Un-Worked Designs: Working Designs minus bullshit difficulty (Latest: Lunar 2)

[FlamePurge]

I absolutely cannot find a Lunar 2 ISO on the internet that has the same MD-5 that your patch calls for,  and whenever I try to patch an ISO that doesn't have the right MD-5 in xdeltaUI, I get the error "xdelta3 not a vcdiff input xd3_invalid_input".

Is there anything that can be done? Like, can you release a patch that converts any Lunar 2 ISO to the proper MD-5 or something?

[Midna]

Everyone is welcome to their opinion, but it is pretty obvious you have an ax to grind for some reason. I am not really sure you would be happy with any translation intended for a mass audience. Having looked over the differences I personally am not bothered by them and I doubt you are going to find anyone beyond real diehards who want everything literaly translated with liner notes being too upset by it.

I have no ax to grind. I am lacking in axes. Fawful is beefless. I just have a tendency to prefer my translations to be as close to the original as possible while still feeling natural in the target language, and adding pop culture references and goofy jokes runs counter to that.

[zuzma]

I absolutely cannot find a Lunar 2 ISO on the internet that has the same MD-5 that your patch calls for,  and whenever I try to patch an ISO that doesn't have the right MD-5 in xdeltaUI, I get the error "xdelta3 not a vcdiff input xd3_invalid_input".

Is there anything that can be done? Like, can you release a patch that converts any Lunar 2 ISO to the proper MD-5 or something?

It's not online but you can make it yourself. All you need is either a bin/cue dump or IMG version of it then you use bchuck to convert it to an iso/wav dump (it's right in the readme on how to do it). I got the TOSEC IMG release then used cdmage to convert it to a bin/cue.

[FlamePurge]

The thing is, I was acquiring these ISOs and converting them to ISO/WAV format with bchunk as the readme outlined, I tried to patch each one anyway. I still got that "xdelta3 not a vcdiff input xd3_invalid_input" error from xdeltaUI. Not sure what I was doing wrong. I was naming the ISO and WAVs how the readme outlined as well.

[Tater Bear]

Again, putting aside whether the new dialogue is "better" or "worse", it rarely represents the original material except in the broadest strokes. I'll just link you to my writeup at TCRF: https://tcrf.net/Lunar:_Eternal_Blue/Regional_Differences#Dialogue

To be honest I found the original Japanese script to be a bit dry and robotic at times. People repeating the same points and characters with little personality. WD script made the NPCs a bit more human and unique. I took a quick glance at your write up and chose a line of yours that clearly demonstrates how you missed the purpose of some of the script changes.

The US version gets the point across, but suddenly not only have we lost the congruence among the different messages, we've invented new details about how this character is a guard who was assigned duties by Rainus/Lunn.

They are repeating nearly verbatim what the prior character said. Who talks like that? It is not natural and makes the people of each area have nearly the same feel... When you visit different areas the people should not mirror each other and should have their own unique characteristics. I love Lunar 1 and 2 (I own every console version), but the truth is that the Japanese scripts are not that well written and should not be treated like masterpieces that can't be tweaked. The games' scripts needed work and thankfully WD made the games a bit more "human", yet kept true to the overall point.

[elmer]

Yeah, your ISO doesn't match mine. Check the readme for information on how to verify the MD5 sum of the ISO -- if it doesn't match, you can try converting it from BIN/CUE using a different program, or using an ISO or BIN/CUE that was produced by a different ripping program.

I suspect that a lot of folks are having problems because the instructions in your readme.txt file suggests using bchunk, and in the case of Windows, points to a specific version of bchunk that's on GitHub.

The problem is that the default bchunk builds have a bug in their calculation of the sector lengths when splitting up the .bin into the separate .iso and .wav files.

I found/fixed this when building bchunk for use with my Zeroigar and Legend of Xanadu translations.

I downloaded a .cue/.bin copy of Lunar, and used the https://github.com/extramaster/bchunk release to extract the contents, and the MD5 of the resulting .iso file does not match the 1fab8a8b2d83627ba0c6591e02b451ce that you show as the correct value.

The extracted .iso file has an MD5 of 8a37f225b440eb6a3791b3ba3c1e3678 instead.

BUT ...

When I use my fixed version of bchunk on the same .cue/.bin, then the extracted .iso has the same MD5 that you list.

So I suspect that the problem that people are having isn't with the .cue/.bin, but it's with the version of bchunk that they're using.

Here's the patch to fix the bchunk 1.2.0 source, and also to allow it to compile on Windows with mingw-w64 & msys2 ...

Code Select Expand


diff -Naur bchunk-1.2.0/bchunk.c bchunk-1.2.0-msys2/bchunk.c
--- bchunk-1.2.0/bchunk.c 2004-06-29 21:42:34.000000000 +0100
+++ bchunk-1.2.0-msys2/bchunk.c 2015-07-02 17:35:51.323496400 +0100
@@ -58,7 +58,11 @@
  */

#include <inttypes.h>
-#include <netinet/in.h>
+#ifdef _WIN32
+ #include <winsock2.h>
+#else
+ #include <netinet/in.h>
+#endif

#define bswap_16(x) \
      ((((x) >> 8) & 0xff) | (((x) & 0xff) << 8))
@@ -279,7 +283,7 @@

printf("%2d: %s ", track->num, fname);

- if (!(f = fopen(fname, "w"))) {
+ if (!(f = fopen(fname, "wb"))) {
fprintf(stderr, " Could not fopen track file: %s\n", strerror(errno));
exit(4);
}
@@ -289,7 +293,7 @@
exit(4);
}

- reallen = (track->stopsect - track->startsect + 1) * track->bsize;
+ reallen = (track->stopsect - track->startsect) * track->bsize;
if (verbose) {
printf("\n mmc sectors %ld->%ld (%ld)", track->startsect, track->stopsect, track->stopsect - track->startsect + 1);
printf("\n mmc bytes %ld->%ld (%ld)", track->start, track->stop, track->stop - track->start + 1);
@@ -332,7 +336,7 @@
sz = track->start;
sect = track->startsect;
fl = 0;
- while ((sect <= track->stopsect) && (fread(buf, SECTLEN, 1, bf) > 0)) {
+ while ((sect < track->stopsect) && (fread(buf, SECTLEN, 1, bf) > 0)) {
if (track->audio) {
if (swabaudio) {
/* swap low and high bytes */
@@ -399,7 +403,7 @@

parse_args(argc, argv);

- if (!((binf = fopen(binfile, "r")))) {
+ if (!((binf = fopen(binfile, "rb")))) {
fprintf(stderr, "Could not open BIN %s: %s\n", binfile, strerror(errno));
return 2;
}
diff -Naur bchunk-1.2.0/Makefile bchunk-1.2.0-msys2/Makefile
--- bchunk-1.2.0/Makefile 2001-08-02 13:51:40.000000000 +0100
+++ bchunk-1.2.0-msys2/Makefile 2015-04-04 12:38:34.870112100 +0100
@@ -33,7 +33,7 @@
BITS = bchunk.o

bchunk: $(BITS)
- $(LD) $(LDFLAGS) -o bchunk $(BITS)
+ $(LD) $(LDFLAGS) -o bchunk $(BITS) -lwsock32

bchunk.o: bchunk.c


The thing is, I was acquiring these ISOs and converting them to ISO/WAV format with bchunk as the readme outlined, I tried to patch each one anyway. I still got that "xdelta3 not a vcdiff input xd3_invalid_input" error from xdeltaUI. Not sure what I was doing wrong. I was naming the ISO and WAVs how the readme outlined as well.

Yep, even when I've actually got a .iso file that matches the MD5 that's in the readme.txt, I still can't get the xdelta patch to apply.

I've tried using DeltaPatcher that's in the Utilities here, and I've also tried using the official XDelta releases 3.0.10, 3.0.11 and 3.1.0, and none of them likes that .xdelta file

However, I finally tried a very old version of Xdelta from 2003 (that even the author recommends that you don't use), and it does work, and the resulting patched .iso has the correct MD5 that is given in the readme.txt.

That ancient version of xdelta can be found at ...

http://www.evanjones.ca/software/xdelta-win32.html


****************

So, I suggest that Supper creates his xdelta patch with a more modern version of xdelta, and that he includes a working version of bchunk in the patch distribution.

I can send him a copy of mine, if he wishes. It's only a 13KB executable. It's the same version that's included in the Zeroigar translation patch that's on this site.

[zuzma]

Ah sorry about that. I should have explained better how I got it to patch, thanks elmer 

[FlamePurge]

Thank you so much, Elmer. I appreciate the directions. I would hope you get in contact with Supper to improve these patches, because I was very vexed and I'm not exactly a newbie to the scene. Take care everyone and thanks for your hard work!

[Supper]

For anyone following along, I've been hammering away at Vay the last few days. Not sure when the patch'll be ready, because WD really did a number on this game: item price boosts, equipment nerfs, massive increases in MP costs, inns that used to cost 4g each now charging steadily rising prices... Not to mention they apparently actually added in some new dungeon gimmicks, or at least the script says that's what happens if I ever manage to get that far. I thought the Japanese version was already unpleasantly grindy, but the US version really takes the cake.

You realize I have already posted links to complete write ups of the differences for all the WD games in this thread? This information has been available for anyone who wants to read it for years. The fact that you are linking me to /another/ writeup shows you aren't actually reading what I'm posting. Why would I need a link to your own personal translation when I already have one from a more trusted source? Also I have to be honest with you, "Maou" vs. "Destroyer" is not the strongest difference you could have gone with there. There's good reason to go with destroyer vs demon king considering the connotations in English. Demon's don't have the same connotation they do in Japanese as they do to an American, largely judeo-christian believing audience.

I was mostly linking my article because it focuses more on the overall differences than just straight "facts" that changed, but I didn't start this topic to argue about this, you've clearly made up your mind already, and honestly, I don't even care about Lunar very much. My interest in it is largely a product of how flabbergasted I am at the sheer enormity of the localization changes. (Can't deny I didn't read much of the article though -- sorry, but after spending weeks proofreading the entire script, I can't take much more of WD's Lunar right now!)

stuff

Oh, geez, I just grabbed the first "xdelta" I saw in the Debian repos. Turns out that that gets you version 1.1.3, and the newer versions are separate xdelta2/xdelta3 packages. Sorry, I'll update the patches in the near future to use something more modern. I'd really appreciate having your patched bchunk too, since apparently there's no other really-actually-free BIN splitter for Windows.

Thanks for taking the time to look into this so thoroughly.

[cj iwakura]

The script rewrites are a completely different can of worms that I don't really want to get into, but having extracted and compared the dialogue from the Japanese and English versions of Eternal Blue, I can tell you that the English version changes or makes up a lot of stuff. It's not just the jokes, it's random details all over the place, even though the overall plot is the same. For example, in the Japanese version, Leo's been ordered to kill the "Demon King Lucia" specifically, whereas in the English version he's hunting merely for a nameless "destroyer". Again, putting aside whether the new dialogue is "better" or "worse", it rarely represents the original material except in the broadest strokes. I'll just link you to my writeup at TCRF: https://tcrf.net/Lunar:_Eternal_Blue/Regional_Differences#Dialogue

Maybe the US version wanted to keep it vague. Regardless, the core plot is intact, and silly jokes are relegated to NPCs, which makes them fun to talk to (IMO). NPC dialogue doesn't need to be slavishly faithful, as long as the major plot elements are. I never minded that.

If you go the opposite route, you wind up with localizations like (shudder) the Namco Legend of Heroes trilogy or Sword Art Online or, nightmare scenario, Super Robot Wars OG.




Looking over the comparisons, I also like how Leo is in the US version. He comes across as more stoic and less excitable.

[PresidentLeever]

Nice work! I know it's the same in the JP version but if someone could also make the scrolling a bit tighter in Popful Mail, I think it would improve the gameplay a lot.

Edit: I really didn't have a problem with the changes to Lunar 2 though, except maybe for the new saving system. And I doubt fans could make better voice acting work.

[Piotyr]

If you want people to act more human in games they would not act at all like the characters in Lunar, they would have PTSD from their adventures not lots of fun times saving the world. In the end I don't understand a lot of localizers want to "Jazz up" everything, not everything has to be the Simpsons, a story can go without a joke every 5 minutes.
I think I had it with working designs old stuff when I read a thing about their voice actors trying to slip in as many fart noises without getting caught as a little game they played.
Also pop culture jokes take me out of the game hardcore.

[travel27]

For anyone following along, I've been hammering away at Vay the last few days. Not sure when the patch'll be ready, because WD really did a number on this game: item price boosts, equipment nerfs, massive increases in MP costs, inns that used to cost 4g each now charging steadily rising prices... Not to mention they apparently actually added in some new dungeon gimmicks, or at least the script says that's what happens if I ever manage to get that far. I thought the Japanese version was already unpleasantly grindy, but the US version really takes the cake.

I was mostly linking my article because it focuses more on the overall differences than just straight "facts" that changed, but I didn't start this topic to argue about this, you've clearly made up your mind already, and honestly, I don't even care about Lunar very much. My interest in it is largely a product of how flabbergasted I am at the sheer enormity of the localization changes. (Can't deny I didn't read much of the article though -- sorry, but after spending weeks proofreading the entire script, I can't take much more of WD's Lunar right now!)

Oh, geez, I just grabbed the first "xdelta" I saw in the Debian repos. Turns out that that gets you version 1.1.3, and the newer versions are separate xdelta2/xdelta3 packages. Sorry, I'll update the patches in the near future to use something more modern. I'd really appreciate having your patched bchunk too, since apparently there's no other really-actually-free BIN splitter for Windows.

Thanks for taking the time to look into this so thoroughly.

Now Vay, that was harder than the Lunar games.  Way too grindy and everything was expensive.  That game could use a tweak.  I beat it, but it was not easy.  Not super hard, but for sure not easy.

[Kallisto]

Wait..what exactly happen to the LOH games? I didn't know there was something bad about those games.

[KingMike]

I haven't yet played them, but I've read even since the release of the first in the original PSP trilogy (localized by Bandai) complaints about a bad or dull translation.

[cj iwakura]

Wait..what exactly happen to the LOH games? I didn't know there was something bad about those games.

It's just... rough. Like it was translated without an editor.

[thainferno305]

Pls help the lunar patch is not working

[Midna]

Are you getting any kind of error message? Is it just not patching the ROM correctly? How are you trying to patch it? You need to tell us exactly what's happening or nobody will be able to help you.

[FlamePurge]

He's probably having the same problem we've been having.

[Lentfilms]

It's just... rough. Like it was translated without an editor.

More like it was translated by a Japanese intern that knew some English, or maybe a computer. I don't think any amount of editing can fix a bad base translation.