Final Fantasy 5

darthvaderx asked me a while ago to msu FF5, and I had a little time to give it a try.

Unfortunately, some stuff is a bit "special", so I do not know how many of disgusting exceptions this game will hold.

Inside the package are 2 patches:
ff5_msu1_2.0MB(J).ips: apply on Japanese 2MB rom (this patch expands the rom to 2.5 MB)
ff5_msu1_2.5MB(E).ips: apply on English 2.5MB rom


Track Mapping
There is an easy way to track all themes!
- start the ORIGINAL rom (without MSU patch applied in Geiger Snes9x debugger)
- make a breakpoint at C0C214 (Execute hook)


Run the game until the intro starts, the debugger will halt when the intro theme 02 is going to play.
- Click on edit registers:


- type in the lower byte of A the theme you want to map, in this example 3e is the meteroid noise pcm -62: change the 02 into 3e (or any other theme you want to map).


Click ok and run. You should hear now the meteroid noise during the intro.

Continue with all other tracks.

------------------------------------------------------------------------------------------

To map the tracks ingame, make a write hook at 2004 in the PATCHED rom

In this example the debugger halts at the title screen with lower byte in A:xx00, so you know the theme is pcm-0

Track Map

Edit: patch_v4 - spc fallback
Edit: patch_v5 - non-loop tracks coded
Edit: patch_v6 - meteor fix
Edit: patch_v7 - theme interrupt fix
Edit: patch_v8 - more fixed
Edit: patch_v9 - resume fixed
Edit: patch_v10 - meteor fixed
Edit: patch_v11 - another mute fixed
Edit: patch_v12 - try to fix pop sound
Edit: patch_v13 - another mute
Edit: patch_v14 - battle resume support
Edit: patch_v15 - clear resume flag on start
Edit: patch_v16 - piano resume
Edit: patch_v17 - mute at track $30 added
Edit: patch_v18 - mute at prison

Ok ok, thanks, I'll try. Could you also show me the tutorial on how to add loop to a track too?
About Earthbound-Sonic MSU-1 I think I will give it up, I was not very pleased with the final result.
Finally the author of this hack asked me how to make it compatible with MSU-1:

http://www.romhacking.net/hacks/4134/

Could you help him too?

Thanks again.

I never dealed with tracks, the only tutorial I can refer to is that from qwertymodo:
https://www.youtube.com/watch?v=9pDXdaWS9-Q

As for the FF4 Ultimate, he can find the asm source code in the patch folder of
http://bszelda.zeldalegends.net/stuff/Con/ff2j_msu1.zip

Probably the both of us use same free space or something, he'd only need to rearrange the addresses in the asm file?

I think I know the problem with ultima, it is the rom expansion which is problematic here (you need to use lunar expand, otherwise the addresses of the msu code are not recognized):
https://www.smwcentral.net/?p=section&a=details&id=4599

- take a fresh 1Mb rom of FF2 (US) no, header.
- use lunar exapnd to make it the size 1.5 MB
- Apply Final Fantasy IV - Ultima v5.41 no header.ips
- REMOVE header (the ultimate patch automatically adds a header)
- apply ff2j_msu1_1.5MB.ips

then it should work

darthvaderx wrote:Ok ok, thanks, I'll try. Could you also show me the tutorial on how to add loop to a track too?
About Earthbound-Sonic MSU-1 I think I will give it up, I was not very pleased with the final result.
Finally the author of this hack asked me how to make it compatible with MSU-1:

http://www.romhacking.net/hacks/4134/

Could you help him too?

Thanks again.

If you are struggling let me know, i can help you, i deal a lot with making tracks. qwertymodo stuff is the best place to start.

As for the patch itself, there's a small warning, I did it in 2 hours, so there should be many bugs. What I also couldn't trace are partial mutes like when rescuing the first girl right at the beginning. I could add a mute here but it was only good guessing.
I would be surprised if this covers all mutes.

Also some themes will not loop where I need to make exceptions, when the track map is finished (like 0a, pcm-10, dying). But darthvaderx and I know this procedure already from ff4. Here I expect less problems

Conn wrote:I think I know the problem with ultima, it is the rom expansion which is problematic here (you need to use lunar expand, otherwise the addresses of the msu code are not recognized):
https://www.smwcentral.net/?p=section&a=details&id=4599

- take a fresh 1Mb rom of FF2 (US) no, header.
- use lunar exapnd to make it the size 1.5 MB
- Apply Final Fantasy IV - Ultima v5.41 no header.ips
- REMOVE header (the ultimate patch automatically adds a header)
- apply ff2j_msu1_1.5MB.ips

then it should work

About FFIV, I'm still working on my set and I'm using this translation this time (which is my favorite):

http://www.romhacking.net/hacks/2337/

However, specific to this translation I found a bug at the game end, in the end credits instead of having pcm 186 (Ending part 3), repeats pcm 93 (Ending part 2).

In case this does not happen in the original rom, it's probably a bug in this translation and you should consult these authors

Or maybe the translation remapped something, oddly enough that it combined with the MSU-1 patch doesn't work on the SNes9x version you sent me, otherwise I would send you a save state.

Better to keep this topic ff5 related to avoid confusion. If you think I can help on ff4 please post in the respective thread. However, I cannot fix all possible combinations of all various patches out there. I hope you understand

Sorry...

Okay, one step at a time.

I sent you a version of FF4 which I could load in snes9x debugger. Using one of your savestates revealed no problem at the endings. Maybe you assigned the wrong pcm? Like pcm part 3 is the same as part 2 ending?

I edited the patch to v2. If loading a savestate made problems (spc played). v2 fixes it (hopefully) find it in first post.

Stupid mistake by me, v3 in frst post

Here's the track map. Anybody willing to help is welcome to complete it

Track Map

I'd be willing to look into this, I should be able to help with track mapping, and PCM creation.

@Conn, which English translation patch should I be looking for to use with this? Is it this one? https://www.romhacking.net/translations/353/

Yes, exactly RPGe was the translation I built it upon, but you need to be careful that the romhacking patch needs a header, but the 2.5MB english msu patch is for without header.

So you need to ensure that your rom has a header before applying the romhacking RPGe translation and then remove it before applying the msu version for the eng 2.5 MB rom. Alternatively you can google for the prepatched eng rom: it is called Final Fantasy V (J) [T+Eng1.1_RPGe].smc
But also ensure it has no header  before applying the msu patch

Much thanks

P.D. if you make a playtest and find a buggy place, I think best is to send me the srm (no savestate) before this bug happens and send me the time of a youtube longplay so I know what to do to trigger this bug. srm have the advantage that I can load them in any emulator (I prefer geiger snes9x to debug, but this one is msu incompatible).

Thanks for the tip about the header, I woulda gone mad trying to figure out the problem given that the romhacking.net page for the mod lists the CRC info for the headerless ROM...

When using the MSU patch without any PCM's present, you get no sound. No way to get it to play the SPC when no PCMs (I know some games can't get that added, but most seem able...)?

I think I was able to code a spc fallback, check first post for v4.

And I've got about 20 tracks mapped... Though I haven't even glanced at which tracks loop (unless I'm missing something, that'll have to wait for an actual playthrough so I can see how the game handles each song).

Track Hex 11 (Dec 17) might work as a mute, if whatever you used for the SPC fallback (That's what it's called! I couldn't remember the term when I made that last post, heh) doesn't work right. At least when I used that trick in the Geiger Debugger you gave in the first post (and let me tell you it got a LOT less tedious when I remembered I could hold Tab to speed through the beginning before the breakpoint triggers, bringing it from like a minute to get there down to a second or two, lol) on track 11 I didn't get any music or SFX.

I've always thought it best to have SPC fallback whenever possible, not only because it makes things easier for testing, but also so all SFX and such work properly (PCM only can lead to 2 PCMs trying to play when it has a SFX at the same time as there's background music, and MSU doesn't like when that happens).

That is awesome
Indeed, looks like 11 is a mute theme, but I still get sfx using it.

I used a different technique to implement spc fallback... let's see how it works, otherwise I will use track 11. You can leave track 11 unmapped (means no pcm) or have a blank pcm track with 5 seconds mute (does not matter).

Also awesome that you already have 20 themes mapped. It is however very important for me to know whether these themes are meant to loop (you can use fast forward to check whether it has an ending or not). I need to hardcode the looping behaviour to each track, so please add it in red in the track map if you have a track that does not loop

And 30ish tracks mapped, including loop info, and I'm done for now. It's 7:45 AM right now, and I should probably get some sleep... That's what you're supposed to do when the sun starts coming up, right? Right?

It's a game, nothing more, nothing less. Take care of your mental healthy (though I know it is hard sometimes) and make a break
At least sleeping is something essential

Much thanks for your effort!

Oh don't get me wrong, I'm not losing sleep doing this; I'm doing this because I can't sleep. Subtle difference there. ;P

And my mental health was shot years ago, it's a lost cause now, lol.

I know these problems too well. There's not much that really helps and is safe to use. I tried zolpidem (ambien) and this is ok in terms of risk/benefit (quarter pill, 2.5mg is sufficient), but still has the danger of addiction and not to underestimate side effects (though still better than benzos/opioids/barbiturates). But better trying with melatonine first

Ah well, let's stick to the topic

Your track map skips Hex value 29. Is there a reason for this, or are all the Dec values after that point wrong?

I added Hex 29 in, but didn't adjust the Dec values, so you'll want to check on that.