A few PAR codes submitted by chakrafire

[justin3009]

Yes, it does give me bad glitches.  That's why I only tested it when I was fighting with Magus.  (REplaced Gato with Magus) and it worked fine.

[FaustWolf]

Justin, I think I just found what these codes affect by digging through your notes. In one of your posts you refer to the data beginning at (unheadered)xD45AC. At that point there's some FFxx values along with lots of other data. The values begin with FF01 (for Cyclone), and you had a value of FF0C pertaining to Marle's Ice, I believe. Curiously Ice falls at an index value of 0B, but Aura appears to be missing in the list Zeality got from Chakrafire. If Aura were factored in at 09 (thus replacing Provoke), that would bump up the index value of the techs, thus making Ice 0C again.

That was a confuzzling description of what Justin already had. Now let's see what happens if I start fiddling with the pointers/index values beginning around (unheadered) address 0xD45AC...

Interesting. Changing unheadered address 0xD45AD from 01 to 03 causes Crono to go through the motions of Lightning, but Cyclone's graphics are still used. Basically the cyclone slash gets dropped on the enemy instead of a lightning bolt. Is that the effect these codes tend to have?

[justin3009]

SUCCESS!  If you change FF01 to FF03!  He learns Lightning instead of Cyclone!  Now let's try somethign else ^^

Edit: YES!  IT WORKS!  The ONLY problem is that it retains the graphics and the attack data from the original tech, not the one you replaced it with.

[FaustWolf]

Whoah, waitaminute! It changes the tech LEARNED as well!!?? I didn't know that because my savestate's screwed up (I've got Magus on Crono's status screen). I'm going to try this out on a fresh game...

EDIT: Trying it right now. Also, I should note that the damage and targeting done by the tech are still the same as originally.

[justin3009]

Yes.  So we just redid the tech learned and the graphics that it does...Now we need to find a way to point it to the actual damage and stuff.

[FaustWolf]

Hm, it doesn't cause Crono to "learn" Lightning as opposed to Cyclone in my game. Did Lightning actually appear on Crono's status screen for you Justin? Also, are we looking at the same data? I'm looking at address xD45A0 in an unheadered ROM, at which point the data reads FF 01 DA EF....

Crono goes through the motions of Lightning after he learns Cyclone, but Cyclone is still the tech listed on Crono's tech status screen. Also, be careful. My ROM freezes occasionally after Crono uses LightningCyclone.

[justin3009]

Yea, it does that because the graphics aren't actually there.  You have to copy and paste the 8 bytes of lightning over Cyclone and it should work.  But it'll still list Cyclone in the menu

[FaustWolf]

Ah, okay, so we haven't found a tech learned pointer just yet. I'm going to do a full data transfer from Lightning to Cyclone and see what happens; I just changed that one byte before. For one thing, I'm a bit confused as to whether the amount of data is consistent for each tech; it looks to me like things work out okay if 7 bytes are devoted to each tech. Cyclone starts out FF01 as the first two bytes; next is Slash with FF02; then Lightning with FF03; then the next tech (Spincut?) gets wonky, starting with 4504. Should have been FF04, so I'm not sure what's up there.

But at least we know that the index values for these codes work as we would expect. I think Aura needs to be inserted somewhere though in Zeality/chakrafire's list. (it needs to be inserted as Aura - 09 I think)

[justin3009]

8 bytes per tech is what it is.  Some of them aren't seperated like that, just always go by 8 bytes and you'll have each tech.

[FaustWolf]

Harumph! You're right indeed. 8 bytes per tech! Lightning only works for me if I get that 45 in there that I thought was the beginnig of Spincut.

Justin, do you know if all the techs are consistently 8 bytes long? It seems I'm running into another tech's index if I keep counting that way.

EDIT: Actually Justin, I'm wondering if each tech doesn't have 7 bytes after all. If you start at address xD45AD, leaving out the FF and begin straight with the 01, and go 7 bytes over each time, the next byte will be the next tech's index like so:

000D45A0                                         01 DA EF
000D45B0  20 01 01 FF  02 E8 EC 21-02 02 FF  03 F1 00 22 03
000D45C0  03 45  04 D6 00 27 04 04-FF 05 F1 00 22 05 03 02
000D45D0  06 E1 02 1B 06 06 1D 07-20 24 2D 07 07 FF 08 FC
000D45E0  00 08 08 08 01 09 E1 E5-1D 09 09 FF 0A 0A 0E 0E
000D45F0  0A 0A FF 0C F4 00 25 3A-0C 44 0B E1 02 1E 0B 0B
000D4600  FF 0D BB 00 3E 0D 0D 56-0F FC 00 25 3A 0C 57 0E
000D4610  E1 02 1F 0B 0E FF 10    

                       

Waitaminute, I'm all discombobulated here with this post. I'll do a proper reply with the hex aligned properly.

[justin3009]

I'm almost certain it's 8 bytes each. 

Edit: Ack, apparently some of them are 7 bytes starting at FF03 is the first 7 byte tech I saw

[FaustWolf]

000D45A0                                         01 DA EF
000D45B0  20 01 01 FF 02 E8 EC 21-02 02 FF 03 F1 00 22 03
000D45C0  03 45 04 D6 00 27 04 04-FF 05 F1 00 22 05 03 02
000D45D0  06 E1 02 1B 06 06 1D 07-20 24 2D 07 07 FF 08 FC
000D45E0  00 08 08 08 01 09 E1 E5-1D 09 09 FF 0A 0A 0E 0E
000D45F0  0A 0A FF 0C F4 00 25 3A-0C 44 0B E1 02 1E 0B 0B
000D4600  FF 0D BB 00 3E 0D 0D 56-0F FC 00 25 3A 0C 57 0E
000D4610  E1 02 1F 0B 0E FF 10                           

Red is Cyclone; Green is Slash; Purple is Lightning. At least, that's what I believe. The FF before Cyclone may not be a part of Cyclone's data at all.

Here's how to test -- let's copy Lightning's 7 bytes in this scheme over Spincut's and see if we get a full Lightning effect.

[justin3009]

Your probably right, when I was doing it, I musta miscounted on some of them.

[FaustWolf]

Well, it was the FF before Lightning that threw us off before I think. Starting directly with the 01 instead of the FF01 gives us the most logical scheme, but I've gotta test this out somehow. Let me see if I can track down a proper savestate. Actually, I have it -- just need to adjust unheadered ROM addresses to headered ones, since I've been behind the times and playing with a headered ROM before I joined up here.

But anyway, if there is a list of relative pointers into the tech data, the pointers should read: 00 00; 00 07; 00 0E... or something if they're relative offsets with 0 beginning at unheadered address xD45AD (which I'm absolutely UNsure of).

[justin3009]

If needed I could host up a zst with every char at lvl 99 max stats and all techs ._.