…are there so many 8-bit synths and composers, but almost ZERO 16-bit FM tools?
Adlib is the game music master race, and i grow tired of the amount of chiptune in games nowaday
Hell, if i want to compose some epic adlib music, the only tools available are for DOS, unless you want to use a FL Studio VOPM plugin, which isn’t the same.
And using MIDI and that lame soundbank is a downer :’(
Videos (actually songs) related:
Rant over.
I really liked the original Doom music. I think I might have been the only person who did in the entire world, though.
Cas 
You are not alone, the MIDI playback is a downer.
v9snl7f5oms
Sudden craving to play the original Doom.
I do see your point. So why not find or make a MIDI soundbank with Adlib sounds, and use that to play your MIDI-tracks with? It’s fairly easy to implement and package with your game.
Sadly its fairly easy to make chiptunes nowadays, since there are so many VSTs available for that. I do have 2 that aren’t completely “chiptuned”. ProtoPSG VSTi is a (free) emulator of the Sega Genesis sound chip and there’s a GameBoy VSTi named PooBoy 2.0 VSTi. They do make other sounds than chipsounds.
The best you can do is probably VOPM VSTi (as you’ve mentioned), which is an emulation of the Yamaha YM2151 (OPM) 4 operator FM sound chip, which is basically the same as the OPL-2, except it has 4 oscillators. It should enable you to tune yourself in on the kinds of sounds they used back then. There’s also UNO FM which is along the lines of the old DX-9. I’ve also been recommended the Nostaljia VSTi, but I haven’t tried it yet.
For software specifically for creating this sort of music, there’s always Open MODPlug Tracker for Windows which also supports VSTs, but if you already know FL Studio, you should probably stick to that.
It’s VERY hard to get exactly the sounds they used in the good old days. Partly because they spent a lot of time making them, but especially because you don’t have the real chip. None of the SID emulators sound exactly like the C64, but they’re very close. If I ever get into VSTi programming, which I plan to when I get some free time, I’ll see if I can make something for this kind of niche-sounds.
I really hope you know you are awesome :clue:
Well, looks like VOPM is better in terms of usability/results, and it works with OpenMPT too! ;D
You’re welcome, and I have put a check in the “Was awesome today” checkbox
One thing I forgot to mention. When using VOPM, try turning off 2 of the 4 oscillators. Since the OPL-2 only had 2 oscillators, this will help you tune in on the sounds of the old SoundBlaster cards.
So you’re using OpenMPT now? Couldn’t you just as well use FL Studio, considering you’re using VSTi’s? I mean, the GUI must be easier in FL Studio than OpenMPT. I only ever used Cubase and Reason, and EVERYTHING is easier than those
checks cubase WHOAH :o
I never said i was using FL Studio, and i’m more used to the tracker module structure, so it makes more sense to me 
Also, now there’s so many options! (even obscure ones) I’d never imagine i could rig MIDIs with OPMs using OpenMPT or VSTs using Synthfont.
Cubase is rather elaborate, and will sometimes “break” for no reason…except that thing you did, and now you don’t know what’s wrong, and it won’t tell you. But when it works, it reeeaally good. Kinda forces you to be more precise in your work.
hint install “ASIO4all” and use that as your driver in Cubase. Should be faster than anything else.
lmms looks pretty good to me. I’ve used it just a bit, but not being a musical person, i didn’t get very far.
I think that the thing with FM synthesis and perhaps the reason it got a bit out of style is that it’s quite an unintuitive way of creating sounds and most of the time creates sounds that people just don’t like much.
It was quite a genius, flexible and cheap way of generating a very wide range of waveforms, but it isn’t as simple and intuitive as taking a sawtooth or square wave and turn some filter knobs until it sounds right.
To get the most out of it actually requires some math to know what sort of overtones you are going to create.
But probably more importantly, FM synthesis almost never sounds ‘natural’ and often it ends up with metallic sterile sounding noises that can sound quite jarring, harsh and alien. You almost always end up with sounds with loud and harsh overtones that don’t mix very easily.
Square waves, triangle waves and pulse waves with a low pass filter with resonance otoh is a very natural sounding effect that mixes easily in music, which FM can’t truly replicate.
Personally I still like FM for some types of sounds (those wonderful ‘blubbery’ bass noises, bell sounds, and rich ‘pad’ sounds and such), but I wouldn’t want to use it for everything and I usually still end up putting it through filters.
This is one of my passions!
I put a lot of time into the DX-9, DX-7, and then DX-7S, in the day, and still have the latter two synths. I’ve had a patch published as a patch of the month in the Yamaha magazine for DX, and sold music cues written with my original voices.
Yes, FM programming it is non-intuitive, AT FIRST. But it gets better if you understand the physics of it, and know a thing or two about musical acoustics. Not many do this, but it combines well with formant filtering (as opposed to low-pass), for creating very naturalistic and ear-friendly sounds.
I have been dabbling with procedural FM synthesis, and among other things, made this controller which is producing 6-operator FM bells, live.
http://hexara.com/SpiderBellDashboard.htm
(The graphics use up more cpu than the sound!)
I am rewriting, making a basic mechanism for polyphonic play that does not require managing separate threads for every note, and creating hooks to allow MIDI triggering.
The goal is to have a selection of nice FM patches (from my library of work done on the DX-7) playable via a Java Sequencer in a game. The patches would match what I can score with on my DAW via a VST. It would then be possible to compose on the DAW, then export the MIDI file to Java and have the Java-FM synth play the file during the game.
The single biggest hitch: I’m having troubles when I chain 3 operators. The “SpiderBell” is three sets of carrier-modulator pairs. When I try to implement modulator-modulator-carrier, all sorts of problems start to occur. Very likely due to accumulating calculation errors? I am looking forward to creating a “simple case” example of the problem and presenting it on this forum, eventually. Would love to have a collaborator, though, to work on this.