Qhimm.com Forums
Miscellaneous Forums => Scripting and Reverse Engineering => Topic started by: cromas on 2007-04-15 14:07:50
-
Sorry if this is covered somewhere else, I had no idea what to search for.
As we've been discussing in thread 6537 (http://forums.qhimm.com/index.php?topic=6537.0), I'm studying in Japan right now and thought it would be nice to play through FF7 in Japanese. Sadly there was never an FF7PC in Japanese, which limits me to the ugly, low-resolution PSX version. Boo.
kini had the idea to extract the text out of the PSX version and load it into the PC version. I could see a few potential issues of this, given the surely-double-byte and possibly proprietary encoding of Japanese text on a PlayStation, but I thought I'd ask. So, does anyone know how I might go about extracting the game script from the PlayStation game discs?
-
One of the problems is that we don't have the Japanese version of the game. I'm going to guess that the whole Japanese text rendering subsystem has been deleted in th U.S. version of the game because the character page space in VRAM is never used. The characters themselves are probably banks in winodow.bin.
What would be absolutly awsome is if you could do some VRAM dumps of the Japanese version using an emulator. Also some TIM rips of an uncompressed kernel.bin and window.bin wold be cool. The words are most likley written in kernel.bin, and in each field file in the text block...
The problem is you can't edit the files because the PSX version has the file locations hardcoded for lookups. Moving one file so much as a byte with break the whole game.
-
The problem is you can't edit the files because the PSX version has the file locations hardcoded for lookups. Moving one file so much as a byte with break the whole game.
Yes, but he wants to edit the PC version and only extract the data from the PSX one, so this problem doesn't apply here. (But there maybe enough other problems. ;-) )
-
Sorry... confused.
There are no Japanese characters in thr PC version of FF7. There is no way to render the text. Even if you inserted the Japanese text, the PC version only works with single-byte charatcers. Sorry ;_;
It would be nice to see how the PSX version rendered the Japanese though... We could use that for Q-gears.
-
Sorry... confused.
There are no Japanese characters in thr PC version of FF7. There is no way to render the text. Even if you inserted the Japanese text, the PC version only works with single-byte charatcers. Sorry ;_;
It would be nice to see how the PSX version rendered the Japanese though... We could use that for Q-gears.
Gemini (translated FF7 to Italian) informed me that the remnants of the Japanese Encoding are in the English version. Some of the codes that didn't seem to be used in the English are in the Japanese. These are MAP codes IE they change the map the symbols come from (or symbol page?) So in summary the Japanese encoding is in there and the japanese symbols removed that's about it.
Cyb
-
Wow lots of interesting comments, thanks everyone. The FF7 Japanese discs arrived yesterday. Here's the contents of kernel.bin and window.bin. I was only able to find three tim files in window.bin, i can't figure out what the fourth 2kb file was. FFText was useless on the kernel.bin files because...well, they're not encoded in english. Anyway, here's a couple of zip files.
http://students.washington.edu/ckovatch/ff7/window.zip (http://students.washington.edu/ckovatch/ff7/window.zip)
http://students.washington.edu/ckovatch/ff7/kernel.zip (http://students.washington.edu/ckovatch/ff7/kernel.zip)
The contents of 2.tim and 3.tim are most interesting, but will be difficult to understand unless you're familiar with kanji. halkun, I'm hoping that your avatar guarantees me this assumption :-D
2.tim has the standard Hiragana/Katakana/Romaji/weapon images/etc. stored on the odd-numbered CLUTs (the first CLUT being #1), but on the even-numbered CLUTs, there's an entirely different Kanji set stored on the upper half. Interestingly, the 11th row of Kanji is cut-off midway. This may be the fault of my TIM viewer more than the file itself.
3.tim is similar, with a Kanji/romaji split on the odd-numbered CLUTs, again with some evidence of overwritten Kanji present. Even-numbered CLUTs have a whole bunch of kanji.
I also checked window.bin from FF7PC, which has only three files, two of which are TIM. The Hiragana/Katakana (Japanese syllabry) is present, but the kanji is not. I'd be interested in looking at window.bin from the US PSX FF7, but my discs are in America :-P
What TIM viewer/converter do you guys use? None of the ones I've found seem to handle the Japanese TIM files properly.