Please login or register.

Login with username, password and session length
Advanced search  

News:

Please use .png instead of .bmp when uploading unfinished sprites to the forum!

Pages: [1]
Fosil [Posts: 54]
  • View Profile
  • share
  • [August 19, 2013, 09:35:14 PM]
So, in my trials of game design, I am currently stuck on how to create a well-balanced, yet interesting skill set for job classes.
I've skimmed the net for resources on how best to do this, but.. uh.. I have not seen much that was helpful.

My best resource for, as it turns out, Skill Tree design, came from the Diablo II Phrozen Keep forum, where a very knowledgeable user gave his input on skill design philosophy. (Link for those interested.) But, obviously, that is Diablo II, and this is Final Fantasy Tactics... I think the basic concepts generally still apply?

That aside, feel free to give example skill sets/jobs from any game or mod. (Diversity is great!)
But FFT-esque compatibles are ideal. :)

I suppose I should offer my favored skillset for "balance": The Squire's Basic Skill skillset. (FFT Arena especially.)
And, as far as skillset "flavor" goes, I would say: Samurai's Draw Out. In addition to how the skills originate in-game--(inspired by a real-world culture as well as drawing power from a weapon)--there is much it can do, and each skill has its uses. (Again, FFT Arena's incarnation is much better.)

Thanks in advance for those of you who participate!
ALL THE THINGS Official Caretaker.
RavenOfRazgriz [Posts: 3030]
  • View Profile
  • share
  • [August 20, 2013, 12:07:03 AM]
Basic Game Design stuff to remember:

If you can cut it, cut it.  "Does my class need X?", where X is a piece of equipment, ability, cost, whatever.  If your class or ability doesn't need it, get rid of it.  "Fluff" sounds like nice variety on paper but all it does in the end is add noise to the primary goal of your class.  Don't worry about your class not having enough skills or anything like that.  For example, in Vanilla FFT, a character like Agrias is fun for most players to use despite "only" having 5 skills which have some noticeable overlap because what she does is concise, clear, and easy for players to understand.  Meanwhile, a class like Vanilla's Oracle isn't quite so well-designed.  It's very versatile because it can add almost every bad status in the book... but at the same time, the Oracle has tons of statuses with overlapping functionality which pollutes decisions unnecessarily.  You could argue it takes "skill" to decide whether the best idea is to Don't Act, Sleep, etc. that enemy unit based on hit %, CT, MP cost, etc... but at the end of the day, these statuses often do the same thing in terms of the flow of the battle.  There's no real difference between this and a more concise skillset that lets you choose between hitting a unit with Blind, Silence, or Undead as a "long term" debuff, Don't Act as a "short-term" but more powerful debuff, etc.  All having the giant book of statuses does is convolute the decision after the player has already decided what they think the best course of action is.  Concise, intuitive skillsets are always going to be more fun to use than ones that give you a false set of "options", so make sure everyone only has what they need to function and nothing more.

The second big thing to remember comes from the first - design space.  "Design space" is basically the amount of room you have to do new stuff before things begin overlapping with each other.  If Skillset A does something, other skillsets doing the same or similar things should come at a premium.  Each class's uniquity is what makes them attractive to use, and in FFT, a lot of fun is combining the uniquity of two classes together in order to create your favorite units.  This comes right from the first point - if each class has concise, intuitive skillsets, making the next class just as concise and unique is that much easier because there's more room to grow.  I mentioned before about the Oracle being an example of a class that's not as well designed, but I did oversimplify some of the differences between some statuses such as Sleep, Confusion, Berserk, and Don't Act, despite them all roughly affecting battles in the same way.  On the Oracle, having all four is wasteful and leads to false choices, but if they're spread between 2 to 4 classes, the overall design of the classes in question can be geared to what makes each of those statuses unique, leading to more thought-provoking and intuitive gameplay while conserving design space.  In FFT, conserving space is doubly important because the amount of Abilities in FFT is fairly limited.  Well-designed and concise classes leave room for a larger number of unique classes and abilities to be implemented overall, allowing you to give bosses and unique units more ways to be unique and memorable.

The third thing I'll mention for now is making it so each class has a "strategy" built-in.  ie, if a class can debuff units, they may want other skills to follow up those debuffs directly to make them self-sufficient.  There's nothing wrong with a utility unit like Vanilla's Monk (and a few classes like this are good for the sake of stability), but giving classes skills that combo with other skills or weapons they use leads to people figuring out how those things work and again overall more fun gameplay.  Ie, if a skill adds Berserk, you might want it on a melee-focused class and the skill itself to only have a range of 2 or less.  While a range of 2 sounds limiting and like a drawback, this ensures the Berserk enemy (usually) goes for the person who gave him Berserk, allowing the strategy for the class to build around itself.  Everything from costs to range restrictions should build into each class' overall strategy whenever possible.  Obviously, at the end of the day, you want the player making their own strategies and feeling good about them, but these kinds of "starter" strategies being built into the skillset gives players a place to start to learn and grow from.

(Also, one thing I'd slightly disagree with after honestly very quickly skimming the linked article, is that you shouldn't really have a 'bible' when making a class for FFT.  Just don't have skills be all over the fucking place, ie a class with 1 skill that's 5 CT but no MP, another skill that has MP but no CT, a bunch of free skills, etc., with the skills themselves having no real cohesion.  That's just terrible and unintuitive class design.)

(Also, remember to aggressively cost JP.  This is what allows the AI to learn more skills.  For example, in Journey of the Five, many skills, abilities, etc. cost relatively small amounts of JP but the player only gains 10 JP per action with no way to modify it or gain more beyond Move-JP UP.  This results in AI human units learning more abilities than they would in Vanilla and be more threatening without any kind of back-screen trickery forcibly making them stronger than they would be normally.)

That's all the time I have for now.  A lot of this applies to general game design as well, it should be noted.  The tl;dr is just to design good, concise things and the rest will fall into place.
Bad code, beware! The NOPs are coming to eat you!
formerdeathcorps [Posts: 1322]
  • View Profile
  • share
  • [August 25, 2013, 10:10:05 AM]
Move this to Help, Elric, and sticky it.  This is important.

Raven gave the basics, but I think some more abstract considerations may be helpful.  If you want more concrete examples, PM me.

一、Every patch should have a theme, some base set of inherently balanced mechanic(s) that determines most of its gameplay.  For Arena, this would be the fury / faith mechanics; for Raven's Redesign or SentinalBlade's Symbols of Rage, this would be how elements were tied to certain jobs and a weakness system designed to add a cost to boosting elements, for my patch, this would be my HP penalties on PA/MA/SP equipment, as well as equipment weight.  Notice that the above mechanics are inherently balanced because they are double-edged; in every case, the optimization of damage increases your exposure to damage, reduces your HP, or makes you more immobile.  Of course, balance needn't always be created on a double-edged sword, but it's often the easiest and simplest way.  Such makes most balancing decisions simply a matter of following the overall framework, and players will appreciate how easy it is anticipate the equipment / spell choices.  Special class skills / legendary items are then simply things that break these rules.
二、Determine the average tempo of your patch as well as the degree in which it will vary.  Tempo is defined as how many turns it takes to kill or incapacitate a unit.  High-tempo options would be 1 or 2HKO attacks, Charm / Don't Act / Petrify / Frog, etc.  Normal tempo options would be normal attacks, death sentence / undead / blind / poison / berserk, 35% weapon break, etc.  Slow options would be bag smacks, slow dance, power break (at 35%), accumulate, vanilla yell / scream, etc.  In general, an increase in variance means the strengths of slower tempo units (usually by boosting their defense) and the weaknesses of high-tempo units (usually by lowering their HP or defenses) will need further exaggeration.  Pretty much, as long as you don't change the average tempo from vanilla (which seems to be what you are wanting), always make sure maximum instant damage does not deal more damage than your beefiest class has when HP is maximized and to always allow for defensive actions to erase the offense, meaning that 1HKOs are undone by raise 2, AoE status are resolved by esuna, etc.
三、Speed of both attackers and spells is always the hardest to balance due to multiple balancing factors that often are contradictory.  Let X be the CT of a spell, Y be the SPD of the target, and Z be the target's CT at the time of cast.  The target will not intercept the spell only when  Y * X + Z < 100.  In general, a charged skill is too slow (X is too high) if an intercept happens when when Z = 20 and Y = average unit SPD, but if the top tier of charged skills are more dangerous than the top tier of instant skills (as they are in vanilla, my patch, Arena), maximum SPD (Y) and skill CT (X) must be high enough that even if Z = 0, the skill will be intercepted.  Pretty much, a patch that has higher maximum charged damage vs. maximum HP will necessarily have speed tiers.  However, fast units over 5-6 turns will end up accruing at least 1 extra turn compared to their average counterparts.  This means their effective tempo over the course of the battle (assuming they live) must be multiplied by their SPD / avg SPD.  Naturally, the adjusted tempo of a fast unit cannot be allowed to exceed the maximum instant tempo, or the fast unit must be denied access to adequate defenses (so they'll die before they can realize their advantages); otherwise, SPD supremacy will become the best strategy (as it was in S5 and S6).  Of course, you can ignore most of these problems if you set everyone to the same speed tier AND balance both charged damage (with lowered CTs) and instant damage to be less than maxHP sets, as Raven did in Redesign.
四、Unless you are making a patch with fixed levels, make sure your formulas support an equal damage progression between types.  Unless you have no improvements in weapons or gear and no PA growth (or no level ups), most physical weapons increase their damage quadratically with level because both PA and WP increase as the game goes on.  Most magical attacks in vanilla only increase linearly because only MA increases while the Y constant stays the same, resulting in spells that are both powerful and impossible to intercept early while being nearly useless late.  Unless you have no SP gear unlocked at different periods of the game and no SP growth, status magic is awful early game and godly late because higher SP increases the worth of each point of CT.  Long-term game balance requires consideration of both early and late-game.
五、Once you have the boundaries (maximum tempo, minimum tempo, range for SPD and charged CTs, maximum healing) and the theme of your patch set, you can then fill in the blanks left in between.  Usually, these are the skills / items that will not be the best in one category, but will have some dual use or some unusual feature that retains usefulness.  Here is where you can really show off your creativity and skill, because everyone has their own style.  ARena's method is to fill every niche in a unique way while making sure overspec'd sets meet their doom in a variety of ways from a variety of units.  Raven's patches tend to create semi-standard gear / attacks where slight differences induce large changes in choices made due to the effects of stacking.  My patch simply plants unique flowers in arctic soil, even if it increases my ASM work five-fold.  Of course, feel free to experiment with this, just make sure your experiments don't exceed the boundaries or the theme; as long as you are able to detect when you've created something too weak or redundant, or too strong or capable of making other things obsolete, or simply too unsuitable for the patch's goals, balance and flavor are easy to maintain.
« Last Edit: August 25, 2013, 10:34:20 AM by formerdeathcorps »
The destruction of the will is the rape of the mind.
The dogmas of every era are nothing but the fantasies of those in power; their dreams are our waking nightmares.
Paheej [Posts: 139]
  • View Profile
  • Nexus Studios
  • share
  • [October 06, 2013, 09:00:42 AM]
Backing up Raven's - "if you can cut it, cut it" - I always use this rule in designing skills and classes:

If you ever have an option which is always the preferred option then you either need to improve other options or decrease the effectiveness of your preferred option.  This goes for classes, skills, innates, etc.  If you have options which are never used or rarely used they should be eliminated.
The Joker
Rouroni Elmdor [Posts: 241]
  • View Profile
  • share
  • [November 14, 2013, 09:33:45 PM]
It's difficult to balance a system that's inherently imbalanced. I found classes like monk and ninja way overpowered (to the point I didn't use them in vanilla because it made the game waaay too easy) whereas classes like thief are underpowered in comparison. The most difficult part is not balancing the classes individually, but looking at your classes altogether and balancing characters with various skills.

Just about every time I played vanilla, I'd end the game as a SQUIRE with heavy armor, 2 knight swords, bloody hamedo.. and the rest doesn't even matter anymore. In many mods, I have seen monks with accumulate, which makes sense but is SO OVERPOWERED!

Most balanced classes? Mediator, samurai, archer, geomancer.. to an extent (overall high stats, but balanced)

Least balanced? Any knightsword class, monk, ninja, CALCULATOR

just my opinion from what I gathered from years of playing fft.
« Last Edit: November 14, 2013, 09:41:35 PM by Rouroni Elmdor »
The bird of Hermes is my name
Eating my wings to make me tame
Hyraldelita [Posts: 194]
  • View Profile
  • share
  • [July 10, 2015, 04:22:43 PM]
I have a bit different point of view on FFT for skills.

i'm building skill for class right now, and what i'm doing is this:

Create skill, try if they work , then get the class and do 2 or 3 fight with it, you'll be able to see your self , what is too OP or what is not.

It take time this way i know, but for me is the 100% way to make it balanced, or at least balanced in your point of view...

Cause balance in a game doens't exist, is just a matter of how much time you can play, how good at games you are, and how do you see balance in general.

for Example ... FFT samurai is totally unbalanced, i don't know why people say he's ok. Or to make it more specific, not the samurai itself, but his skill.

Magic based high damage, which means, put them on a mage and he'll destroy everything in the screen without any problem at all.

what is nice about FFT is that you can mix classes, and that's why balance is not the same for everyone, that's what i think.

a ninja with martial art no katanas is 100% hit Ko every attack

what is balance?

You can get a squire , give him Trow and only trowing chaos blade around.

The point is having a balance in the overall game is very, but VERY hard, cause if you think too much about balancing the sroty flow, the post game will have no strong class, while if you do the countrary the story will seems too easy.

That's why i say test them youself after you made a set of skills.
I'm trying to make every class a it more interesting then they normally are, for example i can give a very high damage skill to a class, but to do that skill it will require a lot of MP and by a lot i mean 50 or 60 which means even if you're high lvl you can't use it too many times, and in the early game you'll be glad if you can reach the mp to use it.

Or a no SP skill with a very long range but low damage, low damage+status.

In my point of view, "Balance" is pretty much a point of view most of the time.  So i'm just trying to make skills funny to use, and not too op.
The Puppet ---Master---
Elric (Overseer) [Posts: 4050]
  • View Profile
  • Youtube Channel
  • http://steamcommunity.com/id/Bones42
  • share
  • [July 10, 2015, 06:39:14 PM]
Why did you necrobump this? This seems extremely unnecessary...

    • Modding version: PSX
  • <@Angel> Teach a man to fish and he'll open up a fishery to compete against yours.
    Journey of the Five Youtube ChannelThe Lion War Current Status
       
    Jot5 Leader :: Eventer :: OtherTLW Leader :: Eventer :: Other
    Hyraldelita [Posts: 194]
    • View Profile
    • share
    • [July 13, 2015, 01:37:52 PM]
    Rofl i didn't even looked at the date XD sorry ahha
    Lockeadon [Posts: 126]
    • View Profile
    • share
    • [November 12, 2015, 07:39:39 AM]
    is it even possible to necrobump a sticky? sticky are stuck at the top because they have important stuff, right? so being at the top makes them more likely to be 'necrobumped' inadventantly by people commenting on the topic without realising how old it is, no? o.o
    The Puppet ---Master---
    Elric (Overseer) [Posts: 4050]
    • View Profile
    • Youtube Channel
    • http://steamcommunity.com/id/Bones42
    • share
    • [November 12, 2015, 07:45:45 AM]
    How can you not realize how old it is when you get a BIG RED WARNING saying that the thread hasn't been posted in in several months??

    Yes it is possible to necrobump a sticky. This is why stickies on -MOST- forums are locked. We keep them unlocked in case someone has something useful to add to them, since they are important threads.

    Your post was not this, and the one before that was unnecessary because he was responding to someone who posted 2 years before hand. Though, since it had a point, it was still more excusable than your post >.<

    • Modding version: PSX
  • <@Angel> Teach a man to fish and he'll open up a fishery to compete against yours.
    Journey of the Five Youtube ChannelThe Lion War Current Status
       
    Jot5 Leader :: Eventer :: OtherTLW Leader :: Eventer :: Other
    Pages: [1]