Insane Difficulty: Started a Legend of Mana mod. - Insane Difficulty

Ok, so I've just completed the first two missions; Lost Princess and Nicolos Bussiness Unusual, and heres how it went:



Here's my thoughts on your changes:

A few considerations real quick:

0. If anyone else wants to contribute, PLEASE FEEL FREE TO! I put the source in the zip file and you might be able to glean some useful information from where my modified code jumps back to.

1. I have no idea if my movement changes for enemies are going to impact some of the bosses that are mostly stationary but either teleport around or have parts that activate/deactivate. Like that face guy in the underworld.

2. I'm going to try to make the hitboxes more lenient, since this will help both the player and enemy connect attacks better. It'd be harder if I selectively made the player's hitbox bigger, but I don't think this would be as fun.

3. I'm pretty sure Tropicallo is going to have nothing but one-hit-KO attacks and I have no idea how to address this. He was pretty much at that point in the original game, though. If I can figure out the data files maybe I can reduce his HP a lot to compensate.

Anyway, @ Raemnant's points.

1.
I actually wanted to do stuff with the heavy and quick attacks, for exactly the reasons you mentioned, except I couldn't figure out how to determine whether the player was doing a quick or strong attack.

Power attack and quick attack varies in usefulness by weapons. The longer weapons just kind of do better with quick attacks than knives. That being the case, yeah, it's more focused on the strong attacks now. By a good margin.

I noticed that knockback and skills have separate movement functions. I think that increasing the knockback from strong attacks might actually offset their usefulness, since if the enemy moves back far enough you can't really chain them as much. Although this isn't going to help with the more stationary bosses.

Are heavy attacks still hitstunning them..?

2.
The stuff you mentioned about the tech move distance is nominally a bug. I guess they use a different position updating function than normal movement.

Although now that you mention it I think the issue with jumping isn't so much that you're moving fast but that you're basically invincible while moving.

Maybe I could make jumping drain the ST bar -- then there'd be some symmetry to how running in SoM emptied your charged bar.

Hey actually making strong attacks drain the ST bar and having quick attacks refill it could be interesting, and would introduce an element of risk and reward.

The question here is just... finding the relevant code in the game for doing this.

If someone could give me advice on how to find the input handling code/find it for me, that would be a HUGE help.

Quote

Not really hitstunning them in the standard sense, but there is a power attack you can keep chaining nonstop to the point where the enemy cannot do anything. They get knocked back a bit, you go forward a bit, and the enemy gets a few frames to act before you reposition yourself back up to chain another one. Like literal deciseconds

Quote

This is a very good idea. Sometimes the Special Techs are a bit too strong, and it would be nice to have things that can take away from the ST pool so its harder to spam them, if you can manage that sort of thing. Now that I think about it, it would actually make it phenomenal. Such a new dynamic would greatly increase the gameplay value

If I had any knowledge of coding and modding games, I'd totally help you out

Okay, good point on the knockback. I'll try increasing the knockback a little so that you can't chain it so much. If the enemy gets pushed back far enough that they move out of range of your next attack, that might help some. I have no idea if you can pin them against the map boundaries or not though.

The dynamic I want is that the knockback is good for breaking up groups of enemies so you don't get surrounded, but not something you can keep spamming at close range.

Just tested it, and you cant really wall lock a single enemy with the down diagonal power attack, but what you CAN do is travel lock them. When you hit them with it, they get knocked back and down a bit, so all you have to do is move down and over and use it again. Its super effective until you get them at a wall, then you have to reposition yourself entirely. In simple terms, you merely hold down and over and spam square to win. An increase to the range that they get knocked back would stop you from quickly chaining them together

New version's up. Use the same link as before. This should fix the bug.

You don't need to start a new game -- just apply the patch to a clean LoM iso, and make sure you boot the game normally and load from the save menu instead of loading from a save state.

To note:

Bud's ridiculously overpowered in the save I'm using. I don't think I did anything that would be causing this but I... don't remember him being able to solo enemies before. Uh... someone keep an eye on this sort of thing.

--- edit ---

Okay, I was testing this with gloves, which /don't/ cause you to lunge forward that much when doing a d,df attack.

Boosting enemy knockback causes weirdness. I want to reduce the amount you lunge forward when doing a d,df attack I think.

Ooooooh.

Okay this is an actual bug. It turns out the function that modifies the enemy's position when they move is the same function that modifies the player's position when they do a plunge attack or jump.

So what you're seeing is not the player tech movement speed being the same. What you're seeing is the player tech movement speed being doubled.

This should be fairly easy to fix.

Cool beans. I'll nab this new one and test the changes, as well as see if Buds OP on my end as well.

Edit: Ill test Bud once um...you know.......I can actually defeat them....

Edit edit: Yea ok, I think the reason Bud is so strong is because of his Frying pan. You did see it has 31 power right? That things a Lorant Silver 2H man. Also, I REALLY have no use for any attack other than this plunge attack. Its kind of all I can do and still get through fights quickly while living lol

Is the movement stuff working better, though? From my own testing I wasn't sliding forward nearly as far when doing d,df attacks.

Also the ST bar stuff is going to take awhile.
- Writing code to decrease the ST bar wouldn't be hard, as there are existing gameshark codes affecting it so I could get the memory location from that.
- Again, the difficulty is finding out what actually triggers the attack to occur.

My guess on how the game, roughly, handles attacking is that it polls for input, when input occurs, some sort of input event is generated, and then handled. This results in an attack, ST, or tech being launched, which gets sent to the collision handler, which propagates further events which would increase the ST bar and do enemy damage, if the attack was successful. This is kind of hard to trace.

I'm thinking I want to start looking into when the ST bar decreases when you do an ST attack. It's possible that the same group of functions trigger the ST attack AND zero out the ST bar, and it might be easier to trace these back to what actually triggers the player's actions. Finding the attack functions from the damage calculation functions hasn't been successful 'cause they're too far removed. The thing that triggers the ST attack and nulls out the ST bar might be a lot closer, though.

Other thing I could try to do for quick attacks is put some sort of waitstun after /all/ of them, not just the standing normal one, but then that's getting closer to the way recharge times worked in SD3, and I didn't like that.

If I get the ST bar thing working, I don't know if I'm going to stop quick attacks from /working/, or if I'll just zero out the damage they do

Yea, you don't slide forward as far now, so the attack isnt AS effective, but its still the most effective thing in the arsenal, and its still spamable and safe to use. The things your saying about the quick attacks, are you meaning that youre going to make them useless? What exactly do you mean by applying waitstun to all of them? Is that the same as when you press X too many times than you can swing, so your character goes into a stasis animation that leaves you vulnerable?

1. The issue you're describing with plunge attacks was present in the base game. I think the original intent was that plunge attacks would be slower and leave you vulnerable, and that quick attacks would be weaker but not leave you vulnerable. This doesn't really work out, since you don't stun when doing a plunge attack unless it's the standing, normal one. So quick attacks would have been useless in the base game anyway, if it weren't for the broken hitstun mechanics.

2. The current plan is to have plunge attacks drain the ST bar, and quick attacks refill the ST bar. When the ST bar is empty, plunge attacks will either do little damage, or will not work at all -- this depends on what I can figure out and what I'm able to change. *This could be fairly difficult to implement.*

3. If this slows down combat too much or makes it tedious, I'll take it out. SD3 introduced length waits after attacking for some characters -- I didn't like that.

4. Alternatively, I'll have the ST bar fill up slowly on its own over time, up to about the half-way point, so that the player isn't forced into a boring attack/wait loop. *This would be fairly easy to implement.*

5. I haven't addressed this at all because it was out of scope for version 1.

6. As noted, what you're describing doesn't apply equally to all weapons. You slide forward a lot when you use a knife. Each of the weapons impacts player movement's in some way, but this isn't really well understood, so I don't know how much I could do about it.

7. I had considered applying some damage reduction to the weapons with excessive range -- spears, bows, and hammers, although knives sound like they're a culprit here too -- but I couldn't find a way to do this.

8. I'll buff enemy HP if I can make the player/enemy hitboxes bigger. Otherwise I'll probably leave the HP as it is.

So, tentative version 2 scope statement:

- Plunge attacks drain ST, and are less effective when ST is at zero.
- Regular attacks gain ST.
- Hitboxes are more lenient when attacking.
* May have ST regenerate up to 50% over time.
* May buff enemy HP.

Ok, I think I was a bit confused there, but that all sounds great

Minor, but promising, progress.

Found the code that adds to the ST bar. If you currently do a quick attack, 10 ST points get added. If you do a plunge, 20 get added.

This behavior would be easy to implement --

1. Jump the current code for adding ST points.
2. If a1 != 80105ae0 (start of player stats), skip new code and do nothing.
3. Otherwise, if V0 (ST to add) == 20:
..3a. If V1 (new ST) >= 33, set V1 to 0.
..3b. Otherwise, subtract 33 from V1.
4. If V1 == 10, then add 1A to V1.

With this, five plunges will drain the ST bar, and eight quick attacks will recharge it. The numbers can be fiddled with as necessary.

This still doesn't actually nerf plunges, but it puts the framework in place. I'll put up a new patch once this is in.

--- edit ---

Actually I think I might drop the quick attack ST gain rate after you have more than half of the bar filled. I have no idea if the increased gain rate will make actual STs spammy or not. Quick attacks put the player at a pretty big risk so that might balance it out.

I might also let the player move faster if the ST bar is full.

Yes, yes, I'm liking what I hear. This would be good. Make plunge attacks reduce ST instead of gain, make quick attacks gain more than before, until half, and you move faster if full, which you can use outside of boss fights to give yourself a decent buff until things get hairy. These should all make the game much better

I think I found the memory locations that store the attacker and target, when doing damage calculations.

Target: 80123b2c
Attacker: 80123b28

80123b30 refers to the player in this context.

Anyway, if this works, all the ST bar stuff I was talking about should be good to go. Currently targeting Saturday as the next release -- I want to have the plunges draining ST and plunges not doing damage when ST is empty in, at least.

This also means it's possible to play around with special rules impacting only NPCs/only impacting the player, to make the AI companions more balanced (although that's not in scope at all right now).

-- edit --

Also, there's a JAL at 8007702C that /appears/ to call the collision detection function for combat. It gets triggered every time the enemy attacks, whether the blow connects or not, and nop'ing it causes hits to not register at all, while still having the enemies do the attack.

Think I can use this to resize the hitboxes, or possibly trace back from it further to find whatever in the enemy AI routine triggers attacks.

The downside is that the function that gets called is huge, so it might take awhile to break down fully and find the actual relevant code sections. Still, progress.

Collision detection for attacks and collision detection with map objects appears to be handled totally separately. As does collision detection with regards to the player walking into enemies.

--- edit 2 ---

For balance reasons, I'm considering the following:

1. For bows, reverse how ST logic works -- quick attacks drain it, and plunge attacks refill it.

2. I might have ST attacks refill some life bar and remove status effects from the player.

Merged in the code for the modified attack logic and started debugging it.

Current problem is this: When you do an ST attack, it zeroes out the ST bar before damage is done. ST is otherwise gained /after/ the attack damage is calculated.

What this means is that my modifications result in ST attacks not doing damage.

I'm still trying to find a way to fix this. Guessing that all pointers for ST attacks are going to be in a certain range of memory. I just need to /find/ that.

https://dl.dropboxus...Patch%202.0.zip

Beta-ish version 2.0.

What's changed:

1. Quick attack damage is scaled based on how far you get in the quick attack combo.
L1 = 1/2 damage.
L2 = 3/4 damage.
L3 = 2x damage.
L4 = 2.5x damage.
L5 = 3x damage.
L6 = 4x damage.

Ideally this makes spamming level 1 attacks less powerful, tones down bows (since they only ever do L1 damage), but shouldn't reduce damage TOO much. One concern I have is that nunchaku are the only weapons, afaik, that go up to L6, and they're quick enough that you might actually be able to get L6 consistently. I don't know.

2. The uppercut, thrust, and flippy jump thing plunge attacks have a kind of cooldown mechanic built in. The first time you use one, it'll do damage but also set a flag such that using the same attack again won't result in any damage being done -- although the player will still move when doing the attack and launch the enemy they hit backwards.

Doing at least a level three quick attack combo OR doing the normal plunge attack will reset this flag.

Also the flag only sets on hitting, so there's no detriment to missing (besides moving yourself right next to the enemy and probably getting attacked).

3. Damage for the normal, uppercut, thrust, and flippyhop plunges was increased because of the new limitations on them. And... well, normal plunges never have their damage turned off, but they leave you super open because they stun you after you do them, so I wanted to offset that.

4. I didn't change the damage for the tech combo plunges because they're not really effective to spam, since they either don't cause knockback, leave you open if you try to spam them, or already induce dizziness if you spam them.

I haven't tested it a lot but this APPEARS to all be working. Source is included in the zip.

Future plans:

- I'm considering an alternate mechanic where plunge attack damage is scaled to whichever level you got to in the quick attack combo. Something like:

1. Using plunges by themselves results in zero damage.
2. Doing a plunge after a level 1 quick attack does 1/4 damage, 1/2 damage after level 2, 1x damage after level 3, etc.
3. Moving or doing a plunge attack zeroes out the multiplier again.

This is similar to what I'm doing but it's finer grained. I don't know if this would make much of a practical difference.

Ideally I THINK that plunges were supposed to be used mostly as combo finishers anyway, and both approaches should accomplish that. The biggest difference is that with what I'm currently, you can save up the plunge attack and do it later. I think forcing you to use it in a combo would result in a harder game but might not be /fair/ feeling, since enemies don't hitstun and can move out of range before you set off the plunge.

- I'd still LIKE to fix the values for which personalities are conveyed by the foods you can feed your monsters but I'm having a hard time getting personalities to get set in game anyway, so tracing this with a debugger is not easy.

Thank you for all of your work, Elegnaim.
I'm just popping in to remind you that there are people who are following your progress and really appreciate everything you are doing. Some of us just don't reply much.

I'll be honest and say that personally I'd rather not try this out until its complete. Which I kinda feel is selfish, but I've been avoiding playing LoM for years even though it is my favorite game. The next time I play it, I want to really go all in and be immersed and finish the game. I'd rather not be beta testing it, and then having to start over once even better improvements have been implemented.

However, if you really need people testing I suppose I could repeat doing the first level or so just to try out the mechanics. But only if you really need more testers.


I wish it was easier to mod. I'd love to make new quests and characters and such, but learning how to get through all of its difficult code for me is a deal closer. I might even say that from my perspective it might be easier to remake the entire game on an easier engine than to try to make sense out of the ps1 files.

I'm following you as well. It's just that I'm a lurker. xD

Seriously though, this is one of the games which I always wanted to see modded.

Kudos for your work!

Quite a few people are following this with interest, myself included.

I wasn't able to get into the original, which is a shame, but this is sizing up nicely.

Keep up the good work, man!

Yeah totally. I think I'll see if I can find my copy of it to make an ISO and probably get started on it as soon as I get more of my gaming backlog cleared.