Sampler modulations - Pre Filter - Post Filter

Ideas & suggestions
1

I have some thoughts around the modulations of the sampler that i’d like to air.

Even though i like the option to have DSPs in my instrument i prefer to avoid using it when i don’'t have to and can get away with using the modulations only, for the simple facts that it makes the instrument less CPU hungry, it supports polyphony in a way that a DSP chain does not and you can play the instrument in several tracks at once.

I find some of the filters sounds pretty good, while others pretty much useless, like the distortion filters, they constantly end up sounding bad for me. Now what if we had an option to have two filter sets, then we could use the distortion as a pre filter and i.e a moog filter as a post filter, which would give the sound an entirely different character.

The way i see it is that the pre filter is processed before or after the standard modulators vol/pan/pitch while the post filter comes last. I think this would make the instruments sound much better and much more versatile.

Not shure which one of these configurations would be most useful, i think the first might sound a bit crazier while the second a bit more “musical”:

Pre Filter -Vol/Pan/Pitch - Post Filter

or

Vol/Pan/Pitch - Pre Filter -Post Filter

I have a hunch that the first configuration is way more CPU hungry though?

Now if there also was made an upgrade to the filter sets, for instance if a simple reverb was added we could very conveniently make a gated reverb to our drums right out of the modulations. Delay, phase modulation, new filters…

I have no idea though, if this is possible without spending way to much CPU or the devs rewriting everything?

2

Yes, this is a good approach: do something harsh sounding - and then tame it…most likely, using a post filter.

I usually manage by sticking a lowpass filter into the effect section - or the track DSP chain, if I need to spread the sound across channels.

The only downside is that it does not allow for filter envelopes that gradually fall off, and stuff like that. That is fine if the instrument is meant to be played monophonically, but otherwise would sound a bit funny.

Pre Filter -Vol/Pan/Pitch - Post Filter

or

Vol/Pan/Pitch - Pre Filter -Post Filter

I have a hunch that the first configuration is way more CPU hungry though?

Not really. If you mean, voices can be processed together when the filter is the first thing? You would only need to add a velocity/key tracker, or random LFO in order to loose the opportunity to optimize - voices would then become filtered slightly differently.

And yes, DSP filters that are processed on a per-voice basis are far more demanding on the CPU, but that’s what could be expected…I’d love to see it happen anyway.