Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
Discussions Rules and Guidelines
5 
Report this post
I played some title of this series real time those days. Still I'm keeping this "Legend Pack" (not STEAM version, original one) in my PC, and it may work if be installed again.
Especially about LUNATIC DAWN II, also I'm keeping strategy guide book (for MS-DOS version) in my bookshelf and it may help something.
I think I can help some of your work only in LUNATIC DAWN II just like translation. I played little on LUNATIC DAWN (I) and III.
- Quest text is stored in luna2win.pac.
- 89F3AE is the starting position for a bit of quest text.
- In the .pac file, you have 14 D8 50 96 1B F4 52 EC.
- Once the game decodes this, you get 8E 64 8E 96 81 48 8C EC.
- So basically, every fourth byte is already encoded correctly, and then the first through third bytes are getting substituted with something.
- I suspect this is a Vigenere cipher with a four-byte long key. I'm not that familiar with how this works tho, so I could be wrong.
- Also UTF-8 encoded text seems to work fine once you get around the cipher.
- Not clear how other text is encrypted. There's a lot of un-encrypted text in the .pac file too (lke item and monster names).
---
Not Vigenere. Looks like it might actually be a XOR cipher?
---
Looks like XOR cipher with this key:
10011010 10111100 11011110 00000000
Or 9ABCDE00 in hex.
Gonna leave this alone for now but I'll try to write a script dumper this weekend. Also wanna see if The Atlas 1/2 do something similar (Neo Atlas 2 is amazing so I really wanna look into the earlier titles!)
1. Line feeds in the input text are represented as 0D0A in hex.
2. As a general rule, once you decode something to 0D0A, you want to restart the cipher key back to the first byte (9A). Note that this is something that DECODES to 0D0A, and not 0D0A existing as such in the luna2win.pac file.
3. One major exception to this: If you get something that decodes to 0D0A, followed by 80 (that is, 80 in the luna2win.pac file, NOT something that decodes to 80), the cipher key goes to the 4th byte, meaning that 80 decodes to 80.
Not 100% sure this is correct but it LOOKS right. I'll make a new thread for the dumped script.
Lunatic Dawn 2 uses Shift-JIS font, right?
Can I change its japanese font graphic?
I just want to translate it to Korean.
(I have all korean script of luna 2)
Ascii font is included in Shift-JIS char set but Korean is not.
Now i'm fining font file with hex editor,
but it is very hard to find.
if you can unpacking luna2win.pac,
please tell me font file is excist in pac.
If the font file is in luna2win.pac, i'll change it to Korean font gradly.
Hi byunja.
Regarding luna2win.pac, the sequence of bytes 0D0A80 seems to indicate the end of the file. You don't strictly need to unpack the .pac file, but if you do, that's what I believe you'd want to break it up on.
All lines of text within a given "file" in the .pac file, just so you know, need to be the same length. in bytes. So if something ends up shorter than the original japanese text, you'll need to pad it with nulls.
https://www.dropbox.com/s/k6i1nfvieowu937/LD2TextDumper.zip?dl=0
Otherwise, here's the text dumper I wrote. It outputs to unicode iirc. Writing something to re-encodes it back into Korean should be pretty simple, though I'd recommend trying to find a character encoding that didn't use more than four bytes per block of hangul.
If you go through the source, you should be able to figure out where I started reading the text from, what the decryption key is, etc.
I never tried reinserting the text back into LD2, but for LD1 and The Atlas 2, the English text was encoded in UTF-8 and it worked fine. I think as long as the encoding is recognized by either Windows or your locale emulator, it SHOULD be fine.
Also doesn't the game already have an official Korean translation..?
https://www.dropbox.com/s/do8dwq0n7tyqcvp/korean%20mojibake.png?dl=0
Here we go.
This is the result of running the game in AppLocale, using Korean as the language. That's ultimately just mis-encoded Japanese text, so meaningless, but it shows that the game can display Hangul just fine without any modification.
https://www.dropbox.com/s/do8dwq0n7tyqcvp/korean%20mojibake.png?dl=0
Here we go.
This is the result of running the game in AppLocale, using Korean as the language. That's ultimately just mis-encoded Japanese text, so meaningless, but it shows that the game can display Hangul just fine without any modification. [/quote]
---------------------------------------------------------------------------------------------------
Thank you a lot!
I'll start translation.
P.S.
Official Korean translation is already excist for DOS version
but it is trash.
1. It has many bugs. It occurs much lags in battle scene so
player can't any battle. Or pay at least 5~10 mins per battle.
(Just only run away from battle!)
2. Korean version is translted from Taiwan version (translated from original).
The client files are very broken in the process of being translated twice.
Some graphics are deleted and images are blining in plays.
3. Translation is very garbage.
Big bird Roc was translated into 'Big Ant'.
Hellhound was translated into 'Medusa'.
In Korea, Lunatic dawn is not famous.
I think the reason is due to disgusting official translated version.
I just want to translate this legend pack LD2 for Korean
and enjoy that.
Not sure if this is still true in LD2, and I'm assuming Korean encodings use something else for space, but if you start seeing line breaks or the text ending where you don't expect them, that's probably why.
Addressing this in a potential English version of LD1 would take some assembly hacking.
Do I have something to do?
As you can see, my English is not good.
I would be glad if I could help.
AppLocale does not working.
Does not work in windows10 environment or
my appLocale setting is wrong.
I'll test in another windows8 PC.