Let me give it a shot. I can probably try to explain it in a way you can better understand.
This is basically what no one ever explained to me and I had to learn on my own:
I'm guessing that you don't really understand the difference in synthesis that well, so I will start from there.
As you know, sound is just vibration of the air that reaches your ears.
Someone found a long time ago, that there is a fundamental piece to sound: The Sine Wave.
Then they found that if you take Sine Waves, at different frequencies, and play them together, you can build other type of waves, like Sine Waves, Square Waves, etc etc.
This is what you'll see referred to as: Additive Synthesis
And when dealing with Additive Synthesis, you'll hear about the fundamental, and the overtones, etc. All that means is the Sine Waves that have to be played in a frequency and volume (amplitude) to generate other type of waves (almost like building a chord out of notes. C Major for example, is C, E, G).
Easiest way for you to understand it is to try it yourself:
https://meettechniek.info/additional/ad ... hesis.html
So the crazy thing is that, with enough sine waves, you can replicate insanely accurate sounds, like an Acoustic Piano's Timbre. I have no idea how many, but I'm guessing probably in the hundreds?
Wasn't practical back then. So before we continue, let's understand Subtractive Synthesis:
This one is much easier. The concept is that you don't build your Square or Saw Waves on your own (like with Additive), but rather the factory gives you a synth with prebuilt ones you can play with (let's say a Juno 106)
So that's what they call an "Oscillator", it's the circuit that is generating that wave. So let's say you select a Saw Wave and play it. You'll hear that typical Saw Wave sound, sharp, brassy and cutting.
So those synths come with another function called the "Low Pass Filter" which is almost literally like a knob that only cuts the treble out of the sound, but leave the bass/low end intact.
If you twist the knob, you'll be cutting, and restoring the high portion of the saw wave in real time, and this will sound like "Neeeeeeeeeeeoooooooooooommmmmmmmmmmmooooooooooooooooooooeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee"
That's what they call Subtractive Synthesis. Now usually you can do more, like add Resonance (which is just a function that loops the sound onto itself creating a resonant frequency, and gives it that squelchy sound (search for acid 303 on youtube), and other functions.
Resonance in particular is very sought after, because it makes sounds have that resonant "phaser" like effect, and sounds very futuristic.
So this is why you'll notice that Subtractive Synths are usually limited to Saw/Square type sounds, at which they excel. All the stuff people go crazy over (Juno, Jupiter 8, 303, etc).
FM synthesis is really very very similar to Additive Synthesis, but the Sine Waves aren't added together, but rather, the second sine wave directly modifies the first one, and the 3rd modifies the 2nd. Instead of being called Overtones they are caller Carriers. So this is not as good as "true" additive because the carriers are not able to control the volume of the sine waves created, but the method is so close that FM might as well be Additive Synthesis' retarded cousin. Still, due to this limitation, everything FM does sounds "metallic" and harsh. Great for Electric Piano sounds, as my favorite sounds of all time are "DX"ish style EP's.
Can additive synthesis recreate the same sounds as FM? Sure, with enough work, because FM is a hammer approach, probably would be 10 times as complex to make the same sound in a "true" additive synth, but not impossible.
Each style of synthesis is trying to recreate the same thing: real acoustic instruments. Each one fails miserably. Additive being the one with the potential to actually be successful, and maybe know they could do it with a VST, on a Core i7 with a gazillion Sine Wave OSC'
At that point, it's no longer called "Additive Synthesis" but rather "Physical Modeling". Technically, since tuning 2000 Sine Waves at the exact right frequencies and amplitudes to recreate a violin, complete with bow attack, and vibrato would be an insane task, instead, smart people have had their software analyze real people playing the violin, and the software learns how to create that exact range of waveforms, and play it back in real time, completely synthesized............................................or at least that's the idea. A few synths can do a very good job, but if someone can point me to a VST that can be played with all the nuances of a violin, I'd love to see it
Subtractive Synthesis, on the other hand, is limited to the waveforms programmed from the factory. Whatever waveform they created from the factory, that's what you get.
You can then slice it any way you want, with the low pass filter (treble cut), and if you're lucky, your synth is advanced, and has a high pass filter (bass cut), a bandpass filter (cut everything except mid frequencies), etc etc.
So you can get 100 variations of a Saw or Square, but they are *still* a saw and square.
So how to get real acoustic sounds then?
This is where the "ROMplers" come into play.
Someone figured out "Why the heck are we wasting time trying to synthesize real instruments? Why not just record someone playing the originals note by note, assign them to a keyboard, and everytime you play a key, it's like you're hitting "Play" on your CD player for that recorded sound?"
And Wavetable Synthesis was born, also known as "ROMPler". Hated by everyone yet everyone needs it, lol.
Wavetable, literally has the recordings of instruments as played by someone in real life, note by note, and permanently recorded onto your synth.
The Korg Triton is a Wavetable Synthesizer. Let's say you select "M1 Piano", or "SG Piano". If you play the note of C, you are playing back the prerecorded sound that Korg recorded originally in studio of a rinky dink (but very cool) piano sound at that note. If you play C major, you are hitting the "play" button for 3 recordings, etc etc.
Because you are just playing back recordings someone else made, there is no real "synthesis" going on, you're just playing back stock sounds someone else made. And this is why ROMPlers are hated by people obsessed with synthesis. You can't really do much with those sounds.
But, lucky for us, a ROMpler like the Triton, doesn't just have the stock recordings (which Korg calls multisamples). It has filters, just like a Subtractive Synthesizer (low pass filter, high pass, notch filter, and ever resonance, but it suuuuuuuucks on the Triton unfortunately), and it has up to 7 individually assigned effects (chorus, reverb, delay, distortion, etc etc)
So, like an old Subtractive Analog Synth, a ROMpler like the Triton can select an "Oscillator" and you can choose which sample to play back as if it was a waveform (Can't remember how many the Triton Extreme has, but I think it's over 1000, pianos, organs, bells, guitars, strings, special FX, etc etc)
So for you to "sound design" on a Triton, what you're really looking to do is how to combine this massive amount of things into cool, usable sounds.
If you go thru most of your Combis, you'll find a lot of them are just 4 or more sounds joined together (like Piano and Strings, or Piano and Bass). Or complex layers (4 or more separate strings connected together, etc)
So in a nutshell that's the basics right there. Let me know if you found it useful.