I want to play octave one saw bass lines on the digitone. From my understanding the problem with bass melodies, the played notes sound different in volume, because the filter messes with the phase of the signal, decreasing some of the sines volume. When i would employ a negative tracking, i would set the filter cutoff frequency to a lower hz value depending on the note played.
I.e if i would center the cutoff around the middle of the octave, i should get the most stable sounding tone out of the used scale. (Higher notes result in lower cutoff values, lower notes in higher cutoff values?)
Is it also possible to modulate the volume with the note number, to equalize the loudness even further?
What is your approach? Or do you stick to post processing? Is there any linearity with these values, or is that an exponential relationship, as the filters phase effect is exponential?
If DN dosent support it, lets asume my example would be serum.
Interesting question! This is not my understanding of how filters work with bass. Maybe someone else is more of an expert.
One problem with perceived loudness of a wide-ranging synth sound (very low and high notes) is that high frequencies are often perceived as louder than a low frequency of the same amplitude. This is, as I understand it, how sound energy works (or rather, the perception of it), rather than anything to do with phase.
Since it’s the Digitone, best to start with no filter, just one or a couple sine tones, write the melody you want, and then decide how to proceed. What sounds wrong to you at this point?
Generally, high positive filter tracking is good for sounds that should be more “realistic” sounding, as the tracking keeps the filter in the same place relative to all the other partials. Lower filter tracking sounds more stereotypically synthetic, and values in the middle are useful because most instruments or real sounds benefit from some high frequency roll off for high notes.
I’m not sure what role phase plays here. What makes you think that phase is your problem? It sounds like a red herring to me.
Saw or sine. Which is it?
Its an FM synth, is filter tracking is a problem for you, you know you dont need to use it right? Make a saw wave sound you like just using the operator parameters, envelopes etc.
Im also puzzled by parts of the question, but DT2 (and I believe DT1, not sure) has a key tracking setting in the sound setup menu. -8.00 cancels out pitch changes in case that’s a helpful reference for other negative keytracking
A saw consists of a series of sines, the filter effects the sine at the cutoff frequency the most, and then has a steepness of lets say 12, 18, 24 db attenuation of the frequencys above it. (LPF)
A way around that would be ofc fm, not using a filter, instead using operators, to intriduce hsrmonics to a fundamental.
I am aware of fletcher munson curve, percieved loudness.
I want to understand filter tracking and filter distortion to the signal a bit better. With distortion i mean the attenuation of the fundamental sine.
If i play f in octave 1, the fundamental is around 84 hz, i want to play all notes of the octave at 100 velocity and the loudness meter should read eg minus 12 db.
Can filter tracking get me there is what i wonder?
Play with it and find out.
I dont think any one can help you with this one. You’ll have to try it and see.
You’re making it way more complicated than it is.
Sure you can describe a saw wave as being made up of sine waves, but it depends what tool you use to observe it. Observe a sawtooth wave with an oscilloscope and all you see is a sawtooth shape. No sine at all.
And what does any of that have to do with what your ears tell you?
I think if the goal is to have something that is musically useful, the working by analogy can help. For a bass voice, you probably want some element of the sound to allow the ear to hear the attack of the note, and some element of the sound to carry the fundamental (so that the bass sound performs its musical / harmonic function).
If you think of an acoustic upright bass as in jazz music, it has a strong fundamental particularly when playing lower notes, and less ringing of the fundamental when playing higher notes. This fits the role of the instrument in most arrangements where lower notes give harmonic context to the music, and higher notes are more ornamental.
You can definitely use filter tracking to arrive at a similarly functioning bass voice. If you were to replace an acoustic bass in a piece of music with a sine wave playing the same notes, you might find that the upper ornamental pitches are overbearing compared to the lower ones, probably for these and other reasons. This doesn’t really have much to do with the sound as a decomposition of sine waves but rather the musical context, including our expectations for how similar music is typically structured.
I think if you want to find sounds that are musically useful, this is the kind of stuff you can include in the patch, via filter tracking and other features, which can make a big difference. But it’s a different point of view from trying to make all the notes hit the same level on a scope, which you could also probably do but in the long run wouldn’t really help you accomplish anything.
Yes, its a more theoretical question, and i havent started to add filter delay, to compensate the fundamental cycle for example, but its the same for music theory, its a theory, but it does not make music by itself. But understanding it can help.
I saw Venus Theory crafting a snare drum from fm8 synthesis, where he applied his understanfing of the actual instrument.(the snare) to his patch. So i just aim to get better understanding, not a perfect solution.
A good read that I found a while ago is a Google Doc called “Digitone Guide to FM Synthesis” which goes into that kind of thinking a bit.
To be honest, I am still struggling to apply it to my own work with the Digitone, which battles me at every turn. But it’s a good start, if you haven’t already found one, for approaching DN with that kind of thinking imo