Chrono Cross Translation Tools by Terminus question

[Psyladine]

I've tracked down the accent files to accents.txt under 40_textes_avant, and some masters on the letter substitutions in auto_accents.txt under 42_tables_avant, but both files make mention of suffixe, which I must assume is French for suffix, and is called for under quite a number of pre-form text files in ScriptC_CD1.txt, but I can find no information on suffixes themselves.

I don't know if Terminus stripped out the information when formatting his accents.txt files for inevitable frenchification, as he only notes in the accents_infos.txt under 41_textes_apres (i.e. after translation) that Leah has "-um" added. No other logic I can see in the files seems to account for this.

The only other forums that were active on this subject were italian & spanish/portugese translation efforts but none of them covered this area, so I feel like I'm missing something important or there is a dead end as regards the suffix translations.

Anyone have any experience with the translator tool or any insight into these processes?

[yugisokubodai]

I was translating this game long ago, but I didn't use the tool.
Yes I translated all the accent, which the files in another laptop of mine.
Let me check the files again.
January 07, 2017, 05:34:01 PM - (Auto Merged - Double Posts are not allowed before 7 days.)
I looked at the old files and here're somethings about the accent:

- In the file 0000.out
- Starts and ends at ($515A3, $515CE)
- Each accent has pointer point to it, from 51572 to51E9C

[Psyladine]

Hey thanks yugiso, Something about that just sounds right. One question for you; what folder is the 0000.out file in on your version? I can't locate it and searching the .out extension is yielding thousands of txt docs from 05_Infos that list pointer directories.

Also do you have a hex editor you recommend? Sublimetext is fine for viewing but not so easy on the eyes with uncommented assembly.

[yugisokubodai]

The data lay in my old laptop and I can't access it right now.
As I remember, this game uses LZ7 compression or something.
After extracting the file, you'll get many folders containing text data, 3d model,... and among them is out.000 for the accents.

Any hex editor out there will work.
I think the hardest part of this game is the compression. If you can solve it, all will be done easily with a little of Asm knowledge without using the Termimus' tool.

Let me know if you have any trouble then.

[Psyladine]

Interesting, I wasn't able to find the file via Terminus' translation tools, but by using the R-Y Tool by Ramsus (developed from the original Yazoo tools which also branched into the Terminus tools I was using), I was able to locate the file, and via hex editor find the offsets as you described.

http://imgur.com/a/vquJr

Unfortunately decoding the compression, as you indicated, is the real challenge. I wish any of the tool makers had put in something like documentation considering the enormous labors they spent in decoding the square compression algorithm.

[Gemini]

From what I can recall, the compression is nothing more than a regular LZSS that encodes everything as bit streams, rather than groups of data with a byte functioning as flags indicating literals and compressed.

[yugisokubodai]

Yes, the compression is not that hard but is enough to keep you busy for a while.

This is my test on the accents.

https://www.youtube.com/watch?v=4mVn7q6jteI

[Psyladine]

Maybe for you :-)

The author of the original tools my thread refers to posted in the chronocompendium on the lSZZ compression in CC:

>CC LZSS is rather different than other implementation of LZSS. Usualy, you have a byte giving you the compression state of the next 8 blocs. In the case of CC, you just have one bit giving you the state of the following bloc.

[Gemini]

Still, exactly the same concept behind LZSS (seek back, read from buffer, write to result). There are several variations of the same algorithm, so no such big deal.