LPB v. LPC with Random LFO

First a little background.
I’m working on a generative music project with a friend of mine. I am trying to use Renoise to generate MIDI data and send it to various softsynths and a few external synths; some of the DSP devices and routing options in Renoise were absent from my primary DAW, so I figured I’d give the Renoise demo a try. Since the majority of my time producing music has been spent in Ableton Live, just opening Renoise for the first time was a bit of a shell shock. ^_^/>/>

Anyhow, I’ve managed to accomplish what I initially set out to do (sort of). Namely, using the random waveform in the Meta/*LFO DSP to modulate various parameters in my synthesizers. My problem is that I cannot seem to figure out the relationship between LPB and the LFO device’s LPC value when the random waveform is selected. I was assuming that the random waveform operates in a sample & hold fashion; if this is the case, there seems to be some discrepancy between the value I have set for LPC and what the oscillator actually does.

Here’s what’s happening:
I started a new “song” and configured an *LFO device to modulate CC19 in an *Instr. MIDI Control device. CC19 corresponds to the filter cutoff frequency of an external synth, and it is very obviously functioning. The filter cutoff frequency jumps all over the place (exactly what I’m looking for) and it sounds fantastic.
I left the LPB setting at 4 since I won’t be using Renoise for sequencing and I figured it won’t affect me until I have to ReWire it to Ableton. My problem is that if I set the LPC in the *LFO to 2.00 (which I figured would correspond to a new value being sent on every 1/8th note, since my LPB is set to 4), the sound changes much more rapidly; perhaps 1/64? It moves so quickly that it’s really hard to tell, even if I decrease the BPM to something low like 60. I can only get the *LFO to change the sound on every 1/8th note by changing the LPC to 16.00, but if I select a different waveform (such as the sine), an LPC setting of 2.00 modulates the CC parameter at exactly right frequency.

Can anyone explain to me how the random waveform functions, compared to the other waveforms in the *LFO device?
Does it operate on the same principle as a S&H circuit, or am I completely wrong about that?
Is there another, better/simpler way to do what I’m trying to do with a different DSP device? Perhaps one that functions more like an S&H circuit?

Thanks for reading the wall of text. Any input is greatly appreciated.

I can understand your perplexity, but before I reply I would like to give a look at an example XRNS file about this issue.

if you rightclick any DSP slider (in your case, the LFO “frequncy” slider), it will write automation for that slider, so you could use this technique to record a song with different frequency values to show the problem

The LFO is a wonderfully simple and powerful device, but it does have a few limitations.
One of these are the rate/tempo, it can go infinitely slow (well, literally it is stopped then), but it cannot go any faster than 1 line per cycle (LPC).

The random mode, however, change it’s stepped value 8 in a rate times of any of the other waveforms (I think it’s random, not S&H)
I guess this is a decision which was made to make it easier to approximate “noise”, without having to raise the overall project tempo.

Check out the “custom” mode if you want full control of the waveform, or higher time resolution

Hello from reddit oscillating! Just wanted to link you to this since I mentioned it:

“Generate” actually puts the random melody on the clipboard for you to paste into Renoise.

Hey, you found me! Thanks for recommending Renoise. It’s very different from what I’m used to (coming from music-making easy mode, Ableton), but I’ve been having a lot of fun with it so far. Kinda wish I’d known more about this program when I was going through my chiptune phase in high school, haha.

…and thanks for the link, too. I’ll be trying out Psymachine as soon as I get back home to all my gear. In combination with the random LFO, this sounds like exactly what I need to get some of the sounds I’m after.

Edit: D’oh. Looks like I can’t use Psymachine. It’s a Windows executable and I’m running OS X. Oh well.
I’ll mostly be running Renoise as a ReWire slave inside of Ableton, and as it stands I can get some pretty interesting semi-random melodies using the MIDI effects that ship with Live.

It’s indeed 8x the speed given by the Lines Per Cycle. LPC 8 gives new value every line. Don’t understand the difference between random/s&h?

Would love if the LFO could go faster.
Any work-around for this?

Could this possibly be done with a Formula Device?

Any help/suggestions appreciated.

Something I’ve done in the past is to use a custom LFO, then draw 2 cycles of the desired waveform into the envelope. This essentially doubles the overall speed of the LFO, so you must also compensate for this when automating the frequency.

For example:
5067 renoise-lfo-lpc-hack.png

Another approach is to simply double your song’s LPB, although your song/pattern structure may not allow for this.

Thanks. Very helpful.