Topic: FF6 Improvement Project

[Poco Loco]

ah ok great I just wanted 2 make sure of that man thank u

[darkmage]

I don't remember when I got it, but I came across a zip file on my thumbdrive with the battle dialogue and battle message text from both FF3us and FF6A! I've spent most of the day moving over the (better) battle messages to FF6I and accounting for space constraints where necessary.

So now the Ultros, Ziegfried and Siegfried fights will make more sense (and hopefully be more amusing), and overall the battle messages will be more coherent and better-paced. The battle dialogue I'm saving until the baseline patch is complete and I can make the dialogue over-patch.

 

[Poco Loco]

nice man, so are u making 2 separate patches then?

I mean because some might not want to change it or go about it their own way while keeping the rest of the improvement going on

[darkmage]

Yeah, this is going into the script/dialogue side of the project for now. I didn't want to make any changes to the game's story in this patch for precisely that reason.

[Poco Loco]

ah nice nice

another thing I wanted to suggest, since I spoke with Lord J about making his editor compatible with dialog that has been edited thru hex he said that might take a while, could u tweak it so that the fixed dialog was edited thru the editor (I know it may be a shit load of a task to re implement the hex data on a fresh rom but he said it would take a good while n in the meantime dialog cannot be edited)

[darkmage]

One thing you and TheNattak could do is something Dragonsbrethren turned me onto before Lord J updated his editor: Use Cartographer to dump the script, make your changes, and then use atlas to insert it back into the ROM. It's a few more steps, but once you get the hang of it it's not bad.

Look back in this thread, like the 2nd or 3rd page, and you'll see where we were talking about it.

[Poco Loco]

ok man I'll check that out

[TheNattak]

I just found another spot that if you edit the text, the game screws up. After beating the tentacles in the Figaro engine room, and not only does Edgar not say his line,

"Edgar: What're ya waiting for, Celes?
Give me a hand!!

But after the battle, one tentacle NPC is still hanging around, and worst of all the screen goes into permanent screen shaking L1 tile earthquake mode for the rest of the game. Every single map, except the overworld. The solution? Re-patching mblock128's Edgar Reveal patch. This one was easy to figure out since it screws up right where that patch does it's "magic" =p Just like his other one. So after re-patching, everything is good again.

Watch the bug in action here: http://www.youtube.com/watch?v=OHNuI9hnzrU
However for some reason Youtube makes it look normal most of the time, but you can still see it happening.

[darkmage]

I suppose I'll have to think seriously about implementing the Free C2 patch just so I can move those fixes from the dialogue pointer region. I wonder why mblock129 would have chosen to put them there...?

[Poco Loco]

hmmm, not sure, but I suppose he did it more for the sake of adding those bytes, its still a huge win, the dialog thing I suppose is a thorn on the side u know

[darkmage]

True - no one else made those fixes. 

So begins my journey into the deeper, darker world of event editing and hex editing.  I've just dumped the event scripts for the clean ROM and FF6I, and when I figure out xkas' syntax, I'll be implementing the Free C2 patch to see if I can move some of those fixes into the newly-created free space.

[darkmage]

I've got a question that I'm gonna throw out:

For event scripts, is there anywhere acceptable to put them that won't conflict with any other information? I was thinking C2, but if that's a battle-related bank it may not work. I'm pretty sure CF is out as well, since we're seeing some issues with dialogue pointers and fixes put into that bank.

Any ideas?

[Lenophis]

You have one more problem than that, the events are relative to bank CA. That means if you want events to go before that, you need to calculate the address of a jump (B2 command, for example) so the program bank wraps. The problem with that is, I don't even know if the hardware will deal with that the way you would want to. I also don't know if that's correct behavior or not.

There's a bunch of free space in bank EE, starting at EE/AF01. You'll still have the same problem of calculating the address, but it'll be much easier in this case.

[darkmage]

Easier meaning less potential problems, in this case. I see what you mean. 

I'm wondering if it could really be as easy as putting in the new bank address? From what I'm seeing, the jump events already have B2s to the subroutines they call...

Do you know if this has been done before, and how much lag such a big jump might add to the script?

[Poco Loco]

looking into some of the data changed within this patch and the not scary enough patch,my friend Angelo26 managed to find the bytes that were changed by the NSE patch

Offset: Bytes:
0000D874 21 --> At this address, 21 bytes of info were modified.
0001032D 4 --> 4 bytes modified
0001FF9C 2 --> 2 bytes modified
000215D0 3 --> 3 bytes modified
000224E8 4 --> 4 bytes modified
00022640 3 --> 3 bytes modified
00025F94 3 --> 3 bytes modified
0002A8CE 104 --> 104 bytes modified
001099FE 13--> 13 bytes modified
00109A0C 5 --> 5 bytes modified
00109A12 11 --> 11 bytes modified
00109A1E 39 --> 39 bytes modified
00109A46 1 --> 1 byte modified
00109A48 1 --> 1 byte modified
00109A4A 3 --> 3 bytes modified
00109A4E 1 --> 1 byte modified
00109A50 1 --> 1 byte modified
00109A52 1 --> 1 byte modified
00109A54 1 --> 1 byte modified
00109A56 3 --> 3 bytes modified
00109A5A 1 --> 1 byte modified
00109A5C 1 --> 1 byte modified
00109A5E 7 --> 7 bytes modified
00109A66 1 --> 1 byte modified
00109A68 1 --> 1 byte modified
00109A6A 3 --> 3 bytes modified
00109A6E 6 --> 6 bytes modified
0010D14A 51 --> 51 bytes modified
0010D17E 7 --> 7 bytes modified
0011F6F4 24 --> 24 bytes modified
0011F70D 16 --> 16 bytes modified
0011FA2C 2 --> 2 bytes modified

some of the patches in here rearranged a handful of data so when trying to look into it, it kid of points towards the floating continent event, so yeah, kinda stuck here =/

just throwing it out there hopefully there might be a way to fix the issue :p