Bug in the credits of Final Fantasy 4?

Paladin Knight · Post by **Paladin Knight** » Fri Oct 30, 2009 3:54 am

There's this site I found about Final Fantasy 4 and it has a link at the bottom of the page for a ROM of the original Japanese Final Fantasy 4 that's been translated into English by a group called J2E Translations. This is the site:

-link edited out-

Just click on the link that says, "About The Game" on the top left and then scroll all the way to the bottom and click on the link in the second-to-last sentence that says, "Downloads section". The second link down on that emulation page it'll take you to is the file I downloaded, which is the original Final Fantasy 4 ROM with the translation patch already applied.

Anyway, the ROM also came with some documentation. In one part near the bottom of the document it comes with, in section four titled "SPECIAL NOTES", it says quote, "Unfortunately due to an error in ZSNES the whole credits are not displayed when completing the game. In order to view the credits, use SNES9X or any other emulator."

I was wondering, has that bug ever been fixed? If so, in what version was it fixed in?

I realize that this is only one version of a translated Japanese Final Fantasy 4 ROM, but I was reading about it and it sounds like this team of translators is probably pretty solid, so I was planning to get to playing this one soon, but I just wanted to ask about that bug in ZSNES first.

Anyways, that's all. ...So was it ever fixed? If so, in what version?

adventure_of_link · Post by **adventure_of_link** » Fri Oct 30, 2009 5:41 am

First off, it's against the rules to link to ROMs.

As to the issue at hand, I dunno. Sorry I can't be of much help.

Gil_Hamilton · Post by **Gil_Hamilton** » Fri Oct 30, 2009 7:45 am

Man, the J2E translation is forever ago.
The most recent version of the translation, 3.21, was released in July 2001. That was back when people still used the DOS version.

ZSNES 1.31b came out the same day, so they hadn't tested on it.
There have been TEN ZSNES releases since that readme was written.

I'm just gonna bet that it's almost certainly been fixed. Assuming the bug was truly in ZSNES and not their code. They don't say if the patch was ever tested on a copier, and there's an awful lot of hacks that don't work on real hardware.

blackmyst · Post by **blackmyst** » Fri Oct 30, 2009 3:21 pm

1.31 came out that long ago? Man... time sure flies.

Paladin Knight · Post by **Paladin Knight** » Fri Oct 30, 2009 7:13 pm

Yes, Gil_Hamilton, I know that that one version of ZSNES came out on the same day and that it's a long time ago and the problem, if it were with ZSNES, probably has been fixed by now, but the reason I was wondering is 'cause I purposely still use version 1.337 of ZSNES, so I was trying to figure out if it were fixed in that version or not without having to play it to that point to find out.

The reason I still use version 1.337 is because I tried later versions, even on different computers, and for some reason, the versions after 1.337 have some problems that aren't there in 1.337. For example, I noticed some slight speed issues in later versions that are not there in 1.337. ...I seem to remember there being some issue in Star Fox, but I can't remember what it was, unless I'm mistaken, but I think there was some problem in Star Fox. I think it was playing slower, at least in certain parts, than it does on 1.337. Also, I noticed that in the latest version, version 1.51, there's a buggy line in the bottom of the screen on Super Mario World. The version right before 1.51 doesn't have that buggy line on Super Mario World, but it does have the other speed issues I mentioned. I also noticed some slowdown in Super Mario World as well, particularly when I was running fast through one of the first areas in Yoshi's Island when I got the invincibility star. All the spikey balls in the water moving around combined with me running fast made it slow down, I guess, but I tested it over and over again on the two latest versions of ZSNES and it slowed down every time and then I tested it several times on version 1.337 and there was no slowdown at all. Again, I don't know why that is.

I don't know why these issues started appearing, and I don't know why no one else seems to have ever noticed them. Like I said, I've tested these emulators on different computers and had the same results, so it can't be just my computer.

Anyway, that's why I still use 1.337, and that's why I was wondering if, and when, the bug was fixed.

Sorry that took so much text to explain, but I had to explain why I'm still using an older version. I mean, I've never had any problems with 1.337. It runs all the games I've ever tried on it just fine, and I've tried quite a few. I guess the only games it has issues on is games that I've probably never heard of and aren't very popular at all, 'cause everything I've ever tried on it works just fine, so I don't see a need to change that. I say if it ain't broke, don't fix it.

I do have the next-to-latest version of ZSNES also, the one that doesn't have the buggy line at the bottom of Super Mario World, just in case I ever do run into any issues with 1.337, but so far I haven't.

But anyways, so that's why I was asking those questions.

sweener2001 · Post by **sweener2001** » Fri Oct 30, 2009 7:41 pm

when you have issues, and find that you are the only one having them, it tends not to be someone else's fault

Paladin Knight · Post by **Paladin Knight** » Fri Oct 30, 2009 8:20 pm

I wasn't trying to blame anyone. And like I said, I've had this problem on different computers, so it's not like it was just one computer that was having the problem.

Gil_Hamilton · Post by **Gil_Hamilton** » Fri Oct 30, 2009 9:20 pm

Paladin Knight wrote:For example, I noticed some slight speed issues in later versions that are not there in 1.337. ...I seem to remember there being some issue in Star Fox, but I can't remember what it was, unless I'm mistaken, but I think there was some problem in Star Fox. I think it was playing slower, at least in certain parts, than it does on 1.337.

Yes, Star Fox does indeed run at about half the speed in current builds as it did in older builds... meaning it now runs at the RIGHT speed as opposed to playing the entire game in fast-forward with all sorts of audio glitches.

I also noticed some slowdown in Super Mario World as well, particularly when I was running fast through one of the first areas in Yoshi's Island when I got the invincibility star. All the spikey balls in the water moving around combined with me running fast made it slow down, I guess, but I tested it over and over again on the two latest versions of ZSNES and it slowed down every time and then I tested it several times on version 1.337 and there was no slowdown at all. Again, I don't know why that is.

An SNES game? Having slowdown issues? That's crazy talk!
The slowdown, again, is proper behavior. 1.337 is in the wrong.

I don't know why these issues started appearing, and I don't know why no one else seems to have ever noticed them. Like I said, I've tested these emulators on different computers and had the same results, so it can't be just my computer.

Because emulation improvements have created a more reliable product that runs more games more like they ran on an SNES.

Sorry that took so much text to explain, but I had to explain why I'm still using an older version. I mean, I've never had any problems with 1.337. It runs all the games I've ever tried on it just fine, and I've tried quite a few. I guess the only games it has issues on is games that I've probably never heard of and aren't very popular at all, 'cause everything I've ever tried on it works just fine, so I don't see a need to change that. I say if it ain't broke, don't fix it.

You just dragged out two inaccuracies in 1.337 as examples of "ain't broke"

byuu · Post by **byuu** » Fri Oct 30, 2009 9:40 pm

There's always going to be people who favor running the game with artificial enhancements.

The only option seems to be to provide a way for them to overclock the processor, and understand that what they're doing can cause serious game freezing issues and bugs.

That way, everyone will still upgrade, and for the most part the emulator will be more faithful and result in less false bug reports.

I'm much more concerned about the false hardware options that result in games only playable when they are enabled, like the VRAM write one. Such an option here would remove any motivation for fixing the games properly.

Because emulation improvements have created a more reliable product that runs more games more like they ran on an SNES.

I will say that the SMW last line glitch is indeed a regression. The last line is definitely rendered by real hardware, and in rare cases it actually is garbage, but SMW isn't one such game. ZSNES is drawing the wrong stuff there for some reason. I'm guessing HDMA isn't being applied, but who knows.

Sadly it has given some the impression that we should limit the output to 256x223, sigh. If anything, I'd much rather see an emulator offer adjustable overscan for all screen edges.