Language…
14 users online: Dennsen86, deported, Dispace, Golden Yoshi, Green, Green Jerry, Jolpengammler, LazyRuns, mio, neidoodle, playagmes169, RicardoDeMelo, ShoopDaWhoop, YuriGamer - Guests: 259 - Bots: 270
Users: 64,795 (2,376 active)
Latest user: mathew

Official SMW Music Porting Help Thread (Read first post!!)

Originally posted by Masterlink
Try experimenting with stronger instruments, like @4 and @6 (if used with strong ADSR like $ed $7e $e0), raising the volume of the percussion and redefine the instrument for every note, or they will lose its pitch (note that this only happens with percussion) after playing the first note. So you have to do this:

Code
r8/o3@21b32r16^32
[@21b32]2r32@21b32/@21
b32r16[@21b32]2r16@21b32
[@21b16r16@28b16r16@21b16
@21b16@28b16@21b16]42


If it still isn't enough for you, maybe using a different midi with more stuff may do? There are some with more than one channel for percussion, and with background strings to make it sound less empty.

Oh, and adding echo may make it sound better, just make sure it isn't too strong. Try with this one:

$ef $ff $20 $20
$f1 $04 $50 $00

Where do I put all this stuff? As I said, I don't understand anything about porting. And I can't really use "a different" midi because I didn't use one in the first place; I can't get tinymm to output anything that sounds like music. I prefer doing it from scratch rather than building up from a tinymm-disaster. I used the original .nsf file for reference.

Anyway, is music porting really about trying every single combination of every command until it gives something half-decent?

Originally posted by Masterlink
If it still isn't enough for you

I would say this is enough. The problem is with the standards. Nowadays it's unthinkable to have a soundtrack lesser than Chrono Trigger. That's why I'm hating hacking more and more. You can't do it for fun anymore, it has to be more than perfect. You have to make a polished Brutal Mario with orchestra music and HD hand-drawn graphics with outstanding level design. I'll stop here, but I could rant for HOURS about this.
sorry

Mod edit: Nuked annoying table stretch.
Originally posted by Sakuya Izayoi
Where do I put all this stuff?


For the notes, just replace your percussion channel with it. For the echo, it goes after a channel define (#0, #1, etc).

Originally posted by Sakuya Izayoi
Anyway, is music porting really about trying every single combination of every command until it gives something half-decent?


Not really, it's just knowing how to use the instruments correctly and balance the volumes. Hex commands are a plus that may/may not make them sound a lot better; they help a lot, yeah, but you don't need them to make a good port.

Originally posted by Sakuya Izayoi
I would say this is enough. The problem is with the standards. Nowadays it's unthinkable to have a soundtrack lesser than Chrono Trigger. That's why I'm hating hacking more and more. You can't do it for fun anymore, it has to be more than perfect. You have to make a polished Brutal Mario with orchestra music and HD hand-drawn graphics with outstanding level design. I'll stop here, but I could rant for HOURS about this.


This isn't the place to discuss this kind of thing, but I'm not sure where you're getting that from. In the past they were close to that, but nowadays the standards are loosen up a lot, especially regarding to hacks. For music, at least me, I accept the port if it meets the following criteria:

- Sounds good → Includes not having extremely high echo values, vibrato, etc, unless it's supposed to. Most of the time I ignore if isn't 100% accurate, but if it doesn't sound like the original at all, I reject it.
- It's well looped → Self explanatory.
- Works on real hardware → I hate this one, to be honest, but I can't ignore it.

And that's basically it. I don't expect anyone to come with Dominator quality port just to accept it.

(Also, Chrono Trigger's music isn't that good and it's way too overrated, in my opinion. Many other SNES RPGs are a lot better when it comes to music).

If you want to discuss this any further, feel free to send me a PM, since I kinda agree with some of your points.
Why do you need positive criticism in order to do something "for fun" if it isn't that good though?
Originally posted by Sakuya Izayoi

I would say this is enough. The problem is with the standards. Nowadays it's unthinkable to have a soundtrack lesser than Chrono Trigger. That's why I'm hating hacking more and more. You can't do it for fun anymore, it has to be more than perfect. You have to make a polished Brutal Mario with orchestra music and HD hand-drawn graphics with outstanding level design. I'll stop here, but I could rant for HOURS about this.</s>sorry


Hey man, your stuff is pretty decent especially if you're just starting. I'd say try octave-doubling some instruments as well since you have more than enough space for it, but reduce the volume on the lower-octave ones so it doesn't sound too boomy. I do this a lot when I'm composing the MIDI version of my own stuff.

I have to agree with you about the lack of tutorial being really obnoxious, I'm sort of planning on trying to create one for addmusicK when I get more time to tool around with composition since composing SPC's is infinitely more freeing than composing to a midi tracker.

Personally I have a couple questions that I could probably answer quickly but I'm trying to study for a big midterm so I don't want to fool around for a while trying every possible combination of thing:
One, can I combine quantization and arpeggios in AddmusicK? I'm trying to replicate that neat delay in arpeggio that the Undersea Palace port has in the quickest/dirtiest possible way.

Two, I'm not sure how pitch bending works. Do I replace the note I'm trying to bend to with the command or keep the note I'm trying to replace? For duration, do I use the length of the two notes combined, or do I use the length of the note I'm trying to bend to? Every combination I can come up with that makes sense seems to produce a rest for some reason. I'm probably going to take a look at Aquatic Ambience as soon as I can to see how Slash Man did it, but I'd like a bit more advice on how to do this right.

Thirdly, I'm using an instrument with a weak ADSR to try to simulate the human voice (@2 $ED $56 $83 ) and when it gets to strings of 16th notes it loses that nice attack it has and sounds like every note is being tongued individually. According to AddmusicK's readme, the Legato command sounds like exactly the wrong thing to use here, but am I barking up the wrong tree with the pitch bending?
Unexpected end tag (</s>) at 527, expected </div>
Originally posted by hebesphenomegacorona
One, can I combine quantization and arpeggios in AddmusicK? I'm trying to replicate that neat delay in arpeggio that the Undersea Palace port has in the quickest/dirtiest possible way.

It should work if you use q before, but you can't actually use it inside the arpeggio command, unless you don't use the command of course.

Originally posted by hebesphenomegacorona
Two, I'm not sure how pitch bending works. Do I replace the note I'm trying to bend to with the command or keep the note I'm trying to replace? For duration, do I use the length of the two notes combined, or do I use the length of the note I'm trying to bend to?


If you're using $DD, yes, you basically replace the destination note in most cases, but you also need to make the first note as long as the sum of the length of the two notes. If that results into something longer than an half note though, you'll have to split up the pitch bend into multiple partial bends, connected by tied notes, like c2 $DD $00 $60 d ^2 $DD $00 $60 e
The duration is basically how much it takes for the bend to finish. If you want it to span the whole duration of the note, then go with that.

Originally posted by hebesphenomegacorona
Thirdly, I'm using an instrument with a weak ADSR to try to simulate the human voice (@2 $ED $56 $83 ) and when it gets to strings of 16th notes it loses that nice attack it has and sounds like every note is being tongued individually. According to AddmusicK's readme, the Legato command sounds like exactly the wrong thing to use here, but am I barking up the wrong tree with the pitch bending?


I'm not sure if I got what you mean, but I guess the ADSR attack doesn't have enough time to enter the sustain state. Legato won't work there, since the notes aren't rekeyed, but maybe you can try light staccato ($F4 $02) which reduces the staccato between notes to a minimum. You could also try changing the ADSR for any 16th note to something with a quicker attack and maybe decay, just so it has enough time to enter the release state.
Hello I'm new in music porting,I sent a custom music here in smwcentral but the moderator tell that is all ok,but I forgot the header and 240ms of silence in my music,I not know how to make this,can someone help me?
How to put the header and the 240ms of silence?
Check out my profile to see a list of SMW Hacks I made and where to find the latest and best version of them.
Thanks for playing and have fun!
Clique no meu perfil para encontrar a lista de minhas hacks feitas até o momento.
Look this thread for some information.

Edit: In my words: The 240ms can be done by 'v000c8' however, c8 depends on the tempo of the music. While a slow song only can have a short delay, a fast tempo Needs a large delay. Keep also in mind that you should also add 'rx' where 'x' the number of the lengh of 'v000c8' is.
Not adding the "standard" method of 240ms shouldn't even be a reason to not pass something through moderation, since the method doesn't work.
You should however always include an echo buffer header at the top of your text file with AM4.
At least it works if you do it properly:

Code
#0
$F1 $XX $YY $0Z		; Fill $XX, $YY etc with your echo settings. 
$EF $FF $00 $00
r8			; use r4 or longer rests if you hear noises
$EF $WW $LL $RR		; same as above
/			; omit the intro directive if there is already an intro


Remember to add rests of the same length at the start of the other channels:

Code
#1 r8/ *music*
#2 r8/ *music*
*etc*


Again, you can omit the intro directive ('/') if you're already using that.
Originally posted by Torchkas
Not adding the "standard" method of 240ms shouldn't even be a reason to not pass something through moderation, since the method doesn't work.
You should however always include an echo buffer header at the top of your text file with AM4.

That's not entirely true because it greatly depends on the echo delay and how quickly the rest gets played, r8 is usually sufficient enough for most general song tempos, you'll definitely need a linger note value for faster songs yes, I was the one who pointed this out to HFD years ago because he proposed on stating we needed "r8" but it's a starting point, the reason I made it a submission guideline is because that crackle in the beginning of a song can be earrape at times in accurate emulators

Edit: Fuck you autocorrect
I think a post layout goes here somewhere...
No, it only works the way Lui described it.
No because it worked every single time I did the "standard" method UNLESS I used more than $05
I think a post layout goes here somewhere...
Hello, I'm somewhat new to porting music, one would rather call it incompetent experience, anyway I wanted to try porting some songs again so I could eventually port some fantastic songs.


Right now, I'm working on Bomberman 2 (NES) Area 1 theme:
TheSPC
The original song

I'm looking to do this without custom samples and I really don't know any of the custom commands offered... currently porting this for 4.05

so any tips or helpful suggestions would be greatly appreciated
Originally posted by TLMB
I'm looking to do this without custom samples and I really don't know any of the custom commands offered... currently porting this for 4.05

so any tips or helpful suggestions would be greatly appreciated

That's pretty good #ab{:)} There are some octave issues though, I'll just list them here along with some other suggestions:
- lower an octave to the first notes on #0 and #1;
- lower an octave on the bassline (the @9 piano portion on channel #0; using another instrument such as @8 or @14 should also help) and raise the volume a bit;
- lower an octave on channel #2
- change a to g+ on channel #2, just from ~00:35 to ~00:37.5
- raise an octave on channel #3, just from ~00:44 to ~00:46
- that segment starting at 00:47 sounds pretty good if you ask me, but you could try using some other instruments to see if it sounds better (you could try putting @6 with a strong ADSR, something like $ED $2E $26 should work as well); you could also try to vary the sounds like in the original (see 1:07).
- there's a hi-hat sound going along with the drums in the original version. You could try to replicate it by adding a @22/@23 note between the kick and the snare; it would require a bit of duration tweaking, since you should halve the kick and the snare to make room for the hi-hat, without making the drums go out of rhythm. For example, if you currently have something like
Code
@21c8@29c8

you could replace it with
Code
@21c16@22c16@29c16@22c16



I just based my suggestions on the original version, so I may not be 100% correct. Doing trial and error helps a lot, because you'll eventually find the best solution. Good luck!
Thank you very much RednGreen, your suggestions were highly fruitful (though I'm going to make it easier for everyone by posting the mml as well, to make it easier that saying at :44, not that that is a problem).

I really did want to switch more insturments around, but I wasn't too familiar with the SMW sample set (and what octaves they work best in). I didn't know the bass would sound bad in octaves greater than o2-o3.

So things should sound much better now, though I know I'm not done until I add ADSR (something I need to test out more to understand how it affects notes). #3 just doesn't sound quite right plus the guitars are a bit fickle in the lower octaves <_>

edit: ADDED ADSR CORRECTLY (with values I custom picked via experimentation). Still feel like channel #3 isn't sounding right, maybe I should pick a different instrument.

Newer SPC: click here
MML: http://serioushax.pastebay.net/1412147
I've got this song here to play, but it sounds very simplistic..

http://bin.smwcentral.net/u/19134/Brothers%2BTake%2B1.txt

Any help? I'm not sure why it doesn't start at the beginning either/
That... is a start, but it still needs a lot of work. It also seems like no looping was done at all. If you don't know, you can also loop r1. For example, r1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1 can become [r1]37. The music itself sounds very dull. The instruments could use a change, too. I haven't heard the original song, though, so I don't think I can help you with that. Good luck!
Thank you so much for the response, and yeah I know it does sound a bit dull. How would I go about making it have more substance? More channels? I am using am4 player to test.

I got it up to speed, but it just doesn't have much impact, I want more channels for it: http://bin.smwcentral.net/u/19134/brothers%2Btake%2B6.txt

Song is Brothers from Fullmetal Alchemist btw.

Thanks again!
Okay, I'm trying to improve my Doomsday Zone port.

I'm really struggling to find the right instrument/ADSR/other effects to mimick the instrument heard here, here, here and channel 3 (0-7 notation) of the port (ignore instrument) I nuked all samples and it sounds better the way it is now, but I can't make this one sound nice. Help?
Hey guys! :-)

I've got a really simple question here. I understand that, as of last time I checked, I can only insert all my music files once; so I want to max it out & never look back.

It appears to me that level music goes from 00 to FE while overworld music goes from 00 to FF (not that I need more than 7); so here's my question:

Is there any reason why I wouldn't be able to fill every unused music address with custom music? - Esp' for levels.

I know it sounds like a lot, but I'm utilizing all level/room addresses in a hack I'd started in 2007.



YouTube Twitter Twitch