[zombero]

I finished a rough draft of enemy deployments for Pogrom Forest. This stage will largely be the opposite of Lake Jannenia, featuring the highest deployment count yet seen in this hack along with the greatest rush distance to the enemy base. It will also, of course, feature undead for the first time! As such, undead balance is sure to be a focal design point for this stage.

Another thing I'm looking at is more ways to optimize and/or customize enemy map movement. Potential new options for enemy behavior would be: waiting on a town for a certain period of time, chasing player units that get within a certain range of them, chasing player units after moving a certain amount of distance down a specified path, and considering additional situations where it might be good to refresh an enemy unit's destination choice. Most of these additions cannot be objectively considered optimizations, but as functionality that can be toggled, it could provide additional options for controlling enemy behavior, creating more variety, and different squads different personalities.

[zombero]

Okay! With enemy movement AI out of the way, I've been busy on other (much easier) things. So I thought I'd give an update on what I've worked on so far:

-Made boss minions immune to the Charm item.
-Adjusted Ali requirements on dragons.
-Fixed Archer's promotion requirement.
-A unit's death no longer resets their XP to 0.

I'm currently working on:
-Tweaking bosses to make them harder to kill while reducing their offense a bit.
-Changing player Best tactic to use Weak tactic to break to ties. Basically, if the enemy has 3 Fighters, and a Fighter is the Best target, Best tactic would keep hitting the lowest HP Fighter until its dead, instead of hitting random Fighters.

[Fidelity]

Maybe Best should target the highest HP character. I've noticed that Best tactics has as much trouble with picking off the weak characters, and it might be too much to have two out of four tactics that do that. Not sure.

[zombero]

Do you mean for enemies? Enemies don't use the same tactics as players anymore.

[Fidelity]

Oh, that's right. How exactly do the enemies' tactics work now? Are they all separate down to the unit, like Mage Leader @ Position 03, Best Tactics, Hellhound @ 06, Strong Tactics, etc.?

[zombero]

Nah, although I suppose I could do that if I wanted to. I'm probably going to keep how exactly enemy tactics work a secret. They have ones that are like Strong and Leader, but not quite the same. And they have one that is like a combination of Best and Weak. They also have Random and a tactic that is kind of a mix of everything. Currently pretty much all enemies are using the tactic that is like Strong, which is actually something I'm going to need to evaluate.

[Fidelity]

Quote

Currently pretty much all enemies are using the tactic that is like Strong, which is actually something I'm going to need to evaluate.

Yeah, if they all have to go easy on us (even all the melee), they're really hitting too hard, and this is coming from me, who doesn't want the game any easier. I already mentioned this, but you really might consider adding high PHYS items, and by high, I mean +20-60 PHYS. It sounds absurd, but run some numbers.

But also, you might be able to use harsher tactics on the later units, the ones who benefit less from the Morale Factor after we're liberated more stages.

[zombero]

I've ran math on defensive EQ before. It's strong lategame, but not so much early game. It would possibly be a good idea to create a Defense stat that counts as both Str and Int for the purposes of damage mitigation only. Wouldn't be too hard to do, I'd just need to convince myself that I want it.

Welp, today I've been working on everyone's least favorite kind of changes: behind the scenes. I improved the process of loading enemy deployments and move routes from the tables so that it'll be easier for me to extend them. I also refined the calculation of enemy kill contributions to stage clear bonus. The number of kills that will contribute to the clear bonus is now equal to the number of enemies in the stage. So the best way to ensure maximizing your kill bonus is to kill every enemy deployment, but you'll also get credit (up to the cap) for killing reinforced enemy units, so destroying every deployment wouldn't be necessary in that case.

[zombero]

Welp, I have little to no coding that needs to be done before I start on stage 5's deployments, just several balance decisions to make.

[zombero]

So I noticed that I was unable to reconcile some of the accuracy and evasion rates I was seeing, and I decided to spend a couple hours stepping through the code... turns out, players receive a + 10 to all accuracy and evasion rates. This effectively translates into a 6.25% increased chance to hit and evade for the player. I got rid of this bonus (hardtype, yo!) but I also raised the base hit rate since the original hit rate was adjusted according to how often the player hit. So basically enemies will miss you a fair bit less often now, but player hit rate should be about the same.

[Fidelity]

It never really felt like that, but I guess it was to compensate for players usually being at lower levels. I guess when we're at higher levels, the enemy does seem to miss us more than we miss them..

[zombero]

I'm currently looking at ways to improve the situation with enemy tactics. I did a lot of analysis on small vs. large characters, on how damage scales into the late game, and on what roles the different types of damage play throughout the game. Based on this analysis, I came to the conclusion that rather than having all early game enemies use their current variant of Strong targeting, it would be more to the point to have their ranged attacks favor the back row. What I'm envisioning as a new early-game tactic for enemies is this: melee will work the same as it does now, but enemy ranged attacks will select 3 targets, choosing the rear-most targets first, and then (if necessary) use their current Strong tactic to pull in additional targets from the front row until the total the equals 3. Among those targets, it could use a mix of Strong and Random to decide who to hit. This would help spread the damage more evenly amongst the front and rear lines and keep the mortality of individual characters closer together. Enemies could still use other tactics at other times, though.

I'm also considering moving away from the single-tactic system for enemies, though. There could be several tactical settings that could be set individually. Examples of these settings could include:
-How prone the enemy is to using Random or even Best in place of Strong in situations where enemies use Strong tactics.
-Whether Int-based attacks will target using Best or prioritize the backrow or possibly use another setting.
-How much (or little) undead targets will be prioritized by attacks that can hurt them.
-Who Pumpkin will hit (highest HP, highest HP%, lowest agility, large units, or at random)
-Who Petrify will hit (same as above)
-Who Charm will hit (same as above, but lowest Int instead of lowest Agility)
-Who Kiss will hit
-Whether Acid or other defense penetrating attacks will go for high defense targets or low defense targets
-Who to target with AoE attacks that choose an element based on their target.

Only the first 3 of these concern me as far as Stage 5 goes, but it would help to have a plan on if and how I'd want to implement all these tactics before I start working on any of them.