Maintaining a synth's "strength" in spite of LFOs/phasers

[prettiestPony]

Not sure how much help I can really expect with a question this vague, but I was wondering if anyone has any good tips/tricks for keeping the "fullness", "fatness" or "strength" of a synthesizer's sound consistent when there's an LFO modulating a parameter or a phaser effect applied to it. I occasionally will come up with a neat sound that works great most of the time, but at certain points in the LFO or phaser's cycle, the sound gets too quiet or thin.

The most recent example that's bugging me is with ymVST, a chiptunes-esque Atari-inspired VST synth. When you enable the "SID" effect, I think it basically enables an internal LFO that controls pulse width or something similar to pulse width. (Maybe it's combined with a filter.) However, often when I get a good sound going with it, the sound gets too weak at some point in the oscillation's cycle--it gets thin and reedy, and the lower frequencies drop out.

Any ideas about what can be done to help that? I don't think there's a way to control the SID LFO's "depth" or "range". Is ymVST basically just too unreliable for this kind of thing? Any workarounds you can think of?

[the4thImpulse]

Any ideas about what can be done to help that? I don't think there's a way to control the SID LFO's "depth" or "range". Is ymVST basically just too unreliable for this kind of thing? Any workarounds you can think of?

I don't have much of an answer for you, I would say use a vsti with more control over the lfo or use another vst specialized in phasing with built in lfo's.

In case you don't know, phasing is when the audio signal is duplicated and one part is played back with a slight delay that goes slower - faster - slower ect.. Depending how much its phasing and at what time the two waveforms will be positive and negitive at the same time which will stack the output signal and when they are not the signals may cancel each other out. From what I understand its easier for the bass frequencies to cancel each other out than the higher frequencies as they are longer. I may not be entirely right but from my own (limited) research this is more or less the problem.

[Captain Ironhelm]

try setting the lfo to a low pass filter, instead of a phaser, and test with where the cutoff is, but don't make it so low that it kills all the sound. Idk if this helps or not.

[Versilaryan]

4thImpulse's explanation sounds right. It's not that canceling happens more with lower frequencies than with higher frequencies -- it's that it's more noticeable because it happens for longer periods of time.

There really isn't much you can do to fix that, aside from just not using that effect. No matter how much you compress it, you're not going to be able to bring the lower frequencies back. What you can do to work around it, though, is layer two separate synths -- one lowpassed with the lower, warmer frequencies, and the original synth, highpassed to only have the higher frequencies that define the sound.

[ph00tbag]

What Versilaryan said is a good suggestion, although another option is to route the same synth through two channels.

[prettiestPony]

Thanks for the advice, everyone.

Captain Ironhelm, that's not a bad thought. Doesn't help with the ymVST problem, but I'll keep it in mind for the future.

I'm thinking, come render time, I may end up rendering/bouncing the synth and then rearranging/looping sections to get it to sound right.

I like the idea of layering the synth with a copy of itself, although there one might run into the problem of their cycles not lining up in the right way.