Loving the game, but struggling to easily understand how positioning of characters within a unit affects the outcome.
It seems like I have to move everyone around, go back to prediction screen, check damages, and repeat a heap of times to find that one specific layout that does 2x the damage of all the others. And it’s not intuitive as to why that’s the case.
It’s fairly early and I don’t really have anything specifically position-based in play, beyond the front/back and automatic ‘attack nearest’ mechanics (which I do understand).
It’s also not helping that when doing mock battles against a merc the damage prediction seems to change from battle to battle.
It sounds like you’re early on I don’t think position is as important at first until you’re customizing orders and equipment. For example if you prioritize “front row” for a buff it always seems to hit the middle front first.
Choosing which units seems more important. There’s more than rock paper scissors. Armored foes are an early problem. magic works, targeting back row works, there’s classes that confer -def to target too
After you win a combat talk to everyone and play those test battles they are really an tutorial teaching you what is good with what. Honestly thought I fast forwarded them
You can start with simple customizations like make attacks target lowest hp or backline
I still don’t know what pp or ap are. Why are there two ap values.
I am playing the test battles, it was in one of those specifically (the one after getting Rolf where you first face a gryphon knight) that I noticed how one certain obscure unit layout doubled damage over any other setup. I couldn’t really see why, and now I’m worried I won’t be able to stop myself from trying to do that for every battle…
I assume they stand for active and passive points, because that’s what they feed: AP are used on abilities that fire when it’s that unit’s “turn”, while PP are used on abilities that fire when some other condition is met (start of battle, end of battle, unit receives an attack, some other unit in the team does something, etc etc). As you get on there can be interactions between abilities from both pools and initiative/turn order that go pretty deep down the rabbit hole.
I’m 20-something hours into this (thanks, demo!) and still discovering new things and really enjoying it. That said, I’m going to be wildly in the bag for any well-done Ogre Battle clone, not objective at all. There are lots of little design decisions that I like, and I have no problem fast forwarding through some surplus Liberation battles that keep the honor system fed when some of the set pieces are so satisfying. I’m on Expert and it pushes back more than a lot of tactical games do, but not in a frustrating way and not in a way that makes me feel like I have to crunch down and really min-max it. Great stuff.
Let’s take the Soldier class for example, which is Chloe’s base class at the beginning of the game. Since Soldiers can pierce (attack) through a whole file (column), their positioning or gambits (conditionals) can sometimes significantly swing how much damage you put out.
Basically, without getting into the whole gambit system that allows you to set up conditions for the targeting of active (AP) and passive (PP) skills, characters will first target anyone directly across from them. So, if the enemy formation has a file (column) with two or more characters in it then you’ll see a damage boost from Chloe by positioning her opposite it, whether in the front or back row of your formation.
There are classes, like Gryphon Knights, who act similarly except they target a whole rank (row) instead of a file (column). Thus, how you array your characters in formation and how the enemy arrays their characters in formation can be a key interaction in the beginning of the game. Avoiding deep files (columns) across from Soldier classes, and positioning Chloe to hit any deep files (columns) in the enemy formation can heavily swing how an engagement will fare.
Obviously, once you start layering the gambit system on top, you can end up with all sorts of interaction affects. Another obvious interaction is having no one else in the front rank (row) if using Travis the Thief, as enemy characters that lack the ability to target the backrow will first have to defeat Travis. Since Travis is a supreme “dodge tank” having no one else in the front rank (row) forces most enemy characters to focus their attacks on Travis, who due to his general evasion and his starting PP skill will dodge most, if not all, attacks.
Hopefully that makes sense. Otherwise I’ll be back with illustrations, hah.
The damage prediction can sometimes change between repeating the same battle due to factors like your hit chance rolls on evasive units like flyers (Gryphon Knights, etc). One time you might be lucky and land a hit, other times you might not roll so well and get a miss instead.
That one position for Alain significantly changes the damage output against that layout of enemy (maybe because it enables 1 extra attack?), and I find it hard to intuit why in a way that allows me to create that setup without the tedium of manually trying a bunch of positions for everyone.
Seems a similar question to this Reddit thread, where someone claims:
Changes to formation and tactics can lead to a reroll for the fights rng, so crits, dodges, etc can be different.
That would definitely be one facet of it. The Gryphon Knight in that enemy unit formation, due to the lack of an archer in your formation, will cause the battle prediction to swing around a lot depending on each seed’s RNG for the hit chance. And if hitting that Gryphon Knight and defeating it is the difference between Josef regaining 1 PP or not, then that will then have a flow on effect upon how much Josef can use the Pursuit skill.
Even something just as simple as whether Josef lands the AP skill Slice can have a snowball effect on your total damage output. Because getting additional uses of the PP skill Pursuit will rack up how much damage your squad can dish out.
The damage prediction is not a Monte Carlo simulation (an average of expected damage) but an exact prediction of how the battle is going to turn out, including random chances.
That’s why stuff like adding nearby support to a tight combat can actually lower the expected results, since it changes the roll order.
At least that’s my hypothesis. I think I’ve yet to see a final combat result that didn’t match the damage prediction.
Yeah, it’s definitely a precise outcome result rather than a Monte Carlo simulation. That is easy enough to notice during some of the Overworld challenge fights that task you with defeating a specific unit in a one-off battle. You can repeat those fights with the exact same conditions and get wildly different outcomes between attempts solely due to all your RNG for hit chance, crit chance, etc… being different.
The number always matches up with what would have happened had you chosen to observe the battle.
I love that way of formulating it. Missed opportunity for the Japanese developer to tag some silly acronym onto it and feed it to Famitsu, something like QuaBaBa Quantum Based Battles.
How do you move characters from one unit to another? I would like to put Joseph into a unit with weaker folks to help level them but cannot figure out how to do that (unless that ability is not available yet).
Edit: Looks like I have to be in the Open World (I think that was my mistake). I will have to give it a try.
I’ve seen multiple times on let’s play videos where the prediction indicated that an enemy unit was killed, but it was instead left with low health. That might have been in the pre-1.0 version that was released to streamers, etc a couple of weeks ago.
