miscellaNEouS files

[lancuster]

On a semi-related note, here is the Shiru Famiclone port of Tetris, as seen on several recent My Arcade consoles. This specific copy was ripped from the Tetris Pixel Pocket; it extracted with its title screen zeroed-out for whatever reason, so I restored the screen using data from the Tetris "Go Gamer" (DGUNL-7029) console. There is also a separate, earlier version of this game designed for the Retro-Bit "Go Retro Portable" systems, which AFAIK is currently undumped.

I haven't written TCRF pages on these two systems yet, though they also have many oddities. Both of them have many unused games, among other plug & play leftovers. The Go Gamer notably includes the original NES version of Tetris and Nice Code's "Bubbles" (VT03 Tetris clone) unused, as well as the Data East games and Inventor hacks from the 2017 Pixel Player.

I should also add that the Go Gamer contains many previously-undumped games, though they feature PRG protection that I can't 100% figure out (seems to involve 2-to-3 bytes being zeroed-out, with added bytes at the end of the ROMs); so the other games remain unripped for now.

It looks like Tetris has some kind of protection, either in the ROM itself. Unfortunately, I have to admit that the game doesn't work 100% well, the "ghost pieces" don't work at all. I haven't really tested it, everything else seems to work.

[yaiba1989]

Can DAIKU NO GENSAN 2 convert Mapper 65 to MMC3?

[lancuster]

The Panda Prince CHR Optimized hack. CHR-ROM was halved from 256 kilobytes to 128 kb. Along the way, all animation frames were reassembled for ease of editing.
In theory, I fixed everything that was needed. The only thing is that it's still possible to enable the Chinese version of the game.

[forgotusername]

Here is an absolutely bizarre set of games ripped from an "unknown VT3xx-based 270-in-1" (BL-867 PCB03), which was recently dumped for MAME. I was genuinely shocked by some of the things included on here, many of which tie up some loose ends with prior plug & play documentation:

*Two games, Pong and Free Way, are from the "ex-Subor" developers. Pong was previously seen on the "Great Boy" emulation handheld, though unfortunately appears to be a bad ROM; Free Way is a port of the Atari 2600 game (Freeway), and was otherwise undocumented. I'm not sure if these are connected to the Gamezone II arcade ports, but they're very similar conceptually.
*Goal Keeper is an earlier revision featuring Elmo, as well as proper music and sound effects. This effectively confirms my theory that Goal Keeper, alongside River Jump and some similar Nice Code titles, were prototype versions of games designed for the Techno Source "Elmo's World" plug & play.
*A number of "educational computer"-like games are included (Apple Chess, Minesweeper, and Solitaire), which do not match any common or dumped revisions. Link is also included, which was documented but I'm not sure if it was dumped.
*The presumably-original revisions of Airial Hero, Care Bear, and Ghost Ship. Airial Hero had its copyright blanked out, so I restored it using the CHR data from the earlier "Aimless" variant.
*A previously-undumped Falling/Burbles variant titled Denger Zone, and a slightly altered version of the Inventor hack Golfer (seemingly to cover up some VT01 color palettes; it doesn't seem to be the original).

Downloads attached below.

[forgotusername]

An addendum to Pong: it appears that the game uses VT02 registers for whatever reason, not running on a standard Famicom/NES. Most emulators do not seem to support VT02 games with a separate CHR file (I don't know if that even exists otherwise...), though NintendulatorNRS does for VT01; so the extracted ROM can be played with some cruddy colors if the header is set to VT01 mode. I imagine a better solution here would be to modify the game to use standard NES registers, though I'm not sure how that would be done.

[forgotusername]

Another additional note: the Apple Chess ROM still retains its CHR-RAM code (as is typical of "educational computer" clones), with a faux-CHR file applied to it. Running the game without the CHR attached results in a working screen when pressing the "?" icon, credited to "Colvin of Shenzhen BHD Electronic Co., Ltd". I guess that's another company to research one of these days...

[NewRisingSun]

Apple Chess was originally written for BBK's keyboard famiclones. Attached please find the "BBK Floppy 1" system ROM and a floppy disk image containing the game. After loading the ROM file in the latest NintendulatorNRS build, insert the floppy disk image via File->Load 3.5" disk. It will boot into a demo application; press ESC to exit it, then type "STA" to run the Beijing XKD 3-in-1 game collection (Solitaire, Tetravex, Apple Chess) from floppy disk. Closing the problem shows a nice 1997 credits screen. The program is exclusively controlled via an emulated PS/2 mouse.

[lancuster]

I wanted to reply in a personal message, but, alas, after writing the message, I received this error in response:

User 'NewRisingSun' has blocked your personal message.

So I will answer here.
Unfortunately, I can't give this version of the cartridge version "Fake News Covfefe" ROM, for the reason that I don't own it myself, it has already been finalized by other people, and also published by other people who didn't share the final ROM with me.
If you're not aware, the hack itself was born after someone on NintendoAge wanted to see Trump's head shooting up the Empire State Building. While everyone was discussing how it would look, I just took an idea and made this hack, and people liked it, they wanted to develop the idea to the end. Replacing the final boss is also their idea. They offered, and then I did. I didn't replace the icons in the HUD and the text in the intro.
The only thing left to do in this situation is to find the people involved in completing the hack and ask them to find the final rom. Alternatively, you can watch the walkthrough of the game and manually transfer the different graphics and text. That's all I can offer in this situation.
PS. if anyone is interested, I'm also simultaneously optimizing graphics in the bootleg "The Dragon". Bao Qing Tian has been temporarily postponed.

[NewRisingSun]

I have not consciously blocked you; instead, the strange forum software, which is highly confusing to me in many aspects, was set to "receive messages from Buddies and Administrators only" because of the many spammers that plagued this forum in the past. You would think that a good forum software would warn you that if you send a message to somebody who is not already on your Buddy list, that somebody cannot reply and so that you need to add him, but no.  Sorry for the inconvenience, in any case.

Thank you for clarifying the situation regarding the Fake News Covfefe cartridge version. I suppose I will just have to find somebody to obtain the cartridge and dump it, then.

[lancuster]

No problem.

By the way, while I was reassembling the graphics in "The Dragon" bootleg, I noticed some bugs and flaws:
1) After the death of the character, the status in the lower left corner isn't restored. Rather, he tries to recover, but literally immediately jumps to another inscription.
2) In "The Dragon", in battle mode, after the death of a character in PAL mode, the HUD breaks.
3) If at the end of level 1-1, having nunchucks, you defeat the last enemy and immediately sit down, then he will freeze in this position, the music will play to the end, and the stage will softlock.
4) After defeating the last boss, there is no final screen, just a transition to the title screen.

[forgotusername]

Here is a custom English translation of Whiz Kid Magic Wand (神童魔棒), a game from the Waixing English CAI 3-in-1 cartridge. Every problem has had its explanation translated at the top of the screen, in addition to menu prompts and a small title logo being added.

Unfortunately, I ran into an issue with the game itself (not as a result of my translation): some of the levels appear to be physically impossible to complete. I was able to verify that at least one puzzle ("Move 6 to make triangles form rhombuses") is broken within the code, going by its mapped-out solution. There are three other layouts that I suspect are unsolvable; though every other level I have verified as beatable. I was unable to find a way to fix or zero-out the broken levels; so if anyone can think of a solution, it would be appreciated.

Download is attached below, alongside information on the broken layouts (with attached FCEUX save states).

[forgotusername]

Apple Chess was originally written for BBK's keyboard famiclones. Attached please find the "BBK Floppy 1" system ROM and a floppy disk image containing the game. After loading the ROM file in the latest NintendulatorNRS build, insert the floppy disk image via File->Load 3.5" disk. It will boot into a demo application; press ESC to exit it, then type "STA" to run the Beijing XKD 3-in-1 game collection (Solitaire, Tetravex, Apple Chess) from floppy disk. Closing the problem shows a nice 1997 credits screen. The program is exclusively controlled via an emulated PS/2 mouse.

Interesting to see and know its origin. I had seen those three game revisions before, but not on one of the proper floppy-based systems. Ironically, I recall having a copy of that credits screen as a .nes file conversion, with nothing else attached to it (probably dug out of NonGoodNES or something); so I had no idea who or what it was actually designed to credit...

[NewRisingSun]

I don't see any VT02 register writes in Pong. I do have problems getting the actual gameplay to start. I can reach the gameplay screen by pressing START, but then, some seemingly random combination of START and A is needed to start playing.

[forgotusername]

Pong only requires the pressing the A button (after the title screen) to start when run as VTxx. This also applies if the game is loaded in a multicart menu; I attached an example using NES Maker. When run as a standard Famicom/NES game, even spamming the A+Start combination results in the game freezing shortly thereafter on my end. I couldn't actually get the 270-in-1 ROM to run, so I have no idea how it behaves on there.

I previously noticed that some of Pong's CHR data is copied from Ding Dong (the "Y2K" Binary Land hack). Curiously, though, the Minesweeper, Dragon, and Solitaire ROMs also have portions of the Ding Dong CHR data; specifically the Hudson Soft byline at the very bottom of the data (Solitaire and Dragon only have the "K" and "E" letters remaining from it). I'm guessing there's some sort of connection here; maybe the "educational computer"-like software was recompiled by the same ex-Subor team as Pong/Free Way?

[NewRisingSun]

I have now spent (too much) time trying to find out why the controller becomes unresponsive in that situation. The answer: The game reads the controller both in the main game loop and in the NMI handler. In fact, the NMI handler's controller read procedure interrupts the main game loop's controller read procedure. Since the controller is read by writing 1, then 0, to $4016, then reading $4016 eight times to get the result of the eight buttons (four buttons and four D-pad positions, actually), this implies that if a controller read procedure from the main game loop is interrupted by a controller read procedure from the NMI handler, the interrupted controller read procedure will read not the 1st-to-8th bit, but a 9th+ bit, in other words, it's basically a garbage read result. The game commits a logical error by having two controller read routines that can interrupt each other.

Now, why does it still work on VT02, but not on regular NES/Famicom? Actually, in NintendulatorNRS, it works on VT02 and with Dendy timing, but not with NTSC or regular PAL timing. Apart from frame timing, NintendulatorNRS will also emulate the controller behavior of a famiclone in VTxx mode and in Dendy mode, while emulating the Standard Controller behavior of a licensed NES or Famicom in NTSC or regular PAL mode. What is the difference in controller behaviors? The difference is that Nintendo controllers return "1" when a 9th bit and above bit is read, while famiclone controllers return "0" when a 9th and above bit is read. The game, despite being buggy in having two controller read routines that interrupt each other, will still work with famiclone (Dendy or VTxx) behavior because it's not bothered by the "0" bit, but is bothered by Nintendo controller behavior, because it sees the garbage "1" bit as a "phantom" START button press.

Long story short: Run the original extracted ROM in Dendy mode, or use the patched ROM that I created, which fixes the double/interrupting controller read behavior.

Edit: Regarding the full 270-in-1, it does not run unmodified because we lack the dump of a 256-byte serial security ROM. When patching out the security checks, it's not a pretty looker, either, as the attached pictures show. The mock-up screenshots in Amazon's item description indicate that my emulated colors are correct.