[PSX/PC] Battle editor - Proud Clod (1.5.0/FINAL)

[secondadvent]

ok, i understand it now. i guess i'll just get back to searching for other things then . what is the initial status thing, it doesn't seem to add the status, or even make you immune to it, so what is it supposed to do?

[nfitc1]

I'm always amazed that the developers thought it would be a good idea to bundle each scene with its own attacks. This wastes storage space as there are MANY duplicates. However, it is simpler than having one file with 1000 attacks worth of data and having that loaded at all times with the limited active memory you're alloted. But I digress....

I got basic script copy-paste functions "working". It's basic, but not too friendly a system. It's probably as simple as it can be and still be practical, and as complicated as it can be and not take very long to perform. It's going to be up to the user to work with it and it probably has a bug or two. I'm not completely ready to release it yet. Maybe tomorrow, then I'll get WM up to snuff with the latest changes I've made in PrC.

I'm still thinking of adding a patching "mod file" that will save the changes you've made along. This is going to be more complicated for PrC because there are more pieces. Oh well. I'm still pleased that PrC is smaller than the file it edits (225K vs 264k) as opposed to WM (+730K vs 22K ). I'm sure it'll grow larger once I add the ability to modify formations.

I'm out for the day. Later.

[secondadvent]

i just got killed by a level 144 MP enemy with 6 attack that did 150 damage a hit XD. im going to have SO much fun with this . i never knew it'd allow over a lv 99 to happen, but i guess i was wrong... now if only i knew where the status immunities and item info is in the ai... if there is any . guess ill have to update my little spreadsheet to accommodate the higher levels allowed...

yay... a level 255 with 255 attack can reach the new 30k limit easily .

[Kudistos Megistos]

i just got killed by a level 144 MP enemy with 6 attack that did 150 damage a hit XD. im going to have SO much fun with this . i never knew it'd allow over a lv 99 to happen, but i guess i was wrong... now if only i knew where the status immunities and item info is in the ai... if there is any . guess ill have to update my little spreadsheet to accommodate the higher levels allowed...

Level 144?

Actually, that doesn't come as much of a surprise. The game allows for any level between 0 and 255; it's just impossible to level up past 99.

BTW NFITC1, can anything be done about the amount of time it takes to write battle dialogue in the AI editors? At the moment I have to type everything byte-by-byte after converting what I want to write with FF7text. Is it possible perhaps to copy and paste text into a single "argument" box? (my apologies for such non-technical terms )

[Izban]

quick question using this would it be possible to make it so depending on your level your enemys level and stats increase too along with the xp/ap received from winning

[Kudistos Megistos]

Stat increase? Yes - Safer Sephiroth does it.

But I'm not sure about increasing exp/ap 

[nfitc1]

BTW NFITC1, can anything be done about the amount of time it takes to write battle dialogue in the AI editors? At the moment I have to type everything byte-by-byte after converting what I want to write with FF7text. Is it possible perhaps to copy and paste text into a single "argument" box? (my apologies for such non-technical terms )

Yes, actually. I don't like the way it is either. I'll see what I can do today before the next version comes out.

quick question using this would it be possible to make it so depending on your level your enemys level and stats increase too along with the xp/ap received from winning

Stat increase? Yes - Safer Sephiroth does it.

But I'm not sure about increasing exp/ap 

42B0|AP
42C0|Gil
42E0|Exp

Modify those to get what you want. The max for AP is 65535 and the max for Gil and Exp is 16777215 each.

[secondadvent]

quick question using this would it be possible to make it so depending on your level your enemys level and stats increase too along with the xp/ap received from winning

actually, i am doing that at the moment (not exp/ap yet, i will see about that later ), and it'll take a good bit of room to do, especially if done to every enemy in the game. as of right now, with the basic outline and only level and hp modified, it took around 90 bytes to do, but adding another stat will not increase it too much more at a time.

edit: heh, for an enemy to increase all stats by a set and random amount (the absolute max size for the level ai, though no error checking yet...), including exp/ap/gil is around 400 bytes, so i will need my compressed ai project as a base . it is doubtful that all enemies will be getting both a set and random amount for everything, but even still it will be up there.

Level 144?

Actually, that doesn't come as much of a surprise. The game allows for any level between 0 and 255; it's just impossible to level up past 99.

actually, it woul be possible to go past level 99 with the main prty if you have the ai for the party make global variables for all stats and exp, have the post battle ai for the characters add the experience value of all enemies with the dead unit flag set, and do the leveling in the post battle as well (though i dont know the exp equation if there is one, so the amount needed may not be too pretty ). the only real problems is that size could become an issue pretty quickly that way (likely psx users will not be able to use it), and i do no know which global variables are available for use, and there would need to be plenty of them.

ok, remind me NOT to fight a lv 99+ MP enemy, cuz it could result in me having -16000 hp left after one hit . i guess that is what happens when testing, especially when random stat increases per level isn't added in yet, i just wanted to get the basic outline for every stat done.

i was glad that there wasn't any slowdown in-game when the ai leveled two enemies up to 99+, but it may not be the same with a group of 6, and could be slower on the psx version. i guess that if it does slow down, that is the price to pay for better balance . i also noticed that sense shows levels over 99 as well, but i guess that isnt a surprise since the game allows it anyway.

[nfitc1]

On the PC version the scripts can run through probably millions of lines of script in a single frame and not experience slow down. The PSX on the other hand only has a slightly better than 33 Mhz processor and could run through a few dozen thousand lines of assembler code in a second. While each script can be quite long and there be no noticeable performance issues (slow-down of video/audio), the only negative side effect is a slight pause before any action takes place. That's not a big sacrifice to the PSX/PSOne users. I'm assuming you put all these level up scripts in the Pre-Battle, right? That doesn't get executed until the battle scene is fully loaded and the opening camera stops moving. The biggest noticeable pause would be between the camera stopping and the ATB starting (this is when "Pre-Battle" is executing). It would take less than a second for probably anything less than Eligor's script to execute. Well, take that for what it's worth since Eligor never runs more than probably 50 of those 1288 lines of script he has during any given turn. The moral of the story is that everyone will just glare at each other a little longer before actually attacking.

The Text editing in codes A0 and 93 are about to get 1000 times easier to use. KM recently reminded me to go back and fix that (I had planned to do that for a while, but I have a habit of not bothering something that works) and it's turning out quite nicely. So, everyone, I'd like to introduce you to Proud Clod 1.1.5 now available on the first post.

[secondadvent]

and i claim the first download . how exactly can you read the A0 script if it required the debugging console? i would really love to have the ability to use it for the pc version (or at least the psx one for experiments), and i found nothing about it anywhere, and i know i could put it to good use .

hmm... the copy/paste is fine for now, but there is a little problem with it, when injecting into a fresh area, because it puts a 73 opcode in automatically, when injecting a whole script, it carries over the new end statement as well, so you either have to delete one, or copy one less byte from the code. also, i thought you were adding the ability to turn off all status inflictions (and maybe the element/target info as well)? not really needed, the copy/paste function is all i really needed right now anyway (kinda held off on continuing my mod since this release would have it ).

let you know if i find any bugs with the copy/paste function, though hopefully there isnt any .

[nfitc1]

and i claim the first download . how exactly can you read the A0 script if it required the debugging console? i would really love to have the ability to use it for the pc version (or at least the psx one for experiments), and i found nothing about it anywhere, and i know i could put it to good use .

A0 is worthless unless you're playing on an emulator that can display the debug window or a hacked ff7.exe. If you find a way to see this, let us know as I'm sure it'd be more than useful for finding out the unknown address values.

hmm... the copy/paste is fine for now, but there is a little problem with it, when injecting into a fresh area, because it puts a 73 opcode in automatically, when injecting a whole script, it carries over the new end statement as well, so you either have to delete one, or copy one less byte from the code. also, i thought you were adding the ability to turn off all status inflictions (and maybe the element/target info as well)? not really needed, the copy/paste function is all i really needed right now anyway (kinda held off on continuing my mod since this release would have it ).

This is intentional and mostly unavoidable. Well, sort of. I can make the default range copy everything except the 73 at the end, but for now it's up to the user to handle that. It'll be in the next update.

let you know if i find any bugs with the copy/paste function, though hopefully there isnt any .

too slow. I already found one. Not with the copy/paste though. I have no idea what's causing it or how to fix it. The animation window is worthless now because it can't close in some situations. It works fine on XP, but the Vista machine I have won't display the close button because the window is too small. Strange.... I'll fix that ASAP.

EDIT:
There. I fixed the Animation Window (although now it looks strange in XP) and made the default copy range not include the script end. It still can if you want it to, though. Let's pretend that the earlier update didn't happen. "LALALALALA THIS IS ME PRETENDING IT DIDN'T HAPPEN! LALALALA~!!"

Seriously though, a little word of warning. Don't try to "outsmart the copy function" by giving it decimal values or anything like that. It'll copy what you tell it to in the format that it's given so just change those six characters. If I wanted it to be super user-friendly I'd make another window to handle this, but I don't feel that it's worth it or necessary.

[secondadvent]

i only use xp, because i hate vista with a passion >_>. so far things seem ok, it is c/p'ing fine so far, other than the end thing. do you know of a way to add something to the core of the game, or like a way to make a global ai? i realized that i will probably have to make the enemies level up ad different speeds depending on the level range (like the party), and that will take a lot of extra coding, which isn't a good thing. as of right now, making a lv1 MP enemy having 100 hp and gaining about 40-50 hp a level is a lot of hacking in the beginning, especially when a lv 1 cloud of mine is doing only around 12 damage a hit.

i kinda doubt there is a way to easily add to the core, if that much is even known, and i highly doubt there is a way to make a global ai (maybe using the link command? but would they have to be in the same scene?) that would function that well, and if it did it probably wouldn't increase the speed in which it levels enemies up, but if you know of anything, let me know .

hmm... the kernel updating thing is still up to it's old tricks, not wanting to stop updating the kernel... i thought you had this fixed? it is still making the test kernel file.

[nfitc1]

hmm... the kernel updating thing is still up to it's old tricks, not wanting to stop updating the kernel... i thought you had this fixed? it is still making the test kernel file.

It's creating a correct kernel at least....

[secondadvent]

meh... i need a break from vii, so i think im gonna play iv-vi, and beat them, before i go for perfect saves (iv and v i always quit right before the final dungeon, but beat vi a couple times... wanna get them off my mind), so i think a short hiatus is in order . i will still be hacking vii when i am not playing them, but it will be less...

i guess my bug checking of PrC will have to wait .

[gjoerulv]

Yay, love the copy function. It was kinda stupid of me suggesting not to copy the AI. It kinda must be changed after you've copied it anyway to not create duplicates.

You planning on making a formation and battle setup editor as well? It shouldn't actually be that hard. I don't know what the data do from offset 0x0058 to 0x0117 does though. I guess it has something to do with camera movement when the battle starts.

[nfitc1]

You planning on making a formation and battle setup editor as well? It shouldn't actually be that hard. I don't know what the data do from offset 0x0058 to 0x0117 does though. I guess it has something to do with camera movement when the battle starts.

Join the club. No one really has much of an idea of what any of it does. That's what's going to be so cool about this. Probably intro camera and Formation ID are in here. I know that Kudistos Megistos has done quite a bit of formation editing and may know a few things about it. After this, everyone'll be able to make changes and test them out to see what things actually are! When work dies down again I'll get on it. For now, however, looks like SceneEdit is your choice for hexediting that data.

[secondadvent]

i dont think the formation id is in there, i looked through the code and only found one of the formation id's (012c), but that is in a couple places through the entire code, so i am betting that the formation id's are automatically set, each scene having four formation id's, making the last formation number in a scene the scene's number (starting from 0) x 4, so i do not think there is a way to directly change that.

more than likely, the camera as said, the battle type (back attack side attack, etc) that can happen, or is the only one that can happen (like the one triple MP fight on the way to the train with cloud after the first reactor, and air buster), stuff like that.

i will look into it a little to see what i can determine.

[Kudistos Megistos]

Join the club. No one really has much of an idea of what any of it does. That's what's going to be so cool about this. Probably intro camera and Formation ID are in here. I know that Kudistos Megistos has done quite a bit of formation editing and may know a few things about it.

A little bit

But I haven't had much luck with that part of the battle setup (since I focused on the bit from 0x0008 - 0x0058), but I do remember that it controls the camera.

[nfitc1]

i dont think the formation id is in there, i looked through the code and only found one of the formation id's (012c), but that is in a couple places through the entire code, so i am betting that the formation id's are automatically set, each scene having four formation id's, making the last formation number in a scene the scene's number (starting from 0) x 4, so i do not think there is a way to directly change that.

Then "formation ID" may be scene_id << 3 + formation index

Code: [Select]

0 1 2 3 4 5 6 7 8 9 A B C D E F
[ - - - - - - - - - - - - ] [ ]
        scene_id            form_index
Looking at Hojo's battle (in scene 214) he points to formation ID 035Ch when his battle is done. This does fit the above example and translates to Scene 215 formation index 0 (Heletic/Heretic Hojo) WHICH subsequently points to formation 0360h that translates to Scene 216 Formation index 0 (Lifeform:Hojo) which points to nothing. So it appears that this is correct. Good eye there, secondadvent. Therefore formation index cannot exceed 03FFh

[secondadvent]

yay, i was useful . i will keep looking through 0x58-0x117 to see what i can learn. so far i have the air buster scene up (looking for ambush data), the initial mp scene (no escape, just checking to see whats all here, and the guard hound battle has back attack capability), as well as the triple MP scene, that holds the fights before cloud jumps onto the train (has two normal, and a side attack fight), just for hopefully pinpointing something.

the air buster scene, which has three air buster formations, only the first of which seems to be used, as it is the only one that doesn't have the generic 10-27-78-ec-70-17-00-00-90-01-2c-01 that appears in most normal battles (possibly normal formation info, there are four groups of 12 bytes, the first seemingly different from the others except in the extra air buster battles, then two which i think are the normal and special formation info, then a 12 byte filler, which may be for a third battle type, havent seen it in use yet). going to see what all happens when i change a group of this for the first fight in the game, since it is the easiest to reach, and is guaranteed), to see if the second two groups are camera data, and may help to find out the first group's data.

if you guys know anything, even something small, that'd be helpful . lets figure this thing out .

im beginning to see somewhat of a pattern, but it is too soon to know for sure. if you made a little addition to PrC to just display the raw data for now, or a simple program to make a hexdump of every scene in the scene.bin, then i know it would be much easier to decipher. i will see what i can do, with what i have now by testing the things in-game, since i cannot really sleep right now, considering my bed is soaked from it raining >_>.

my bad, deleted because of idiocy on my part *faceplam*

ok, from what i can tell, the info stored is only the camera data for the three types of battles, normal, the special formation (back attacks and such), and pre-emptive special formations. i noticed that the camera starts in the same original position each time, and moves however the battle setup 2 data tells it to, so the main battle setup holds the starting camera data, while the second one "should" just be camera data, i noticed nothing else different. here is a couple pics taken using different camera layouts:

safer/seraph sephiroth:


lifeform hojo:


guard scorpion:


sorry the pics are small, they shrunk much farther than they were <_<.

gonna mess around with the settings to see if i can narrow down the different axis and such, i think the victory scene's camera data is there as well, since it looked different than normal for the lifeform hojo battle, panning to view all three people, even though only cloud was there. get back with hopefully some decent results.

ok, i have some findings on the workings of the camera, they aren't extremely accurate as of yet, but more than i have seen anybody else put up :

0x1 is the zoom, lower makes it zoom less, and higher zooms far in, and if far enough the camera will flip back around as it zooms out on the other side, showing the party on the left, enemy on the right

0x3 will lower the camera and tilt the angle upwards, giving a possible view from under the ground when the value is low, moves up high and angles downward for an aerial view when high.

0x5 pans the camera to the right, facing inward when low, and to the left facing in when high.

0x7 low barely zooms and turns the camera around, high zooms more, not turning around

0x9 moves the camera down, facing down when low, high facing up when high

0xb moves the camera right, facing right when low, left facing left when high

as you can see, there are two versions of each camera movement, ones that move a certain way turning to the center, and the oes moving the same way that pan to the outside.

0x0/2/4/6/8/a seem to be smaller tweaks to the ones next to them, because i can't really tell much of a difference when they are used. they may be the degree in which the camera curves for each following movement, but i havent tested yet. having all 00 or ff is the same, the camera moves directly forward, and into the ground, where you cannot see anything. i will test more tomorrow, i am kinda tired right now .

all that really needs checked is the even numbered ones, and the extra formations, to see which is bad special formations, and which are good specials. guess some of the more fun stuff is in the main battle setup .

[nfitc1]

ok, i have some findings on the workings of the camera, they aren't extremely accurate as of yet, but more than i have seen anybody else put up :

0x1 is the zoom, lower makes it zoom less, and higher zooms far in, and if far enough the camera will flip back around as it zooms out on the other side, showing the party on the left, enemy on the right

0x3 will lower the camera and tilt the angle upwards, giving a possible view from under the ground when the value is low, moves up high and angles downward for an aerial view when high.

0x5 pans the camera to the right, facing inward when low, and to the left facing in when high.

0x7 low barely zooms and turns the camera around, high zooms more, not turning around

0x9 moves the camera down, facing down when low, high facing up when high

0xb moves the camera right, facing right when low, left facing left when high

as you can see, there are two versions of each camera movement, ones that move a certain way turning to the center, and the oes moving the same way that pan to the outside.

0x0/2/4/6/8/a seem to be smaller tweaks to the ones next to them, because i can't really tell much of a difference when they are used. they may be the degree in which the camera curves for each following movement, but i havent tested yet. having all 00 or ff is the same, the camera moves directly forward, and into the ground, where you cannot see anything. i will test more tomorrow, i am kinda tired right now .

all that really needs checked is the even numbered ones, and the extra formations, to see which is bad special formations, and which are good specials. guess some of the more fun stuff is in the main battle setup .

So you're saying that the camera's controlled completely by those numbers and it's NOT pre-set indexed movements like the attacks, huh? Fascinating.

[secondadvent]

from what i have seen, i may be wrong, as i just tested random things, but i know trying the same number in a different section had different effects, so either each specific section has predetermined camera movements, or a specific route for the camera is already taken, and the number just directs it in the direction, and distance the camera would move, making the first 128 combos one direction and distance, while the next 128 is the other direction.

i will test more for this here soon, i just want to fiddle with the first battle setup as well, because i want to know where the chance for specific encounters lie, and it is either in the scene file, the field files, or a list somewhere with specific settings for each and every scene. if you do know where that is located, then i guess i dont need to look for it then .

i can upload more pics based on the normal battle camera movement for that battle, and just tweaking one of the values to show it's effects if you want me to. i need to do more testing anyway for the even offsets, but i do think it is just minor tweaking of the values. also, does the victory part of the battle have set camera movements as well, because i noticed that it didn't seem to alter that at all, though i did see at least three different camera movements used.

all i can say is that even with full control of the camera, without a visual representation of the current movements, it would be hard as hell to get a good camera movement made from scratch. a simple little program to show a basic idea of the camera movement may be possible for somebody to do sometime in the near future, so that the starting camera could be changed at will very easily.

i found out something about one of the unused flags for attacks, the one right under 0002h will IGNORE magical immunity (probably the attack type's immunity, so phys ignores phys immunity). i am tinkering around to figure out some things ^^.

added effect 22 causes quadra magic damage, so it seems to be split to about half strength, or is /32 instead of the normal 16, with all, i did ~950 f my normal 2k damage with blizzara, will see what happens with different damage formulas. yep, even when using an exact strength x 20 formula, the damage was pretty much halved (should have done 640 damage, but did 660 instead, so not quite halved. this is when the modifier was 255 though, still need to check for different amounts. ok, the effect modifier has seemingly random results, while 32 was causing a set damage attack that did 1280 normally do 500, a mod of 16 caused 660 again, like 256. im not quite sure how this works exactly, but it does cut the damage by some amount.

[nfitc1]

i found out something about one of the unused flags for attacks, the one right under 0002h will IGNORE magical immunity (probably the attack type's immunity, so phys ignores phys immunity). i am tinkering around to figure out some things ^^.

added effect 22 causes quadra magic damage, so it seems to be split to about half strength, or is /32 instead of the normal 16, with all, i did ~950 f my normal 2k damage with blizzara, will see what happens with different damage formulas. yep, even when using an exact strength x 20 formula, the damage was pretty much halved (should have done 640 damage, but did 660 instead, so not quite halved.

So did you read my latest update on WM topic that said I noticed that Unused 0004h actually does something but is never active?
I just tested this ability and it did not work. So maybe you're meaning something that I'm not understanding. I understand this to be if the enemy is immune to magic (it's 4029 is set to 1) then the attack will do damage if this flag is set. This didn't work when I made the Midgar Zolom immune to magic this way. I'd like to know if 0002h has ANY effect at all. I mentioned once that Pandora's Box once oddly affected a Grangalan if this was off, but I couldn't get it to do that again.

Most Additional Effects are still not known either so continue to test them. 22h could be called "halve damage" or something.

[secondadvent]

against the mp enemy, with the mag immunity set to one, the unused 1 caused normal damage, i dunno if you tried unused 2 instead, but it is unknown 1 that breaks this immunity, and i am testing with different elemental resistances as well, to see if it causes normal damage no matter what the resistance (ignore resistance maybe?).

well it doesnt stop an enemy from absorbing, but it does kill mag immunity, i will see if this works for any other resistances (doubtful), and if it works the same way for phys immunity.

making them immune to ice (using blizzara still ) still causes 0 with it on, so it must just pierce through the ai flag. testing phys now. nope, it seems to only ignore magic immunity, maybe another flag ignores this?

hold on, i made some errors on what i just posted... removed >_>.

ok, nothing seems to ignore the physical immunity flag, but the magical one is passable so far.

what exactly happened with the grangalan that you said screwed up in some way? it could have been something in it's ai that caused issues.

speaking of grangalan, i looked at it's script, and i think i will use what it does to activate one of the grangalan jr.s, after turning an enemy invisible, to see if it is possible to bring it back in this way. nope... still nothing <_<. seems like moves will have to be used then to do this, at least until it is figured out completely.

ok, an update on the unused 1 flag... it seems that it will cause magical damage to be 0, and status effects to miss if the magical immunity is not on, but if it is, it will act as iff it were off. i guess this is a way to make an attack damage magical immune enemies, that only affects those enemies. haste missed when this unused flag was enabled, but worked when immune to magic (at the base 255% accuracy), so this is what it appears to do.

does pandora's box always cause a weird graphical bug, where there is a transparent orb in the center right after the attack, then a weird hovering shadow effect shortly afterwards on the pc version, or is it just the custom driver?

do you have any idea why enemies with the shoot element on their attacks do not do long range damage, while anything within the kernel will do long range damage when it's element is shoot? is it just something to do with the scene.bin itself, since it is not the main place the game gets it's attack info from, causing it to miss out on the shoot =  long range clause?

ok, the long range materia does NOT change the element to shoot, so something else has to be triggering the long range effect, because cloud 1hko'ed himself, while absorbing the mp's gunfire (which is shoot, but not long ranged).

[nfitc1]

against the mp enemy, with the mag immunity set to one, the unused 1 caused normal damage, i dunno if you tried unused 2 instead, but it is unknown 1 that breaks this immunity, and i am testing with different elemental resistances as well, to see if it causes normal damage no matter what the resistance (ignore resistance maybe?).

well it doesnt stop an enemy from absorbing, but it does kill mag immunity, i will see if this works for any other resistances (doubtful), and if it works the same way for phys immunity.

making them immune to ice (using blizzara still ) still causes 0 with it on, so it must just pierce through the ai flag. testing phys now. nope, it seems to only ignore magic immunity, maybe another flag ignores this?

Ah.. I see what I did wrong. I set both physical and magical immunity to on. 0004h forces physical damage and has nothing to do with immunity.