Wizardry DS pointer/table help needed

Discussion in 'NDS - ROM Hacking and Translations' started by AshenPhoenix, Apr 14, 2010.

Thread Status:
Not open for further replies.
Apr 14, 2010
  1. AshenPhoenix
    OP

    Newcomer AshenPhoenix Member

    Joined:
    Mar 14, 2009
    Messages:
    27
    Country:
    Sweden
    Hi

    I was curious how easy this game would be to translate since all the text is in ordinary text files, but I'm having trouble figuring out
    how the pointer files are built up.

    Here's a text example:
    [​IMG]

    and here is the table file for it:
    [​IMG]

    Hoping someone can see some connection between the files [​IMG]
     
  2. rastsan

    Member rastsan 8 baller, Death Wizard

    Joined:
    May 28, 2008
    Messages:
    963
    Location:
    toronto
    Country:
    Canada
    check out the rom hacking wip here on gba temp or type pointer in the search on hte top of the page, I know not easy to follow but you will get it. Basicly its a number that points to the text so relatively speaking where is the text in the rom or in the single file sometimes the pointers will be in the file with the text and sometimes not so...
    so look in hex for a number that pints to the location of the text
     
  3. AshenPhoenix
    OP

    Newcomer AshenPhoenix Member

    Joined:
    Mar 14, 2009
    Messages:
    27
    Country:
    Sweden
    Mmm... I've read that long ago, and I've worked with pointer tables and made a text extractor/inserter with pointer recalculation for another game.
    I'm just stumped as to how this table worked. But thanks for trying to help [​IMG]

    And to clarify, these are two separate files, and their entire contents are as seen on these screen shots.
     
  4. FAST6191

    Reporter FAST6191 Techromancer

    pip
    Joined:
    Nov 21, 2005
    Messages:
    21,746
    Country:
    United Kingdom
    Very odd table "unicode" 0D0A-> 000D 000A for what I guess is a new line.

    Pointers, I find it easier if you change the hex editor so the column width is approximately the same as the pointers (I would say exactly but sometimes the header means the pointers can end up on different lines).

    Still trying to see the end of line/section marks in the text. I assume it is not compressed or it has already been uncompressed. Several candidates in the the new line, 00 09 00 or my personal favourite is 5c 00 (in your text conversion it comes out as a yen sign with a subscript p) as it seems to correspond the roughly arabic numbers counting up (I count nine such incidences) and when they do not appear the text is fairly short (all able to appear on one screen?), I will also note the " appearing alongside the yen sign and in some of the smaller sections as well.

    The pointers are not plain leading me to call either relative or offset type and more likely relative. By the looks of the repeating numbers the pointer file may also house some text formatting/ownership as well (another reason for lining things up).
    Going the other way and looking at the pointers the text file finishes at 2CD, I see nothing like that in the end of the text


    0086
    00CE
    0128
    0163
    01AA
    0203
    0244
    0270
    02BA

    Relative differences between them are
    48
    5A
    3B
    47
    59
    41
    2C
    4A
    Nothing obvious there.


    Ignoring the header (ASCII section) of the pointers gives about 170 (C5-1B) to play with- assuming 168 that means for the 14 "sections" in the text there are 12 per pointer although this can narrow if there are other things in there.

    I shall have to play around with this one further.
     
  5. AshenPhoenix
    OP

    Newcomer AshenPhoenix Member

    Joined:
    Mar 14, 2009
    Messages:
    27
    Country:
    Sweden
    Here's a shot at how the txt file looks in JWPce:
    [​IMG]

    Saving in JWPce ruins the file though, since for some reason it converts spaces to what the game uses for linebreaks or something. ie 000A is turned into 000D 000A. Trying to run the game after this will only give you the first line of text, then a bunch of empty text boxes until the next text file. Adding text manually with a hexeditor works just fine though. Not sure if CT2 can insert/remove bytes? Wanted to try adding characters manually since JWPce is messing up text formatting. And all text files end with an empty section like section 14 in this file.
     
  6. rastsan

    Member rastsan 8 baller, Death Wizard

    Joined:
    May 28, 2008
    Messages:
    963
    Location:
    toronto
    Country:
    Canada
    try emeditor for editing it then it has an autodetect function and usually doesn't mess thing up.
     
  7. AshenPhoenix
    OP

    Newcomer AshenPhoenix Member

    Joined:
    Mar 14, 2009
    Messages:
    27
    Country:
    Sweden
    Hmmm, not sure really what those table files are for. I left the table file unedited and went ahead and "translated" the beginning dialog.
    I don't know japanese though so it's google translate and some editing.
    I thought it would crash but as long as I keep the same amount of lines, it works.

    [​IMG]
     
  8. Faramir

    Newcomer Faramir Newbie

    Joined:
    Apr 17, 2015
    Messages:
    8
    Country:
    Russia
    Nobody engaged in the translation? Maybe it is better? ;)
    [​IMG]
    [​IMG]
    [​IMG]
    [​IMG]
    [​IMG]
    [​IMG]
    [​IMG]
     
  9. BORTZ

    Global Moderator BORTZ wtf, nintendo

    Joined:
    Dec 2, 2007
    Messages:
    10,652
    Country:
    United States
    Thread died 5 years ago, please be mindful of where you post.
     
    nxwing likes this.
Thread Status:
Not open for further replies.

Share This Page