[Super Famicom] Final Fantasy IV & Final Fantasy IV (Revision A)

[layzee]

For those of you who don't know/forgot (just about everybody in the emulation/rom scene; outside it, everybody), Final Fantasy IV [SHVC-F4] for the Super Famicom was released in Japan twice. Apparently, the first release was unsatisfactory enough to the extent that it was quickly replaced by the second release (Revision A, or if you prefer Version 1.1). According to this website (http://superfamicom.org/info/final-fantasy-4/), the initial release was or is "extremely rare". According to the person who found and "dumped" the game (transferred the FFIV cart into a digital format), it was a one in a hundred effort (i.e. it took him 100 carts to find the non Revision A).

I have two copies of the game: one with the "A" imprint on the back, and one without. Without having the tools to open up the cart and without planning to get the tools in the future (I'm not a dumper and don't plan to be one), is there a way to confirm early in-game whether the latter cart is for sure, the initial release or not? Apparently, the initial release if FFIV had bugs or something similar. If true, then identification of the respective carts will be easier once the bugs itself are identified.

[JLukas]

It's possible to use a Game Genie as a version check tool. The US version of FF2 is a great example - Galoob printed two separate sets of codes for "Version A" and "Version B". The different versions can be confirmed by codes working on one cart but not the other.

[Edmond Dantes]

But isn't the US version, itself, a different revision of the game?

[layzee]

It's possible to use a Game Genie as a version check tool. The US version of FF2 is a great example - Galoob printed two separate sets of codes for "Version A" and "Version B". The different versions can be confirmed by codes working on one cart but not the other.


But isn't the US version, itself, a different revision of the game?

Unfortunately, I do not have a SFC Game Genie/Action Replay device (and do not wish to get one); however, even if I did have one, I would probably also need codes specific to the Japanese version ROMs. Also, would there not be compatibility issues between non-Japanese Super Nintendo (SNES) GG/AR hardware with Super Famicom (SFC) consoles? So not only do I need to spend time finding and spend money on a Japanese GG/AR, I also need to spend time on finding Japanese version game genie codes.

The above isn't impossible, but I prefer someone with prior existing knowledge to specifically point out the bug/differences so that I can just play through the game and to personally encounter and confirm them myself. And it's more interesting this way too.

According to someone on a different forum, a/the bug may be found in the final battle with Zeromus fight. However, this is unconfirmed.

EDIT: Actually, I just remembered that someone on this very forum, a few years ago, was selling this specific cart. Once I find his thread, I'll send him a PM to see if he knows anything.

EDIT 2: Cool, nice to see that my memory of something I saw in early 2008 was not wrong (http://www.digitpress.com/forum/showthread.php?t=116194). And the game list was in an attachment, no less. PM Sent.

[layzee]

Thanks to a PM from fellow forum member JLukas, it has just come to my attention that my question has been answered (by the same person no less):

http://tcrf.net/Final_Fantasy_IV


Version Differences
Bug Fixes
Japanese version 1.0 to version 1.1

A significant bug exists in the Japanese 1.0 release which occurs during battle initialization. The programming beginning at SNES address 00896B checks memory addresses 7E1701-7E1702 to get the 16-bit value of the player's current location. If it's location 015A-015C or 0167-017E (all lunar dungeons), the game sets a flag to load a different table of enemy related data. However, the current location isn't accurate on a world map because of a shared memory byte.

On world maps, 7E1702 will continue to store the second byte of the last location visited, but the other half in 7E1701 is now used to store which world map you are on (00 = overworld, 01 = underworld, 02 = moon) The problem starts when you leave location 005A-005C or 0067-007E. These location values include the Misty Cave, Mt. Hobs west entrance, Antlion cave, Toroia basement rooms (use Exit spell), and the water dungeon caves between Kaipo and Damcyan. If you take the airship to the underworld, the result is you now have a current location in the 015A-015C or 0167-017E range even though you're not inside a lunar dungeon.

When you get into a battle on the underworld map, the game uses the wrong table and the enemy AI for most formations is completely broken. In many cases, the enemies cast Stone on themselves over and over until successful, and will never attack your party.

http://tcrf.net/images/9/99/Final_Fantasy_IV_SNES_JPN_10_battle_bug.png

This bug was corrected in the Japanese 1.1 version by first checking if 7E1700 is 03, which signals the battle was started inside a dungeon and not a world map.

No other changes were made between Japanese version 1.0 and 1.1 besides the bug fix.

Now to check if I can reproduce it with my own cart.