Posted by infidelity

---

bnash222005, I wasn't being rude, sorry if it came off like that, cause that's how I'm perceiving your response.

Thanks Ice Man, I'll get on that when i can. Sorry everyone.

(But the all-stars that's in the uploader, the leaves blow in smb2j correct? Want to make sure I uploaded the right all-stars patch)

---

Posted by infidelity

---

Alright np.

OK the link should be fixed now. But I'll re-upload the .ips tonight, I'm out on the road.

---

Posted by stgiga

---

Hi! I just wanted to point out that, without a header like it is on cart, your hack is the exact size of the SNES release without a copier header. With a copier header, it has the same file size as a trained copy of the hack with no ines header. The PC-10 data banks bring that number closer to the all-stars+World multicart. I wonder if Nintendo recompiled the 6502 code in old-style snes code recycled from the old roms in the Abe's release? In fact, game genie codes can be ported, and many debug or physics code stubs/bugs are in both versions... They probably used snes-colorized mmc5-8x8 graphics in the same graphics resolution as the nes of 256x224/240. The skid sample was also upsampled, along with the tubular bell. If one wanted, they could create a patch that when applied to all-stars on snes (preferably sfc), it demakes the game into your hack, so that those who have the wrong version of SMB3 can still play, and so those who have bad patchers can still play due to the differences and file sizes being closer to identical (more so if using the US version/Eu version.)

Anyhow, good work, and in my short free time I have not found any bugs. I think bnash22005 had something to say, but did not say it, which makes me think he had a bug that fixed itself, but that was just a guess. I have personally seen no bugs though, so good work! This is stgiga signing off, so good luck on your new projects!

---

Posted by stgiga

---

Posted by bnash222005

---

Posted by stgiga

---

Hi! I just wanted to point out that, without a header like it is on cart, your hack is the exact size of the SNES release without a copier header. With a copier header, it has the same file size as a trained copy of the hack with no ines header. The PC-10 data banks bring that number closer to the all-stars+World multicart. I wonder if Nintendo recompiled the 6502 code in old-style snes code recycled from the old roms in the Abe's release? In fact, game genie codes can be ported, and many debug or physics code stubs/bugs are in both versions... They probably used snes-colorized mmc5-8x8 graphics in the same graphics resolution as the nes of 256x224/240. The skid sample was also upsampled, along with the tubular bell. If one wanted, they could create a patch that when applied to all-stars on snes (preferably sfc), it demakes the game into your hack, so that those who have the wrong version of SMB3 can still play, and so those who have bad patchers can still play due to the differences and file sizes being closer to identical (more so if using the US version/Eu version.)

Anyhow, good work, and in my short free time I have not found any bugs. I think bnash22005 had something to say, but did not say it, which makes me think he had a bug that fixed itself, but that was just a guess. I have personally seen no bugs though, so good work! This is stgiga signing off, so good luck on your new projects!

---

The only bugs I seen were the leaves animation which infidelity fixed and the title screen music and the SMB 1 and SMB2J world music was doing that weird skipping but I have not encountered it again since that one test run. Other than that I do not think there are any more bugs and I hope there are not any bugs. I just hope infidelity can get the irq thing fixed and I hope that the Super Mario Bros 2J ending song plays on an actual NES console with no problems.

---

Whoops, sorry for the mistake. Evidently you found another hack.

---

Posted by infidelity

---

My universal flasher finally came in the mail. I was successful in erasing, and reflashing the chips on the development cartridge Ben made for me.

Just have to add my custom debugging to Mario 3, then I'll begin testing.

---

Posted by infidelity

---

Posted by PKstarship

---

Today I was playing SMB3 on the Ice World, and I got a game over, but when I got a game over I got this tile error for some reason. One thing worth noting is that I got a treasure ship to spawn, and that it only seems to have this error when the treasure ship is onscreen for me so far. (I'm not sure if this is also true for the normal Hammer Bros though)

---

Thanks for posting that. I had to eliminate the exclamation point icon, so that I could free up a single tile. (for the life of me I cant remember why I did it) So I had to rewrite the nametable writes to that section for the GAME OVER.

Yes if you have the save file, please send it here for me to check out, and just let me know what emu you are using.

EDIT

Figured it out. It is because the game uses 2 vram routines for the world map. It draws the game over screen 2 different ways, it will draw the game over screen at a different location on the nametable, when you have initiated the scroll of the world map. So for simplicity, in World 2, if you were to get a game over where the map is at Morton's castle before the 3rd warp whistle, the nametable would draw the game over routine over a different address location. I never knew that, and I was able to locate the 2nd nametable values. I will upload a new version down the road, but for those that want to manually edit this using a hex editor, here are the locations in the rom.

$541A6 = FE
$541AD = 6B

On a different topic aside from PKstarship's find, I need to send back the MMC5 dev cart Ben made for me. I think I may have damaged one of the prg slots he setup, while I was removing the chips consistently. I received the Famicom from him, so I gotta wait again before I can begin my debugging.

---

Posted by infidelity

---

Just got around to mailing back the mmc5 devcart for repair. :-) Once I get it back, I'll resume work on finishing the irq fixes.

---