11 March 2016 - Forum Rules
Started by markus-projects, September 07, 2020, 01:39:37 PM
Quote from: erinnk on September 16, 2020, 11:16:45 AMMarkus, can you share your extraction code?Without having looked closer, I'm thinking perhaps the text is simply encoded as indices into the font you found. That would make sense. The way to locate the intro dialogue, especially where there's substantial repetition, is to find repeated bytes, I think... especially lines like this: "勇者さま 勇者さま......" There's going to be four bytes (or double bytes) followed by something different, followed by those same four bytes.
Quote from: Guadozoku on November 22, 2020, 01:02:56 PMGotta disagree with using the mobile version text. They translation with its phonetic accents is annoying to read. Either translate from scratch or use the NES version as a base and translate the party chat.
Quote from: markus-projects on November 22, 2020, 07:51:01 AMHi,it's been some time now since I worked on the hacking because of actual work.After I finished DQ III in my leisure time, I thought for a moment why not playing the DS or Mobile version of DQ IV.However, I came this far and I want to continue this hacking journey.I have some important updates in my blog post.The new sections are:Use Dragon Quest VII to find the textText-BlocksThe text decoding routineTranslated Text from Dragon Quest IV Mobile VersionYou can read the details here http://markus-projects.net/dragon-hackst-iv/.TL;DR: I used DQ VII (en + jp version) to identify the block that contain the text; analysed and parsed the text blocks in IV; debugged the text routine in the PSX-EXE of DQ IV and as a sideline found the dialogs (en+jp) in UTF-8 text files in the Mobile version of IV.Next steps and help or advice neededThe text is compressed (no wonder we didn't find the dialog text in Shift-JIS) and it needs to be uncompressed, then translated and again be compressed. I looked into the assembler code to understand what has to be done, but didn't understand it completely. Does this 0x80 pointer mechanism (dataED part) look familiar to someone? Is it some kind of LZ variant or another encoding (see e.g. https://en.wikibooks.org/wiki/Data_Compression/Dictionary_compression#dictionary_algorithms). I guess it should output DAMI- (engl. Dummy) for this example.Any help on this is appreciated.
Quote from: CoolCatBomberMan on November 22, 2020, 03:17:54 PMOr, y'know, rewrite the mobile text to remove the accents. e.g. wee bairns > children
Quote from: nejimakipiyo on March 08, 2021, 05:34:29 PMWould love to be part of working on an accurate and faithful translation. I'm in the end stages of another DQ related project now, and would happily work on this game next. All I would need to get started is an extracted version of the Japanese text, like Chicken Knife mentioned. My only skill is with the translating itself, but I'm sure between the lot of us here we could find a way to get this off the ground.
Quote from: spookdy on March 08, 2021, 12:14:54 PMIf anyone wants to do this work, like I said before, I am more than willing to set you up. You shouldn't need any technical knowledge.
Page created in 0.071 seconds with 20 queries.