Please login or register.

Login with username, password and session length
FFTA Long Night 0.64 release!  (Read 23809 times)
Pages: 1 2 [3]
dck [Posts: 77]
  • View Profile
  • share
  • [April 13, 2017, 11:21:52 PM]
Re: FFTA Long Night 0.64 release!
« Reply #40 on: April 13, 2017, 11:21:52 PM »
Lots of feedback! I'll try to address things individually then explain a bit why some things work the way they do.
  • Un deux is a weird skill in that the game just takes the basic attack, halves its damage and uses it twice per weapon without much intervention on what you can make it do. I'll look more into it but at a glance I can't tell at all why it ignores the skip reaction flag for Dam > MP while it works on everything else.
  • Although for the most part the AI can use new skills, there's a decent bunch it can't and some like cultist's slipped through. A friend of mine ran into that same moogle as well so hopefully I can track down what's wrong without removing it from the AI skillset.
  • Healing power branches from the core issue with magic users, which is that starting out their stats are not appropriate for what their skills require early on due to being forced to start at level 1. More on this later down the post.
  • I remember finding out Prayer interacted like that with auto-life against some dragonheart dragon encounter :P. Not a conscious design choice though and it's not right to mislead the player into thinking something will work when it won't, at least by now it's been changed to dispel first (dispel can be resisted by immune monsters though).

Alright, so I figure it'd be a good idea to clear up what this hack's initially set out to do and how it relates to how it currently is.
I've always felt the base game is too bogged down by tedious tasks that didn't add much in the way of fun but did enable a great deal of benefits; such as walking aimlessly around the map as your units did deployment missions, stealing everything of value, grinding classes to unlock advanced ones or rechecking laws before fights and walking a few extra times if they happened to be inconvenient.

Another problem I had with vanilla is that its variety in skills is questionable, but could probably be good enough, if only many of them weren't rendered completely useless by others being straight upgrades. That, and the fact you have completely overkill damage from infinite sources that outclass skill usage outright.
There's nothing inherently wrong with OP, but I think the best type of OP is the one that can be used from both sides without the game becoming unplayable- which would be the case if you were to redo melee enemy units to be dual wielding chocobo armor clad monstrosities and gunners that ultima charge you from nine tiles away.

So with those two main things to overcome, you end up with a game that would benefit from having enemy scaling so new encounters are guaranteed to maintain a set threat level; this, I think, still keeps a sense of progression as the gear updates are very significant but lets random encounters stay engaging since the skillset is still somewhat viable even if their gear is no longer up to par.

Here is when technical limitations come in. Due to the implementation of the level scaling hack (necessary to not overlevel story missions and not to get level 25 enemies with a clan of 6 level 50s and a ton of level 1 recruits), it's no longer possible to specify a level on unit generation. This includes the starting party and given the way the stat restructuration works, makes them basically really bad at their specific jobs since they have no bonus from them.

To make matters worse, I found several ways to stop xp gain, yet couldn't find any way to multiply the gained exp at all. This is a pretty big deal, as playing the storyline without level bonuses to enemies meant the xp gain was so meager that finishing a test run up to Exodus left my Marche at maybe level 12.
This is of course unacceptable since the units only really begin working well with their roles at around 8+, so as a workaround a level bonus was applied to random encounters and sidequests- the extra levels do make them tougher, which I'm not against in non-mandatory missions, but in my experience don't really make that much of a difference compared to them normally outgearing you or having strong abilities for you to work around. However the main point of these (up to 3) extra levels enemies have is to kickstart the player's own xp gain and hasten the progress of his units so they begin to behave appropriately as soon as possible.

As you can imagine from the above, yeah the idea of doing away with levels altogether has been considered before :D.
It's a fine course of action, I think. The very good thing it does is letting units behave as intended from the get go, then it also enables all classes to best use their available skills from the first battle (debatable whether pro or con)- however from here on you have severe downsides like starting units being hardlocked to their starting roles with no chance of ever changing them, secondary skillset decision becoming a lot more fiddly since your starting class may not have stats for it, generally a large deal of loss when it comes to customization.

Not all is grim though, an idea I've been toying with for a bit is leaving experience gain alone and just reduce the level cap to far lower, perhaps 10. This way the initial level the starting party has makes a far greater impact and should stop units from being in that awkward zone I've seen testers struggle with in the early game. All while retaining a decent amount of build flexibility and still having a progression that leads to being stat-capped around the end of the current patch, even if one were not to do any sidequests at all.
Similarly, it would allow to take off the extra levels on enemies that are currently used to give the player more exp. These encounters are already plenty hard in some situations, such as Frosty Mage that you bring up yourself and "Gelato's cat", which is how some of the lads referred to the panther that was initially wiping them pretty hard.

Tying in with your concerns about encounters in general. I do set them up with the intention that they create scenarios in which players' choices are important and you can screw up bad enough that you throw the mission. I've lost a significant number of fights myself, and I don't think there's an issue with this so long as it isn't unfair in an unfun way. What I mean by this is that human players have a significant advantage by grace of not being a gba game AI, so they can deal with pretty stacked scenarios as long as they are given a fighting chance.

Now, I had the good luck to have a bunch of friends who raced each others on the main release and could see how they adapted to problems and how they dealt with stuff. Since they all shared tips with each others there were hastes and due/zea skillsets as far as the eye could see, but some things started directly walling that type of approach so eventually I could witness them splitting into their own thing each pretty much. Overall, I was quite happy with the result.
Some things were toned down (a lot*) since they were dumb and forcing stupid approaches like a team of 95 evasion onion knights; other than that, spacing to stay out of range and some debuffs like slow and berserk might be a bit too general solutions that work really well, but as far as I know none of them were using savestates during battles or otherwise gaming the hitchance RNG and could make it through most things. One of them even killed the roaming naiads which is something I wasn't completely sure was possible.

Of course, some of this is hard (adrammelech, tower ruins when first available) and some of it is unfair (naiads), but I don't really mind that as long the chance to fight back is there (which is very questionable when it comes to naiads)- and this is primarily why development has become stunted lately, because the main contributor to unfun wipes are monsters that get turn 1 moves they're not meant to get, and are capable of doing tremendous damage that doesn't leave the player with enough resources to fight back. There's a world of difference between being able to get hastes, shells and grace up before sprites start throwing clay imps around and titanias get a move, and just getting four of your unbuffed guys whacked by a holy before they move.

As you can see, walls of text are not an issue here :P, I rather enjoy talking about the game and spitballing can lead to some neat improvements. For example as consequence of giving the level 10 cap some more thought I realized I have artifacts already mapped, so I could easily give access to the ones that let you choose the type of recruit you want after battles. The artifact system is pretty much a dead mechanic at this point since the list was trimmed down and their battle effects altogether removed, so this could be a neat way to flesh it out and give it a real purpose.

*a lot means solar flare used to hit five units, gelato's cat used to have rend, totemas used to gain speed as they leveled up (so adrammelech could actually be faster than some units), fae used to be made to act first instead of only getting it on accident, etc etc. We've come a long way :D
Mickey B [Posts: 25]
  • View Profile
  • share
  • [March 16, 2018, 07:39:21 AM]
Re: FFTA Long Night 0.64 release!
« Reply #41 on: March 16, 2018, 07:39:21 AM »
Still working on this? Been about a year...
dck [Posts: 77]
  • View Profile
  • share
  • [June 04, 2018, 11:08:10 AM]
Re: FFTA Long Night 0.64 release!
« Reply #42 on: June 04, 2018, 11:08:10 AM »
Hi, unfortunately the project was dragged completely into a halt some time back due to real life in part, but mostly indecision on how to continue it:

Basically I can either ignore the issues with speed and leveling systems as they are and continue adding story content and sidequests or I can mostly gut the leveling system so there's no need for enemies wildly superior in terms of stats, which will be better but won't allow real finetuning of speed (and would be a bit messy looking since monsters would need speed equipment too); there's also killing leveling altogether which -would- make for rather elegant speed handling and no wild stat disparities in general, but would obviously kill crossclassing and customizing characters stat-wise. This sounds bad but bear in mind it would place a focus on the recruitment system with early artifacts to control it, which could make for interesting reasons for doing missions and progressing in general.

Without being able to iron out the random stat gains I'm fairly sure the last option is inevitable, a very minor level range would actually be catastrophic because most of the stats should be gained through levels if they would exist at all, and that would make randomly generated units that are one level higher almost unkillable powerhouses.

So yeah, lifestyle changes in the future might prevent me from finishing this whatsoever, so I've been retreading it a bit and thinking about giving it some closure, but for that I'd have to decide what base change to instate if any at all.

For what's worth I've had good ideas for the rest of the content in these two years anyway so I think if you liked what was here already, well, you'd probably like what's to come as well. Or not so much.
Adrammelech might not look that bad in the end.

PS: the new sorting of editing tools and mods on the forums looks great btw, thanks for that.
Chrone's Dome Fanatic
Zeke_Aileron [Posts: 189]
  • View Profile
  • share
  • [June 04, 2018, 12:33:36 PM]
Re: FFTA Long Night 0.64 release!
« Reply #43 on: June 04, 2018, 12:33:36 PM »
PS: the new sorting of editing tools and mods on the forums looks great btw, thanks for that.

You're welcome if this is referring to the new stickies i made.

    • Modding version: Other/Unknown
    • Discord username: Zeke Aileron™#0606
  • dck [Posts: 77]
    • View Profile
    • share
    • [July 20, 2018, 05:01:45 PM]
    Re: FFTA Long Night 0.64 release!
    « Reply #44 on: July 20, 2018, 05:01:45 PM »
    Since I've had this on the back of my mind for a bit now here's some skill changes I just whipped up:


    Blind Archery:Retains random damage element (normally a boost anyway) and now dispels and reflects the target (which makes it mostly impossible for him to get rebuffed).


    Fireshot:Now a fixed range 4, 5 AoE fire attack. All shots keep same physical base and go to 24 mp.
    Boltshot:Now fixed range 5, piercing lightning attack.
    Iceshot:Now fixed range 3, 13 AoE ice attack.

    Forest Ranger:

    Boldness:Can now be cast on allies up to 4 tiles away.

    Black monk's 3 0.5x damage skills are probably going to be thrown away or buffed significantly since they don't make much sense as they stand, and red mage is in an awkward spot and might well be remade from the ground up, although I do realize some people like the entire fullscreening thing.
    It might be obvious but working on this doesn't really take a very high priority for me right now, however if you have any comments on skills you find useless or other stuff you think might be worth looking at feel free to mention or shoot me a pm.
    dck [Posts: 77]
    • View Profile
    • share
    • [July 31, 2018, 10:38:57 PM]
    Re: FFTA Long Night 0.64 release!
    « Reply #45 on: July 31, 2018, 10:38:57 PM »
    Annd my drive to work on LN has actually kept up so far! Ability design always was my biggest draw to this game, so I took some time relearning and fixing things for a bit. The first post is now updated with the changed or new skills marked in red, it's not that much but it's stuff that always bothered me because it wasn't as polished.

    Also after finishing Ezel's skillset (two years late) I have the way paved to the rest of the unique characters, for which I have much clearer concepts. Next up is Llednar and from there one more mission content until Mateus, which I'm really looking forward to as well!
    dck [Posts: 77]
    • View Profile
    • share
    • [August 19, 2018, 02:44:12 AM]
    Re: FFTA Long Night 0.64 release!
    « Reply #46 on: August 19, 2018, 02:44:12 AM »
    And as biweekly update, the level cap, monster speed changes and general speeding up of the early game's power curve is underway. As two people noticed, yeah the first post is being edited almost daily since I actually use it as a resource when comparing certain skills. It's also generally more complete since it's a good reference point for balancing things.

    Towards the end of the week I got sidetracked with the numbercrunching for monster speed brackets etc and started making more skills because that's a lot more fun, so now (a few) vampires can resurrect the entire map as zombies or charmed zombies depending on team. Also zombify temporarily KOs the unit because that's a more interesting mechanic than just charming your units right away.
    Mind you, this resulted in the cultist's Ritual ability being turned into a mini-mass resurrect too, covering an invocation area around him and with the same charm properties for enemies. Ritual was always cute but impractical and not really niche either, so this should be a lot more useful.

    Also, I bit a very old bullet and started actually triple checking how the AI uses certain skills. Some I can't or simply won't fix as they're not worth the time, but I'm actually having a lot of success even getting them to nail you with reckonings -every time- they hit critical hp, which is cool. In general most of the "fancy" skills are actually AI friendly by now.

    Next progress update sometime soon (maybe)!

    EDIT: Testing more and more of how the AI works in detail is really interesting, it's not particularly cowardly but it has no way to judge the value of options if they only affect the user, so if it sees a negative effect it gets spooked and discards the idea of using stuff like Block Pain no matter what. What's funny is this changes immediately the moment it gets hit by astra lmao.

    EDIT2: Actual AI edits are beyond the scope of this hack, but understanding this more is making it really easy making small changes so the skills are properly recognized and used. It was always an annoyance for example that enemy cultists couldn't really use most of their skillset, but now they can turn and charm/petrify your units without any problem at all! \o/
    « Last Edit: August 21, 2018, 04:09:25 PM by dck »
    • Modding version: Other/Unknown
    • Discord username: adri#1824
  • Pages: 1 2 [3]