11 March 2016 - Forum Rules

Main Menu

Japanese text limitations

Started by gloomspirit, July 11, 2014, 11:55:31 PM

Previous topic - Next topic


I'm trying to translate a psp japanese game, and I have found that it uses two hex values for each character instead of one. Tha game supports basic latin so I was able to translate some parts of it, but now the problem is when I want to put larger text than the original.
What would be a way to solve this problem?
Thanks in advance.


You might be running into two slightly different problems

1) Games do not know where new lines or sections of text start, end and otherwise. Typically this means pointers are used and you have to edit them. I usually describe pointers when it comes to text handling like the contents page in a book.
2) You are running out of memory to copy the text into, common enough as most languages we see tend to be slightly longer winded than Japanese.

1) really does amount to figure out the pointer system, screen limits (number of characters per line and the like) and sort it all from there.

2) can be slightly trickier. The first thing most try is typically to shrink their text if they can. The second is an 8 bit conversion. It is exceedingly rare for games not on the PC to have such a well developed text engine that you can do this easily, however if it was to happen anywhere then the PSP would be a place I would expect to see it. More likely though you get to edit the text decoding/handling to use 8 bits. Unless the encoding is put along with the font then you get to play assembly hacker to sort this, it is probably not as hard as the related class of hacks where you put in a variable width font but I would not call it simple either.


The game uses standard SJIS encoding probably. Most people end up converting that to 1 byte ASCII. If you don't know how to debug and hack assembly code, this task will be impossible.


I'm not sure that resorting to ASCII is needed on the PSP, as I expect most games to not fill-in the UMD (once dummy.bin is removed). I'll let the author clarify, but I believe "1" from Fast's response is the most appropriate. We shouls be able to elaborate once confirmed.