be the relevant point here.

Generally I'm okay with erring low for levels anyway (if someone needs extra levels to perform, obviously they look worse than someone who doesn't), but generally you should acknowledge the range of level possible for any point in the game. Party composition, how much backtracking one does, and RNG luck with randoms all have a significant effect. Same with JP, which has extra variables like propositions to consider: does the player do them? Just the ones on adjacent blue dots? Does he/she know how to get the most JP out of them (something I don't even know, but apparently this is documented now). Basically, if a class or setup is "just missing" a speed checkpoint or JP checkpoint, you should try to acknowledge that, and vice versa: if a class or setup is just over one of your checkpoints, acknowledge they may fall short.

Erwin Schrödinger will kill you like a cat in a box.
Maybe.

Dark Holy Elf

  • DL
  • Denizen
  • *
  • Posts: 8161
  • Well-behaved women seldom make history
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #547 on: October 16, 2015, 07:44:34 AM »
Quote
Is it though?  Like...I'd hazard a guess that Throw outdamages Earth Slash...certainly throughout Chapter 2, and probably during a decent chunk of Chapter 3 (I'd have to run the numbers).  And like...it's pretty similar range (6 range; 7 with Battle Boots.  But not height or line restricted).

Idly running numbers:

Early C3: Assumed base 7 PA, 8 Speed

Earth Slash using PA+3 (Twist Headband and one of Judo Outfit or Diamond Armlet): 75
Throw Ball using Green Beret: 72

Other stuff exists but is more pricy. So yeah, similar.

Orbonne checkpoints: Assumed base 8 PA, 8 Speed

Earth Slash using PA+4: 108
Throw Ball using Green Beret: Still 72

This is where Earth Slash starts putting in a meaningful advantage. Quadratic PA formula makes this the best possible case for using a Bracer... in fairness, you can buy four Slashers with that money.

Earth Slash using PA+7: 154
Throw Slasher with Sp+2: 156 (144 with Germinas Boots for max range)

So that's actually fairly comparable, though obviously the cost of thrown Slashers will catch up with you in a big way if you use this often.


Generally I do think Earth Slash is a decent push from mid C3 on (I find thrown axes a bit overcosted, and balls don't really keep up once you have Power Sleeve) but yeah other skillsets do compete as mentioned. The more other skills you can pick up (Wave Fist for the extra 13-17% damage, Chakra, Revive) the better Punch Art looks, but there's some lost time of course.

Erwin Schrödinger will kill you like a cat in a box.
Maybe.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4378
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #548 on: October 16, 2015, 08:47:20 AM »
Quote
I do think your levels are erring slightly on the low side for what it's worth. You're proposing that we gain only 5 levels over the course of chapter 3, so that's... 45 Exp a battle per PC (assuming no randoms, which isn't really reasonable given the crossing of Zeklaus Desert thrice and potential backtracks for equipment), in a chapter where average enemy level is like 26 (i.e. every action which targets an enemy should be netting you almost a quarter of a level, so the average battle ends in 1-2 rounds?

Mmm...maybe I should go for something more organic and assume all battles take 3 rounds (except for obvious assassination missions like Roof and Wiegraf 2).

Quote
The more other skills you can pick up (Wave Fist for the extra 13-17% damage, Chakra, Revive) the better Punch Art looks

Oh well for sure--the point of Punch Art to me is variety.  Like...it's a mild downgrade from item in terms of healing, but it has damage moves.  It's a big downgrade from options like Jump, but it has healing and revival.

But this still leaves me skeptical about the "learn only Earth Slash, it'll be fine" plan.

Grefter

  • Villain.
  • DL
  • Denizen
  • *
  • Posts: 10386
  • True and Honest. Smarter. More aggressive.
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #549 on: October 16, 2015, 12:53:19 PM »
Mmm...maybe I should go for something more organic and assume all battles take 3 rounds (except for obvious assassination missions like Roof and Wiegraf 2).

I in no way want anyone to actually do this, but this statement pushed the methodology part of my brain that went "Well we have all these studies of "regular" play throughs in LPs, I wonder what the data in a meta analysis shows your actual average level is."

But fuck that data gathering exercise.  I am dead certain you are both estimating way under what regular gamers end game levels are even as you try to take into account that you are better than average and tend to favour super efficient play.

My knee jerk is that you want to take SCCs as a point of reference here.  Not because they are a great data point for average plays, but the SCC rules are a pretty damn good reflection of a moment in time where FFT was a wonderful new mystery to people and your "average" player was fairly decently represented.  My knee jerk was to say 3/4 of the level "caps" of the challenge would be your target, but even that I am sure over targets it.  2/3 is a bit more in line with what I think of as "average"  (which is also about 10 levels a chapter, a touch lower in Chapter 1).

From a player psychology perspective I think you both tend to underestimate how even knowledgable players will get trapped in a completionist mindset and try to get stuff on the way up a tree that they don't need and actually slows down their progression.  "If I just stick around in Squire a little longer I can get Move +1 after JP Up which will be really good, Oh and maybe I should get accumulate too?"  Repeat for half the classes on the way.

Not that this should really impact the actual root assessment at play here of what is most powerful though...
NO MORE POKEMON - Meeplelard.
The king perfect of the DL is and always will be Excal. - Superaielman
Don't worry, just jam it in anyway. - SirAlex
Gravellers are like, G-Unit - Trancey.