K-sama wrote:

The isos don't seem to have been properly dumped. They should have a size multiple of 2048. This issue makes them incompatible with the other emulator, yabause.
Anyways i posted the fix here in last post: http://yabause.org/forums/index.php?act … ;page=-1#8

Just in case someone wants to play utena with yabause, which is multiplattform.

I have the same problem you mentioned (also on a Mac) but I haven't been able to figure out how to use your suggested solution! I got a hex editor, but I've tried editing the files with extra numbers as your suggestion, and nothing seems to get it to work.  I know it's been a few years, but hopefully you have reply notifications set up for this thread...I could really use some some help.

Thanks so much Giovanna for posting the isos!  I really hope I can managed to get them to work. It's been 6 years that I've been trying to find these files! XD

I'm suprised too that not one's attempted to translate this game yet but hopefully I can remedy that as I'm a fan of Utena and would love to see this game get localized.  I came across the Let's Play posts on here and I think this is the best site to post to to ask if anyone with knowledge of Japanese would like to help out in translating this.  If nothing else, dalbun, would you mind if I used your let's play notes to translate this game?  I"m posting here rather than a wanted ad on Romhacking.net as I figure there'd be more interest here.

So far I've ran through and dumped all the scripts for each scenario plus the testing scripts they left on the CD (which may or may not be interesting to translate, figured why not though since I dumped everything else).  I haven't extracted the videos yet but will look at doing that later on.  I'll admit now the videos could be the harder part for me as graphics are not my strong point at all but I figure they'll probably be a format that can be extracted already.  Worse comes to worse I'm quite familiar with assembly hacking so can probably code something from there as well as looking at Yabause's code.

If anyone's curious, here's a sample of the first script extracted from the *.SCN files on the CD:
DAY_A0.SCN.TXT
I haven't found out the convention for displaying the names yet so no names are added to the script.  In all honesty though if it's too much of a pain to link then I probably won't add any.

And my code that can be used by any programmers to extract this as well : utena_extract.cpp
(note, my newest code is uploaded yet as I'll be doing that later on tonight, along with the rest of the SCN files.

And later on when everything is uploaded you can check this folder for it all : revolutionary_girl_utena
(The repository all of this is hosted on is open to the public, it's run by myself and another where I do the hacking and and he the translating.  He may or may not look at this one eventually based on time and his own interest.  If your curious to see what else is in there feel free to go to transprojects)

So yeah, if anyone's interested let me know.  As stuff is translated I'll post everything in the above repository along with partial patches.  Or if we want to host the files elsewhere to be translated wiki style as some projects do, that'll work as well.

Yup, already tested.  I can post pics as well if anyone's curious.

The only problem like I said may be the videos which will need to be dumped and subtitled.  Subtitling isn't hard, just the conversion to something editable such as an avi and then back to the proper format may be but I enjoy a good challenge so should be fun

Awesome!  That means that it should be a pretty conventional format which should hopefully mean easy conversion between the two.  So no worries there then.  The only thing also I need to look at is the buttons for title screen which need to be dumped but those shouldn't be too hard.  Thanks!

Hey, awesome, esperknight. The samples look great, lack of formatting notwithstanding. I think the only reason nobody's tried this before is because we've never had a convergence of someone who's up for translating, and somebody who was confident working with Saturn ROMs. The Utena fan community isn't that huge, unfortunately...

But yeah, I'd be happy to help translating the rest of the game. It was pretty fun, and half the work is probably done already. Besides, I'd want to go over my Let's Play translations once more before you used them, anyway - I was pounding them out without a whole lot of proofreading, and made distinct errors sometimes. (For instance, I totally misunderstood the Shadow Play the first time through.) Also, my translated dialogue tends to sound rather stilted, so maybe I could smooth it out a little.

Having all the scripts ripped will make things much easier. Not knowing who's saying what could be annoying, though - hopefully it will be clear from the context, but if you can figure out how the names are coded that would be cool. And I don't suppose there's some easy-to-understand way that the lines of the script are connected to the sound files, such that we could pull up a given line to hear the characters' tone of voice? That's probably asking too much.

I'm willing to tackle the rest of the script myself, but that might make me the bottleneck on the project - I'm not certain how much time it would take to work through. If other translators want in, no doubt we could work out some division of labor!

As for uploading the translations - I guess it would be easiest to just take the script files, and add in the English translations under each section of Japanese? That way the original would still be there for reference until you go back and format them properly.

Thanks again for setting this up. Hopefully we can make it happen.

Esperknight's friend here. I'd be happy to split the script translation duties with you. I know what it's like working on a long script and I'm happy to collaborate. The only issue with multiple translators is sometimes a little more editing work trying to get the tone consistent.

One thing that may help with that is a translation aid/tool that I coded that we could enter character/place names into and things like that for consistency. Once they are entered into the translation memory, they come up each time you look-up a line with the same words in it again so we'd only have to make one canonical list and enter it once.

Also I'm working on a bunch of other projects but I should be able to work this in between. I probably owe EK another high priority title once Sailor Moon R is finished.

If you need the video subtitled I might be able to give you a hand. Even though there were several hundred individual video files, only a few contained speech. One thing to keep in mind is that the video is blurry and small and adding subtitles might make it look worse. I'm also not sure how to convert the AVI to the CPK that the game used. I'll try poking through all of your information but my programing is not very good.
edit
oops
turns out there are about 20 on the first disc and 100 on the second. Miscounted the total number of files.

Last edited by gpink (06-18-2011 08:56:26 PM)

filler wrote:

Esperknight's friend here. I'd be happy to split the script translation duties with you. I know what it's like working on a long script and I'm happy to collaborate. The only issue with multiple translators is sometimes a little more editing work trying to get the tone consistent.

Also I'm working on a bunch of other projects but I should be able to work this in between. I probably owe EK another high priority title once Sailor Moon R is finished.

Cool, I'd appreciate it. Don't push yourself too much, though, if you have other projects. If you want, I could go through the script files and insert in the translations I've already done, revising a little as I go, and then we can see what's left. I should have a fair bit of time I could set aside this week. Alternately, we could just to something simple, like I take the first disc and you take the second disc? I've never collaborated on a translation before.

One thing that may help with that is a translation aid/tool that I coded that we could enter character/place names into and things like that for consistency. Once they are entered into the translation memory, they come up each time you look-up a line with the same words in it again so we'd only have to make one canonical list and enter it once.

I don't completely follow how one uses the program, but a tool to coordinate terms sounds good to me! I guess we also have to make some basic decisions, like what order to put given and last names in, and what to do about honorifics... (I tend to leave them in as-is, but then, I'm lazy. )

esperknight wrote:

Far as the conversion goes, does it already look blurry on the Saturn?  I haven't see any videos except the beginning which looked fine I believe...

Yeeeah, I'm afraid they're kind of grainy. They don't look as good as Sakura Taisen, for instance.

And wow that's a lot of videos!  Thankfully there's only a few we gotta do then.

If I remember right, there aren't that many with dialogue - a voiceover when the protagonist arrives at Ohtori; Chigusa and Utena exchanging a few words before their duel; Anthy summoning the Sword of Dios; and maybe Akio and Anthy have a line each during their brief transitional scene. Maybe Touga introducing the Akio Car. Most of those files should be dialogue-less duel segments, character-introducing panning shots, and the like.

On top of those, there are at least two songs that are matched to video - Zettai Unmei Mokushiroku, and the opening song. We can probably get away with not subtitling ZUM if we don't want to, since fans know it well enough. I'd say we should probably do the opening song if we can, though.

I'm not sure if there's any way to subtitle the duel songs. In most of the character paths, at least, Gertzen's Head is sort of a background music, that isn't synched to the video. It might be different with the Utena ending? Or maybe that's where The Inversion of Me and My Room comes in? I haven't actually played the game that thoroughly. Anyway, if we can't subtitle them, we can just include translations, Verthandi's or our own, in a translation notes text file.

Last edited by Dallbun (06-19-2011 09:11:54 AM)

Cool, I'd appreciate it. Don't push yourself too much, though, if you have other projects. If you want, I could go through the script files and insert in the translations I've already done, revising a little as I go, and then we can see what's left. I should have a fair bit of time I could set aside this week. Alternately, we could just to something simple, like I take the first disc and you take the second disc? I've never collaborated on a translation before. emot-smile

It's all good. It's not like anyone is standing over me telling me I need to finish X project by Y date. That's why it's a hobby.

I think you have the right idea about splitting the script. Actually, what I did on the last script I split was to break it into quarters and assign each (of 2) people 2 alternating quarters. So person 1 takes quarter 1 and 3, and person 2 takes quarter 2 and 4. That way if they each start with their first quarter the first half of the game will be translated by the time they are done, then they do the second half. I don't know if that matters but it seemed like kind of a neat waypoint to set. These scripts are so big, sometimes it really helps to have intermediary goals and some feeling of accomplishment as you go.

I don't completely follow how one uses the program, but a tool to coordinate terms sounds good to me! I guess we also have to make some basic decisions, like what order to put given and last names in, and what to do about honorifics... (I tend to leave them in as-is, but then, I'm lazy. emot-redface)

I'm kind of evangelizing the tool right now. I'll probably make a video about how to use it but it's still in alpha so it's a little quirky but I think it's pretty easy to use. I use it for all my translating and it's really handy. I think you'll appreciate it. I'll give you the quick explanation.

It's kind of like a cross between a translation memory and Google translate. The work flow is to copy paste maybe 3 to 20 lines of text into a text area and hit submit. Then the program generates a suggested English translation for each line. You edit the suggested translation via a combination of choosing definitions from dropdown lists, selecting words to use that the translation program may not have chosen properly, and manually editing the line. Then you submit and the translated output is displayed in another text area. The output is recombined with any non-Japanese text that appeared in the original, so it can be really handy if there is a lot of control codes to save you time and headache on formatting. Copy paste your results back into your file and repeat.

The cool thing is that since it's a translation memory too, you enter in your translations for words or phrases to help you build your translated sentence. The more you enter in, the less you have to translate in the future because it will bring your translations up again when you translate that same word or phrase. Like Google translate, it comes up with comical translations much of the time, but a few moments tweaking the translation and you're good to go, with the benefit of having made a deposit toward your future translations. It saves a lot on doing look ups in dictionaries and the more you use it the better it becomes, especially for a given script. The genre and writing style tends to lean towards a certain lexicon and once you've translated the first 1/4-1/3rd of a script, you notice that it starts getting even more accurate on the remainder.

Ok, I guess that was a little long. ; If you PM me an email address I'll set you up an account so you can check it out.

As for honorifics I'm fine either way. We should probably default to community standards. If Utena fans tend to like honorifics, keep them in. I mostly use Western order of given name, family name, but again, I'd look to community standards. If everyone calls her Lina Inverse, changing it to Inverse Lina probably isn't a good idea (Sorry, Slayers fan).

You're right then, it may be a bit harder to translate it if it comes in game like that then.  I'd say best thing for that one is to just have it included in a text file.

The players is marked as <Player> I think... but I do have those marked.  It just happened to be that for some lines they went beyond the range of what I had found as I thought I had them all.  It looks like I may be only missing 3 I think... but not sure.  I'm hoping that these also don't change between days or discs but we'll find out

For the choices... you know I didn't even realize they didn't get dumped.  Interesting... could you email me some samples if you don't mind?  If they take a bit to get to though we'll grab them after inserting some of the scripts then.  I have a feeling I skipped them by accident as they probably get tacked onto the sentence somehow... we'll see though.

For the formatting, don't worry about it.  The notepad issue is due to the line endings I use.  I think I normally use Unix style while Notepad wants Windows style or I may mix them by accident...  I use EditPad Lite for all my editing and notepad gets used for scratch stuff.  As long as you can open it though it shouldn't matter for us. 

Thanks once again for working on this!

Please don't put any effort into extracting the video. I've already done it using a batch file and some manual work to combine the video and audio together. The source for the programs was included for one of them. I don't remember specific detail but I think I posted them in this thread. I have the video as .avi files. The files should be the same quality since I avoided re-encoding. If the link I posted in this thread is dead please let me know and I'll get them to you.

For the avi > cpk I think the only tool I found was on the official SDK disc provided to Saturn developers. If I can find the source of an cpk > avi program you might be able to come up with a program.
try this thread
http://www.segasaturno.com/portal/panze … t4145.html
this link appears to have specifications
http://segaxtreme.net/community/topic/6 … vi-to-cpk/

The tool that had source code is called "CPK Viewer". It is several different tools. The source code contains Japanese text and seems to be encoded in shift-JIS. I would recommend that you take a look at "acpk2avi.cxx".
This page will explain more
http://www.satakore.com/news132,,Dezaem … Codes.html

edit geez
just found exactly what you need
AviToSaturn.rar
contains the tool and source however the program has a flaw

One big problem with AvitoSaturn: it compresses EVERY frame as Key frame!! As you need a video with an average bitrate of about 120-140kb/s max with this configuration if you don't want any flickering when running on the Saturn (no matter the screen resolution and the refresh rate), your video must be tiny as hell (mine were 144x120 at 15fps with a compression rate of 2000 in AvitoSaturn).
So I hope some good programmer could modify the source code of AvitoSaturn program, or someone could make a virtual mac os for the MovieToSaturn_J program to be run on the PC.

I can try getting a Macintosh virtual pc running but I don't think I have the skill to modify the program.
How do you want me to send you the file?

Last edited by gpink (06-21-2011 06:04:01 PM)

Ok I managed to do some subtitling tests and it looks like the quality hasn't been hurt much if at all. Due to the low resolution I suspect. I only tested using my computer I don't know how to convert to the Sega format. I assumed that re-compressing to cinepak (the original codec) at the same resolution would be a decent test. Seem like my subtitling skills aren't too bad. Given a text transcript I can match it to the video. I could upload a sample if desired but I'm not sure how it would look on a video streaming site.

Very awesome!  If nothing else you could upload the subtitled videos to megaupload or mediafire.  I wouldn't think it'd be too big to redownload but not sure.

Thanks for the links, I grabbed that huge save game manager rar that had the programmer SDKs and the source code.  Sadly the specifications linked in one of those posts didn't work but the good thing is I could find it in the wayback archives.  So just in case anyone is looking for the specifications of Cinepak I have them hosted here : cinepak.txt.  I'll post it to RHN as well later.  As well as the tools you linked too.

It's sad for the Saturn there's so little for it... it's odd though as I remember way back in the day that there were a lot more tools around... maybe my memory is off though.  I swore segaxtreme had a lot more listed though.  Now it's just mostly a forum and a small wiki.

Also, the AviToSaturn was in that huge archive on http://www.satakore.com/news132,,Dezaem … Codes.html so I have that one now.  I'll look into maybe making it better but my video programming skills go along with my graphics one so could be a long while   I'm curious what a key frame is/does though as Google is not helping much and why it would slow it down.  If worse comes to worse, I guess a virtual mac machine will be the way to go and use that official tool for the Mac since that one seems to work the best I guess.

Thanks for your help!

Edit :

Doing a bit of research it looks like FFMPEG can also decode SEGA FILM/CPK files http://trac.perian.org/browser/ffmpeg/l … segafilm.c.  I'm curious now if it can also convert to them as well... I did see it can convert to Cinepak so may be it's possible...  Hopefully I can get a chance to look at it tonight but if not, I'll look more into it this weekend.

Last edited by esperknight (06-22-2011 06:59:07 AM)

Thanks!  That makes sense then why it stutters so much and what was meant by it encodes everything as a keyframe.  For the P-Frames are they just used for small changes I'm guessing?  Something like.... someones lips moving?

I'm guessing to fix it then it when converting between the two it should use the same P-Frames that the video does, correct?

Also found this explanation of the code as well which pertains more to the Saturn than the one I linked before : SEGA FILM.

Thanks for the sample also!  When I get home tonight I'll definitely check it out.

Edit :
I just realized you were talking about mixing the audio and video streams.  For Utena does it encode the video separately from the audio then?

Edit Again :

Looking more at the AviToSaturn I think I understand why it does what it does.  From what I can see it uses one of the standard windows libraries to decode the AVI.  The function it uses returns a BMP of the frame he's decoding so it's not going to return the P-Frames at all.  From looking at it on MSDN I don't see a way of getting the I-Frames at all so not sure how easy this would be to fix (besides switching to a different library maybe).

Last edited by esperknight (06-22-2011 08:12:42 AM)