Deprecated: mysql_connect(): The mysql extension is deprecated and will be removed in the future: use mysqli or PDO instead in /storage/content/49/145849/famitracker.com/public_html/forum/classes/dbHandler.php on line 29
FamiTracker
arps. But you can also do both and just let the track play the loudest note. since in real life situations the note that will be heard is the loudest one, that's why triangle kicks and triangle snares work so well.
Usually chords unless I want a Sunsoft sounding track. A release envelope with an echo part instead of silence dfinitely helps, something like 15 8 / 2
[quote=Xyz_39808]Double channel echo with the melody line being those complex enough to outline the chord structure implicitly[/quote]
This would probably net the best results. As far as channel cancelling on real hardware is concerned, it's usually better if some echo is cancelled rather than something more important like the chords.
Perhaps you might want to have a look at tssf's Megaman 10 cover to see how he does echoes (especially since he restricted his music to using pure 2A03 without samples; he'd either have to use the DPCM channel, or use that along with an expansion chip for additional sound effects).
Track 46 has a simple single-channel echo, while track 12 has a dual-channel echo; you'll see why he'd benefit from using arpeggios in track 12 when you play it.
_______________________
Technology: the one thing that's hated & cursed at by all engineers, technologists, scientists & technicians!
I know the nes is very constrained in terms of availble sound channels. Either you use the two squares and sacrfice a channel(most nes games), or use the DPCM for sound effects( I only know of what game that does this, Earthbound).
For sometime like the Mega drive, you're basically have the FM chip and the gameboy sounding chip. The other doesn't sound too good and is very limited for anything other than simple melodies and soundeffects. That leaves all the heavy pushing on the FM chip, which has six channels, but that gets pretty limited very quick with two channels dedicated to SFX. I've would personally limited echos to leads and use at least two note chords. I could possibly fit drums on the same channel with the bass, but not many composers actually did that. That would leave me with 4 or five channels not counting the square channels, with no PCM being used.
Developing a Mega Drive homebrew with a friend and only having access to limited pre-made software for audio I had to make something good sounding with a very limited setup. I basically have 6 FM channels, no samples so the drums are all FM (Feedback noise + sine wave to provide enough thickness in the sound) and all the sound effects are made with the master system sound (since there's no SFX interruption in the playback engine).
I basically have one or sometimes two channels for drums + hi hats/cymbals, 1 bass channel, 1 "faked" FM square wave playing arpeggio chords and 2 or 3 channels for melody and whatever. It made me go with a sound that's really in-between NES and regular Sega music.
If you want to limit yourself, try making music like that. YM2612 only, no samples.
One time I tried making the YM2612 sound like a 2A03, the result is the vgm file attached.
Sorry for reviving the thread, but I find the Mega Drive pretty flexible in saving sound channels. It's 4op, which mean you can use at least two waveforms at the same time(if that's worded correctly). You can do things like chords and bass + noise on the same channel, as demostrated with this short loop. It's a lot like the C64 in that regard.
Firstly, thom, your explaination of how 4op works is inaccurate; it will operate with all of the operators at once, if that makes sense. But, because of operation algorithms, you can technically have up to 4 instruments playing at one time in one channel, not two, though each one will be playing the same note, and be a sine wave.
Secondly, I actually prefer a method of compromise that I don't think is mentioned here. Basically, use your echo, until you need the other channel for something else. Make sure the echo is quiet, and not unique-sounding. Also add a bit of echo into the echoed channel itself. As long as you used the second channel for echo before repurposing it, there will be the illusion present that the channel is still being echoed. I'd provide an example, but it'd need a fair amount of instrumentation, and I'm currently at school. I could possibly make one when I get home, though no guarantees.
EDIT: Done. Observe how it still sounds like there's an echo at the very end.
[quote=Thom]Doesn't sound like there's an echo at the end to me.[/quote]
Darn, it's meant to. I must've done something wrong with the illusion.
The idea, though, is that the introduction of the echo, combined with the insignificance of such a light echo, and the slight echo on the main channel itself, is supposed to make it so that, with enough coverage, the echo can be entirely removed from the song after being established without too easy of an audible difference. It really mostly relies on coverage, because coverage is magic. When I was but an accountless lurker, I saw a module that (seemingly-)unintentionally made a musically-unintelligible row of timpani hits sound like a wolf's howl, thanks to the magic of coverage.
EDIT: One of the democompo modules (that I don't think made it) used the magic of coverage to try and make a switch from double-channel to single-channel echo without it being much of a big deal. It was one of the 2a03 modules, and, henceforth, could not get sufficient coverage
) 