Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.

Commander Keene posted:

Good to see this back. Also, I was just thinking that this game didn't do as much to make its dragons seem like the alien monstrosities like in the 2020 LP, and then you go and fight the Rainbouroboros. :allears:

Still not caught up though, I haven't read the most recent update, though.

A lot of that has to do with this being the first game and the later games going in a completely different direction. That being said, the batshit stuff you see in the later games does have its roots in this game given stuff like the freefall fight, killing a goddamn mountain, and a rainbow. But the Imperial in this game aren't really like the later games Imperials in that they're really just bosses sitting around until you kill them. Which is really more due to the more traditional fantasy JRPG structure of the game, and how mid-game is just very open ended.

Granted in this game, the dragon apocalypse finished a long time ago and you're basically just rebuilding a ruined world from there, and picking up the pieces. Where in some of the later games, you're in the middle of a dragon apocalypse and dealing with that mess as it happens. (And then there's VFD, which is frankly a very weird case.)

Adbot
ADBOT LOVES YOU

Omobono
Feb 19, 2013

That's it! No more hiding in tomato crates! It's time to show that idiota Germany how a real nation fights!

For pasta~! CHARGE!

In VFD, the team time travels to various dragon apocalypses to hunt down imperial dragons for fun and profit. They also have a dragon apocalypse they have to deal with in their own time.

VFD classes start at what would be endgame level here or in 2020 and go up from there.

Kinu Nishimura
Apr 24, 2008

SICK LOOT!
...

...

Maybe this one finally broke her.

...

I mean... it is a rainbow.

...

Ah, I see. Okay. It's just an odd form of wyrm.

Huh?

It's a rainbow, Koron.

No, you see, it has a flat, paper-like body—like a tapeworm, you see—using the undulations of its body to propel itself through the air, as it obviously appears to be less dense than air. This is also how it's partially translucent—actually, a simple trick to keep up an optical illusion, do you see the bit where the rainbow curves in to meet itself? I'd wager anything its thought center and whatever necessary internal organs it possesses are behind that point, being blocked from sight. This would fall under the theoretical designation of a ouroboros, a being whose body meets itself, forming an infinite loop.

It's not a dragon, but it makes sense within the larger draconid taxonomy.

...Well, poo poo. I can't joke about that.

Get better jokes.

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.
I should probably note if you play VFD first, you can end up with a hilariously wrong idea of how the series works. (Which gets swiftly corrected by the other games in the beginning moments.) Which works mostly fine within the context of VFD itself, but you can end up with some rather big culture shock if you move onto the other games. Or end up confusing the poo poo out of a friend who has only played the first three games as you recount what goes on in VFD to them, which is what I ended up doing. :v:

Oh yeah there's an LP of 2020 going on. You can probably redirect most 2020 focused discussion to that thread if needed.

Araxxor fucked around with this message at 00:23 on Jul 1, 2021

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.
Ark gets a change in career.

Oh and we start an Imperial Dragon extermination for some jerk.

Araxxor fucked around with this message at 06:42 on Apr 24, 2022

McFrugal
Oct 11, 2003
Dagger strong.

Chuu
Sep 11, 2004

Grimey Drawer
> Oh yeah, if you’re using a Dagger Rogue (or your strategy is reliant on ailments in general), you’ll really want to make sure to clear as much of the Bloom around a dungeon as much as possible to increase the ailment multipliers of the dragons and bosses in there. A 130% multiplier increases odds quite a bit, and dealing with 65% ailment multipliers instead of 50% is a pretty big boon. More so if you have ways of increasing the chances of ailments landing, such as the Axe Fighter’s Status Down skill.

I completely missed this somehow. How does the bloom effect ailment multipliers and/or resist?

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.

Chuu posted:

> Oh yeah, if you’re using a Dagger Rogue (or your strategy is reliant on ailments in general), you’ll really want to make sure to clear as much of the Bloom around a dungeon as much as possible to increase the ailment multipliers of the dragons and bosses in there. A 130% multiplier increases odds quite a bit, and dealing with 65% ailment multipliers instead of 50% is a pretty big boon. More so if you have ways of increasing the chances of ailments landing, such as the Axe Fighter’s Status Down skill.

I completely missed this somehow. How does the bloom effect ailment multipliers and/or resist?

I made a post before about this, though the details were rather vague.

Araxxor posted:

Yeah it's fun to see what was thrown at the wall and what did and didn't stick in such an era. DS era EO was also gloriously messy until the 3DS era started straightening things out.

And speaking of weird mechanics, turns out there was yet another surprise mechanic I discovered while working on the next update. Or more specifically, I discovered how exactly a previous surprise mechanic worked.

A lot of this is flat out wrong. What actually happens is that there's a 0.95x to 1.3x multiplier applied to the dragons' damage and ailment multipliers, in increments of 0.05. (So 0.95, 1, 1.05, 1,1, etc.) What this depends on is how much of the overworld Bloom is gone in an area relative to the dungeon the dragon resides in. Basically, it's the shop price mechanic, only applied in such a way it makes killing dragons easier if you engage in it. The more of the Bloom around the dungeon that's gone, the easier it is to kill all the dragons in it. Bloom in the dungeons doesn't matter for this. Though that being said, the dungeon does check a really large portion of the continent around it, and sometimes the ocean as well, so you can't always get that 30% damage and infliction rate boost even if you do your best to destroy the Bloom, especially in the earlygame where your transportation methods are limited.

The short version is, lawnmow the continents and the ocean, and the dragons get easier to kill.



It's a good idea to clean up the Bloom too.

Turns out this was indeed hinted at in the game. I just didn't pick up on the hints until I realized how the mechanics worked. :shepface:

This information is incomplete and I snagged the actual details on this mechanic.

Upon loading an area with dragons in it, the game will check part of the overworld and measures how "Bloom infested" it is. Upon measuring, it will then load up a multiplier to apply to any dragons you fight. This multiplier will be applied to their damage and ailment multipliers. (Non-elemental damage is affected as well.)

More than 64%: Not supposed to happen. If it does, the game will multiply all of a dragon's multipliers by 0. Their other stats (LIFE, ATK, etc.) will glitch out and be loaded improperly.
64% or less: 0.9x
48% or less: 0.95x
36% or less: 1x
24% or less: 1.05x
12% or less: 1.1x
0%: 1.3x

The first case is fortunately impossible to see without hacking.

Gilgamesh255
Aug 15, 2015

Kinu Nishimura posted:

...

...

Maybe this one finally broke her.

...

I mean... it is a rainbow.

...

Ah, I see. Okay. It's just an odd form of wyrm.

Huh?

It's a rainbow, Koron.

No, you see, it has a flat, paper-like body—like a tapeworm, you see—using the undulations of its body to propel itself through the air, as it obviously appears to be less dense than air. This is also how it's partially translucent—actually, a simple trick to keep up an optical illusion, do you see the bit where the rainbow curves in to meet itself? I'd wager anything its thought center and whatever necessary internal organs it possesses are behind that point, being blocked from sight. This would fall under the theoretical designation of a ouroboros, a being whose body meets itself, forming an infinite loop.

It's not a dragon, but it makes sense within the larger draconid taxonomy.

...Well, poo poo. I can't joke about that.

Get better jokes.

Meanwhile, in a more Monster Hunter themed universe:

"What...what the hell am looking at? Is that a freaking sentient RAINBOW?! This...this is beyond our understanding--"
"Lol, just slap the Elder Dragon tag on it, murder the gently caress out of it, study it's corpse, then make pretty clothes out of it's corpse bro!"
"Oh poo poo, yeah, that is how we generally handle weird poo poo like that, lol"

theshim
May 1, 2012

You think you can defeat ME, Ephraimcopter?!?

You couldn't even beat Assassincopter!!!
so what you're saying is

4 dagger rogue party

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.

theshim posted:

so what you're saying is

4 dagger rogue party

That would blow up pretty much everything in the game, yes. That being said that party would have absolutely no defense and have tougher times against fights that punishes glass cannons.

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.
Things die. Lots of things die. Said things are dragons. Also Ark is a loving monster, goddamn!

Araxxor fucked around with this message at 06:42 on Apr 24, 2022

Blaze Dragon
Aug 28, 2013
LOWTAX'S SPINE FUND

https://lpix.org/4066835/DragonB1FL.gif isn't embedded, it's just a link.

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.
Fixed.

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.

Araxxor posted:

Addendum to the soft cap mechanic!

Yeah, so I was wrong in my previous post. I found out how it actually worked when I looked into things more. And it's not so much a soft cap so much as the game switching over to a different damage formula in certain circumstances.

Damage = [(User's INT / 2) * Spell Power + (Base Damage * Mastery Power) – (Target's DEF / 2) – (Target's INT / 2)] * Damage Multipliers * Concentrate Factor * EX Factor

It's a little more complex than the more straightforward physical damage formula, but everything works as you would expect. Concentrate and EX stack fully here.

However, if at any point the result from this formula is equal to or greater than INT * 10, the game swaps over to this formula instead:

Damage = (User's INT * 8) + [(Spell Power * (INT / 10)) + (Base Damage * Mastery Power / 5) – (Target's DEF / 10) – (Target's INT / 10) * Damage Multipliers * Concentrate Factor * EX Factor]

I overcomplicated this. As it turns out the check is pretty simple.

Softcap Mechanics:

If Damage <= INT * 10, take no further action.
Else, Damage Difference = (Damage - INT * 10) * 0.2
Softcapped damage = INT * 10 + Damage Difference


Also discovered an... interesting design decision. When taking damage from INT based attacks, the target's INT is capped at 100 if it's above that. No such thing happens for DEF. This basically means Mages could not care about the target's INT past a certain point (cause it does start getting ridiculously sky high later) and remain very consistent damage dealers throughout the game.

Araxxor fucked around with this message at 07:03 on Jul 15, 2021

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.
We deal with Flame Eater once and for all. And of course, there's a party vote in this one for the next mission.

Araxxor fucked around with this message at 06:42 on Apr 24, 2022

BisbyWorl
Jan 12, 2019

Knowledge is pain plus observation.


Let's go with Dia, Nila, and Buront.

Jade Rider
May 11, 2007

All the pages have been censored except for "heck," and she misread that one.


Haru, Dia, and Nila.

Chuu
Sep 11, 2004

Grimey Drawer
Dia, 2x LPer's Choice

NyoroEevee
May 21, 2020
Dia, Nila, and Ark.

It's scary to see someone actually get on the bad side of the perpetual ray of eternal sunshine that is Stella; to wit, not even Dragons trip on that nerve as hard as the innkeeper has...

I bet that Stella's going to plan an amazing welcoming party for Arietta though~

Eeepies
May 29, 2013

Bocchi-chan's... dead.
We'll have to find a new guitarist.
The three lowest level characters

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.

Eeepies posted:

The three lowest level characters



That's how things are shaping up at the moment.

Though uh, the entire guild is admittedly, very very overleveled for the next dungeon. The suggested order had it placed as the second Imperial Dragon you dealt with in mid-game. And the random encounters are level 46, while the dragons are levels 42 and 43, with Dreadnought being level 48. Not helped by the Tower of the Gods having a whopping 54 dragons to kill, so that'll be even more exp (despite the scaling nerfing the payouts for overleveled parties.)

I guess I could see how the guild fares in the Tower of the Gods, and toss some Oblivion Pearls their way if they do too well.

quote:

Targets' INT stats are capped to 100 maximum when being attacked with magic.

Addendum to this statement. This only applies to Mages attacking enemies. If an enemy tries to attack a 200 INT Mage with magic, all of that 200 INT will work to reduce the damage, instead of being hard capped to 100 like it is for enemies! A lot of things are really tipped in the Mage's favor! :shepface:

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.
Votes so far:

Dia: 5
Haru: 3
Nila: 3
Ark: 2
Buront: 2
Edward: 1
Stella: 0

Lowest Leveled Characters: 3
My choice: 2

NyoroEevee posted:

It's scary to see someone actually get on the bad side of the perpetual ray of eternal sunshine that is Stella; to wit, not even Dragons trip on that nerve as hard as the innkeeper has...

I bet that Stella's going to plan an amazing welcoming party for Arietta though~

The game provided an opportunity for her to go off, given how it shows your lead character actively intervening against Jen in the actual game. And I figured it was a good time as any to actually show her pissed off. :v:

The dragons are more of an expected obstacle on the journey to save the world. And only the Imperials can really talk anyways. Even then they don't really say much before trying to kill you.

Rawkking
Sep 4, 2011
Edward, Haru, Buront

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.
Voting is closed! Final results:

Dia: 5
Haru: 4
Nila: 3
Buront: 3

Ark: 2
Edward: 2
Stella: 0

Lowest Leveled Characters: 3
My choice: 2



Funnily enough, the entire top 4 are actually the lowest leveled party members in the guild, so that secures their place.

I will try to get an update out before the 26th, as there is a lot releasing on that day. However I make no guarantees, and if I don't get a non-gameplay update out before then, uh, don't expect one for a while.

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.
Game Mechanics: Turn Order

During a battle, the game has to decide what order the units in the battle will attack, with faster units being more likely to get the first strike in. How the game calculates this is pretty simple, so let’s just get right into it.

Turn Speed

Turn speed is the statistic the game uses to determine who gets to move first. The game calculates it in this way:

Turn Speed = SPD * 20 * Skill Speed Factor * Mastery Speed Boost * Random Factor

The units with the highest turn speed gets to move first, and the game moves things along like that to the next highest turn speed until it reaches the unit with the lowest turn speed.

This formula does result in some rather comically large numbers. A unit with 80 SPD and using a skill with a speed factor of 100 will end up with a possible turn speed from 144,000 to 176,000. Oh well, whatever works. As for what these factors actually mean...

SPD: The user’s SPD stat. This is directly affected by SPD buffs and debuffs. The only SPD buff players have access to is the Samurai’s Iai Arts, which only affects the user, and they have no SPD debuffs they can inflict. Enemies on the other hand, have quite a few ways to buff their own SPD, or lower the players’ SPD stats.

Skill Speed Factor: The Speed Factor from the skill the user is casting. A Speed Factor of 100 is generally on the high end of things, with very few skills going above that value. Most player skills tend to have Speed Factors of 80 or 100, averaging out at around 92. And most enemy skills tend to have Speed Factors of 80 and 90, averaging out around 88.

With those statistics in mind, 80 is generally the average Speed Factor, with skills that go below it being rather slow, and skills that go above it being pretty fast.

Using a regular attack, swapping rows, using items, and escaping all have a Speed Factor of 100. Guarding technically has a factor of 100 as well, but more on that a bit later.

Mastery Speed Boost: Some mastery skills come with a speed boost that increases the casting speeds of skills tied to that mastery. Enemies don’t have such a boost, so this is always set to 1 for them. This can make learning masteries give quite the speed advantage, especially since there’s only 1 buff the player can use to boost their turn speed.

Random Factor: A random number between 0.9 and 1.1 inclusive, in increments of 0.01. So you can get multipliers of 0.93 or 1.07, but no nonsense such as 1.034243 or anything like that. This pretty much makes it so that units with close speed stats and speed factors aren’t guaranteed to have one unit outspeed the other in all cases. You’d have to have a big enough speed advantage to be guaranteed to outspeed something.

If there happens to be a tie in turn speeds, then the game just picks the units with the same turn speeds to act in a random order.

That being said turn speed isn’t the only factor that determines how fast a unit acts. Some actions have priority and allow a unit to act sooner, regardless of turn speed.

Priority

Priority is the other part of turn order calculations, and can let units with bad turn speeds act first if needed. The mechanics of priority are as follows:

If any unit is guarding, the guard will instantly take effect at the start of the turn. The actual message saying that the unit is guarding has a speed factor of 100, but it doesn’t matter as the actual effect has absolute priority and cannot be stopped or disrupted.

Then the game checks for setup skills. Setup skills are things like defensive skills like Shield Front, counterattacks, or even the Fighter's Link skills. The priority and speed of any active setup skills are calculated as well. Though no setup skill has innate priority, but can be given priority through the use of EX.

Priority brackets are checked for next. There are 4 priority brackets checked for, with units in earlier ones getting to act sooner than units in later brackets. Anyone within each priority bracket will have their turn speeds checked for. Then the game checks to see if anyone within that priority bracket got Stunned, and will have them act last. Then the game goes onto the next priority bracket to repeat the process, until all the priority brackets have been checked for.

Then all of those steps are repeated, this time with the units that didn’t have any setup skills active. If you had a little trouble keeping track of that, here’s a little chart that illustrates how the game goes about checking the order of actions.

Active Setup Skills
-Priority Brackets
--Turn Speed
---Stunned Units
No Setup Skills
-Priority Brackets
--Turn Speed
---Stunned Units

Now as for the priority brackets themselves...

1: Priority
2: EX Active
3: Default Speed
4: Goes Last

No enemy in the entire game has access to priority brackets 1 and 2, as they don’t have any skills or mechanics that allow them to be placed there. So that’s a pretty big advantage player units have over enemy units. That being said, enemies do have access to setup skills, so they can outspeed player units using priority in that regard. Something to be very mindful of if you’re up against an enemy that can use counterattacks.

Priority: These skills will act first. They can only be outsped by setup skills, so you can’t bypass counterattacks or other defensive skills with these. Only Rogues, Samurai, and Princesses have access to skills with priority. No enemy in entire the game has access to a skill that has access to this priority bracket.

EX Active: Using EX will give skills priority, though not as much as skills that have innate priority. The effects of EX does not stack with priority brackets 1 and 4, and in fact, are overridden by those. Only skills in the priority bracket of 3 can be placed in this priority bracket of 2. Enemies obviously don’t have access to EX, therefore they don’t have access to this bracket.

Default Speed: Basically what most skills will be placed in. Most enemy skills are stuck in this bracket as well. You can raise the priority level of these skills by using EX.

Goes Last: Pretty self explanatory, but it’s not really a priority bracket that sees a lot of use. Only 2 skills in the entire game are placed in this bracket (Out of a whopping 670, mind you.) Said skills are the failed attack from the Rogue’s Sonic Shot if the initial setup fails to counter anything, and one skill belonging to an enemy.

Now as for the remaining mechanics that can interact with turn order...

Hustle is a skill that has innate priority and forces the party member the Rogue targets with it to immediately act after they cast this skill, no matter what priority bracket they are placed in. This is the only skill that has the ability to bypass priority brackets entirely, essentially dragging any skill in the game into the priority bracket of 1, and can let a party member who doesn’t have EX active that turn to outspeed a party member who does.

Stun is a status effect that doesn’t deny a unit their turn like in Etrian Odyssey, but instead forces them to act last within a given priority bracket. Stunned units are not shifted down a priority bracket, as Stunned units in the priority bracket of 3 will still act before any non-Stunned unit in the priority bracket of 4.

Stunned units don’t even have a turn speed rating. Whatever turn speed was calculated for them before is completely discarded in this step. And no, that doesn’t mean turn speeds are set to 0 for them. A non-Stunned unit with a turn speed of 0 will still act before a Stunned unit. Instead after all non-Stunned units have acted, the order that Stunned units act are prioritized from left to right. Stunned party members are also prioritized first and will always act before Stunned enemies.

Araxxor fucked around with this message at 21:18 on Jul 20, 2021

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.
Sorry for not having an update out for a while (on the plus side, I've updated more times this year compared to last year? :confuoot:) NEO TWEWY came out and ate all my free time. I'm done with that now, so back to this LP I go. I don't have an actual update ready yet, so have a quick writeup about some unused stuff in this game while I work on that.

Resources: Unused/Debug Items

7th Dragon has quite a few items that are in the data, but never really got used in the final release. Some of these items appear to be scrapped concepts, while others appeared to have been debug items to test certain mechanics that never were intended to actually see serious use. We’ll be looking through said items in this update.



Items are coded as skills, so most of these effects were dug out by looking through the skill data. However, these were never actually implemented as items, and attempting to hack them in your inventory will just grant you an unusable “Junk” item that comes with no description, and cannot be used at all. The only thing that can be done with said items is selling them. Despite that, I’ll list out the item ID for each given item in case someone actually wants to do something with this information. Said item IDs will be in hexadecimal. (Instead of the numbers 0 through 9 being stored in one digit, it functions as 0 through 15 in one digit, with the numbers 10 through 15 being represented by A through F instead. This allows for a far greater range of numbers to be stored in fewer digits, like 256 numbers can be stored in 2 digits instead of only 100 like in the decimal system.)

Major Mend
Item ID: 002

Restores 200 LIFE to one party member.

First cut item and it’s basically a heavy tier single target healing item. Given that LIFE stats don’t get particularly high in this game, going up to a base of 300 at most with a max level Knight, it’s pretty easy to see why this got scrapped, given that Miracle Medicines essentially do the same thing with their full heal properties and with the item availability in this game. However, as to why this was made in the first place...



Here’s a prerelease screenshot of the game. A few things stick out. One is an impossible enemy formation that does not actually exist in the final game. The main thing I want to bring attention to is the LIFE and MANA totals. Just look at the sheer amount they have! You can’t get anywhere near that much in the final game, so the power curve most likely changed at some point in development, resulting in smaller numbers overall, and thus, no need for this heavy healing item. The difference between 200 LIFE restored and all LIFE restored would have been far bigger in that environment, but it doesn’t exist anymore.

Major Mend All
Item ID: 005

Restores 175 LIFE to all party members.

Medium Mend All
Item ID: 006

Fully restores all party members’ LIFE.

And next on the docket are these cut AOE healing items, which left only the Heal Aerosol as the sole AOE healing item you could use in the game. Who knows why these were cut, but they may have figured incredibly easy AOE healing on demand would be pretty strong. And might have obsoleted the Healer. Which can be a problem in certain games, where healing items can make a medic type party member useless to drag along because they were too strong. That and well, LIFE totals in this game doesn’t get really high compared to the beta version. The middle tier version would be fantastic even at endgame, so the issue of these items being overkill was also likely a factor in their removal.

Also yes, the item names are reversed in Japanese as well. Not really sure what happened there. The names themselves are also rather blunt in their function, so it’s likely they designed the functionality before deciding on how the item names and descriptions would work out flavor wise, which does give a small bit of insight as to the order things were developed in.

Major Mana
Item ID: 008

Restores 40 MANA to one party member.

Medium Mana
Item ID: 009

Fully restores on party member’s MANA.

Mana Water was supposed to have stronger versions as well, however we got stuck with only the lowest tier version. The higher tier versions were likely cut for the same reasons the other higher tier restoration items got cut. Considering that the actual raw MANA pools in this game don’t get particularly high, and that MANA costs in general are pretty low in terms of raw numbers, these items would have likely been overkill in the final game. They were most likely meant for when the party members had much bigger stats, but 40 MANA restoration is just incredibly efficient and powerful, let alone a complete restoration. Combine the fact that healing springs are frequently seen throughout your adventures, and MANA is basically a complete non-issue. The 15 MANA restore (which can be doubled to 30 with EX) is honestly enough for the entire game.

Minor Refresh
Item ID: 00A

Restores 80 LIFE and 20 MANA to one party member.

Medium Refresh
Item ID: 00B

Restores 160 LIFE and 40 MANA to one party member.

Now here’s an interesting bit of info. Silver Water was meant to be the final product of a line of simultaneous LIFE and MANA restoration archetypes, instead of being the only item of its archetype. We can only speculate as to why these got cut. The Medium Refresh likely got put on the chopping block for being too powerful due to the lower numbers in the final game. And the Minor Refresh would have completely obsoleted Mana Waters due to the higher Mana restore and functioning as a substantial heal on top of that.

Minor Revival
Item ID: 00D

Revives one party member at 1 LIFE.

Medium Revival
Item ID: 00F

Revives one party member at full LIFE.

In this case we didn’t get stuck with the lowest tier consumable, but the middle tier. The Hypno Crystals are pretty powerful as is, even in the endgame where 100 LIFE is still a respectable amount of LIFE to be revived at. As for why the Minor Revival was removed, they may have wanted to limit how available revivals were, as you can’t revive on the field early and if someone goes down, that’s a forced backtrack to a clinic since healing springs can’t revive party members. Also again, adds more value to having a Healer on a team since they come with revival skills in their skillset.

Minor Revive All
Item ID: 010

Revives all party members at 100 LIFE.

Med Revive All
Item ID: 011

Revives all party members at full LIFE.

Now here’s an archetype of items that did not have a single item of its kind make it into the game. AOE revivals. Those of you familiar with the second entry of the Etrian Odyssey series (not the first for once) and the severely powerful Nectall items should probably be well aware of why such an item was very likely to be cut.

For those that aren’t, let me put it this way. Imagine having most of your party members taken down, with only 1 or 2 party members hanging on, and then just on demand as long as you had one of these babies, you revive all of them and reverse a disastrous situation instantly. Yeah. Also would have taken away from how powerful the Healer’s Miracle Cure skill is.

EX Gauge: 1
Item ID: 012

Restores 1 segment of the EX Gauge.

Debug version of the Dragon Egg. (The Japanese name for the record, is EX Recovery: 1.) It does come right before the Dragon Egg in the item ID and skill ID list. So maybe the Dragon Egg was originally intended to be a full EX restore or something stronger, which would have been completely absurd if that was intended to be the case. Though we can never say for sure.

Status Cure
Item ID: 016

Dispels Confusion from one party member.

Curse Cure
Item ID: 019

Dispels Curse from one party member.

Fear Cure
Item ID: 01A

Dispels Fear from one party member.

Skillseal Cure
Item ID: 01B

Dispels Skill Seal from one party member.

Status Cure was Confusion Cure in Japanese. Not that a mistranslation here really matters since you can’t use these items.

Most of the other ailment curing items had unused versions in the code. Interestingly enough, Sleep and Charm are the only ailments that specifically do not have an item that cures them. At any rate, perhaps all these got cut down, and they let Somanels take on the role of dispelling the rest of the ailments to reduce clutter.

Resistance Burst
Item ID: 02E

Specific data unknown. Would have increased one party member’s resistance to ailments for 5 turns.

In Japanese, this was known as “Status Resistance Up” if you’re wondering about the rather nondescript name. But anyways, this is pretty much the first of several skills that has undefined skill data, which is why I said that specific data is unknown for this and several of the later items. It flat out doesn’t exist. The only things I have to go on are the skill parameters, and the item name, which is the only indication of how the item functions. Not sure why this item got cut, but there really aren’t that many sources of ailment resistance you can really use in this game, so they may have just wanted to cut that down.

Caustic
Item ID: 035

Specific data unknown. Would have had a chance of inflicting Blind to one enemy.

Total Paralysis
Item ID: 036

Specific data unknown. Would have had a chance of inflicting Paralysis to all enemies.

Total Sleep
Item ID: 037

Specific data unknown. Would have had a chance of inflicting Sleep to all enemies.

Total Poison
Item ID: 038

Specific data unknown. Would have had a chance of inflicting Poison to all enemies.

Total Envenom
Item ID: 039

Inflicts the Venom Boost effect on all enemies.

Total Curse
Item ID: 03A

Specific data unknown. Would have had a chance of inflicting Curse to all enemies.

Total Skill Seal
Item ID: 03B

Specific data unknown. Would have had a chance of inflicting Skill Seal to all enemies.

Caustic was “Blind One” in Japanese. I’m not sure if these were intended to be usable items that could inflict ailments on demand, or if they were just a way for the dev team to test how ailments worked, as their skill data was entirely stripped from the ROM. Certainly would have made Masked Pain builds far stronger if every class had the opportunity to inflict ailments though.

Fruit
Item ID: Unknown

Restores 10 LIFE to one party member.

Yep. That’s the final item and skill in the data. Fruit. Might’ve been an earlier version of the Paro Fruit item, but that’s all I can really say about it. I don’t even know what the proper item ID for this item was supposed to be.

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.

quote:

Sorry for not having an update out for a while (on the plus side, I've updated more times this year compared to last year? :confuoot:) NEO TWEWY came out and ate all my free time. I'm done with that now, so back to this LP I go. I don't have an actual update ready yet, so have a quick writeup about some unused stuff in this game while I work on that.

As it turns out more video games came out and continued to devour my free time. But back to the LP.

We start dealing with Dreadnought's roost.

Araxxor fucked around with this message at 06:41 on Apr 24, 2022

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.
We unlock some very powerful music for the Two-O Duo.

Araxxor fucked around with this message at 06:41 on Apr 24, 2022

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.
Update 37.5: Tower of the Gods Bloom Seed Exploit



You can consider this update to be a part of update 37, though I cut this section out into its own update in case anyone wants easy to access instructions on how to replicate this.

7th Dragon’s RNG for whatever reason, tends to be very static and predictable. You can read the update on the Random Number Generator I wrote to see more details on why it works. Because it’s so static and predictable, we can exploit it to get the results we want. In this case, generating a guaranteed Bloom Seed encounter for easy level ups. For those that read the update, RNG 1 will be the function we’ll be exploiting here. Let’s get started!

Video: Tower of the Gods Bloom Seed RNG Exploit

Here’s a video on how to this, though I’ll write this out in screenshot form as well.



There are a few preparations you’ll have to make. Make sure you have ways to reduce the encounter rate on hand. Either with Bamboo or Ronam Flutes, or the Healer’s Invisible skill. The Samurai’s Evil Eye skill doesn’t work for this. If you’re playing on the USA patch, you don’t need those tools.

Also make sure to save at this save point on the 4th floor of the central tower of the Tower of the Gods. It’s also very much ideal to have killed all the Tyrannosauruses on the floor below, though it’s optional. Doing so just severely reduces the likelyhood of this exploit going wrong.

If the Tower of the Gods is free of Bloom, you can just use a Bloom Seed item to restore the Bloom to allow the use of this exploit.



7th Dragon does not have a soft reset function, but it doesn’t matter. A fresh boot will accomplish the same thing as 7th Dragon’s RNG functions don’t check for the system date and time to get its RNG seed, and the seed is always set to a value of 0 on booting.





Upon loading the game, immediately use an encounter rate reducer. The RNG advances every time the encounter rate reaches 0. Unfortunately, because the RNG seed starts out at 0, the encounter meter is always set to 1500, the lowest possible starting value, and it’s not possible to reach the stairs heading down in time before you get into an encounter. Evil Eye doesn’t work here because the RNG will advance the moment the encounter meter hits 0, even though it can prevent a battle.

If you’re playing on the USA patch, this isn’t needed as the encounter rate is lowered enough to not need a reducing item or skill.

Also do not try to use the staircase leading up to reset the meter, as that advances the RNG a different number of times than the staircase leading down, screwing up this exploit.





The next step is to head on over to this staircase and walk down and up the stairs 8 times each. Or witness 16 screen transitions.

This is the reason I recommended that you kill every last Tyrannosaurus, as their movements advance the RNG. If you immediately head up upon heading down you can avoid this, but it’s just better to have that room for error.





After doing all that, walk down the stairs one last time. Don’t walk back up here.



Then immediately head over to this Bloom tile and trample it.







It will produce a Bloom Seed battle.



Use EX here, or some other form of priority, as it’s very likely the Bloom Seed will run on the first turn.









And presto! An easy kill with lots of experience awarded to our party!



You can then head on back up to save and immediately repeat this process for very fast level ups. Rinse and repeat as needed. Since Bloom Seeds always give out a maximum percentage of the experience needed to reach the next level, you only need around 2 Bloom Seed kills per level to level up.

This results in very easy ways to power level if you want. This isn’t the only known Bloom Seed exploit, but it’s one of the easier ones to pull off. I’ll cover the other ones when we get to those.

Now next time, we’ll finally take on Dreadnought.

Omnicrom
Aug 3, 2007
Snorlax Afficionado


I'm always fascinated by RNG manips like these. I think my favorite one was the person who came up with a semi-consistent way of getting Seitengrats in FF12IZJS on the PS2. I never found out if the PS4 version as a similar method.

Also, holy poo poo Angel Cage.

Omnicrom fucked around with this message at 04:36 on Oct 4, 2021

BisbyWorl
Jan 12, 2019

Knowledge is pain plus observation.


My team ended up being Knight/Healer/Mage/Princess, so end game I just threw up Angel Cage and just shredded anything in my way with Venom Boost/Mana Bullet/Punishment.

Team DPS gets insane once everyone gets their full kit.

Omobono
Feb 19, 2013

That's it! No more hiding in tomato crates! It's time to show that idiota Germany how a real nation fights!

For pasta~! CHARGE!

And here I thought Impure Reach in Persona Q 1 was utterly broken.

The hell heaven were they thinking?

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.
Ahoy there! Time to take down Dreadnought!

Omobono
Feb 19, 2013

That's it! No more hiding in tomato crates! It's time to show that idiota Germany how a real nation fights!

For pasta~! CHARGE!

That old concept art is straight up the Yamato charging up the Wave Motion Cannon, isn't it?

Omnicrom
Aug 3, 2007
Snorlax Afficionado


The Dreadnought as a whole is a Yamato reference, and they were not even bothering to hide it.

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.
No more Imperial Dragons left to kill, so it's time to just sit back, relax, and clean up the rest of the dragon infestation!

HE ARRIVES.

BisbyWorl
Jan 12, 2019

Knowledge is pain plus observation.


I really like the way 7D handles this scene.

In any other apocalyptic game, there would be some catch to the Giant Fuckoff Soul Cannon. Maybe the King is using this as a chance to deal with undesirables and the poor, maybe there's a Secret Other Way the heroes learn five minutes prior that would make it unneeded. It would end up with some climatic battle on top of the thing where the party makes a grand declaration about how they won't sacrifice their humanity to save humanity.

Here? The Giant Fuckoff Soul Cannon is absolutely needed, and the King himself walks in to entrust things to the next generation. And when Emille gets Pissed and wants to sacrifice all of humanity to kill a single True Dragon? She just gets arrested. She's a thinker, not a fighter, so at the end of the day she can't do anything if no one listens to her.

Oublietteer
Jul 30, 2018

Is that you, Edea Lee?
Yay, it’s back!

And it’s getting good! This is a pretty memorable couple of scenes, and there’s more to come. Can’t wait!

(Also, for a literal god, Niara looks dumb as hell :v: )

Adbot
ADBOT LOVES YOU

Araxxor
Oct 20, 2012

My disdain for you all knows no bounds.
Hey it's been a while. I don't have an update ready yet, as I've been busy with quite a few things and it might not calm down anytime soon. However one thing I was busy with is relevant to this game and it's been something I've really been trying to look into for a long while.



I was finally able to give a butterfly a machine gun.

Which is really to say I finally located the drat AI files for enemies and locating these drat things sure wasn't easy!

Unlike in Etrian Odyssey where everything has clearly defined directories and sorts everything into neat little files, 7th Dragon DS... does no such thing.



That's the entirety of the extracted game data, stuffed into a bunch of bigass random files, so finding what you want is like searching for a needle in a haystack unless you know exactly what you're looking for. (The AI files were located in btldata.bvb)

Anyways this isn't going to be a formal update on the subject, as I'm still looking into how it works, but I've pinned down a lot of the important stuff, enough of it to share right now.

7th Dragon's AI files are coded as data tables. Yeah for those of you familiar with Etrian Odyssey 1's inner workings, it's nearly the same format. The exact bits of coding are different but the design principles are almost the same. Each AI has 2 data tables that handles how an enemy should behave.

The first data table defines a bunch of conditions in a list. The moment one of them is true, the game is told to hop to a specific routine in that AI.

The second data table defines the routines, such as what actions the enemy should take, and the chance of them taking that action. Each routine can contain 5 functions, and each function follows this format:

quote:

Action ID, Param 1, Param 2, Target type, ???, Probability to activate function.

Action ID is the type of action the enemy should be doing. Which are as follows:

quote:

0: Do nothing. Display message: "[Enemy] seems addlepated..." Due to how the AI handler reads the data, this is not actually accessible normally, and must be accessed by reading an invalid action ID. (Values of 0 are ignored and trying to make the handler read this Action ID simply causes it to skip to the next possible routine.)
1: Use a basic attack.
2: Use a skill. Param 1 defines skill ID, Param 2 defines skill level.
3: Escape.
4: Guard.
5: Invalid ID. Typically used to access switch case 0.
6: Do nothing. Display no message.
7-9: Summon help? (Doesn't seem to be functional...) Notably this uses the text "[Enemy] shouted for help!" but this text is never used in-game.
10: Call for help. Param 1 defines enemy ID, Param 2 defines number of that enemy to summon
11: Call for help. Can summon 2 different enemies. Param 1 defines enemy ID 1, Param 2 defines enemy ID 2
Any Invalid ID: Default to case 0.

quote:

0: Do nothing. Display message: "[Enemy] seems addlepated..." Due to how the AI handler reads the data, this is not actually accessible normally, and must be accessed by reading an invalid action ID. (Values of 0 are ignored and trying to make the handler read this Action ID simply causes it to skip to the next possible routine.)
5: Invalid ID. Typically used to access switch case 0.

:psyduck: That's certainly a way of getting around that limitation but okay.

If there are no valid functions in the AI that can be followed, then the enemy will simply do nothing and display that message.

And yes, for those of you familiar with EO1's infamous AI bugs, the same thing can theoretically happen here where the wrong action ID defined can cause an enemy to use a basic attack instead of a skill, or worse, skip their turn entirely. I say theoretically, because as far as I know 7th Dragon's enemy programmers were more careful to avoid that sort of behavior happening in this game, and so far I haven't seen a single enemy that suffers from that same bug that EO1 enemies do.

As for examples of enemy AI;

Rabi posted:

Conditions:
-If HP is at or below 50%, jump to routine 1.

Routine 0:
Use a basic attack (Default targeting).
Routine 1:
-60% chance to use a basic attack (Default targeting).
-20% chance to cast Sand Kick (Default targeting).
-20% chance to escape.

Nutshell posted:

Conditions:
-If HP is at or below 70%, jump to routine 1.

Routine 0:
-60% chance to use a basic attack (Default targeting).
-10% chance to guard.
-30% chance to cast Nutball (Default targeting).
Routine 1:
-50% chance to use a basic attack (Default targeting).
-25% chance to guard.
-25% chance to cast Nutball (Default targeting).

Routine 0 is the default routine followed if no conditions defined are true. Though the game can forcibly tell the AI to jump to that routine if necessary, which some enemies do.

Bloom Seed posted:

Routine 0:
-15% chance to use a basic attack (Default targeting).
-85% chance to escape.

Not all enemies have conditions defined, and are rather simplistic. Bloom Seeds for example can rarely do attacks but 85% of the time they'll just run, so just kill them straight away if possible.

As for some of the bigger enemies...

Invisible posted:

Conditions:
-If self cast Chandelle last turn, jump to routine 2.
-If the turn count is divisible by 2, jump to routine 1.

Routine 0:
-50% chance to cast Shock Bolt (Default targeting).
-50% chance to cast Bolt Wing (Default targeting).
Routine 1:
-75% chance to cast Chandelle (Default targeting).
-25% chance to cast Bolt Wing (Default targeting).
Routine 2:
-Cast Wind Cutter (Targets the front row).

Invisible has a rather simple AI, but the fight itself doesn't need much as it has a simple tactic of forcing your rows to swap, and sniping your front row with a hard hitting physical.

Earthshaker itself is a bit more complex:

Earthshaker posted:

Conditions:
-If HP is at or below 10%, jump to routine 1.
-If HP is at or below 35%, jump to routine 4.
-If self cast Soul Press last turn, jump to routine 3.
-If self cast Mow Down last turn, jump to routine 2.

Routine 0:
-50% chance to cast Grand Crush (Default targeting).
-15% chance to cast Soul Press (Default targeting).
-35% chance to cast Mow Down (Default targeting).
Routine 1:
-50% chance to cast Tail Guard (Default targeting).
-50% chance to cast Soul Press (Default targeting).
Routine 2:
-25% chance to cast Soul Press (Default targeting).
-25% chance to cast Tail Guard (Default targeting).
-50% chance to cast Grand Crush (Default targeting).
Routine 3:
-Cast Mow Down (Default targeting).
Routine 4:
-25% chance to cast Grand Crush (Default targeting).
-50% chance to cast Soul Press (Default targeting).
-25% chance to cast Mow Down (Default targeting).

Yeah so that challenge video where it spammed Tail Guard constantly? Turns out that was because it was at low HP. Also turns out until that point, you can basically avoid Tail Guard if you never attack after a Mow Down turn. And between 10% and 35% HP it can't cast Tail Guard at all.

Its AI does not change with how many body parts you take out, however it does change if you accept Jake's help. But as 7th Dragon doesn't use flags to determine AI, what it instead does is...

Earthshaker posted:

Conditions:
-If HP is at or below 10%, jump to routine 1.
-If HP is at or below 35%, jump to routine 4.
-If self cast Soul Press last turn, jump to routine 3.
-If self cast Mow Down last turn, jump to routine 2.

Routine 0:
-50% chance to cast Grand Crush (Default targeting).
-15% chance to cast Soul Press (Default targeting).
-35% chance to cast Mow Down (Default targeting).
Routine 1:
-50% chance to cast Tail Guard (Default targeting).
-50% chance to cast Soul Press (Default targeting).
Routine 2:
-25% chance to cast Soul Press (Default targeting).
-25% chance to cast Tail Guard (Default targeting).
-25% chance to cast Grand Crush (Default targeting).
-25% chance to cast Distracted (Default targeting).
Routine 3:
-75% chance to cast Mow Down (Default targeting).
-25% chance to cast Distracted (Default targeting).
Routine 4:
-25% chance to cast Grand Crush (Default targeting).
-50% chance to cast Soul Press (Default targeting).
-25% chance to cast Mow Down (Default targeting).

Replacing the entire AI file with a nearly identical file with some routines slightly changed. There are only 2 points where Earthshaker can get distracted and once it falls below 35% HP, it can no longer get distracted.

Beetle posted:

Conditions:
-Undefined. Jump to routine 1.

Routine 0:
-50% chance to use a basic attack.

Now here's a weird one for you. You remember those Beetle enemies way back in the beginning of the game that can only use regular attacks? Well this file seemed to indicate something more was planned for them, but got hastily cut out. In practice all it can do is use regular attacks. The fact that it only has 1 function means it will do it every time, despite the "50% chance" which is functionally useless. Which would explain why those enemies were bizarrely lacking in skills.

Now for one final bit of weirdness. Every enemy actually has a duplicate of their AI in the files right next to the original one. There's actually a very bizarre reason for this. See, that function that lets enemies attack twice in 1 turn? It's used for this. The first action reads from the first AI file, and the 2nd action within that turn reads from the 2nd AI file. You can actually hack the hard coding to allow enemies to act 3 or more times in 1 turn, but anything past 2 actions will have the enemy do nothing because there's no "3rd brain" to read. This function is actually used for 1 enemy in the game as its gimmick, as the 2nd file for that enemy is not actually a duplicate of the first, but for whatever reason all other enemies just have copies of their brains in the game data that go completely unused. Yeah. :psyduck:

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply