Golden Sun: "Alternate Elements" game concepts

[Misery]

Here are my ideas for a Golden Sun hack I started long ago. It's currently no more than a game balance adjustment with aesthetical changes, but if map editing through the editor becomes available it's likely I'll try to make something of it. If you're interested/patient enough to read through this, all criticism and opinions are greatly appreciated. Everything I plan to implement so far has been successfully tested, save for enemy resistances. Expect "new" skills, animations, equipment, classes, enemies... the works.

I've distributed stat bonuses to the classes, and I'm currently working on ability strength, distribution and PP cost, which I should be able to provide some good initial values for by using expected character stats for reference. Abilities, equipment and enemies will need a lot of testing of course, so nothing is really finalized, and this will mostly serve as an introduction (I doubt the class descriptions would fit anyway). If I haven't managed to scare you away by now, read on!

First of all, the four elements have new themes (except for earth) and slightly different properties, and are made to be more unique in terms of abilities. Note that I haven't changed any text or plot points to reflect this, though I suppose I could remove a lot of dialogue since it serves little purpose for this hack. This game normally has about two hours worth of unskippable cutscenes.

Elements:
[spoiler]
Earth
Makes the sturdiest fighters - attack, defense and HP are all high, and earth-based attacks are certain to hit those high HP monsters where it really hurts. However, PP gets a penalty, making them poor magic users, and they tend to be slow. They can raise defense and slow down enemies. They can give life (revive) or bring death. Lastly, they can break away the opponent's stat boosts.

Spirit
Has the strongest healing abilities, high luck and PP, good hp and defense, so they're ideal for keeping the party alive. They can protect the party from magical damage with wards and lower the offensive power of enemies. Other abilities are sealing away an opponent's magical abilities, purging the body of status ailments and drive out evil spirits. They can also summon spirits to their own aid. Their magic is powerful against undead monsters, but is limited to that, and their overall offensive power is the worst out of the four elements.

Chaos
The element of disorder, their high PP and agility coupled with good offensive power and wide range of spells means they excel at destroying their opponents. Their defense, HP and luck are all unimpressive though, making them prone to suffer damage themselves, and while they are the most powerful damage dealers, healing is nearly impossible for them.
They can increase the attack power of their allies and make enemies less resistant to elemental damage, and some of their attacks can cause paralysis.

Nature
Drawing their power from plants, animals and the moon, users of the nature element generally have a balanced set of stats (only agility and luck being above average), and a wide range of abilities. These abilities include healing, curing poison and status ailments, inflicting sleep, delusion and poison, HP and PP draining, regeneration, making allies faster, and reducing the defensive strength of enemies. They can also summon a variety of monsters. However, they aren't notably proficient in any field.

Note that when I say an element is effective against a type of enemy, that assumes I can edit resistances. I know enemies share sets of resistances, but I've also noticed in the editor that enemies have elemental levels, whatever purpose that might serve, so maybe it goes beyond that.

Anyway, what I have in mind for resistances is:
-High-defense monsters have proportionally high earth resistance.
-Chaos resistance is inversely proportional to defense - lower defense, higher resistance.
-Undead and "dark" type enemies have 0 Spirit resistance (if possible), others have 200.
-Nature is more or less neutral against most opponents.
[/spoiler]With that said, what I have in mind for this hack is:
-Make battles against normal enemies (and thus grinding) more interesting (more challenge, greater variety, more rewarding)
-Make all classes useful for the full range of levels, and rebalance the difficulty without changing the game's core mechanics. Ultimately achieved through extensive playtesting.
-Climb through the full range of levels, making the average player about level 80 at endgame/postgame. Increase the significance of levels.

I don't intend to make a "hard mode" type of hack, but rather change how the game is played. But of course, these changes include stronger and more numerous enemies, reduced healing ability, and other things to magnify the consequenses of the player's actions, so I guess it's technically harder.

Some stuff regarding the classes/game balance:
[spoiler]
-At some point I was going for a different organization in the class system, but unknown parameters kept screwing me over. ALL classes in this hack are currently edits of existing ones, the only change I've done to the class system is to remove a stage from the aligned element dual classes (Brute and Hermit lines in the original).

-Since there's no stat used exclusively for calculating magical damage, the effectivity of spellcasting-oriented classes is mainly decided by how frequently they can use their special skills. PP, in other words. There will be no PP restoring items. Expect more attack based spells even for mage-type characters, since the attack stat scales with level. This doesn't transfer well to wide area spells though, since the attack based add damage mod loses the diminishing damage effect for multiple targets, so they will still have to rely on base damage.

-Some higher classes have more powerful versions of their own psynergy series, based on the expected level the PCs will have upon reaching those classes.

-Bind uses the same effect as psynergy sealing djinn, giving it a perfect success rate.

-Pure stat down psynergies are not considered an asset to a class since they aren't even guaranteed to be successful, and even if they were, they would barely (if at all) contribute to a statistical advantage. As such, they are almost exclusively seen as secondary effects.

-No special djinni abilities. ALL djinn function as an EPA with 0.8x multiplier. That way you're still allowed to make djinn kills, but it's a bit harder than usual.

-No summons, there's simply no good way to make it balanced without completely changing the way they work, which I'm trying to avoid. Summon animations are used for fancy one-shot nuke style class exclusive killer moves instead.

-Unleashes are reduced to flashy critical hits with little impact on the game balance (much like in the first GS). They generally won't have a secondary effect, even most artifacts won't have them, and unless the weapon has an elemental affinity, the unleash will be elementless as well. After all, why use designated attack skills if you can get
them for free?

-The second party will not join you - you're stuck with four party members. On the flip side, there's no djinn blast/storm, since you can't switch characters out. 36 djinn will be obtainable. You will be playing as Felix, Jenna, Ivan and Mia (until I can make custom characters, if I haven't finalized the patch by then).

Some stuff on enemies and item drops:

-There will be many instances of enemies dropping non-artifact equipment, and other items that are mainly intended to be sold.

-Enemy experience/money rewards and item drops will be distributed depending on how close to a healing spot they appear. So enemies appearing on the overworld, for common instance, would be more work for less reward.

-No artifacts will appear *exclusively* as enemy drops (bosses are the exception).

-Enemy groups will generally be bigger. Also, several humanoids will be used as enemies.
[/spoiler]Coming up next: class stats and abilities.
Thoughts so far? I think it would be kind of hard to get an impression before seeing the classes...
anyway, stay tuned!

[Salanewt]

This seems like it will be a fairly interesting hack. I suggest that you continue with what you have been doing. There are just one or two things that I should bring up, even though you probably already know.

Since there's no stat used exclusively for calculating magical damage

This is sort of true, but magical damage is done a bit differently. Enemies (and PCs) have individual elemental levels, powers, and resistances. If I recall, they use these for calculating damage. While it isn't a stat like Attack or Agility, these values can manipulate damage. When it becomes easier, modifying the formulas for these should be able to make a really big difference in damage; if that is what you are after, that is.

You will be playing as Felix, Jenna, Ivan and Mia (until I can make custom characters, if I haven't finalized the patch by then).

Cool, those are nice characters to work with. If you plan on having custom PCs in the future, what would you name them? You could probably make these four placeholders for your other four if you want.

To add on to the latter response, I might be able to help you edit cutscenes to an extent - sprite swapping for the most part - if you want. I think the PC sprites are usually stores within the sprite/object data of each room, so it would be a matter of finding the values that coorespond to your PCs and swapping them. Kraden too, I think.

[leaf]

Hm... final fantasy approach? Or something similar, at least.

Judging by your djinn/summon revamp, it seems you want to pretty much completely obliterate djinn used as combat tools. A single target 0.8x multiplier is all but useless in combat if it doesn't at least set up something else, which pretty much renders djinn solely useful for class switching in your hack. This is a pretty radical change, and definitely steers the game away from being GS-like.

As for balancing summons without completely changing how they work... it's possible, just not doable with the current editor. The core problem with summons is the HP% damage being too high. Reducing that to be 2%/djinni instead of 3%/djinni should in theory yield relatively balanced summons. Until then, probably the best you'd be able to do is to completely strip them of their base damage component, and do strictly HP% damage, although that's just a band-aid fix and far from ideal, since it renders them near useless against normal foes, and still extremely powerful against very high HP foes.

Making the game span a larger level range is something you'll need to be careful with; with the current stat growth rates, you might have difficulty making a boss that can challenge the party at endgame. From my experience, it takes 600+ atk with powerful EPAs just to make a lv40 party feel threatened with the current system. With weaker healing, you shouldn't need as much, but it's still something to keep in mind - you might need to lower stat targets with a higher target level. Although this seems to go against your goal of making levels more significant, since typically, in low level games, levels are more significant than in high level games, since you earn levels slower and they have to deliver more. On the other hand, in high level games, you gain experience at a much higher rate, so while you may be quite a few levels behind in an area initially, it doesn't take long to catch up.

Overall, some interesting ideas here, even if I can't agree with your djinn/summon solution. Unless the whole point was to make a non-GS-type game, in which case it's probably fine.

[Misery]

Enemies (and PCs) have individual elemental levels, powers, and resistances. If I recall, they use these for calculating damage. While it isn't a stat like Attack or Agility, these values can manipulate damage. When it becomes easier, modifying the formulas for these should be able to make a really big difference in damage; if that is what you are after, that is.

Elemental levels are less than ideal for actual spell power, they already have a slightly different purpose in determining the effectivity of different elements and I want that function to stay intact. Adjusting enemy resistance depending on where they appear could sort of work, but it would prevent me from what I'm trying to accomplish. If I could use a stat to modify spell damage, I would prefer to use either be PP or level.

Cool, those are nice characters to work with. If you plan on having custom PCs in the future, what would you name them? You could probably make these four placeholders for your other four if you want.

I haven't given it much thought yet, I have some characters I was going to use at first, but now they're meant to be the main cast of another game :P But yeah, I should at least come up with new names for these, their current build is pretty close to what I want to use though.

To add on to the latter response, I might be able to help you edit cutscenes to an extent - sprite swapping for the most part - if you want. I think the PC sprites are usually stores within the sprite/object data of each room, so it would be a matter of finding the values that coorespond to your PCs and swapping them. Kraden too, I think.

That's great to know. I think I have a pretty good impression of what can be done with scripts and know roughly how it would be accomplished, but I haven't really experimented with it yet so chances are I'll run into problems if I try to do anything more than change the characters who join. I realize script editing will be necessary to keep too many djinn from joining you as well.

@leafgreen: Yes... now that you mention it, it's very final fantasy o.o
I really enjoyed the first FF (haven't played much else yet), this will be quite similar in both class concepts and the presence of resource management.

As for nerfing the djinn I'm a bit reluctant to do that, they're a very interesting device unique to Golden Sun after all. However, as I mentioned I want to focus more on normal enemy battles, djinn and summons stubbornly disagree with me, so I have no choice but to boot them out. Since djinn can be reset after every battle they function as non-exclusive special abilities that don't expend any PP. Since you have four characters, you effectively get a free level 4 summon for every battle, and that still leaves three characters for attacking, allowing players to easily bulldoze their way through powerful enemies at lower levels. Djinn kills can still be used for added experience and money if you're strategic, I'm not going to expect it from players though.

By replacing summons with mega-powerful (and expensive) psynergies, they can be used to wipe out threatening enemy groups every now and then, and chip away a good amount of health from a boss when used by an attacking character who doesn't use PP for much else. More conventional skills/spells will deal much more damage per PP though.

I'm aware of the problems that arise with a high-level party, adjusting the game accordingly seems like one of the biggest challenges with this. I won't reduce stat growth rates, but experience requirement and gain will rise exponentially to have a bit of control over levels, and encourage players to progress through the game. I'm not going to attempt to modify the HP cap, boss difficulty with have to come from something other than raw stats. This game offers some interesting options for bosses - Sentinel is immune to psynergy, Star Magician can call new allies indefinitely, Doom Dragon is fought in three stages, and most late-game battles are fought against multiple opponents.

Anyway, you bring up good points. I'll try to provide some more reference soon, perhaps it would be best to hold of further discussion until then...

[Rolina]

So instead of elemental resists, it's resistances to schools of power or different power sources?  Not bad.

Also, keep in mind that eventually we should be able to add three new elemental types - so if you wanna look into "derivative elements", you can try that as well.  For example, one that uses Average Overall Power vs Average Overall Resist,  or that looks at the average of two symbiotic types for offense/defense.  There's quite a few things you can do, once we get the tools to do it.

[Misery]

Well put, couldn't have said it better myself.
I'll have to retract what I said earlier about using attack for spells though, I had forgotten that pure base damage psynergy has a higher multiplier for power vs resist, reaching double damage under ideal conditions (whereas attack can reach 1.5x at best). I'll probably have spellcasters learn two series of psynergy, one for use at lower levels and one for higher.

I've also been replaying the games a bit and fully realized why I want to handle the djinn as I plan to, earlier I've found myself using them all the time in every battle instead of using psynergy, simply because it's the most effective way to play, and it just becomes boring. Playing through the original Golden Sun without ever using the djinn command felt much more rewarding. But I think I'll give them a 1x multiplier instead of 0.8x, so they act as a regular attack, but elemental.

Also, here's a screenshot of an edit of what seems to be pretty much the game's only uncompressed graphics, along with the font. What would normally be the Mercury element uses the same color as the background as its darker color, so it looks completely transparent, and has potential to look pretty weird if you choose a bad background color. Not sure what would be a better alternative.

For now I've made it so Ivan joins instead of Sheba when you talk to Echo, and Mia joins where Piers normally would. I'm not sure how to change which djinn come with them (as you can see), but I assume they're assigned when you start a new game, after choosing if you wanted to enter a password. Oh well, I'll try to handle that later. There was potential for some pretty hilarious cutscenes by entirely replacing Sheba and Piers with these characters, but I think I'll just have the latter act as "Kradens" - only ever coming out during cutscenes and not doing any fighting.

[Salanewt]

According to Teawater, the djinn that Piers starts with are defined shortly after he joins; basically selecting which unit and which djinn will join your party. I don't even think they are written into RAM until he joins you for the first time either.

As for the others, I believe it is written into RAM as soon as the transfer data stuff is taken care of; either after you enter the codes, or after you say 'no'. However, we know how to change this to an extent... If we play around with it, then we can sort of get what we want from them. However, aside from Flint and Fizz, they are mixed up and spread evenly throughout Party 1.

Here, this topic should help you in terms of editing the djinn that join Party 1: http://forum.goldensunhacking.net/index.php?topic=1421.msg28366#msg28366

If you need any help with figuring out the basics, please let me know.

[Misery]

According to Teawater, the djinn that Piers starts with are defined shortly after he joins; basically selecting which unit and which djinn will join your party. I don't even think they are written into RAM until he joins you for the first time either.

That's not very surprising, but good to know nonetheless. Anyway, I was referring to the GS1 party, Piers won't join in this game.

However, aside from Flint and Fizz, they are mixed up and spread evenly throughout Party 1.

This makes sense as well since Flint and Fizz are supposed to be unskippable, so they wouldn't want to include their info in the password. And the random spreading is probably to avoid having to record which characters the transferred djinn are assigned to.

I had a look into the topic, guess I'll just hijack Isaac's code to give Mia an extra djinn instead, since Teawater so conveniently posted the address. Then I'll somehow have to make the game skip the code that keeps giving them djinn.

If you need any help with figuring out the basics, please let me know.

Yeah, some day you guys will have to teach me how you do to find this stuff! But I suspect it mostly consists of using VBA's memory viewer/disassembler and having enough knowledge of the game mechanics to recognize patterns in instruction sequences. Much of it is still a mystery to me though.

[Misery]

Justified double post, I've finally written down all of the class and ability concepts! They're written somewhat in the style of the game mechanics guide that can be found on gamefaqs, due to this nature I've included them as text files, which I will update if I make any major edits. Most magical skills are just copypasted from original GS, new abilities have a lot of unknown power/cost values and even the ones I've already written down may be subject to change. However, this should give you a pretty good idea of where I'm going in terms of stats/abilities for classes. So, this is just a first draft, but if something looks particularly weird or broken to you, just say so.

Things could get confusing with the classes since I've used a lot of old names and messed around with them, putting them in places that seemed more appropriate to me. I've also not listed elemental requirements for classes, but since they're all edits of existing classes, anyone who knows their class mechanics will easily be able to figure it out. I think.  On the bright side, shared class names are nearly nonexistant (if there are any yet).

I haven't covered the item classes here, but I have an idea of what I want to do with them:
- a pure attacker class series, with high stats but PP, defense and luck set to 0
- a "puppet master" class series based on animism, manipulating armors in this case
- a class series that uses musical skills to fill a variety of roles

Now I can finally move on to enemies and equipment! I won't post any details about that, just some general ideas maybe, if they're different enough from the original. Don't want to spoil ALL of the game for you.

Anyway, I've been thinking a bit about if character/map editing would be added to the editor... then I might want to do something more in line with the original GS after all, so I'm not really sure where I'm going with this. But that's probably ways off still, and I think I would go with a similar build to this even if I used the original elements. So I'll carry on with this.