In the current route of Pokémon Yellow No Save Corruption,
we use a sound loading trick that works on the original versions of the game, that works on the BGB emulator, but that doesn't work on Gambatte for Bizhawk.
However, I love Emuhawk, is there a way to load another "kind" of emulator (not Gambatte) for BizHawk ?
Unless you want to run in SGB mode, no. If you or someone else knows how to fix this sound loading trick in Gambatte to match original hardware behavior, that patch would be accepted.
Well, I don't know how exactly.
That's sad I can't use my favorite emulator.
I did that run a bit with BGB, and apparently, we load a special value that matches with the special stat of the last pokémon met into a sound value with the use of the trainer-flying glitch.
I don't know if that could help you, however I hope you'll do a patch to fix that :)
I know we had a pokeymans issue once that someone identified the actual issue in the gambatte code, and it was fixed, so it's possible that someone else might know...
After we discovered this speedrun route works on console and BGB but not Gambatte/Bizhawk we investigated a little and found out that the issue is caused by Gambatte's implementation of bankswitching for MBC5 specifically.
MBC5 is capable of bankswitching to bank 0 (having bank 0 accessible from 4000-7FFF as well as 0000-3FFF) but (lib)gambatte assumes that any bankswitch to bank 0 retrieves bank 1 instead like in earlier MBCs..
This particular glitch setup causes Pokemon Yellow's audio engine to bankswitch to bank 0 and read invalid data from there. The invalid data in bank 0 leaves the game in a more stable state where we can proceed to catch Missingno, while the equivalent data in bank 1 instead softlocks the game.
TL;DR: this inaccuracy is caused by an inaccuracy in gambatte's MBC5 emulation, to fix it adjust Gambatte's MBC5.adjustedRombank function to allow bank switches to bank 0.
After we discovered this speedrun route works on console and BGB but not Gambatte/Bizhawk we investigated a little and found out that the issue is caused by Gambatte's implementation of bankswitching for MBC5 specifically.
MBC5 is capable of bankswitching to bank 0 (having bank 0 accessible from 4000-7FFF as well as 0000-3FFF) but (lib)gambatte assumes that any bankswitch to bank 0 retrieves bank 1 instead like in earlier MBCs..
This particular glitch setup causes Pokemon Yellow's audio engine to bankswitch to bank 0 and read invalid data from there. The invalid data in bank 0 leaves the game in a more stable state where we can proceed to catch Missingno, while the equivalent data in bank 1 instead softlocks the game.
TL;DR: this inaccuracy is caused by an inaccuracy in gambatte's MBC5 emulation, to fix it adjust Gambatte's MBC5.adjustedRombank function to allow bank switches to bank 0.
After we discovered this speedrun route works on console and BGB but not Gambatte/Bizhawk we investigated a little and found out that the issue is caused by Gambatte's implementation of bankswitching for MBC5 specifically.
MBC5 is capable of bankswitching to bank 0 (having bank 0 accessible from 4000-7FFF as well as 0000-3FFF) but (lib)gambatte assumes that any bankswitch to bank 0 retrieves bank 1 instead like in earlier MBCs..
This particular glitch setup causes Pokemon Yellow's audio engine to bankswitch to bank 0 and read invalid data from there. The invalid data in bank 0 leaves the game in a more stable state where we can proceed to catch Missingno, while the equivalent data in bank 1 instead softlocks the game.
TL;DR: this inaccuracy is caused by an inaccuracy in gambatte's MBC5 emulation, to fix it adjust Gambatte's MBC5.adjustedRombank function to allow bank switches to bank 0.
After we discovered this speedrun route works on console and BGB but not Gambatte/Bizhawk we investigated a little and found out that the issue is caused by Gambatte's implementation of bankswitching for MBC5 specifically.
MBC5 is capable of bankswitching to bank 0 (having bank 0 accessible from 4000-7FFF as well as 0000-3FFF) but (lib)gambatte assumes that any bankswitch to bank 0 retrieves bank 1 instead like in earlier MBCs..
This particular glitch setup causes Pokemon Yellow's audio engine to bankswitch to bank 0 and read invalid data from there. The invalid data in bank 0 leaves the game in a more stable state where we can proceed to catch Missingno, while the equivalent data in bank 1 instead softlocks the game.
TL;DR: this inaccuracy is caused by an inaccuracy in gambatte's MBC5 emulation, to fix it adjust Gambatte's MBC5.adjustedRombank function to allow bank switches to bank 0.
Fixed in SVN, thanks.
Wait, this is fixed yet ? Where can I download it ?
You're just fucking stupid, everyone hates you, sorry to tell you the truth. no one likes you, you're someone pretentious and TASes only to be on speed game, but don't have any hope, you won't get there.
You're just fucking stupid, everyone hates you, sorry to tell you the truth. no one likes you, you're someone pretentious and TASes only to be on speed game, but don't have any hope, you won't get there.
Can I request to update GBhawk core to latest Gambatte?
I thought GBhawk core is updated at every major updates, but it's not likely so.
A major bug on the joypad irq are fixed in April. it effects to memory glitches of DQ3, DQM2 and other games for sure.
Joined: 3/31/2010
Posts: 1466
Location: Not playing Puyo Tetris
tetora_X wrote:
Can I request to update GBhawk core to latest Gambatte?
I thought GBhawk core is updated at every major updates, but it's not likely so.
A major bug on the joypad irq are fixed in April. it effects to memory glitches of DQ3, DQM2 and other games for sure.
Do you know what version fixes this issue? Are there other issues that this new version introduces?
When TAS does Quake 1, SDA will declare war.
The Prince doth arrive he doth please.
Can I request to update GBhawk core to latest Gambatte?
I thought GBhawk core is updated at every major updates, but it's not likely so.
A major bug on the joypad irq are fixed in April. it effects to memory glitches of DQ3, DQM2 and other games for sure.
I manually patched in some joypad irq changes. Would you be willing to test them and tell me if they fix the expected stuff?
