What's Shenmue English translation

What's Shenmue English translation RELEASED

Hi there, newbie on the forums here.

My name's IlDucci and I'm currently preparing a What's Shenmue English translation. The original post was in Shenmue Dojo, which you can find here.

Right now the translation process is pretty advanced (Can't say exact details for now), but I've encountered a couple of problems in the meantime.

The first one is, I pretend to release this project via PacoChan's Patcher (The same one I've used for my Shenmue I and Resident Evil 2 Spanish translations), but first I have to set the CDDAs properly. I'm a total noob when it comes to doing releases, and it seems the only one who has achieved this is HyKan. I haven't been able to make the CDDAs sound right, the farthest point I could go is getting the first track on the Press Start Button menu... But only the last half can be heard.

The other thing is... it seems the GDi I ripped from my own disk seems to differ from what's on TOSEC. My disc code is <610-7179-0133A>H1723E. The fourth value is different from what's there. Is this normal or are we looking at two versions of the same demo?

Thanks in advance.

EDIT 01/09/2014: After almost half a year, we've succeeded into releasing the What's Shenmue translation patch. The delivery format is an small patcher that will read an original GDI backup and turn it into a full MDS image. Also, the translation is available in three languages: English, Spanish & German, with some extras as well.

You can see the full details of the patch here: http://www.shenmuedojo.net/forum/viewtopic.php?f=37&t=46933

 

My disc code is the same.
Maybe the famitsu version (orange cover) has another disc code.

 

Well, according to the TOSEC version, their disc code is [610-7179-0133A 11M1, 11MB1]

All I know about my disc is that I bought it in a Retro fare.

Here's a photo:

 

IlDucci: it's actually pretty common for there to be variants that don't show up in TOSEC. For example, my copy of Crazy Taxi isn't in TOSEC.

 

Hopefully I can clear this up for you without creating more confusion...


The ringcode is a code that can basically be used to identify if the discs are from the same batch (when manufacturing/pressing). If two discs have the same ringcode, they are from the same batch; if they have different ringcodes, they are from different batches.


For data tracks, the MD5 should be the same across all batches. If the MD5 is not the same, then that means we have a different version of the game -- perhaps bugs were patched, or something new was added. This is uncommon with Dreamcast titles (In the USA, for example, I think only Midway titles, like Ready 2 Rumble Boxing, Hydro Thunder, etc. had multiple public builds).


For audio tracks, the MD5 may or may not be the same across all batches, even if the source audio track is the same exact audio. This is because during the manufacturing process, bytes are added to the beginning of the audio tracks, and a different amount is done depending on the factory/batch. This is called the "audio offset." Audio tracks from the same batch/ringcode should all have matching MD5s, but audio tracks from different batches/ringcodes may or may not have different MD5s.


Discs with different audio tracks but the same data tracks are called "ringcode variants" because the contents are practically the same, but bit-by-bit they are technically not. They are always less of a priority to dump because they are not truly unique.


Luckily, all PAL titles have the same offset, so there's no ringcode variants for PAL titles making a 100% complete PAL set pretty obtainable! For NTSC-U and NTSC-J, there are some titles with a dozen or more ringcodes, so we haven't gotten around to dumping all of them just yet--if ever!




So, back to your original question, which tracks do not have matching MD5 sums? The audio tracks, or the data tracks? At the moment, we are aware of at least one What's Shenmue? ringcode variant, but we only have one dump of it, so it's considered unverified and thus not in our public dat files. So if you wouldn't mind, we'd appreciate you posting the MD5s for your disc in that thread and you could possibly help us verify another variant!

Also, I'm looking forward to the What's Shenmue translation!

 

Thanks a lot, darcagn!

 

You dedicated some time to answer in full detail, so I guess I'll have to reply everywhere: After checking MD5s and such it seems that only the audio tracks change. The data tracks are exactly the same (So that's one problem down, another one pending).


INFO:
SEGA SEGAKATANA
SEGA ENTERPRISES
D690 GD-ROM1/1
J 0799010
6107179 V1.004
19990720
1ST_READ.BIN
SEGA ENTERPRISES
WHATS SHENMUE

GDI FILE:
9
1 0 4 2352 track01.bin 0
2 600 0 2352 track02.raw 0
3 45000 4 2352 track03.bin 0
4 358492 0 2352 track04.raw 0
5 359081 0 2352 track05.raw 0
6 370791 0 2352 track06.raw 0
7 380787 0 2352 track07.raw 0
8 388253 0 2352 track08.raw 0
9 389344 4 2352 track09.bin 0

CRC-32:
track03.bin cd135ffd
track04.raw 3f1592f2
track05.raw 9c97ad60
track06.raw d7893dc7
track07.raw 39e2a81e
track08.raw 1ed5c3c0
track09.bin 8468eadc
track01.bin a0a8594e
track02.raw 0de03f29

MD5:
6f5441137379f0b4f2aa582ba5c1ed23 *track04.raw
d84891aa4bf130ed049fa631ea849bf6 *track05.raw
7542a63d468a446c74ae4f283e9925a6 *track06.raw
8945ee4a8f2ce16bd33b3b9b198c0206 *track07.raw
0107eed43958e627f78f1fe720eafcfd *track08.raw
ab7c43fb628f9f77c46bce05a27e2a46 *track09.bin
0885c04b770af89826d72c8418afef88 *track01.bin
d09be539a7074a058159505b7f5fdcb3 *track02.raw
8171f4ebf9711cb7a692c630ef56e23b *track03.bin

SHA-1:
324a506197623c29bcd7d468bb4eb97d575ba321 *track06.raw
193872e0ffc3ed1e3df690777394e0fdcb45126b *track07.raw
c04bae30247250f7a9e99384d3b2c48a7529b9d1 *track08.raw
82071f7ac4a2a9c05ae7b144b7314a0689fee889 *track09.bin
d1edcea373c3a939a802c7e5154e6d74277b7930 *track01.bin
db680939aae4bf61d0e588414c72a47c6dd92cdd *track02.raw
15205fb89746214209c82e275a051270471587a8 *track03.bin
ebfa52b7038846e5a2522014e29cd9c4ce23c42f *track04.raw
d126db86c9ac110c90413b5d10171d6f8da1f0d3 *track05.raw

 

Thanks for posting your hashes! It appears to be just a ringcode variant. The version number in the header is the same as the one we have verified already, and only the audio tracks are different. They just have a different audio offset, that's all. You very likely have a good copy of the disc, at least data tracks wise you definitely do.

 

CDDA playback issues, you can modify the 1st............

 

Thanks for the information.

About the CDDAs, well, this is what I've got: I have the real CDDAs at the top of the track list (done with MDS4DC) and another track down below to keep the 45000 LBA. With that, I only get partial sound when going to the Press Start Button menu. I've tried doing the nine CDDA tracks trick (With tracks before and after the real ones) that HyKan did but it seems I'm missing something else.

 

Can you post the part of hykan's ip.hak from the beginning up to the HD TOC?

Also, have you tried to use his release as a basis for yours? I mean, using his ip.hak and bins and cd structure but changing the Japanese files for translated ones?

 

I just tried to get all his tracks, keep his 1ST_READ.BIN and IP.BIN, switch the lbacalc to his, but I couldn't make it boot. Oh, and the 1ST_READ.BIN is completely untouched (Except for the scrambling).

I'm sorry, from where to where?

 

From start (0) to offset 0x02FF, or decimal offset 767 inclusively.

Would like to see both original and hykan's.

Also, what do you mean by switch the "lbacalc to his"?

If the bootbin has been scrambled on disc for his release, then of course the ip.bin will be different as it now doesn't require the code for scrambling back to unscrambled state the would-have-been-de-unscrambled bootbin (if that makes sense?). Scrambling as always been confusing to write about comprehensively.

 

His CDDA tracks weigh more than mine, so I assumed I had to recalculate the lbacalc.

Here's the content you asked and my "original" IP.BIN.
http://www.mediafire.com/download/9ig39cm39iyhs92/IPs.7z

And yeah, the scrambling/unscrambling now makes sense to me.

 

The bins don't seem to have the "TOC" modified, I figure that since this is a beta it might have acutally used it even if no game I know actually uses it. Else I guess the difference is the scrambling-or-not but I can't tell for sure.

Why don't you used his CDDA tracks? They should be the same right? Even if they don't hash-match, as darc told earlier.

Also, keep in mind that if you're using Sizious' lbacalc, you should use RAW audio tracks, not WAVS. Same goes for mds4dc I think.

 

Yeah, I'm always using RAWs. Well I suppose I'm missing something really stupid.

They are not the same: The total amount from HyKan's larger, and his real tracks are also larger than mine. Seems like there's more space before and after the actual song. (Either that or I ripped them wrong).

That's strange. HyKan's release seems to be calculated at 45000 LBA, but the tracks (Ripped with GD-ROM Explorer) weigh more.

 

Couldn't you simply clone his first session and add your 2de session?

[EDIT]
Maybe the best way would be to forget about mds4dc and simply use Hykan's release as a basis. Burning the same 1st session as him, then using the scrambled bootbin and associated ip.in in the building of the 2de session then burning it also. This should make a working cd-r(w) using whatever he figured out to work. Then just make a image of the CD-R(W) using discjuggler and you're done!

Sometimes it's worth it to burn a disc, especially when there's CDDA involved.

 

Quote I found in my archives from someone I know that worked with Hykan to make the what's shenmue release:

Could you post the working release explicit TOC and your version's TOC too?

 

That version you are speaking of was not done by Hykan it is just mislabled as I spoke to him myself

 

Well, Hykan helped and the person who actually did it doesn't like recognition, so he released under Hykan's name to be nice. I spoke to him too.

[EDIT]

You should really try to make it wihtout mds4dc, just do it manually.

1- Extract the 9 audio tracks from the hykan-russian release
2- Convert them into wavs using raw2wav.exe
3- Burn them all to a cd-r via cdrecord. with session opened.
4- Check that the 2de session LBA is really 45000 with msinfo
5- Burn the translated data track.

This should work no matter what.

FG