ROM Hack Help with Neko Atsume

[Lycan911]

So I'm trying to "translate" Neko Atsume for 3DS to English, but I'm running into a problem.
Graphics aren't the issue and @DarkSynopsis already did a handful of them, but the text is where the struggle begins.
The file that contains the text, "evt00_data.evt", actually contains Japanese texts, as expected, but it also contains a fully localized English script which is unused in the game.
So what I'm trying to do is replace the Japanese text with the included English text.

Buuuut (there's always a but ), it isn't going that smooth.

It is possible to view all of the text in Notepad++ or a hex editor that supports Shift-JIS, but when I tried making a small edit and running it using Luma's LayeredFS, I got an error on the bottom screen as soon as the game went past the "Nintendo 3DS" logo. NTR was the same. So I'm assuming that simply editing the text using Notepad++ or a hex editor isn't enough, so that's why I'm making this post, as I'm not very good at this stuff.

If someone could write a tool that allows for a simpler (and working) way of replacing the text, that would make my life so much easier.

Here are some screenshots or what I did:

Spoiler

Shift-JIS in Notepad++ (no edits shown):

File in the LayeredFS directory:

Crash:

 

[ngtunganh]

Did you try ask the guy that did "Yugioh saikyo card battle" translation?
That game also got fully localize english inside game file too

 

[Lycan911]

Did you try ask the guy that did "Yugioh saikyo card battle" translation?
That game also got fully localize english inside game file too

Uhhh... that game definitely didn't have an official translation in the files, at least not a complete one, as the translation contains some words like "faggot", "lol" etc. Anyway, I'm getting help from Kuriimu's dev @IcySon55, so hopefully this will get sorted soon enough. For now I'm focusing on graphics, which will probably be fully translated by tomorrow.

 

[ngtunganh]

Uhhh... that game definitely didn't have an official translation in the files, at least not a complete one, as the translation contains some words like "faggot", "lol" etc. Anyway, I'm getting help from Kuriimu's dev @IcySon55, so hopefully this will get sorted soon enough. For now I'm focusing on graphics, which will probably be fully translated by tomorrow.

wow that quick ,nice!

 

[Lycan911]

wow that quick ,nice!

3AM
Graphics: CHECK

Spoiler

Yup, all of the graphics are now 100% in English! Gonna release the current progress tomorrow as a beta patch, and hopefully a full patch will follow soon™

 

[justneet]

Aah, so you've gotten the game? I'm glad!! I can't wait for this patch, and I will be looking around for it!!

 

[ylylhl]

Hi! Sorry for bother you after such a long time, could I ask how did you solve this problem? I have the same (or similar) problem when I try to edit this file by Notepad++ (;w;

The difference is that I got this file in another game which is also made by Hit-Point and very similar with NekoAtsume. Besides, I can see original Japanese text by UTF-8, not Shift-JIS (though some words are still not displayed properly), but I don't think encoding is the reason

Thanks a lot! <3

edit:
I do some test and notice 1 char can only be replaced by 1 char (take 2 bytes, means 2 english letters), but can't be replaced by more (or less) than 1 char, why? Is there any way to make it can be replaced by more or less chars? Or, is there any tools can help to add padding characters?

Now my work is to translate them into less or same chars and thinking why I'm doing such a boring and strange work, lol
But how can I deal with it if this word have to be translated more than the original length?
I'm still trying to find ways to reduce my workload XD If you know something, please tell me! thank you =D

 

[Lycan911]

Hi! Sorry for bother you after such a long time, could I ask how did you solve this problem? I have the same (or similar) problem when I try to edit this file by Notepad++ (;w;

The difference is that I got this file in another game which is also made by Hit-Point and very similar with NekoAtsume. Besides, I can see original Japanese text by UTF-8, not Shift-JIS (though some words are still not displayed properly), but I don't think encoding is the reason

Thanks a lot! <3

edit:
I do some test and notice 1 char can only be replaced by 1 char (take 2 bytes, means 4 english letters), but can't be replaced by more (or less) than 1 char, why? Is there any way to make it can be replaced by more or less chars? Or, is there any tools can help to add padding characters?

Now my work is to translate them into less or same chars and thinking why I'm doing such a boring and strange work, lol
But how can I deal with it if this word have to be translated more than the original length?
I'm still trying to find ways to reduce my workload XD If you know something, please tell me! thank you =D

In the end, @DarkSynopsis reached out to me and offered help by writing the tool to edit the files. He didn't share the program with me though, so you could ask him. Good luck with your project!

 

[DarkSynopsis]

In the end, @DarkSynopsis reached out to me and offered help by writing the tool to edit the files. He didn't share the program with me though, so you could ask him. Good luck with your project!

It wasn't a very elegant tool and was still lots of manual hex editing so wasn't really much to share.

If I recall correctly since this game had most of the english text still in the file it was mostly a case of swapping the english text with the japanese text and vice versa to keep the file size the same amount of bytes, very hacky way of doing it.

I guess whatever game is being worked on might not have that luxury, best I can say is try keep the file the same amount of bytes, so translate something to english and cut out some japanese else where, though in the end there might not be enough to fully do that.

 

[ylylhl]

To DarkSynopsis:
I get it! Thank you very much!!

To Lycanroc:
Thank you for your advice! It seems that I can't ping or reply because I'm a newbie... haha. please give my thanks to DarkSynopsis ; )


A new finding (to make this post won't be treat as a spam, why it will be thought as a spam? ) :

in this file, the name chars begin with hex '00 xx 00 xx' (the first xx is a English letter's Ascii, arranged in order) and end with '00 0c'. Besides, it's img file name is followed. for example, strings ".N..xxx..img_x" 's hex is '00 4e 00 xx aa bb cc 00 0c 69 6d 67 5f xx', in which "aa bb cc" is the name's hex, which we should translated. And the next name chars will show as ".O..xxx..img_x", for 'O' is next to 'N'.

(sorry my English is not so good, I hope I express it correctly)

Anyway, I hope this finding can help some of my work. No matter I can finish my work successfully or not, I think I really did learn something : )