Advertiser Content
Conversation Between ghoulslash and colonelsalt
1 to 14 of 14
  1. ghoulslash
    February 14th, 2019 8:28 PM
    Glad you found a solution! I hope it wasn't too much of a pain :tongue-poke:
  2. colonelsalt
    February 13th, 2019 1:16 PM
    Cheers for the follow-up -- my dirty fix ended up being just having all backsprites use the same palette, which was sufficient for my hack. Good reminder that I should mention you in the credits for writing the fix, though ;)
  3. ghoulslash
    January 27th, 2019 4:02 PM
    Hey, I recently found a potential solution to the backsprite hack problem you were having - depending on the emulator you are using, the backsprite switch fails because of jpan's usage of SWI statements without BIOS functions. I rewrote the function to exclude any dependence on these functions if you are still looking for a solution?
  4. ghoulslash
    September 10th, 2017 3:11 PM
    I believe the mov r8, r8 was just used to keep the routine the same length, rather than having people re-compile and re-insert, but I could be wrong, I don't have much experience with VRAM either. Sorry I couldn't help out more, but I hope you find an easy fix!
  5. colonelsalt
    September 10th, 2017 2:07 PM
    That's really clever. (Was wondering what good a "mov r8, r8" could possibly be doing...) The issue happens in wild and trainer battles both, so a simple battle check branch won't quite do it.
    But thanks so much for the in-depth pointers and explanations, man. :D I will keep fiddling along those lines, and will no doubt stumble on a fix somewhere.
  6. ghoulslash
    September 10th, 2017 10:45 AM
    The original routine forces the opposing trainer to inherit the player's backsprite palette. Karatekid's fix replaces a 'swi 0xB' with mov r8, r8 to fix this issue. But then I was getting an issue that caused the player's backsprite to inherit the default hero's palette on every frame except the first one. Removing the trainer palette fix resolves this issue, so my routine just checks the frame number; if the first frame is loaded, it applies the trainer palette fix. For all other frames, it performs the original routine without the trainer palette fix.

    I would play around with the original routines and each fix to try and isolate what is causing the HP bars to change. It's strange that I'm not getting the same error.. :/

    edit: these issues only come about for trainer battles. Does the same problem happen during those? If not, we could add a battle flag check to keep the original routine if you're in a wild battle.
  7. colonelsalt
    September 8th, 2017 5:55 AM
    Hmm... That routine doesn't seem to matter, and I don't think there's any issue with the tool's application of the routine (it's identical to JPAN's source).
    I can't say I'm perfectly confident with the workings of the VRAM, but I've tried playing around with your code. When I remove your "player pal fix section" (i.e. remove the branch to "Other" if loading frames 1-4), the problem disappears (but obviously the wrong player palette is shown for frames 1-4). What exactly is your extra branch changing? Might it be related to the palette of HP bars?
  8. ghoulslash
    September 7th, 2017 4:01 PM
    I have heard the tool is buggy, so that might be the problem? I'm not sure if the other routine I'm talking about is relevant, but the pointer is at 6D724.
  9. colonelsalt
    September 7th, 2017 2:35 PM
    May or may not be reassuring.
    I used JPAN's Hack Applier tool, so my backsprite switch routine is at 0x172C9C, and so there was no routine 0xB0 steps down to repoint (original routine was surrounded by a block of freespace).
  10. ghoulslash
    September 7th, 2017 1:22 PM
    hm, I don't seem to be having that issue. The backsprite swap routine is at 161700 but the fix increases the routine length, so did you repoint the routine at 1617B0?
  11. colonelsalt
    September 7th, 2017 12:42 PM
    Pretty bizarre. Don't know if a temporary variable is changed from using the PC that affects battle palettes, because the change seems to persist until the game is reset (demo).
  12. colonelsalt
    September 7th, 2017 12:32 PM
    Ah! Found a bug in your routine after all. Whenever I use the PC and then return to battle, the HP bars of both Pokémon change to red for every frame of the backsprite animation after the first. Should be simple enough to reproduce.
    Do you have a suggestion for a fix here?
  13. ghoulslash
    August 31st, 2017 8:09 PM
    No worries, glad it helped you out! I gotta give karatekid most of the credit for the initial routine fix though :P
  14. colonelsalt
    August 31st, 2017 5:33 PM
    Dude, I'm eternally grateful. Your fix for JPAN's backsprite switch has really saved my hack from probably weeks of delay. I don't know where I would have even begun without finding it. Thanks so much, and keep up your ASM-ing!