Jump to content

frc_

Members
  • Posts

    59
  • Joined

  • Last visited

1 Follower

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

frc_'s Achievements

Fuwa Regular

Fuwa Regular (3/10)

46

Reputation

  1. I'm dropping this project. My apologies to anyone who was looking forward to it. Because the common route hasn't been edited, I won't release a partial patch (it wouldn't be enjoyable to read). Conversely, the editing in the heroine routes is so heavy (often replacing entire sections of narration by original text) that I also won't release the translation for others to pick up. If this game ever gets a full translation, it should be a consistent one.
  2. Not your browser, but your antivirus is blocking the download. And yes, false positives are to be expected with a program like KrkrExtract - after all, the very feature which makes it work so well (messing with the game's code to make it decrypt its own files) is similar to what viruses do. You can either temporarily disable your antivirus or ask someone else to do the extraction.
  3. None of the standard obfuscation methods then. Have you tried extracting with KrkrExtract instead of GARbro?
  4. Does the file start with "TJS2" when you open it in a hex editor? In that case, it's not encrypted but compiled. As far as I know, there's only a disassembler for this format (giving you the list of virtual machine instructions), not a decompiler (giving you an approximation of the original .tjs script). That said, only few Kirikiri applications compile their TJS scripts; the only ones I know of are Nekopara and Emote Movie Maker. If the file starts with the bytes "FE FE", though, it's a plain text script that got scrambled using one of the three algorithms supported by Kirikiri. I have a program that can reverse these, so let me know if you need it.
  5. Finding the .ybn decryption key is trivial: if you know the correct offset, it's right there in the file. No need to calculate or bruteforce anything. Here's a zip with the scripts in an Excel file and a tool that'll put the translated text back into the .ybn files: https://www.dropbox.com/s/lbhwyrt7lio0zd0/Yome juu tools.zip?dl=0 The patched files get placed in the "patch" folder. Copy those to a new "ysbin" folder in the game and you're all set. Some more remarks about the tool: It handles word wrapping automatically by inserting line break instructions. You may need to adjust the "MonospaceCharactersPerLine" setting in tool\textpatch.exe.config to get this working correctly. If you're planning on working on this translation with multiple people, it's also possible to create a translation patch directly from a Google Docs spreadsheet rather than an Excel file. PM me for details on this.
  6. Yep, the binaries are available here: https://github.com/xmoeproject/SiglusExtract/tree/master/binary
  7. Shigure is probably just biding her time. In ten years or so, the feline competition will disappear all by itself. Ahem Coconut is pretty cute I guess.
  8. @Ulysses My own converter uses FreeMote to go from PSB to Excel/Google Docs and back without a JSON step inbetween. I've found the library works on scenario files which other tools crash on, so if you ask me, it's the best .scn patching solution available. The converter is currently not public, but is being actively used in a number of translation projects - I'll probably throw it on Github at some point.
  9. Hi, I'm the one person who's too stubborn to stop working on this project. As you can see at the bottom of the progress page Fred linked earlier, it's indeed still alive. In fact, I've been working on it almost every day since I joined it in September 2016. (The reason the progress page currently says "Last updated 7 days ago" rather than "yesterday" is that I only upload my changes when I complete a script) Progress is indeed slow, but you have to take into account that apart from translating, I'm also already doing a quite thorough editing pass. It takes time to come up with sentences that are natural, clear and concise while also having character, and it takes even more time if you're not a native speaker. Sometimes I'll do fifty lines in an hour, and other times I'll only manage three . This is also the reason I don't normally post in here - seeing as it'll take several more years for this project to reach completion, there's really no point in trying to keep people excited. I'd rather have them forget about it and then surprise them when it's done. If you want to chat with me on Discord, you can usually find me on the Visual-Novel.info server where I'm known as "arc_".
  10. It's an introductory course aimed at first-year students who may not have any programming experience at all. Putting something scary like "reimplement assembly code in C++" in the description would risk driving away potential applicants, right? For their first foray into reverse engineering, it's better to make them acquire reference source code and express the knowledge they gained in a language suited for beginners. Why not all of them at once? The MC joins the reverse engineering course, poses for (or with) his artist girlfriend, and helps his literature geek sister with her homework.
  11. 1) Visual Novel Reverse Engineering 101 2) The most common x86 instructions, how registers and memory work, how to use a disassembler, examples of file formats found in VNs. 3) Field trip: breaching the head office of a visual novel company and lifting their source control server while holding the CEO's daughter hostage. (Different company each year) 4) Writing a converter for a proprietary image format in Python. 1) Restoration of Hand-drawn Japanese Media 2) Methods for accurately estimating how a given damaged area originally looked like and faithfully restoring it. Examples of particularly tricky restoration works carried out by famous experts in the field. 3) Getting naked and acting out scenes in groups to take pictures for tracing. 4) Retouching a lightly damaged artwork. 1) Applied Localization of Bonding Segments in Visual Novels 2) Refresher on the various tropes and clich├ęs found in Japanese romance scenes. Examples of common inaccuracies and exaggerations to be aware of. Techniques for improving upon overly repetitive or graphic descriptions. 3) Playing a segment on the projector and discussing how it could be rewritten to be more realistic and engaging to read. 4) Screwing a sibling of choice and recording the act for reference when translating intrafamilial bonding scenes. (Only children can ask their parents for help)
  12. That's an easy question with an easy answer.
  13. I'm glad you specifically mentioned "essence", as in some cases, the author is simply not very good (at least by Western standards) and you do need to make quite some changes to get an enjoyable result. The visual novel I'm currently translating suffers from, among others, the following deficiencies: Repetition of random words like "city" up to six consecutive times. Narration that's redundant with preceding dialogue (e.g. a character saying "I don't know" followed by narration saying "Apparently she doesn't know"). Narration that's redundant with itself (two consecutive lines that say the exact same thing in a different way). Thoughts and dialogue that jump back and forth between topics without any sort of transition. Overly vague lines ("All of this happened because he's like that." Like what?) Overly detailed lines ("I hug her from the front"). Lacking vocabulary, especially in H scenes (everything is "happy" and "feels good"). It's problems like these that make you realize the source text is not holy and can - no, should - be rearranged, rephrased, and yes, rewritten to a certain extent. Clarify those vague parts. Cut those redundancies. Reuse space taken up by pointless filler to insert a badly needed topic transition. Even some actual events in the story may need to be patched. Imagine the following: you're nearing the end of the route. A huge crowd has assembled on the central square to celebrate the MC and his girlfriend. In the middle of thunderous cheering, they draw closer and... she kisses him they kiss each other. Seriously, you can't have such a scene end on a one-sided kiss, especially not if the whole point of the route is that the characters become equals.
  14. Oh, I forgot my biggest gripes: Less invisible characters. Not drawing so much as a portrait for a character the MC talks with for ten minutes is plain lazy. Less background reuse. If a route ends with a huge gathering of people in the streets, showing me the same sparsely populated road that I already saw on day one is lazy and disappointing.
  15. Less faceless characters in general. They freak me out. Less MCs with dicks the size of their upper arm and balls with the capacity to coat a girl in semen five times over. Less overly long H scenes that describe every last thrust, squelch and splort in excrutiating detail.
×
×
  • Create New...